<%= category %>

\1

Hello
Hello
"
                traceback.print_exc()

Notes: The assignment to sys.stderr is needed because the traceback
prints to sys.stderr.  The print "\n\n" statement is necessary to
disable the word wrapping in HTML.

If you suspect that there may be a problem in importing the traceback
module, you can use an even more robust approach (which only uses
built-in modules):

        import sys
        sys.stderr = sys.stdout
        print "Content-type: text/plain"
        print
        ...your code here...

This relies on the Python interpreter to print the traceback.  The
content type of the output is set to plain text, which disables all
HTML processing.  If your script works, the raw HTML will be displayed
by your client.  If it raises an exception, most likely after the
first two lines have been printed, a traceback will be displayed.
Because no HTML interpretation is going on, the traceback will
readable.

When all else fails, you may want to insert calls to log() to your
program or even to a copy of the cgi.py file.  Note that this requires
you to set cgi.logfile to the name of a world-writable file before the
first call to log() is made!

-- 
Patricia

Successfully verbose user of speech recognition software.

From tjreedy at udel.edu  Tue Mar 28 18:00:29 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 28 Mar 2000 18:00:29 -0500
Subject: __getattr__ / __setattr__ - infinite recursion!?
References: <0yMD4.624$HG1.20784@nnrp1.uunet.ca>
Message-ID: <8brdgr$7qn$1@news.udel.edu>

"Warren Postma"  wrote in message
news:0yMD4.624$HG1.20784 at nnrp1.uunet.ca...
>
>
> Okay my first attempt at a "static type checking object" for Python blows
> up, probably
> a stack overflow, probably due to infinite recursion or some such
thing....

Yes.  Within __setattr__(), you should use explicit references to
self.__dict__[attribute]
instead of self.attribute, which may lead to recursive calls to
__setatttr__.  Probably ditto for __getattr__.

> # "type objects"
> int_type       = type(0)
> string_type    = type("X")
> float_type     = type(0.1)

You can also use 'type' module ('import type' or 'from type import *'),
which supplies IntType, etc, including some types that most people would
have difficulty creating innocuous instances of.

>     def keyname(self):
>         return self.keyname

self.keyname will always directly reference the attribute (fine by me)
instead of this 'get' method.

Terry J. Reedy

From drew.csillag at starmedia.net  Mon Mar  6 09:54:24 2000
From: drew.csillag at starmedia.net (Andrew Csillag)
Date: Mon, 06 Mar 2000 09:54:24 -0500
Subject: Supporting #line directives 
Message-ID: <38C3C6A0.456900CB@starmedia.net>

I've lately been using python as a target language for various purposes
and I was wondering if anybody has thought of supporting something like
C's #line directives which would make it possible to (without hacking an
exception renderer to badly) show the lines from the original source
file and not the generated python code.  Looking at the existing code it
would appear to be nontrivial to implement (or a patch would have been
attached ).

Or am I the only one?

TIA,
Drew
-- 
print(lambda(q,p):'pmt:$%0.2f'%(q*p/(p-1)))((lambda(a,r,n),t:(a*r/
t,pow(1+r/t,n*12)))(map(input,('amt:','%rate:','years:')),1200.0))

From exarkun at flashmail.com  Thu Mar 23 12:32:00 2000
From: exarkun at flashmail.com (Jp Calderone)
Date: Thu, 23 Mar 2000 17:32:00 +0000
Subject: xmllib default value question
Message-ID: <38DA5510.75C77425@flashmail.com>

I'm using the xmllib toread in configuration files, but I don't 
quite understand how default values are supposed to be accessed.
I've created the tag_ dictionaries, and I'm pretty
sure the XML is correct, but I keep getting KeyErrors when I
try to look up the name of the value in the attribute dictionary.
Here's the basic code (and xml):

class ConfigParser(xmllib.XMLParser):

   tag_network = {'default': 'no'}
   tag_access = {'order': 'allow deny'}
   tag_allow = {'host': '*', 'port': '*'}
   tag_deny = {'host': '*', 'port': '*'}

   def start_allow(self, attr):
	self.value = (attr['host'], attr['port'])

	...
]>

 As far as I can tell, this should invoke the start_allow()
method with a dictionary something like 
{'host': 'somehost.com', 'port': '*'} but the single line in
start_allow raises a KeyError on port.  Anyone know why/how
to fix this?

 Thanks in advance, jp

-- 
Where a calculator on the ENIAC is equipped with 18,000 vacuum tubes and
weighs 30 tons, computers in the future may have only 1,000 vacuum tubes
and
weigh only 1.5 tons.    -- Popular Mechanics, March 1949
--
 5:26pm up 16 days, 23:55, 3 users, load average: 0.14, 0.06, 0.01

From cpr at emsoftware.com  Thu Mar 30 12:18:13 2000
From: cpr at emsoftware.com (Chris Ryland)
Date: Thu, 30 Mar 2000 17:18:13 GMT
Subject: parsing CSV files with quotes
References: 
Message-ID: 

Warren--

Since I've written a CSV/TSV scanning module (in C) for one of our products,
I'd recommend a simple state machine scanner. The rules are more complex
than meets the eye, if you really want to handle all the random cases that
dumb programs like Access or Excel output.

I've thought about providing a Python wrapper for this scanning wrapper (and
will eventually). If you want to do it yourself, I'd be happy to send you
the C code.
--
Cheers!
/ Chris Ryland, President / Em Software, Inc. / www.emsoftware.com
- - -
"Warren Postma"  wrote in message
news:BDLE4.1642$HG1.47883 at nnrp1.uunet.ca...
> Suppose I have a CSV file where line 1 is the column names, and lines 2..n
> are comma separated variables, where all String fields are quoted like
this:
>
> ID, NAME, AGE
> 1, "Postma, Warren", 30
> 2, "Twain, Shania",  31
> 3, "Nelson, Willy",  57
> 4, "Austin, \"Stone Cold\" Steve", 34
>
> So, the obvious thing I tried is:
>
> import string
> >>> print string.splitfields("4, \"Austin, \\\"Stone Cold\\\" Steve,
> 34",",")
> ['4', ' "Austin', ' \\"Stone Cold\\" Steve', ' 34']
>
> Hmm. Interesting. So I tried this:
>
> >>> print string.splitfields(r'4, "Austin, \"Stone Cold\" Steve", 34')
> ['4,', '"Austin,', '\\"Stone', 'Cold\\"', 'Steve",', '34']
>
> I'm getting close, I can feel it!
>
> The Rules:
>
> 1. All integer and other fields are output as ascii.
> 2. String fields have quotes. Commas are allowed inside the quotes.
> 3. Quotes inside quotes are escaped by a backslash
> 4. Backslashes are themselves quoted by a backslash
>
> Is this complex enough that I basically need the "parser" module of
Python?
>
> Problem is I'm scared of it. Anyone got any Parser Tutorials Howtos/Links?
>
> Or is this beasty solveable by judicious use of Regular Expressions?
>
> While I'm taking up bandwidth, I'll ask another silly question:
>
> Is there a "compressed dbShelve" out there anywhere? In this case I just
> want to store arrays and dictionaries of built-in Python types, in a
> compressed manner, in a bsd database. Anyone heard of something like this?
>
> Warren
>
>

From gerrit at nl.linux.org  Wed Mar  1 09:30:16 2000
From: gerrit at nl.linux.org (Gerrit Holl)
Date: Wed, 1 Mar 2000 15:30:16 +0100
Subject: Tkinter vs. wxPython (was: Which GUI?)
In-Reply-To: <200003011407.JAA06150@smtp2.andrew.cmu.edu>; from geek+@cmu.edu on Wed, Mar 01, 2000 at 09:07:23AM -0500
References:   <5l3dqjckt5.fsf_-_@eric.cnri.reston.va.us>  <200003011407.JAA06150@smtp2.andrew.cmu.edu>
Message-ID: <20000301153016.C4873@nl.linux.org>

> Then  writes: spoke up and said:
> > For fun with buggy user code in Tkinter, try:
> > 
> > import Tkinter
> > b=Tkinter.Button()
> > b.pack()
> > l=Tkinter.Label()
> > l.grid()

Guido did *not* write this. Can you try to quote better in the future,
please?

regards,
Gerrit.

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

From dogboy777 at my-deja.com  Fri Mar  3 09:41:08 2000
From: dogboy777 at my-deja.com (dogboy777 at my-deja.com)
Date: Fri, 03 Mar 2000 14:41:08 GMT
Subject: Path Puzzles
Message-ID: <89oiu3$ggk$1@nnrp1.deja.com>

I've started using Steve Purcell's wonderful PyUnit test framework, and I've
run into a problem with paths and modules.  I want to store all of my test
modules, test files,  in a subdirectory.  What that means is that if, for
example, I have a module called command.py and an associated testing module
called tcommand.py, I need to import command. py into tcommand.py. 
tcommand.py needs to know to look for command.py in tcommand.py's parent
directory.

So far, I've kludged the problem by setting PYTHONPATH in DOS before I run
the script from DOS or Linux (we use Samba to map our Win95 clients to our
Linux server).	This works, but it means that every time I create a new
project which will have its own test subdirectory, I've got to change
PYTHONPATH.  There's got to be a better way, right?

The other wierd little problem I ran into was that when I created parser.py
and commands.py, their testing modules freaked and refused to call them: 
self.commands = commands.Commands()  would return an "attribute error:
Commands".  However, when I renamed commands.py to jcommands.py and parser.py
to jparser.py, it worked fine.	Is this because Python already has modules
further up the PYTHONPATH called parser.py and command.py, or did I goof up
in some other way?

Thanks,
Anders Schneiderman
National Journal Daily Briefings
Technology Production Manager

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

From cgw at fnal.gov  Mon Mar 27 12:56:03 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: Mon, 27 Mar 2000 11:56:03 -0600 (CST)
Subject: __getattr__/__setattr__ - infinite recursion
Message-ID: <14559.41139.403304.673925@buffalo.fnal.gov>

There are several problems in the script you've posted, below find a
corrected version with the changed lines marked with ##Change

Here's an explanation of the changes:

Change 1:  You raise "Error" but don't define it anywhere.  You must
           create it either as a string, or as an Exception object.

Change 2:  You had "self.reset_to_defaults(self)", one self too many.

Change 3:  This is the main point.  You have no way to init your
           class, because setting the initial attributes - "attrs",
	   "values" and "keyname" - goes through the validation
	   mechanism, which doesn't make sense for these attributes,
	   since the validation mechanism can't be used until
	   these attributes are set up.
	   In general, if you have a setattr hook and need to set 
	   some instance attributes while avoiding this hook, you can do
	   self.__dict__[key] = value, instead of self.key = value

Change 4:  I just hate to see "except:" in code... you should always
           name the exceptions you're looking for, if at all possible.
	   Otherwise debugging becomes impossible - you have a
	   mismatched paren or something, instead of getting a
	   SyntaxError you fall into the unqualified "except:" clause.

#---------------------------------------------------------------------
# Why does this blow up Python?
#---------------------------------------------------------------------

import os, sys, string, pprint

# "type objects"
int_type       = type(0)
string_type    = type("X")
float_type     = type(0.1)

# required field flags?
required = 1
optional = 0

Error = "Error"  ### Change 1

# Class with strict type checking and a single key:
class zdb_rec:
    def __init__(self, pattern_dict=None, keyname=None):
        "create an object with a strict pattern of required values to make up a type"
        self.set_valid(pattern_dict)
        self.keyname = keyname
        self.reset_to_defaults()  #####Change 2

    def set_valid(self,pattern_dict):
        self.attrs = pattern_dict

    def set_keyname(self,keyname):
        self.keyname = keyname

    def reset_to_defaults(self):
        # fill in default values for each attribute
        self.values = {}
        for x in self.attrs.keys():
            self.values[x] = self.attrs[x][2]

    def keyname(self):
        return self.keyname

    def key(self):
        return self.values[self.keyname]

      # setattr does typesafety check:
    def __setattr__(self,name,value):
        "custom __setattr_ with typesafety check" 
        if name in ('attrs','keyname','values'):  ##Change 3
            self.__dict__[name]=value             ##Change 3
            return                                ##Change 3

        try:
            if (value == None) and (not self.attrs[name][1]):
                self.values[name] = None
            else:
                if (type(value) == self.attrs[name][1]):
                    self.values[name] = value
                else:
                    raise Error, "Invalid type in assignment"+`type(value)`
        except KeyError: ######Change 4
            raise Error, "Invalid attribute "+`name`

    def __getattr__(self,name):
        try:
            return self.values[name]
        except KeyError: ######Change 4
            raise Error, "Invalid attribute "+`name`

    def attr_count(self):
        return len(self.attrs)

    def validate(self):
        "raises an exception if validation fails, ie requird field not provided"
        for x in self.attrs.keys():
            if self.attrs[x][1]:
                if self.values[x] == None:
                    raise Error, "validate() failed - required field not set"

# "Person Database Schema":
person_attributes = {
 # attribute    type          required?   initialvalue
     "name":    (string_type,    required,  ""),
     "address": (string_type,    required,  ""),
     "age":     (int_type,       required,  0 ),
}

x = zdb_rec(person_attributes,"name")

print "done"

From adustman at comstar.net  Tue Mar  7 15:37:52 2000
From: adustman at comstar.net (Andy Dustman)
Date: Tue, 07 Mar 2000 20:37:52 GMT
Subject: mysql for win98
In-Reply-To: <5Ybx4.6$i17.7664@news.nyc.globix.net>
References: <5Ybx4.6$i17.7664@news.nyc.globix.net>
Message-ID: 

On Tue, 7 Mar 2000, Meshulam Gill wrote:

> Where can I find mysql module for win98?

It's not supported, but supposedly if you follow the README.windows
instructions, it does work.

http://starship.python.net/crew/adustman/MySQLdb.py 

-- 
andy dustman       |     programmer/analyst     |      comstar.net, inc.
telephone: 770.485.6025 / 706.549.7689 | icq: 32922760 | pgp: 0xc72f3f1d
"Therefore, sweet knights, if you may doubt your strength or courage, 
come no further, for death awaits you all, with nasty, big, pointy teeth!"

From tuttledon at hotmail.com  Tue Mar 21 22:59:50 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Wed, 22 Mar 2000 03:59:50 GMT
Subject: Does Guido Have A Will? (The legal kind )
References:  <38d80737.2362399@news.uswest.net>
Message-ID: 

[Andrew Pann]
> Believe it or not, this is actually addressed, sorta, in the (a) FAQ.
> Check it out:  http://www.python.org/doc/FAQ.html#2.7

Thanks!  I'd forgotten I'd read this.  After re-reading the various FAQ's
and reading between the lines, it's reassuring to know that Guido's
consciousness will live on forever in a military supercomputer at Langley.
<0.5 wink>

Don

From phd at phd.russ.ru  Wed Mar  1 11:06:53 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 1 Mar 2000 16:06:53 +0000 (GMT)
Subject: MySQLmodule synthax error
In-Reply-To: <38BD3724.B5E35F7E@scripps.edu>
Message-ID: 

On Wed, 1 Mar 2000, Olivier Roche wrote:
> MySQL.error: You have an error in your SQL syntax near '-dock

   I guess MySQL does not like table names with hyphen. Replace it with
underscore.

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

From moshez at math.huji.ac.il  Sat Mar 18 10:44:26 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 18 Mar 2000 17:44:26 +0200 (IST)
Subject: Tokenizing a string
In-Reply-To: <38D38FC2.E7C211BA@hampshire.edu>
Message-ID: 

On Sat, 18 Mar 2000, Michael Dartt wrote:

> I've got a string I'd like to tokenize, but it's not in a file, and it'd
> be rather inefficient to write it to a file just to tokenize it.  Is
> there any function I can use to pass this string to
> tokenize.tokenize()?  

Use StringIO or cStringIO, or just use split('\n') on it, and write a
small class to get out the lines.

--
Moshe Zadka . 
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com

From embed at geocities.com  Thu Mar 30 11:50:27 2000
From: embed at geocities.com (Warren Postma)
Date: Thu, 30 Mar 2000 11:50:27 -0500
Subject: parsing CSV files with quotes
Message-ID: 

Suppose I have a CSV file where line 1 is the column names, and lines 2..n
are comma separated variables, where all String fields are quoted like this:

ID, NAME, AGE
1, "Postma, Warren", 30
2, "Twain, Shania",  31
3, "Nelson, Willy",  57
4, "Austin, \"Stone Cold\" Steve", 34

So, the obvious thing I tried is:

import string
>>> print string.splitfields("4, \"Austin, \\\"Stone Cold\\\" Steve,
34",",")
['4', ' "Austin', ' \\"Stone Cold\\" Steve', ' 34']

Hmm. Interesting. So I tried this:

>>> print string.splitfields(r'4, "Austin, \"Stone Cold\" Steve", 34')
['4,', '"Austin,', '\\"Stone', 'Cold\\"', 'Steve",', '34']

I'm getting close, I can feel it!

The Rules:

1. All integer and other fields are output as ascii.
2. String fields have quotes. Commas are allowed inside the quotes.
3. Quotes inside quotes are escaped by a backslash
4. Backslashes are themselves quoted by a backslash

Is this complex enough that I basically need the "parser" module of Python?

Problem is I'm scared of it. Anyone got any Parser Tutorials Howtos/Links?

Or is this beasty solveable by judicious use of Regular Expressions?

While I'm taking up bandwidth, I'll ask another silly question:

Is there a "compressed dbShelve" out there anywhere? In this case I just
want to store arrays and dictionaries of built-in Python types, in a
compressed manner, in a bsd database. Anyone heard of something like this?

Warren

From kc5tja at garnet.armored.net  Mon Mar 20 08:16:49 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 20 Mar 2000 13:16:49 GMT
Subject: Forth like interpreter
References: <158801bf8bdc$08516040$6401a8c0@dorb>  <53YA4.637$jX3.10074@typhoon.nyroc.rr.com>  <425A5D00B0F916D1.7920112131B1A88A.D3428C8A56515B00@lp.airnews.net>
Message-ID: 

In article
<425A5D00B0F916D1.7920112131B1A88A.D3428C8A56515B00 at lp.airnews.net>, Tres
Seaver wrote:

>Forth was the first "brain-bender" language I learned, and the one which
>caused me to learn assembler.  I still credit Forth with making me Lisp-
>impaired, via a kind of hysteresis.

Odd.  Forth is precisely what enabled me to learn Lisp.  Forth is a reverse
Lisp of sorts:

(defun cmpint (x y)
  (cond ((< x y) -1)
        ((= x y) 0)
	(T 1)))

: cmpint ( x y -- f )
  2DUP < IF
    2DROP -1
  ELSE
    = IF
      0
    ELSE
      1
    THEN
  THEN ;

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

From just at letterror.com  Wed Mar 15 04:51:38 2000
From: just at letterror.com (Just van Rossum)
Date: Wed, 15 Mar 2000 10:51:38 +0100
Subject: Debugging?
In-Reply-To: 
References: <9JFz4.19$GJ3.1756@news7.onvoy.net>
Message-ID: 

At 9:39 AM +0000 15-03-2000, Mark Hammond wrote:
>I believe the only other graphical debugger available for Python today
>is Pythonwin.

Hey! Not true! The MacPythonIDE has had a (good!) graphical debugger for at
least three years ;-)
(yeah yeah, not relevant to the original poster, but still...)

Just

From godzilla at netmeg.net  Sat Mar 18 13:39:39 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sat, 18 Mar 2000 13:39:39 -0500 (EST)
Subject: need fast parser for comma/space delimited numbers
In-Reply-To: <1258719576-20520305@hypernet.com>
References: 
	<1258719576-20520305@hypernet.com>
Message-ID: <14547.52587.788622.143477@gargle.gargle.HOWL>

> Um, I hate to tell you, but... oh, nevermind .

what .... he was refering to black beauties?????? (what is speed
called on the street these days?????)

> Absolutely. This cheap trick is frequently worth 30% to 60% in a
> moderate to tight loop.

this is great. probably worth more to me than this stinking csv parser
;-)

> I *think* I recall that int / float were somewhat faster, but I
> can't be sure. The major thing here is to get away from extra
> function calls. Flatten your loop as much as possible - function
> calls are (relatively) expensive.

yeah. tahts next on  the list.

les

From hnowak at cuci.nl  Mon Mar 20 17:59:28 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Mon, 20 Mar 2000 23:59:28 +0100
Subject: Reverse of hex()
In-Reply-To: 
Message-ID: <200003202300.XAA22330@dionysus.fw.cuci.nl>

On 20 Mar 00, at 11:43, Gaetan Corneau wrote:

> Hi, 
> 
> I'm looking for the reverse of the hex built-in function: is there any?

You can use eval:

>>> eval('0xf')
15

Or maybe:

>>> string.atoi('0xf', 16)
15

HTH,

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://www.hvision.nl/~ivnowa/newsite/
Python questions? See http://tor.dhs.org/~zephyrfalcon/snippets
You call me a masterless man. You are wrong. I am my own master.

From jamarijr at hotmail.com  Thu Mar 30 15:37:20 2000
From: jamarijr at hotmail.com (Arint�)
Date: Thu, 30 Mar 2000 20:37:20 GMT
Subject: Function override?
Message-ID: <8c0dtu$lr3$1@nnrp1.deja.com>

I have a class similar to this:

class over:
   somefun(self, bakeint):
      ...
   somefun(self, bakeStr):

If the user uses a string then I want to use the 2nd one, else an
Integer then use the 1st one?

How can I get this?  And what is this call in c, the word escapes me
right now.

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

From aa8vb at yahoo.com  Thu Mar  2 13:47:34 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 2 Mar 2000 13:47:34 -0500
Subject: XML-RPC
In-Reply-To: <89i4eq$snm$1@nnrp1.deja.com>
References: <88aka3$mn4$0@216.39.162.232> <89i4eq$snm$1@nnrp1.deja.com>
Message-ID: <20000302134734.A3900361@vislab.epa.gov>

Lisa:
 |In the page http://www.pythonware.com/products/xmlrpc/, there is a
 |mention that 0.9.9 of xml-rpc would have been out by dec.99, and I would
 |like to know if anyone has any info as to where to find that thing?
 |
 |BTW, is there any info online, - anywhere, - that can link me to more
 |info regarding xml, the standard, the tutorial, and such?

For XML in general, you can pull alot out of search engines real quick.
But here are a few I had bookmarked to get you going.

-- 
Randall Hopper
aa8vb at yahoo.com

     http://www.python.org/sigs/xml-sig/
        Python XML SIG
     http://www.python.org/topics/xml/
        Python.org Recommended XML resources

     http://xml.org/
        XML.ORG
     http://www.xml.com/
        XML.COM On-line Magazine
     http://www.w3.org/XML/
        W3.ORG:  XML
     http://msdn.microsoft.com/xml/default.asp
        MSDN Online: XML Developer Center
     http://www.schema.net/
        SCHEMA.NET: The XML Schema Site
     http://www.oasis-open.org/cover/xml.html
        Oasis: XML
     http://www.xml.com/xml/pub/axml/axmlintro.html
        XML.COM:  The Annotated XML Spec
     http://pdbeam.uwaterloo.ca/~rlander/XML_Tutorial/
        Generally Markup: XML Resources
     http://pdbeam.uwaterloo.ca/~rlander/Introduction_XML/
        Introduction to XML (York University)
     http://www.xml.com/xml/pub/98/10/guide1.html
        XML.COM:  What is XML?
     http://www.sun.com.au/sjug/pres/xml/JavaAndXML/seminar.html
        Java & XML: 1 + 1 > 2
     http://www.wdvl.com/Authoring/Languages/XML/Tutorials/
        WDVL: XML Articles and Tutorials
     http://www.oasis-open.org/cover/
        The SGML/XML Web Page - Home Page
     http://www.stud.ifi.uio.no/~lmariusg/linker/XMLtools.html
        Free XML software
     http://www.zdnet.com/devhead/stories/articles/0,4413,2138258,00.html
        XML Authoring Tools
     http://www.xmlhack.com/
        xmlhack: developer news from the XML community
     http://www.ibm.com/developer/xml/
        developerWorks : XML : overview
     http://www.xmlbooks.com/
        Charles F. Goldfarb's XMLbooks.com
     http://metalab.unc.edu/xml/books.html
        XML Book List
     http://www.xml.com/xml/pub/1999/01/3dml/3dml.html
        3DML
     http://www.web3d.org/news/x3d/index.html
        X3D Documents
     http://www.w3.org/Graphics/SVG/
        W3C Scalable Vector Graphic s (SVG)
     http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=509294232
        NETNEWS:  GIS and XML
     http://www.informatics.sintef.no/UML2XML/
        GI Browser (Prototype GIS XML import/export facility
     http://www.lists.ic.ac.uk/hypermail/xml-dev
        XML-Dev Mailing List Archives
     http://xdev.datachannel.com
        DataChannel XML Resource Center
     http://www.flixml.org
        Flix B Movie Guide (created in XML with HOW-TO)
     http://www.stg.brown.edu/~sjd/xlinkintro.html
        Introduction to XML Linking
     http://metalab.unc.edu/xml/books/bible
        "The XML Bible" book, on-line
     http://www.textuality.com/xml
        Textuality XML Pages (w/ FAQ)
     http://www.w3.org/XML/Activity.html
        XML On-going Activities

From jimn at minusen.force9.co.uk  Mon Mar 27 14:02:59 2000
From: jimn at minusen.force9.co.uk (Jim Nicholson)
Date: Mon, 27 Mar 2000 20:02:59 +0100
Subject: Newbie - help in opening and reading a file from >>>
References: 
Message-ID: <39bob8.eq9.ln@minusen.minusen.net>

Dan Howard wrote:
> Hi
> I've just installed 1.5 on my Windows 98Se system
> I'm trying to execute the following-
> >>>f=open('C:\Autoexec.bat')
f=open('C:\\Autoexec.bat')
> and I get a no such directory or file error
python was looking for a file with the BEL character in it '\A' which of
course doesn't exist.  Remember you need to escape your backslashes to
prevent them being interpreted as beginning a character code.

Jim

From R.Brodie at rl.ac.uk  Wed Mar 15 07:26:33 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Wed, 15 Mar 2000 12:26:33 -0000
Subject: read NT registry with python
References: <8alo1n$tdc$1@nnrp1.deja.com>  <8ansn6$ece$1@nnrp1.deja.com>
Message-ID: <8anvhr$1756@newton.cc.rl.ac.uk>

 wrote in message news:8ansn6$ece$1 at nnrp1.deja.com...

> [HKEY_LOCAL_MACHINE\SOFTWARE\WOF]
>
> [HKEY_LOCAL_MACHINE\SOFTWARE\ABC\PlusSoftware]
> "AdminPassword"="Admin"

import win32api
from win32con import *

k=win32api.RegOpenKeyEx(HKEY_LOCAL_MACHINE,
'SOFTWARE\\ABC\\PlusSoftware',0,KEY_QUERY_VALUE)

win32api.RegQueryValueEx('AdminPassword')

From te at first.gmd.de  Thu Mar 30 03:57:28 2000
From: te at first.gmd.de (Thilo Ernst)
Date: Thu, 30 Mar 2000 10:57:28 +0200
Subject: Rational-Rose and Python ...
References: <8bt5fh$p2t$1@news.urz.uni-heidelberg.de>  <8btgsg$fnf$1@news1.xs4all.nl>
Message-ID: <38E316F8.C90C23A8@first.gmd.de>

Boudewijn Rempt wrote:

> In sum, I think that an IDE which offers a class browser, an object

> browser, a simple diagramming tool, an editor that allows rich text
> comments and hyperlinks in the source, and a decent debugger are all
> that's needed. Basically, give me many views of the source, and I'm happy

Sounds like BoaConstructor, doesn't it? ClassBrowser, UMLView feature, ...

- Thilo

Return-Path: 
Delivered-To: python-list at python.org
Received: from mail.inhh.de (mail.inhh.net [62.96.208.2])
	by dinsdale.python.org (Postfix) with ESMTP id 84D921CD52
	for ; Wed, 29 Mar 2000 11:02:43 -0500 (EST)
Received: (from ms at localhost)
	by mail.inhh.de (8.9.1a/8.9.1) id SAA04572
	for python-list at python.org; Wed, 29 Mar 2000 18:05:42 +0200 (MET DST)
Date: Wed, 29 Mar 2000 18:05:42 +0200 (MET DST)
From: Marek Schneider - AIS 
Message-Id: <200003291605.SAA04572 at mail.inhh.de>
Subject: How to build shared Python binaries under Solaris 7 (sparc) ?
Content-Type: text
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta1
Precedence: bulk
List-Id: General discussion list for the Python programming language 

Hello ;)

Solaris 7 (sparc) / gcc 2.95 / Python-1.5.2

I have tried my luck es follows:
---
#!/bin/sh
./configure --prefix=/www/local/python --enable-shared --disable-static --with-solaris
gmake
gmake test
(some messages follows:
 test test_zlib skipped -- an optional feature could not be imported
 test test_gzip skipped -- an optional feature could not be imported
 . .  .
 42 tests OK.
 19 tests skipped: test_al test_audioop test_bsddb test_cd test_cl test_crypt test_dbm test_dl test_gdbm test_gl
 test_gzip test_imageop test_imgfile test_nis test_rgbimg test_sunaudiodev test_thread test_timing test_zlib

 zlib and gzip are installed in /www/local or /usr/local
)
---

and further:
---
gmake install
(some messages follows:
 . .  .
 cd Modules;  gmake OPT="-g -O2" VERSION="1.5" \
                prefix="/www/local/python" exec_prefix="/www/local/python" \
                sharedmods
 gmake[1]: Entering directory `/www/ssamp/Python-1.5.2/Modules'
 gmake[1]: Nothing to be done for `sharedmods'.
 gmake[1]: Leaving directory `/www/ssamp/Python-1.5.2/Modules'
 ./install-sh -c -m 644 Modules/config.c /www/local/python/lib/python1.5/config/config.c
 ./install-sh -c -m 644 Modules/python.o /www/local/python/lib/python1.5/config/python.o
 . . .
 cd Modules; gmake \
                OPT="-g -O2" \
                VERSION="1.5" \
                SO=".so" \
                LDSHARED="ld -G" \
                CCSHARED="" \
                LINKFORSHARED="" \
                DESTSHARED="/www/local/python/lib/python1.5/lib-dynload" \
                prefix="/www/local/python" \
                exec_prefix="/www/local/python" \
                sharedinstall
 gmake[1]: Entering directory `/www/ssamp/Python-1.5.2/Modules'
 for i in X ; do \
                if test $i != X; \
                then ./../install-sh -c -m 555 $i /www/local/python/lib/python1.5/lib-dynload/$i; \
                fi; \
        done
 gmake[1]: Leaving directory `/www/ssamp/Python-1.5.2/Modules'
 . .  .
)
---

ldd python1.5 (in /www/local/python/bin) showsas follows:
$ ldd python1.5
        libsocket.so.1 =>        /usr/lib/libsocket.so.1
        libnsl.so.1 =>   /usr/lib/libnsl.so.1
        libdl.so.1 =>    /usr/lib/libdl.so.1
        libm.so.1 =>     /usr/lib/libm.so.1
        libc.so.1 =>     /usr/lib/libc.so.1
        libmp.so.2 =>    /usr/lib/libmp.so.2
 So far it seems to be OK ... 

BUT conf.pdflib.sh
---
      #!/bin/sh
      ./configure --prefix=/www/local \
                  --enable-shared --disable-static \
                  --with-perl=/www/local/perl/bin/perl \
                  --with-py=/www/local/python \
                  --with-tifflib=/www/local/lib \
                  --with-zlib=/www/local/lib \
                  --with-pnglib=/www/local/lib
---

could NOT recognize the presens of python1.5 ;(((
---
checking for Python header files... /www/local/python/include/python1.5
checking for python... no
. .  .
_____________________________________________________________
For your convenience, here's a summary of configure's results
(none of these features is required for using PDFlib with C):

Support for shared libraries:           yes
C++ language binding for PDFlib:        yes
Java language binding for PDFlib:       yes
Perl language binding for PDFlib:       yes
Python language binding for PDFlib:     no
Tcl language binding for PDFlib:        no
TIFF image support for PDFlib:          yes
PNG image support for PDFlib:           yes
Zlib compression support for PDFlib:    yes
---

Do you have ANY Idea why ?

thank you in adwance !

best regards from Hamburg,
Marek Schneider

From wlfraed at ix.netcom.com  Fri Mar 24 00:10:07 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 23 Mar 2000 21:10:07 -0800
Subject: Object oriented design question: classes with mulitiple functions or  multiple classes?
References: <38D446D8.1B1E2801@exceptionalminds.com> <200003232202.RAA06052@tor.dhs.org>
Message-ID: 

On Thu, 23 Mar 2000 17:02:00 -0500, Blake Winton 
declaimed the following in comp.lang.python:

> long.  Then it was a total mess.  So the company refactored it so
> that each function was a "Command" (a separate class, containing a
> single "execute" method), and since then, whenever the client has
> wanted to add new functionality, all we had to do was add another
> class, and test it, instead of making our "Blob" class even bigger,
> and having to re-test everything.
>
	Sounds a bit like what happened in the OO design class the
company put a lot of us through a few years ago. One exercise was to
OO-design (can one use that as a verb?) a calculator.

	Somehow, I managed to persuade the team I was on to 1) use RPN
notation, and 2) put the logic to do any operation into the "key"
itself, instead of creating an ALU that did everything in response to
"key messages"

	The design meant that adding a new "key" only required adding
(or subclassing) a class for the operation. Even the numbers were done
this way -- ask display/register for current content, append key-value
to content, return to display/register...

--
 > ============================================================== <
 >   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 r.barendregt at village.uunet.be  Mon Mar 20 15:47:00 2000
From: r.barendregt at village.uunet.be (Robin Barendregt)
Date: Mon, 20 Mar 2000 20:47:00 GMT
Subject: Question about instance creation
Message-ID: <38d68c10.41363791@news.pandora.be>

Hi all,

This is my first ever post to a newsgroup, so if I'm breaking every
rule, please forgive me :)
I'm a python newbie and am trying to create instances of a class that
can have a changing base class. A code example should clarify
(questions follow after code snippet):

File class1.py contains:

from class2 import *

class Class1:
    def __init__(self, arg = "no arg given"):
        print 'Class1 init'
        print arg
        self.var1 = 'this is var1'

    def method1(self):
        print "this is method1"		

class Class2:
    def __init__(self):
        print 'Class2 init'
        self.var2 = 'this is var2'

    def method2(self):
        print "this is method2"

class Class3(Class1, Class2):
    def __init__(self):
        Class1.__init__(self)
        Class2.__init__(self)
        print 'Class3 init'
        c = DynamicClass(Class1, "arg given")
        print c
        self.var3 = 'this is var3'

    def method3(self):
        print "this is method3"

if __name__ == '__main__':
    c = Class3()

And file class2.py contains:

class DummyClass:
    def __init__(self):
        pass

BaseAlias = DummyClass

class DynamicClass:
    def __init__(self, newAlias, *args):
        global BaseAlias
        BaseAlias = newAlias
        apply(BaseAlias, args)

class DerivedClass(BaseAlias):
    def __init__(self, *args):
        apply(BaseAlias.__init__, (self,) + args)

I want the var c in Class3.__init__ to contain and instance of the
class given as 1st argument to DynamicClass. I tried using return in
the DerivedClass.__init__ but I get an exception saying I should
return None in an __init__. How can I accomplish this?

Oh, and thanks for reading :)

Robin Barendregt.

From newsbunny at noether.freeserve.co.uk  Sun Mar  5 12:04:01 2000
From: newsbunny at noether.freeserve.co.uk (Martin P Holland)
Date: Sun, 5 Mar 2000 17:04:01 +0000
Subject: my first real program blows up, can someone tell me why?
References: <38c28b4f$0$40286@news.execpc.com>
Message-ID: 

On Sun, 05 Mar 2000 10:26:05 -0600,
Jerry F. Davis  wrote:

>	x = sys.argv[1]
>	print 'now doing fib(', x, ')'
>	fib(x)

x is a string not an integer, so in the while loop
in your function you are asking if an integer is <
a string.

fib(int(x))

will fix it but you might want to use a try: except:
to catch when the command line isn't an integer.
Play with the interpreter to discover which exception
you should catch.

atb
Martin

-- 
http://www.noether.freeserve.co.uk
http://www.kppp-archive.freeserve.co.uk

From ezra at pagic.net  Mon Mar 27 07:15:54 2000
From: ezra at pagic.net (ezra)
Date: Mon, 27 Mar 2000 20:15:54 +0800
Subject: binary to decimal conversion
References: <8b0r07$r5u$0@216.39.162.232> 
Message-ID: <38DF50FA.60ABAA04@pagic.net>

>

I write three functions:
import string

#-------------------------------
__doc__ = 'Data convert function.   written by ezra.\012'

def Int2Bin(di):
     chars = ['0','1']
     do = ''
     q = 0
     while di >= 2:
          q = di % 2
          do = chars[q] + do
          di = di / 2
     do = chars[di] + do
     if len(do)%4 >= 1 and len(do)%4 <= 3 : do = '0'*(4 - len(do)%4) + do
     return do
# End of def Int2Bin(di)

def Int2Oct(di):
     chars = ['0','1','2','3','4','5','6','7']
     do = ''
     q = 0
     while di >= 8:
          q = di % 8
          do = chars[q] + do
          di = di / 8
     do = chars[di] + do
     if len(do)%2 == 1 : do = '0' + do
     return do
# End of def Int2Oct(di)

def Int2Hex(di):
     chars =
['0','1','2','3','4','5','6','7','8','9','A','B','C','D','E','F']
     do = ''
     q = 0
     while di >= 16:
          q = di % 16
          do = chars[q] + do
          di = di / 16
     do = chars[di] + do
     if len(do)%2 == 1 : do = '0' + do
     return do
# End of def Int2Hex(di)

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

--
???(Wei-Neng Hung)
?????????????,??????????
High-Speed Communication and Computing Laboratory,NCU
E-mail : ezra at csie.ncu.edu.tw

From rmuschall.fih at t-online.de  Fri Mar 31 14:41:51 2000
From: rmuschall.fih at t-online.de (Ralf Muschall)
Date: Fri, 31 Mar 2000 21:41:51 +0200
Subject: Philosophical long integer question...
References: <38E23160.B16860C2@bioreason.com> <38E3C327.96C077C5@t-online.de> <38E48F00.2D9BD07E@bioreason.com>
Message-ID: <38E4FF7F.26DC54F0@t-online.de>

Brian Kelley schrieb:

> I hadn't thought about this approach.  Currently python will not promote

This is acceptable for me, and probably gives the fastest code.
Haskell uses the same approach (and it is disputed there as well).

> unless forced.  What if the procedure was reversed?  Promote unless
This might be easier for newbies. Lisp uses this approach (almost
-- fixnums and bignums are diferent types, and promotion causes a
type change when necessary).

> this what you mean?
It is a possibility. Whether it is needed depends on whether python
is slow enough to justify the introduction of C-based int32 etc.
Otherwise just making a class int32 and having the constructor
(and maybe other methods) bitwise_and_ing it's argument with
2^32-1 would suffice.

> certainly would affect a lot of code...)
It might break code which silently assumes that ints live in
the ring Z/(2^32*Z) -- that's why I mentioned crypto hackers.

Ralf

From victor at idaccr.org  Wed Mar 22 11:45:24 2000
From: victor at idaccr.org (Victor S. Miller)
Date: Wed, 22 Mar 2000 16:45:24 GMT
Subject: AOL PFC's?
Message-ID: 

Does anyone have any programs that can read (and maybe write) AOL
PFC's (personal filing cabinets)?  These are the things that store
mail folders, address books, and who knows what else.
-- 
Victor S. Miller     | " ... Meanwhile, those of us who can compute can hardly
victor at idaccr.org    | be expected to keep writing papers saying 'I can do the
CCR, Princeton, NJ   | following useless calculation in 2 seconds', and indeed
    08540 USA        | what editor would publish them?"  -- Oliver Atkin

From tchur at bigpond.com  Sun Mar 26 07:07:01 2000
From: tchur at bigpond.com (Tim Churches)
Date: Sun, 26 Mar 2000 22:07:01 +1000
Subject: Project proposal: The Gestalt System (using Python)
Message-ID: <38DDFD65.F660FECB@bigpond.com>

The Gestalt System project aims to develop a free, Open Source,
object-oriented software framework which makes use of other free, Open
Source software tools for the purposes of managing, analysing, and
reporting on quantitative data in a manner which is suited to modern,
networked computing environments in which processor, memory, and storage
resources are abundant but time is scarce and requirements are forever
changing. 

The Gestalt System may be of particular interest to Python users and
developers because we propose to use Python as both the main development
and end-user language for the project.

Full details of the project proposal can be found in the Gestalt System
Manifesto which is available at http://gestalt-system.sourceforge.net 

We welcome any comments or suggestions you may have on the project
proposal - please send them to
gestalt-system-discuss at lists.sourceforge.net 

You might want to subscribe to this list or to the low-volume
gestalt-system-news list in order to receive announcements. Please see
the project Web page at http://gestalt-system.sourceforge.net for
details of how to subscribe.

If you wish your feedback to be kept private, you can respond to the
authors of the project proposal: Tim Churches at tchur at bigpond.com and
Karsten M. Self at kmself at ix.netcom.com

Please note that the Reply-To address of this message is set to
gestalt-system-discuss at lists.sourceforge.net which means that your reply
will by default be distributed to all subscribers to this list. You may
wish to change the address to tchur at bigpond.com and/or
kmself at ix.netcom.com if you prefer that your reply remain confidential.

Regards,

Tim Churches (tchur at bigpond.com)
 and 
Karsten M. Self (kmself at ix.netcom.com)

From Richard.Jones at fulcrum.com.au  Thu Mar 16 19:22:49 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Fri, 17 Mar 2000 11:22:49 +1100
Subject: Pointers 
In-Reply-To: Message from Curtis Jensen of 2000-Mar-16 15:54:13,
            <38D17425.C983EE23@be-research.ucsd.edu> 
Message-ID: <200003170022.LAA03719@envy.fulcrum.com.au>

[Curtis Jensen]
> I need a slightly different twist now.  Somthing like:
> 
> >>> d = {'A':0,'B':0}
> >>> c = {'A': -> (Pointer to) d['A']}
> >>> print c['A']
> 0
> >>> d['A'] = 10
> >>> print c['A']
> 10

   You are asking far too much of the standard dictionary type :)

   You'll either need to use one of the hacks described by other posters which 
involves using a mutable proxy entry for the dictionary items - such as a list 
or an instance or a proxy class of your own rolling. The other option that's 
immediately obvious to me is for you to extend the dictionary class (via 
UserDict) to provide the semantics you're after.

   Actually, the proxy class is a kinda neat idea and was hinted at by another 
poster (though I've lost that post now):

>>> class Pointer:
...  def __init__(self, dict, item):
...   self.dict = dict
...   self.item = item
...  def __getattr__(self, attr):
...   if attr == 'value':
...    return self.dict[self.item] 
...   raise AttributeError, attr
...  def __setattr__(self, attr, value):
...   if attr == 'value':
...    self.dict[self.item] = value
...   else:
..    self.__dict__[attr] = value
... 
>>> d = {'A':0,'B':0}
>>> a = {'A': Pointer(d, 'A')}
>>> print a['A'] 
<__main__.Pointer instance at 80d8e60>
>>> print a['A'].value
0
>>> d['A'] = 10
>>> print a['A'].value
10
>>> a['A'].value = 20
>>> print d['A']
20
>>> 

    At this point, I'd be taking a serious look at your actual requirements and 
debating whether builtin dictionaries are really what you want to be using. 
You've got complete freedom to create any type of class you need - don't limit 
your thinking by sticking to the builtin types!

        Richard

From BgPorter at NOacmSPAM.org  Fri Mar  3 17:31:36 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Fri, 03 Mar 2000 22:31:36 GMT
Subject: Python advocacy
References: <8wQv4.33688$pN1.381485@news1.rdc1.nj.home.com> <38BEE429.179D4865@prescod.net>  <38BF1990.CAA71DE6@prescod.net> <38BF6169.46FC6A66@netgate.net> <89p2kd$1arc$1@nntp6.u.washington.edu>  
Message-ID: 

"Fredrik Lundh"  wrote in message
news:VNVv4.40$y3.187028992 at newsb.telia.net...
> Brett g Porter wrote:
> > Obviously, Paul is entitled to his view that the cost of proficiency in
> C++
> > is too high. As a proficient C++ programmer, I have to disagree. I can't
> > imagine writing industrial-strength apps solely in Python.
>
> what if Python would grow a type system?
>
> and an industrial-strength compiler, on top of that?
>
> is there *anything* you can do in C++ that you
> couldn't express in a typed version of Python?
Hmmm, interesting.

At this point I'm still biased by 10 years of C / 5 years of C++ as opposed
to ~1 year of Python.  I'd certainly like to give it a try. I'd be happier
right now if I had a tighter/more transparent Python->C++->Python connection

If I could generate the same quality of code in Python, and had the same
caliber of development/debugging tools, maybe.

From moshez at math.huji.ac.il  Tue Mar  7 00:56:20 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 7 Mar 2000 07:56:20 +0200 (IST)
Subject: HELP: restore my faith in Python
In-Reply-To: 
Message-ID: 

On 6 Mar 2000, Neel Krishnaswami wrote:

> > Ahh, it's time to include extensions of the rational field. Let's get
> > more abstract. Abstraction is beautiful! And we need to represent 
> > transcendental numbers consistently aswell.
> 
> Is this possible? I've forgotten most of my analysis, but I was under
> the impression that you could divide the continuum as follows:
> 
>   o Rationals -- eg 2/3. Countably infinite.
>   o Algebraic irrationals -- eg sqrt(2). Countably infinite.
>   o Transcendental numbers -- everything else. Uncountably infinite.

Yes, but you can only "talk about" a countable number of them: the
constructive reals.

Definition:
A real number x is constructive (wlog assume that 0<=x<1, otherwise take
x%1) iff there's a recursive function f s.t.

	f(n)==0 iff the n'th bit in the binary expansion of x is 0

Or, in other words:
     oo
x = sigma [(1/2)**n]*f(n)
     1

Exercise: find an example of a non-constructive real. (It's easy: there
are uncountably many of them, as opposed to a countable number of
constructive reals)

> but the full transcendentals are just plain impossible. 

Constructive trancendentals are possible. If you don't mind never knowing
if a calculating some bit sticks your program in an infinite loop (halting
problem anyone?)

--
Moshe Zadka . 
http://www.oreilly.com/news/prescod_0300.html

From jean.hemmi at arakne.com  Wed Mar  1 05:36:39 2000
From: jean.hemmi at arakne.com (Jean Hemmi)
Date: Wed, 01 Mar 2000 10:36:39 +0000
Subject: Using sequence indexes greater than 2^32 on 64 bit computers
Message-ID: <38BCF2B7.7C7CE947@arakne.com>

Hello,

We're using a custom python sequence type on which we need to pass index

values greater than 2^32 (64 bit integers)...

We modified the definition of the sequence fonctions, so that they'd
accept
a long as parameter, instead of an int, but the interpreter still
truncates
the index values to the first 32 bits...

We made a workaround with special methods (slice_get(), slice_set(),
item_get(), item_set()), but the limitation prevents us from using the
build-in features (both the[] notation and the operator functions - e.g.

operator.getitem().

I believe this is a portage problem.
Shouldn't the 'off_t' type be used, instead of 'int' for all parameters
and
types describing index values ?

64 bit computer and the use of multi-terabyte systems / multi-gigabytes
files are coming around...
Is there any plan for fixing the problem ?

Thanks,

Frederic Giacometti
Arakne
fred at arakne.com

From echuck at mindspring.com  Wed Mar  1 16:48:07 2000
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Wed, 01 Mar 2000 16:48:07 -0500
Subject: Most common #! format
Message-ID: <38BD9017.4A2B5941@mindspring.com>

What's the most common #! format?

#!python
#!/usr/local/bin/python
#!/usr/bin/env python

or something else?

-Chuck

From torsten.hoermann at telekom.de  Thu Mar 30 08:59:52 2000
From: torsten.hoermann at telekom.de (Torsten H�rmann)
Date: Thu, 30 Mar 2000 15:59:52 +0200
Subject: extending python: c++ problem
Message-ID: <8bvmko$gki1@mailgate1b.telekom.de>

Does anyone know about the following problem and have a solution?
I've written a python extension using c++. No Problems with msvc, but on
HPUX with aCC. When importing the , the python interpreter can
not find the "initmod" function in the created lib. Error message:

>>> import mod
Traceback (innermost last):
  File "", line 1, in ?
ImportError: dynamic module does not define init function (initmod)
>>>

For compling and linking i have used the makefile.pre.in from the
python-distribution (no errors/warnings while building the mod.sl)
My Lib is a little more complex, but the following small lib demonstrates
this problem (it is a example from earlier posting in this newsgroup):

/*testmod.cpp*/
#include 

#include  // this include causes the error message; without,
there is no problem

static PyObject * mod_getsub(PyObject * self, PyObject * args)
{
    printf("worked\n");
    Py_INCREF(Py_None);
    return Py_None;
}

static PyMethodDef modMethods[] = {
    {"work", mod_getsub, METH_VARARGS},
    {NULL, NULL}
};

extern "C"
void initmod()
{
   (void) Py_InitModule ("mod", modMethods);
}

I have spend some hours with trying some ways to avoid this include, but now
i dont have any more ideas.
Can anyone help me????

Thanks Torsten

From embed at geocities.com  Fri Mar 31 09:35:18 2000
From: embed at geocities.com (Warren Postma)
Date: Fri, 31 Mar 2000 09:35:18 -0500
Subject: examples for HTMLgen
References: <14564.28003.931035.44380@gargle.gargle.HOWL>
Message-ID: 

> I know there is Zope, but that's too much. So what about picking the files
> into a directory like list, then sort that list in different ways and
> display it with HTMLGen???

  Hmm. I sense .... Confusion.

I think ZOPE is the niftiest thing since sliced bread. IN the off chance
that you haven't actually played with it, I'll just urge you to give it half
a chance.

Especially on Win32 using it's own built-in web-server it's extremely easy
to install. Even on my RH 6.1 Linux Box where it involved installing a few
RPMS and tweaking Apache's configuration to get it working through PCGI, it
took me less than half an hour to get it going.

After that, learning some basic DTML stuff took a few hours. It was worth
it. It was just like discovering Python all over, except now I have two
tools, Python for normal client side development, and Zope for generating
HTML, which uses Python objects in the background to gather and format the
data, and the DTML layer is just a thin presentation layer with ability to
handle HTML forms for gathering input from users, and so on.  If you haven't
tried it already, please do, it's so cool it's addictive.

Sorry, don't know any HTMLgen example links.

Warren

From rhicks at nospam.rma.edu  Fri Mar 17 19:31:28 2000
From: rhicks at nospam.rma.edu (Robert Hicks)
Date: Sat, 18 Mar 2000 00:31:28 GMT
Subject: Anyone working on IDLE or PythonWin for CE?
References: 
Message-ID: 

I have no problem running the demos on mine. I have a NINO handheld.

Bob

"Warren Postma"  wrote in message
news:qouA4.35130$Jz3.342199 at nnrp1.uunet.ca...
> Is Python for CE 1.5.2 too broken (unicode support, etc) to support an IDE
> onboard a Windows CE machine?  I can't even get the CE Task Bar Demo to
run
> on my machine (HP 620 LX). I'd like to mess with Python for CE more but it
> seems pretty much 'alpha quality'.
>
> Anyone?
>
> Warren
>
>

From dmost at magna.com.au  Tue Mar 14 00:02:53 2000
From: dmost at magna.com.au (dmost at magna.com.au)
Date: Tue, 14 Mar 2000 05:02:53 GMT
Subject: Job Offer: Python/Zope developer required in Manhattan
Message-ID: <8akh5s$1t3$1@nnrp1.deja.com>

We are looking for an experienced developer to work in Manhattan.
Applicants should have approximately 5 years experience working with OO
concepts in langauges such as python, c++ or java.
Applicants should be knowlegable in Python programming and in the Zope
application server, although a talented applicant willing to learn Zope
will also be acceptable.
Applicants should be comfortable with SQL and relational databases.
The job will entail work with Zope, but not exclusively so. There may
also be work involving java servlets, xml, xslt, and other technologies.
Pay is competetive.
--------------------
Damien Morton
Senior Programmer
Dennis Interactive
(212) 372 3829

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

From gherman at darwin.in-berlin.de  Thu Mar 16 05:01:37 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Thu, 16 Mar 2000 11:01:37 +0100
Subject: Tagging engines like PyFontify for other languages?
Message-ID: <38D0B101.1683912@darwin.in-berlin.de>

Hello,

I'm wondering if there are any Python tagging engines like 
Just van Rossum's PyFontify for languages other than Python?

That is, I'm looking for a Python module that is able to scan 
source code in, say, C, C++, Java, Perl or SQL (DDL/DML) and 
return a list of tags and positions in some way similar to 
what PyFontify is doing for Python.

Thank you in advance for any hints!

Regards,

Dinu

-- 
Dinu C. Gherman
................................................................
"The thing about Linux or open software in general is that 
it actually tries to move software from being witchcraft to 
being a science," [...] "A lot of the programs you see today 
are actually put together by shamans, and you just take it and 
if the computer crashes you walk around it three times... and 
maybe it's OK." (Linus Thorvalds, LinuxWorld 2000, NYC)

From lenny at squiggie.com  Thu Mar 16 19:17:35 2000
From: lenny at squiggie.com (Lenny Self)
Date: Thu, 16 Mar 2000 16:17:35 -0800
Subject: Cookie module
References:  
Message-ID: 

Oops. I goofed on the last post.

Your code should read:

ck = Cookie.SmartCookie()
ck.load(os.environ["HTTP_COOKIE"])
L = ck("ogin").value

Sorry.

Good luck, once again.

    -- Lenny Self
        lenny at squiggie.com

"Lenny Self"  wrote in message
news:BOeA4.932$pR.37158 at news.uswest.net...
> I believe your problem is that you do not the the 'get' in your call.  The
> call should look like this:
>
> ck = Cookie.SmartCookie(os.environ["HTTP_COOKIE"])
> L = ck("ogin").value
>
> Good luck.
>
>     -- Lenny Self
>         lenny at squiggie.com
>
> "Akhar"  wrote in message
> news:JjeA4.200$x24.24089 at weber.videotron.net...
> > I tried your cookie module it works fine under the python interpreter
but
> > when I use it on my web site it gives me an error I am trying to use:
> > ck = Cookie.SmartCookie(os.environ.get("HTTP_COOKIE",""))
> > L = ck.get("ogin").value
> > but it tells me
> > httpd: [Tue Mar 14 17:46:03 2000] [error] [client 192.168.0.2] Premature
> end
> > of
> > script headers: /var/www/cgi-bin/mail/msg.py
> >   File "/var/www/cgi-bin/mail/msg.py", line 20
> >     L = ck.get("ogin").value
> >     ^
> > SyntaxError: invalid syntax
> > httpd: [Tue Mar 14 17:50:43 2000] [error] [client 192.168.0.2] Premature
> end
> > of
> > script headers: /var/www/cgi-bin/mail/msg.py
> > So how can I get the info stored in a cookie from one form to another??
> >
> > Regards,
> > Akhar
> >
> > Note: do you of any good crypto module to encrypt and decrypt passwords?
> >
> >
> >
>
>

From ImranG at btinternet.com  Wed Mar  8 17:13:08 2000
From: ImranG at btinternet.com (Imran)
Date: Wed, 08 Mar 2000 22:13:08 +0000
Subject: Python (Mildly Humourous)
Message-ID: <4gjdcs846qb75s9pfeh9f29nqlul2jpgri@4ax.com>

I've been trying to get my college to install Python seeing as they already have
Basic, Pascal, Cobol, and VB, but when you've technical support like this what
do you do :-)

(Names and Domain deleted to protect the guilty)

Read the message bottom up as the reply is top-posted.

> From: *** ******** <*********@staff.**********.ac.uk>
> To: '10077583' <10077583@**********.ac.uk>
> Subject: RE: Interpreters on the Network
> Date: 07 March 2000 15:55
> 
> Hello,
> 
> To interpret what?
> 
> 
> *** ********
> 
> Technical Support Manager,S**** D**** College
> 
> 
> 
> -----Original Message-----
> From: 10077583 [mailto:10077583@**********.ac.uk]
> Sent: 07 March 2000 15:51
> To: *********@STAFF.**********.AC.UK
> Subject: Interpreters on the Network
> 
> 
> Hi,
> 
> Is there any chance of getting a Python and/or a Tcl/Tk interpreter
> available on the Network ?
> 
> Thanks,
> 
> Imran Ghory

Any suggestions how such as reply should be responded to ?

Imran
-- 
BTInternet FAQ v2.0 (Unofficial)
http://redrival.com/btifaq/

Sign the Linux Driver Petition
http://www.libranet.com/petition.html

From andy.georges at rug.ac.be  Thu Mar  2 03:20:03 2000
From: andy.georges at rug.ac.be (Andy Georges)
Date: Thu, 02 Mar 2000 09:20:03 +0100
Subject: A comp.lang.python code snippet archive?
References:  <200002281844.TAA19752@dionysus.fw.cuci.nl>
Message-ID: <38BE2433.8B4B033A@rug.ac.be>

Hans Nowak wrote:
> 
Hello,

> I would like to know:
> * if it's still worthwile to continue this site

Yes, please do. 

andy

==========================================================================
Dept. of Applied Mathematics and Computer Science, University of Ghent
Krijgslaan 281 - S9, B - 9000 Ghent, Belgium
Phone: +32-9-264.47.66, Fax: +32-9-264.49.95
E-mail: andy.georges at rug.ac.be 
==========================================================================

From gerrit at nl.linux.org  Fri Mar 31 16:27:26 2000
From: gerrit at nl.linux.org (gerrit at nl.linux.org)
Date: Fri, 31 Mar 2000 23:27:26 +0200
Subject: Advice requested: GUI project beginning
In-Reply-To: <8c0dpv$djc$1@nntp6.u.washington.edu>; from owen@astroNOJNK.washington.edu.invalid on Thu, Mar 30, 2000 at 12:35:09PM -0800
References: <8c0dpv$djc$1@nntp6.u.washington.edu>
Message-ID: <20000331232725.A3807@nl.linux.org>

> What Python windowing environment do you recommend for a new project? 
> wxPython? Tkinter? Java (via JPython)? (Is there anything else?). I'd 
> prefer something that is easy to use and scalable.

http://www.nl.linux.org/~gerrit/gui.html

regards,
Gerrit.

-- 
1011001 1101111 1110101 1110010 1110011 0101100
1000111 1100101 1110010 1110010 1101001 1110100

From alan.gauld at gssec.bt.co.uk  Mon Mar  6 07:25:38 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 06 Mar 2000 12:25:38 +0000
Subject: C++ / Java / Python (was Status of C++)
References: <38bf4389.272576531@news.mindspring.com>
Message-ID: <38C3A3C2.A692AF05@gssec.bt.co.uk>

Just to pick up a small point...

> >What I am concerned with is that C++ would be relegated to 
> the COBOL level, with lot of maintenance, migration to do 
> but with very little new development. 

This is plain wrong.
There are lots of new COBOL projects around and a recent 
Gartner(?) survey showed that the majority of the new code 
being written worldwide is still in COBOL - something 
like 25% was the figure I saw. (Remember that every 
single COBOL project produces a LOT of code!)

India is the second biggest software producer in the 
world and an awful lot of the stuff done there is sub 
contracted COBOL... BTW They can do pretty mean C++ too 
if that's what you need but they go with the market, 
there's very little language snobbery there :-)

Alan G.
-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.

From morbid at doesnotlikespam.inet.net.nz  Fri Mar 31 07:06:12 2000
From: morbid at doesnotlikespam.inet.net.nz (Morbid Curiosity)
Date: Fri, 31 Mar 2000 12:06:12 GMT
Subject: lambda
References: <20000329063753.20711.cpmta@c000.sfo.cp.net> <38e1d27a.27069487@news.inet.net.nz> <9sj4esgeh2i9emc4p249ksrj04u1bdm7ur@4ax.com> <8bv7k3$h7e$1@readme.uio.no> 
Message-ID: <38e49499.1173061@news.inet.net.nz>

On Thu, 30 Mar 2000 16:26:43 -0500, Joey Gibson
 wrote:
>||| > It took me a while to 'get' lambdas at first too. They are most useful
>||| > (IMO) when paired with map(), filter() and reject(). These functions take
>||| 
>||| Where did you find reject()?
>
>	D'ooh! I work in Smalltalk also, and that is one of its collection
>messages. Sorry. Python doesn't have one.

You're not thinking of reduce(), are you?

Morbid

From orlov at diasoft.ru  Wed Mar  1 08:57:44 2000
From: orlov at diasoft.ru (Oleg Orlov)
Date: Wed, 1 Mar 2000 16:57:44 +0300
Subject: Win32 KeyboardInterrupt
References: <38BD0AF8.50097EA4@nipltd.com>
Message-ID: <89j7da$2utb$1@gavrilo.mtu.ru>

import sys

def Worker(argv):
  while 1:
    pass

try:
  if __name__ =='__main__':
    Worker(sys.argv)
except KeyboardInterrupt:
  print "Saving state..."

Works for me. May be Worker() spawn another thread and sleep?

-----------
Oleg Orlov

From rob at hooft.net  Fri Mar  3 03:02:01 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: 03 Mar 2000 09:02:01 +0100
Subject: Tkinter vs. wxPython (was: Which GUI?)
References:   <5l3dqjckt5.fsf_-_@eric.cnri.reston.va.us>  
Message-ID: 

>>>>> "FL" == Fredrik Lundh  writes:

 FL> Rob W. W. Hooft  wrote:
 >> For fun with buggy user code in Tkinter, try:
 >> 
 >> import Tkinter
 >> b=Tkinter.Button()
 >> b.pack()
 >> l=Tkinter.Label()
 >> l.grid()

 FL> not to mention this highly annoying
 FL> Python bug:

 FL> while 1:
 FL>   pass

I've been bothered by that one as well :-), but at least a Control-C will
get you out with a nice stack trace to help you locate the problem. The
buggy user code I cited is a bit more involved:

 - It hangs, but Control-C won't kill it, so you can't get a stackdump....
 - b and l might be instantiated by completely different objects in
   different modules in the program, and just happen to have the same
   parent window.

These two features together make this very hard to debug.

Rob
-- 
=====   rob at hooft.net          http://www.xs4all.nl/~hooft/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========

From alex at somewhere.round.here  Wed Mar 29 11:40:42 2000
From: alex at somewhere.round.here (Alex)
Date: 29 Mar 2000 11:40:42 -0500
Subject: Incompatibility between socket and httplib in python CVS build.
Message-ID: 

Hi.  Running python httplib.py currently gives the following error
message:

arc% python httplib.py
testing HTTP...
Traceback (innermost last):
  File "httplib.py", line 272, in ?
    test()
  File "httplib.py", line 238, in test
    h.connect(host)
  File "httplib.py", line 112, in connect
    self.sock.connect(host, port)
TypeError: connect requires exactly 1 argument; 2 given

My python version is

'1.5.2+ (#7, Mar 29 2000, 11:20:40)  [GCC egcs-2.91.66 19990314/Linux
(egcs-1.1.2 release)]'

And emacs gives the CVS versions of httplib.py and socketmodule.c as
1.14 and 1.98

Alex.

From phil at river-bank.demon.co.uk  Tue Mar 28 12:03:46 2000
From: phil at river-bank.demon.co.uk (Phil Thompson)
Date: Tue, 28 Mar 2000 18:03:46 +0100
Subject: ANNOUNCE: PyQt/PyKDE v0.11 (Python Bindings)
Message-ID: <954263081.1503502440@news.demon.co.uk>

I've put v0.11 of my Python bindings for Qt and KDE in the usual place at:

http://www.river-bank.demon.co.uk/software

The main change is support for the new Qt v2.0.x widgets.  PyQt will also
compile with Qt v2.1.0-beta3, but the new Qt v2.1.x widgets are not yet
supported.

There are no visible changes to PyKDE - KDE 2 is not yet supported.

Hopefully, RPMs will follow soon.

Phil

From JGRAVES3 at austin.rr.com  Fri Mar 17 10:24:10 2000
From: JGRAVES3 at austin.rr.com (Jay Graves)
Date: Fri, 17 Mar 2000 15:24:10 GMT
Subject: Java editor for Python
References: <8atgsd$ava$1@vg170.it.volvo.se>
Message-ID: 

JJ wrote in message <8atgsd$ava$1 at vg170.it.volvo.se>...
>Hi!
>Anyone who knows where to find an editor written in Java with syntax color
>highlighting?
>It's a VERY big plus if there is a java bean and/or source code available.

http://jedit.sourceforge.net/

hth
Jay

From jstok at bluedog.apana.org.au  Tue Mar 21 01:22:17 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Tue, 21 Mar 2000 17:22:17 +1100
Subject: what is easier to learn first?...
References: 
Message-ID: 

Collin Greene wrote in message ...
>C+/+ or python or perl, I need an unbiased opionion here.

You expect an unbiased opinion on comp.lang.python.

>I have  NO
>programming experiance. thanks for your time.
>greene at hctc.com

Haskell.

Oh, Ok, Python.  Without doubt.  Perl and C++ are both baroque, complicated,
dense languages.   Python code is far cleaner.

From eq3pvl at eq.uc.pt  Mon Mar 20 05:41:01 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Mon, 20 Mar 2000 10:41:01 +0000
Subject: mod_perl and mod_python, is perl better for that task?
Message-ID: <38D6003D.7DF3AE89@eq.uc.pt>

Hi,

The concept of mod_perl is great for building sites with huge traffic,
the only really disavantage is
perl :-)

After looking (at a glimpse) for a similar mod_python I found PyApache
but it seems not to be very
used in the community.

Is python not well suited for webserver embedding? Or is just because
mod_perl has a better web
maintainer :-) ?

pvl

From mfletch at tpresence.com  Wed Mar  1 13:00:57 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 1 Mar 2000 13:00:57 -0500 
Subject: Newbie: Reading textfiles
Message-ID: 

First approach:

file = open( filename )
line = file.readline()
while line:
	doprocessing( line )
	line = file.readline()

Second approach:

import fileinput
file = fileinput.FileInput( filename )
for line in file:
	doprocessing(line)

HTH,
Mike
-----Original Message-----
From: Lutz.Schroeer at altavista.net [mailto:Lutz.Schroeer at altavista.net]
Sent: Wednesday, March 01, 2000 11:18 AM
To: python-list at python.org
Subject: Newbie: Reading textfiles

Hi,

after looking for a"clever" solution for reading a text file line by line 
from the beginning to the end for the whole day, I have to ask this 
ridiculous newbie question:

How do I read a textfile line by line and catch the eof?

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

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

From bparsia at email.unc.edu  Sat Mar  4 00:37:06 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Sat, 4 Mar 2000 00:37:06 -0500
Subject: Python advocacy
References: <38BEE429.179D4865@prescod.net> 
Message-ID: <1e6xz1s.uuanez7qwj3oN%bparsia@email.unc.edu>

Don Tuttle  wrote:

> Paul Prescod  wrote in message
> news:38BEE429.179D4865 at prescod.net...
> > O'Reilly asked to publish something I had written for my homepage about
> > Python.
> >
> > http://www.oreilly.com/
> >
> > I'm somewhat nervous about the flamage I'll get, but c'est la vie.
> 
> I bow before man with big, big gonads! 
[snip]

I'm afraid you're mistaken, since only last month I ran Paul through a
gonad reduction program...

They-don't-call-it-*Small*talk-for-nothing-ly y'rs,
Bijan Parsia

P.S. With apologies to Paul. And his gonands, of course, but I do
presume that they're, er, "man enough" to take a joke ;)

P.P.S. Please notice my manfully refraining from commenting on Paul's
comments on Smaltalk in that article. Especially since they were, to a
quibbler like myself, not *exactly* correct. But hey, it's an advocacy
piece. And he certainly did try to be nice :)

P.P.P.S. I notice that my newsreader (the very fine MacSoup) will
complain if my sig's too large or I have to much quoted material, but
cares not a whit if most of my article is in the P.S.s!

P.P.P.P.S It was certainly a much better article than that wretched Ruby
one, though I do detect a slight tension between these quotes:
        "C++ and Perl only make sense if you have a particular 
        programming background" 
and 
        "Python merges many of Smalltalk's ideas with a syntax that is 
        more likely to appeal to C, C++ and Java programmers".
I rather suspect it appeals more to Pascal and Modula-2 programmers :)
And poor Dylan, which sold it's soul to mixfix only to be doomed to
obscurity, in the general world, and scorn from its progeniters.

From mikael at isy.liu.se  Fri Mar 24 05:35:37 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 24 Mar 2000 11:35:37 +0100 (MET)
Subject: Newbie: White space
In-Reply-To: <8beg15$fdp$1@yeppa.connect.com.au>
Message-ID: 

On 24-Mar-00 Ryan Abrahams wrote:
 >  They should really put out a list of these, cause all newsgroups seem to
 >  have their own version of them :) ahahaha

You probably will enjoy the acronym finder at 

  http://www.acronymfinder.com/

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson 
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    24-Mar-00
Time:    11:33:37

This message was sent by XF-Mail.
-----------------------------------------------------------------------

From yawyi at gmx.de  Thu Mar  2 13:52:14 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Thu, 2 Mar 2000 19:52:14 +0100
Subject: Get dictionary-keys used to format a string
References: <38BE41F9.F2E9297A@inka.de>
Message-ID: <38beb8c4$0$18180@personalnews.de.uu.net>

Michael Str?der  wrote:

>'%(language)s has %(count)03d quote types.' % vars()
>
>I would like to extract the keys in the formatting string to a list.
>Well I can parse it myself but is there a built-in function to
>extract a list/tuple of the dictionary keys, e.g.
>['language','count'] in this case?

Try something like this:

------------
class GrabKeys:
    def __init__(self):
        self.keys = []
    def __getitem__(self, name):
        self.keys.append(name)
        return 0 # 0 is apparently compatible with all % format characters

gk = GrabKeys()

'%(language)s has %(count)03d quote types.' % gk

print gk.keys
------------

HTH,
Johannes

From daniels at mindspring.com  Wed Mar 15 21:22:05 2000
From: daniels at mindspring.com (Alan Daniels)
Date: 16 Mar 2000 02:22:05 GMT
Subject: Pointers
References: <200003152244.JAA27546@envy.fulcrum.com.au>
Message-ID: 

On Thu, 16 Mar 2000 09:44:45 +1100, the infinitely wise Richard Jones
(Richard.Jones at fulcrum.com.au) spoke forth to us, saying...

[snip...]
>In Python, the things you have come to know as variables are best
>thought of as "labels" for objects. In old-fashioned
>pointer-language-speak, they're about as equivalent to references or
>pointers as you're going to get.

For the benefit of the original poster, I'd like to also add that in
Python, whenever you invoke any of the comparison operators, you're
comparing the objects underneath, *not* the pointer values. Example:

>>> a = [1, 2, 3]
>>> b = [1, 2, 3]
>>> id(a)
135065392
>>> id(b)
134921328
>>> if a == b: print "True."
...
True.

See? Even though a and b are different objects (as shown by calling
"id" on them and getting different numbers), they evaluate to the
same, and that's why "True" prints out. Hope this helps.

-- 
=======================
Alan Daniels
daniels at mindspring.com
daniels at cc.gatech.edu

From deirdre at deirdre.net  Tue Mar 14 13:27:50 2000
From: deirdre at deirdre.net (Deirdre Saoirse)
Date: Tue, 14 Mar 2000 10:27:50 -0800 (PST)
Subject: Wash. D.C. Area: DCPIGgies, Mar. 13
In-Reply-To: <5qvu4.8333$al3.110244@newsc.telia.net>
Message-ID: 

On Mon, 28 Feb 2000, Fredrik Lundh wrote:

> > Can we add to the agenda:
> >
> > * Changing the name from "piggies" to something more respectable.
> 
> old topic:
> 
> http://www.deja.com/=dnc/viewthread.xp?AN=514428490
> 
> (start with John Landahls post -- the stuff
> above that one is from alt.pets.guinea-pigs ;-)

(somewhat belated in that I hadn't been reading on this topic)

One wonders, if John Landahl was so interested in forming a Python group
in San Francisco, why he never showed up to BayPIGgies (afaik) or signed
up for the mailing list.

Anyhow, my apologies to anyone without a sense of humor for using PIG, but
I was also thinking of the old Forth Interest Groups, aka FIGs.

-- 
_Deirdre   *   http://www.linuxcabal.org   *   http://www.deirdre.net
"During my service in the United States Congress, I took the initiative in
creating the Internet." -- Al Gore
"If Al Gore invented the internet, I invented spell check." -- Dan Quayle

From jpet at eskimo.com  Wed Mar  1 03:43:57 2000
From: jpet at eskimo.com (Jeff Petkau)
Date: Wed, 01 Mar 2000 08:43:57 GMT
Subject: Phython as In-Game scripting language
References: <20000222211406.41743.qmail@hotmail.com> <891g7l$cf8$1@nnrp1.deja.com> <893u76$46l$1@nnrp1.deja.com> <38b66215@news.xtml.co.uk> <20000225153511.19694.qmail@hotmail.com> <1e6rnn8.j4tz1uj8dlv4N@gershwin.bio.ic.ac.uk>
Message-ID: 

> > Shawn LeBlanc  wrote:
> > As a slightly off topic question, which other languages like
> > Python could be embedded efficently in a game project? I've
> > seen that several games use Lua (Grim Fandango), and there's
> > got to be other games that use embedded languages. What
> > would be the advantages and disadvantages of using Python
> > in a game?
>
> IIRC, the first-person shooter set in the Star Wars universe (the name
> escapes me - Dark Forces?), uses internal scripting extensively.

A couple of years ago I was working on a commercial RPG
where we tried using Python for the script language. Eventually
we switched to Java for a couple of reasons. My experiences:

* Python was very easy to integrate with C/C++, which is a big
deal for game coding. Java was a pain. [And Scheme is both
slow and painful, and I really wish I didn't have to use it on
my Visor 'cause it's really hard to write a closing paren in Graffiti,
it always comes out as a question mark. Um, where was I...]

* Java is pretty fast, so we were able to do almost everything in
it, and only had to go to C++ for collision detection and rendering.
Python is pretty slow; it would be best to limit it to scripting the
high-level actor behavior.

Incidentally, if stackless Python or microthreads had existed
back then, we probably would have stuck with Python. I'm still
waiting for the day when there will be a programming language
as good for describing actors and state machines as Simula was
25 years ago. (Scheme has the language power but was disqualified
when a blood test found high levels of parentheses.)

I actually considered implementing continuations in Python, but
then we switched to Java so I implemented them there instead.
One of my silliest experiments ever.

--Jeff

ps. We switched to Java for three reasons:
- Speed

- We wanted to standardize on a language internally, across some
  other projects, and the other projects didn't like Python because
  of the whitespace thing.

- I really wanted something like microthreads, and implementing them
  in Java as a .class file postprocessor seemed like a nifty project.

pps. Eventually the game got cancelled, no doubt because I
spent too much time fooling with script languages instead
of actually making a game. Sigh.

I'll stop rambling now.

From aa8vb at yahoo.com  Mon Mar  6 06:56:40 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 6 Mar 2000 06:56:40 -0500
Subject: IDLE and Tkinter
In-Reply-To: 
References: 
Message-ID: <20000306065640.B2115@vislab.epa.gov>

tf at malcolmsmith.net:
 |IDLE and Tkinter
 |
 |I'd love to try IDLE, but am having a bit of trouble with Tkinter.  I
 |*tried* to follow the instructions on the site under "Tkinker
 |Troubleshooting". I can get to the "make install" stage, when I get this:
 |
 |/python: error in loading shared libraries
 |libtk8.3.so: cannot open shared object file: No such file or directory
 |make: *** [libinstall] Error 127
 |
 |Ok, I'm not editing something correctly... 

What flavor of UNIX are you on?

The short answer is that Python wasn't built with the path to one of its
shared libraries (libtk8.3.so).  The loader doesn't find this library in
its default search path, so it can't load python.

If you're on a UNIX that supports it (most do), a quick fix (hack) is to do
a:

   setenv LD_LIBRARY_PATH /usr/local/lib

or wherever your libtk8.3.so lives.  (Note that if your shell is sh'ish
[ksh/bash/etc.] you'll use:  "export LD_LIBRARY_PATH=/usr/local/lib").
Then run python.  If it works, that's your problem.

The solution on most OSs is to compile in the path to libtk.  On other OSs,
you're responsible for adding that directory into the system ldconfig path.
Send us your OS and folks on the list can figure out which it is.

-- 
Randall Hopper
aa8vb at yahoo.com

From embed at geocities.com  Tue Mar 28 13:16:20 2000
From: embed at geocities.com (Warren Postma)
Date: Tue, 28 Mar 2000 13:16:20 -0500
Subject: Python - Next Release Questions
References:    
Message-ID: <6I6E4.934$HG1.29212@nnrp1.uunet.ca>

>
> UTF-8 != ASCII.
>
> (and I doubt people in the US and the UK make up
> 99% of all Unix users -- the rest of us tend to use
> 8-bit character sets that are definitely not UTF-8
> compatible...)

Thou hast indeed caught me flat footed!

I confess I thought UTF-8 was a clever renaming/co-opting of ASCII. You
know, embrace and extend, kinda like Microsoft....

Warren

From jeffp at crusoe.net  Sat Mar 18 14:51:37 2000
From: jeffp at crusoe.net (Jeff Pinyan)
Date: Sat, 18 Mar 2000 19:51:37 GMT
Subject: id() trickery
Message-ID: 

Well, I've broken my brain again.  Could someone explain this to me?

>>> a = id(1)
>>> a, id(1)
(732968, 732968)
>>> b = id(1)
>>> b, id(1)
(732968, 732968)
>>> a,b,id(a),id(b)
(732968, 732968, 732236, 732212)
>>> a = b + 0
>>> a,b,id(a),id(b)
(732968, 732968, 732188, 732212)
>>> a = b
>>> a,b,id(a),id(b)
(732968, 732968, 732212, 732212)
>>> a = b + 0
>>> a,b,id(a),id(b)
(732968, 732968, 732188, 732212)

Could someone tell me why the value returned by id(), which is just an
int, says type(), is acting a little fruity for me?  If a and b hold the
same integer, shouldn't they have the same value for id()?  And why do

>>> a = b # and
>>> a = b + 0

have two different effects?

-- 
MIDN 4/C PINYAN, NROTCURPI, US Naval Reserve             japhy at pobox.com
http://www.pobox.com/~japhy/                  http://pinyaj.stu.rpi.edu/
PerlMonth - An Online Perl Magazine            http://www.perlmonth.com/
The Perl Archive - Articles, Forums, etc.    http://www.perlarchive.com/

From gregj at pobox.com  Tue Mar 21 04:59:26 2000
From: gregj at pobox.com (Greg Jorgensen)
Date: Tue, 21 Mar 2000 01:59:26 -0800
Subject: Request for suggestions: a Workflow application
References: <6D8A17398E28D3119F860090274DD7DB4985A6@pces.cadlab.it> <8b1ehn$e6t$1@bunyip.cc.uq.edu.au> <38d53171.56301831@news.telus.net> <8b7gci$j4p$1@bunyip.cc.uq.edu.au>
Message-ID: 

"Chui Tey"  wrote in message
news:8b7gci$j4p$1 at bunyip.cc.uq.edu.au...
> There was a project around involving serverless version control. Basically
> using email/other mechanisms. I don't have the link anymore sorry.

I think you mean Code Co-op from Reliable Software.

    http://www.relisoft.com/co_op/

I've been using it at work and so far it's worked as expected.

Greg Jorgensen
gregj at pobox.com

From effbot at telia.com  Sat Mar 11 16:19:16 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sat, 11 Mar 2000 21:19:16 GMT
Subject: event loops and Python?
References:   
Message-ID: 

Kragen Sitaker  wrote:
> >just call mainloop(), and off you go.
>
> Looks like that calls a mainloop() method on a Tk toplevel.  This is
> bad if you're not running in X.

sorry, failed to parse that.  Tkinter uses X windows on Unix,
native toolkits on Mac and Windows.  the event loop works
in the same way on all platforms.  what's the problem here?

From godzilla at netmeg.net  Sat Mar 25 09:09:58 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sat, 25 Mar 2000 14:09:58 GMT
Subject: Where to install C extensions?
References: <8bfvsi$u44$1@netnews.upenn.edu> <38DB8AD7.92AF1B8C@abo.fi> <8bglb5$vbn$1@netnews.upenn.edu>
Message-ID: 

/usr/lib/python1.5/site-packages holds all my 3-rd party
extensions. its also in my my sys.path.

les schaffer

From prestonlanders at my-deja.com  Thu Mar 30 13:21:00 2000
From: prestonlanders at my-deja.com (Preston Landers)
Date: Thu, 30 Mar 2000 18:21:00 GMT
Subject: Python and Java
References: <38e26893@wwwproxy3.westgroup.com>
Message-ID: <8c05to$cg8$1@nnrp1.deja.com>

Give it a look:

http://www.xmlrpc.com/

In article <38e26893 at wwwproxy3.westgroup.com>,
"Chris Lada"  wrote:
> We have been writing an application in Python and are faced with a
delimma.
> We have to talk to a Java-based interface. Is there any alternatives
to
> writing yet another interface in JPython ?

--
|| Preston Landers  ||

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

From claird at starbase.neosoft.com  Sat Mar 25 22:39:40 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Mar 2000 21:39:40 -0600
Subject: Python Journal
References: <8bj8uf$3to@journal.concentric.net>
Message-ID: <88F1FCE07EF80A9B.CEA57A7F1805869F.CA1DBF693A16AD2C@lp.airnews.net>

In article <8bj8uf$3to at journal.concentric.net>,
Andres M. Hidalgo  wrote:
>Anyone know of any magazine publication dedicated to the python community
>like the ones
>dedicated to perl (Perl Journal) or Linux (Linux Journal) ???.
			.
			.
			.
Something with a name like *Python Journal* ?
-- 

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

From gmcm at hypernet.com  Wed Mar 15 09:51:39 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 15 Mar 2000 09:51:39 -0500
Subject: How to make a *.pyd or a DLL of several modules?
In-Reply-To: <953124408.460317@newsmaster-04.atnet.at>
Message-ID: <1258990592-4218892@hypernet.com>

Harald Angelow writes:

> I have several Python modules and want to put them into a PYD or into a DLL.
> Freezing only supports executeables, so how can i make one DLL of these
> modules?

You want to freeze them? I'm sure you can do it with the 
frozen code in a dll, but frozen modules need to be listed in a 
table in the executable, so having the code in a dll won't help 
you any.

Maybe you want something like:
http://starship.python.net/crew/gmcm/archives.html

> Importing that DLL should be the same like importing all Modules.

I suppose you could play games with "from x import *", but I 
believe you'd regret it. If they are all user-accessible, then 
each should be imported explicitly by the user, or one should 
play the role of gateway to the others.

- Gordon

From rwgk at my-deja.com  Tue Mar  7 22:36:33 2000
From: rwgk at my-deja.com (rwgk at my-deja.com)
Date: Wed, 08 Mar 2000 03:36:33 GMT
Subject: PyArg_ParseTuple() & Py_DECREF()
References: <89ph17$7si$1@nnrp1.deja.com>
Message-ID: <8a4hs1$p9v$1@nnrp1.deja.com>

For the records: The answer is actually here:
http://www.python.org/doc/current/ext/parseTuple.html
I am not sure why I missed it before.
Ralf

In article <89ph17$7si$1 at nnrp1.deja.com>,
  rwgk at my-deja.com wrote:
> Consider the following two lines of source code at the
> beginning of a C extension function:
>
>   PyObject *O;
>   if (! PyArg_ParseTuple(args, "O", &O)) return NULL;
>
> Is it necessary to call Py_DECREF(O) before leaving the
> extension function?
>
> Thanks!
>         Ralf
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

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

From aahz at netcom.com  Mon Mar  6 00:51:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Mar 2000 05:51:31 GMT
Subject: A comp.lang.python code snippet archive?
References:  <200002281844.TAA19752@dionysus.fw.cuci.nl>
Message-ID: <89vh13$6t5$1@nntp6.atl.mindspring.net>

In article <200002281844.TAA19752 at dionysus.fw.cuci.nl>,
Hans Nowak  wrote:
>
>* what kind of improvements I could make (a search engine would be cool but 
>I don't know much about CGI)

You don't need CGI.  See http://www.searchbutton.com/

(And, yes, I do work for them.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

I wonder what they do with all the fat they harvest from fat farms

From nobody at nowhere.nohow  Fri Mar 31 18:24:18 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Fri, 31 Mar 2000 23:24:18 GMT
Subject: Paid to work in Python? (was: Re: Choice Of Language)
References:  <8ajmib$elo$1@nnrp1.deja.com> <8c3ats$b5f$1@nntp.Stanford.EDU>
Message-ID: 

In article <8c3ats$b5f$1 at nntp.Stanford.EDU>, Amit Patel wrote:

>I do 100% of my professional development in Python, too.  Google's
>using a lot of Python and C++, and fortunately I get to do my work in
>Python.  :-)
>
>I should post Google's Python-related job openings to
>          http://www.python.org/Jobs.html

Pointless tanget: Is it just me, or is Google really the last usable search
engine?

-- 
Grant Edwards                   grante             Yow!  Is it FUN to be
                                  at               a MIDGET?
                               visi.com            

From toth.22 at osu.edu  Thu Mar 16 19:46:25 2000
From: toth.22 at osu.edu (CT)
Date: Fri, 17 Mar 2000 00:46:25 GMT
Subject: Project opinions please
Message-ID: <38d1801d.356312540@news-server.columbus.rr.com>

Hi folks,

I've been thinking lately of creating a python program that does a
number of automated tasks. Currently, I have to do all these manually.

1) Anyalize common logs for irregularties (errors, hack attempts,
etc...)

2) Check  programs on the server that access the network and generate
a log of their activity

3) Monitor ports to see what kind activity is occuring (hack attempts,
probes, etc...)

The user could tell the program what to look for, where to look, and
to how often to look.

I would like the agent/daemon to all the above on a user-defined
scheduled basis, and then email the results to the user.

Now, I realize this has been done before, but I want to do this myself
to help me learn some aspects of the python language and also because
I think it would be fun.

What I need, is some advice on how to approach this project, what type
of issues will I need to consider, where should I start(#1 question
acutally) and what other things I will need to learn...

Thanks for reading,
CT

From godzilla at netmeg.net  Sun Mar 19 20:20:39 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Mon, 20 Mar 2000 01:20:39 GMT
Subject: need fast parser for comma/space delimited numbers
References:  <87vh2ic2v6.fsf@lumen.med.iupui.edu>  <87wvmy7fjw.fsf@lumen.med.iupui.edu>
Message-ID: 

Mike Miller said:

> It checks that the number of fields in each non-comment line is
> constant.  I think that part would be fairly simple to remove.

sorry, i probably wasnt clear.

what i meant was that i have to read in a few hundred files at a time,
each file has equal length rows, but the number of columns varies
__from file to file__.

les

From timo at alum.mit.edu  Sat Mar 11 00:22:43 2000
From: timo at alum.mit.edu (Timothy O'Malley)
Date: Sat, 11 Mar 2000 00:22:43 -0500
Subject: Cookie module segmentation fault
References: <8a91fh$2kg$1@nnrp1.deja.com>
Message-ID: <110320000022433928%timo@alum.mit.edu>

hola.

In article <8a91fh$2kg$1 at nnrp1.deja.com>, Dan wrote:
> I recently ran upon a case where the cookie module was seg faulting on
> me.  
[..]
> --------------------------------
> import cookie
> a = 'Searches="jszXm!\\011Search#952590835.437"'
> cookie.Cookie(a)
> --------------------------------

As other posters note, the segmentation fault comes from the cPickle
module.  If you are using literal strings as your cookie values,
Cookie.py (now at version 2.24 or so) includes a way to skip the
cPickle parsing step:

     cookie.SimpleCookie(a)

See the Cookie.py module documentation for more information.

From gidon at norealaddress.ch  Mon Mar  6 05:40:00 2000
From: gidon at norealaddress.ch (Gidon Friedman)
Date: Mon, 06 Mar 2000 11:40:00 +0100
Subject: compile on aix 4.3
Message-ID: <38C38B00.3E44D373@norealaddress.ch>

hi to all

i get this problem with my first time installation of the latest python
source, i'd appreciate any hint:
-------------------------snip------------------
# ./configure
creating cache ./config.cache
checking MACHDEP... aix4
checking CCC...
checking for --without-gcc... no
checking for gcc... cc
checking whether the C compiler (cc  ) works...
------------------and now it hangs forever----------

thankx

gidon

From amused at webamused.com  Tue Mar 21 22:55:23 2000
From: amused at webamused.com (Joshua Macy)
Date: Wed, 22 Mar 2000 03:55:23 GMT
Subject: Embedding Python in a MUD
References: <38D7EAA8.6863DEE3@classic-games.com> 
Message-ID: <38D84126.BE5EB4EB@webamused.com>

Will Ware wrote:
> 
> Greg Miller (gmiller at classic-games.com) wrote:
> > I'm considering Python as an in-mud scripting language.
> 
> I'm interested in this. I want to do something similar. 

 You might want to check out MOOP in the Vaults of Parnassus.  MOOP is a
MUD/MOO that not only used Python as its scripting language, but is
written entirely in Python.

  Joshua

From greg at cosc.canterbury.ac.nz  Tue Mar 21 23:17:40 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Wed, 22 Mar 2000 16:17:40 +1200
Subject: binary to decimal conversion
References: <8b0r07$r5u$0@216.39.162.232> 
Message-ID: <38D84964.7B479DD@cosc.canterbury.ac.nz>

Michal Bozon wrote:
> 
> def btoi(num, o="0", i="1"):
>     def powL(a, e): # can handle BIG numbers (long)
>        ...
>     c = [o, i]
>     output = 0
>     for i in range(len(num)):
>         output = output+c.index(num[-i-1])*powL(2, i)

Calculating 2^n from scratch for each bit is a tad wasteful.
How about

  def btoi(num):
    result = 0L
    bit = 1L
    n = len(num)
    while n:
      n = n - 1
      if num[n] == "1":
        result = result + bit
      bit = bit << 1
    if result <= sys.maxint:
      return int(result)
    else:
      return result

-- 
Greg Ewing, Computer Science Dept,
+--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+

From pj at sgi.com  Wed Mar  1 17:34:53 2000
From: pj at sgi.com (Paul Jackson)
Date: 1 Mar 2000 22:34:53 GMT
Subject: Most common #! format
References: <38BD9017.4A2B5941@mindspring.com>
Message-ID: <89k5ud$e8fsj@fido.engr.sgi.com>

Don't know which is the most common, but:

    #!python

is a likely candidate for the least common.  

That's because this '#!' interpreter syntax requires the
full path (or at least a path relative to the current working
directory) of the interpreter.

That is, it doesn't to $PATH (execvp or execlp) path searches.

--

See also a thread in comp.lang.python, late April 1988,
entitled:

    Advantages of #!/bin/env python?

which discusses the pros and cons of these alternatives.
-- 

=======================================================================
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 spamfranke at bigfoot.de  Tue Mar 14 20:38:03 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Wed, 15 Mar 2000 01:38:03 GMT
Subject: Should Python documentation for __class__ be improved?
References: 
Message-ID: <38d1e896.8004769@news.btx.dtag.de>

On Tue, 14 Mar 2000 18:28:51 -0500, Tom Funk
<_spam_sux_tdfunk at _spam_sux_nettally.com> wrote:

>  o  when called it creates a new instance of the class  
>     i.e., 
>
>         c=Class(arg,...) 
>
>      is the same as 
>
>         c=self.__class__(arg,...)
>
>(Pretty cool stuff, Guido! :-)

Hmm, is this case, yes, but not generally: 

  c=self.__class__(arg,...)

creates an instance of self's class even in derived classes,
(descendants of 'Class') whereas 

  c=Class(arg,...) 

creates only instances of 'Class' itself. This trick makes Guido's
implementation even a bit cooler than you'd figured out by now..

>So, what __class__ *really* is, is: 

a reference to a *class object*, which is a first order object like
integers, strings and functions too. That means they can be assigned
or passed as parameters/return values or computed by an expression
like any other value in Python.

Regards,
Stefan

From pinard at iro.umontreal.ca  Mon Mar 27 12:19:39 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 27 Mar 2000 12:19:39 -0500
Subject: TrueType file handling
Message-ID: 

Hi, people.

I barely started writing Python code for handling TrueType files.  I'm
telling you in hope that we avoid duplicating efforts.  If you know that
this has already been done, please tell me and give me pointers.  If you
are interested in what comes out of this, tell me, and I'll keep you posted.

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

From jim_kerr at agilent.com  Thu Mar 16 13:46:15 2000
From: jim_kerr at agilent.com (Jim Kerr)
Date: Thu, 16 Mar 2000 10:46:15 -0800
Subject: problems using COM with Visual C
Message-ID: <8ara42$pks$1@nirvana.sr.hp.com>

  I'm trying to use Python's COM facilities to control a build in Visual
C++.
I've run into some problems that don't make sense to me at all (not
surprising,
since I don't know COM).

  Here's the script I have so far:

     # Open and manipulate a Visual C++ project

     from win32com.client import Dispatch
     vc=Dispatch("MSDEV.Application")
     vc.Visible=1
     vc.Documents.Open(r"C:\src\Workspace.dsw")
     projlist = vc.Projects
     ap = vc.ActiveProject()

  So far, so good; ap has the value 'MainApp', as it should. But if I
try to set the active project (from PythonWin)

    >>> vc.ActiveProject = 'Persist'

... it doesn't fly:

Traceback (innermost last):
  File "", line 1, in ?
  File "C:\Program Files\Python\win32com\client\dynamic.py", line 413, in
__setattr__
    self._oleobj_.Invoke(self._olerepr_.propMap[attr].dispid, 0,
pythoncom.DISPATCH_PROPERTYPUT, 0, value)
com_error: (-2147352567, 'Exception occurred.', (61837, 'Microsoft Visual
C++', 'Unable to write read-only property.',
'C:\\PROGRA~1\\MICROS~3\\COMMON\\MSDEV98\\BIN\\MSDEV.HLP', 455053, 0), None)

  The MSDEV documentation sure implies that ActiveProject is writable, so
this
error seems kind of confusing.

  So anyway, I'm doing something wrong... I can accept that. But here's the
odd thing:
from this point on, no matter what I type at the PythonWin prompt, a dialog
appears
that says:

  'The topic does not exist. Contact your application vendor for an updated
   Help file (129)'

  The only cure that I've found for this is to quit and restart PythonWin.

  Has anyone out there slayed these dragons before?

  FWIW, here are some OS and Python specifics:
    NT 4.0, Service Pack 5
    PythonWin build 125

  Thanks for the help!

-Jim

From herzog at online.de  Mon Mar  6 12:08:18 2000
From: herzog at online.de (Bernhard Herzog)
Date: 06 Mar 2000 18:08:18 +0100
Subject: HELP: restore my faith in Python
References: <38C00D46.8CC5BA67@americasm10.nt.com> <38C373C4.3EBB62D8@sightreader.com> 
Message-ID: 

grant at nowhere. (Grant Edwards) writes:

> If you want to round to the nearest integer, then use
> int(x+0.5).  That's been idiomatic in programming for longer
> than I've been around.

Unfortunately, it might produce incorrect results for negative x, at
least on some systems which truncate towards 0 and towards -infinity:

Python 1.5.2 (#1, Nov 13 1999, 12:17:58)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> x = -0.9
>>> int(x + 0.5)
0
>>> x = +0.9
>>> int(x + 0.5)
1

A better solution is to round() first.

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

From hanche at math.ntnu.no  Fri Mar  3 13:33:26 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 03 Mar 2000 13:33:26 -0500
Subject: C++ / Java / Python (was Status of C++)
References: <38bf4389.272576531@news.mindspring.com>
Message-ID: 

+ dickjoe_nospam_ at mindspring.com:

| You, dear Sir, are sadly misinformed.
| 
| I (with others) have written an industrial-strength image server
| completely in (portable) C++. [...]
| 
| Java you say? I laugh in your face. C++ is one of the most sought-
| after skills for "serious" programming (see below). [...]
| 
| C++ is pretty much the default choice for "serious" software, and
| in that I incude "system" and "application" programming.

I am glad you write "serious" in quotes.  You see, I have a little bit
of a problem understanding what people mean when they use that word,
but mostly it seems to mean large, complex, time consuming.

To help me in my own research, I tend to write a bunch of small
programs.  My latest one, to help me explore some PDE solutions, is
less than a thousand lines of python, using numpy and a small
extension written in C, plus a Tkinter piece for the visualization
bit.  It's a bit slow, but it works well enough for my purposes.  By
the time the program has served its purpose (hopefully having helped
my understand what's going on with this equation), it will probably
have used at most a few hours of CPU time and a few weeks of my time.
This program may not be "serious", but I assure you it's serious.

Trying to write this beast in C++ in order to cut the CPU time used
from hours to minutes would likely have increased my own time
expenditure from weeks to months - clearly an extremely bad bargain.

I wonder how much software out there is of this type - serious but not
"serious"?  So this is my wish for the day:  That people would stop
using the word "serious" merely to distinguish big, complex, CPU
intensive projects from those where programmer time is by far the most
important constraint (which also includes myriads of system
administration scripts, the other big category of small programs that
I find myself writing).

I am not saying that you don't understand this - on the contrary, it
seems that you do, at least if one reads between the lines a bit,
particularly as you *do* advocate python for some uses.  But I see the
idea of C++ as the *only* solution to *every* problem bandied around
too much, and I think that does a great disservice to the large (?)
number of people who, like me, need to write bunches of little
programs to solve their very real (and serious, thank you very much)
problems.

| Anything that runs on a server should not use java.

And why not?  Not every server carries a heavy load.

| Java is NOT an alternative to C++.
| Python IS an alternative to Java (except for lack of browser support)

On the contrary, I think both can be an alternative to C++, but you
need of course to consider the tradeoffs between programmer time and
other resource constraints.

| Lisp (its myriad flavors) has it adherents

Indeed, Common Lisp combines advantages in terms of programmer
productivity with the potential for very good efficiency (all modern
commercial CL systems can compile to native machine code).
Unfortunately, it has lost the popularity war, at least for now.

| I am not interested in language flame wars

Neither am I, and I hope I haven't contributed to one.  And I don't
really have any quarrels with you either.  I just wish you and others
would avoid that s-word.

serious-but-not-deadly-so-ly y'rs,
-- 
* Harald Hanche-Olsen     
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon

From alex at somewhere.round.here  Tue Mar 28 18:30:15 2000
From: alex at somewhere.round.here (Alex)
Date: 28 Mar 2000 18:30:15 -0500
Subject: Exec with loops
References: <8breac$39i$1@nnrp1.deja.com>
Message-ID: 

> Anybody know how to use exec to run code containing arbitrarily
> complex loops? Thanks.

This'll probably do the trick:

>>> s = '''for i in 10 * [None]:
  print 'Hi there'
'''
... ... >>> 
>>> t = compile (s, 'string', 'exec')
>>> exec (t)
Hi there
Hi there
Hi there
Hi there
Hi there
Hi there
Hi there
Hi there
Hi there
Hi there
>>> 

Alex.

From shogan at iel.ie  Tue Mar 14 10:32:26 2000
From: shogan at iel.ie (Shaun Hogan)
Date: Tue, 14 Mar 2000 15:32:26 -0000
Subject: what is the equivalent of scanf in python?
Message-ID: <021501bf8dca$80530d60$2801000a@iel.ie>

this may seem like a easy/strange question, but for the life of me i cant
remember:
what is the equivalent of scanf in python?
what code would i need to ask a user to input 2 numbers
then output the result?
Bewilderdly yours
Shaun

From ez4glNOezSPAM at hotmail.com.invalid  Tue Mar  7 17:01:19 2000
From: ez4glNOezSPAM at hotmail.com.invalid (ez4gl)
Date: Tue, 07 Mar 2000 14:01:19 -0800
Subject: Python scripting engine for ASP
Message-ID: <0721e130.1fa10121@usw-ex0103-023.remarq.com>

I may soon be going to work in a Microsoft shop (not that I'm in
an open source shop now), but want to sneak Python in as my ASP
programming language -- can anybody direct me to where I might
find the necessary scripting engine?

Thanks in advance

George

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!

From tom-main at REMOVEME.home.com  Sat Mar 25 16:54:42 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Sat, 25 Mar 2000 21:54:42 GMT
Subject: Python tutorial/comparison for C++ programmer
References:  <8bicur$jnt$2@news1.xs4all.nl>
Message-ID: 

I tried the link but it didn't work.

Tom.

"Boudewijn Rempt"  wrote in message
news:8bicur$jnt$2 at news1.xs4all.nl...
> Tom  wrote:
>
> > I'm a C++/MFC programmer and I'm interested in learning Python.
>
> > Are there any Python language resources geared to C++ programmers?  I
guess
> > I'm looking for something that would compare the syntax, concepts, etc.
of
> > the two languages, eg. how do the object models differ.
>
> > (I've already found Mark Hammond's Win32/MFC Python stuff.)
>
> Well, it's probably not directly relevant, but I took a C++ tutorial
> and translated it into a Python tutorial. The original was meant for
> beginning C++/Qt developers. Taking a look at:
>
> http://www.valdyas.org/python/tutorial.html
>
> can't hurt ;-).
>
> --
>
> Boudewijn Rempt  | http://www.valdyas.org

From cobrien at Radix.Net  Mon Mar 27 16:41:50 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 27 Mar 2000 16:41:50 -0500
Subject: Zope
References: <69sudsokeqsk8iaes7jdrjpd24nfkm9r61@4ax.com>
Message-ID: <8bokiu$akp$1@saltmine.radix.net>

In article <69sudsokeqsk8iaes7jdrjpd24nfkm9r61 at 4ax.com>,
Anders M Eriksson   wrote:
>Hello!
>
>I have some questions about Zope and I thought I ask them before
>spending alot of time reading the documentation.
>
>
>
>
>Can I install zope on a apache system WITHOUT being root? My ISP is
>running apache on a Solaris 5.6 system.
>

Yup.  It will be easier if python >= 1.52 is already installed.  If
not you will have to build that in your own directory.  Check the
./configure --help output to see what flags to set.

Zope, by default, binds to unprivileged ports, so you can run it
just fine.  

>I want to create a system that lets any user create his own User Data
>Record. The user data record will contain data about the user, e.g.
>Name, email address, etc. This record should only be editable by the
>user (and admin) with somekind of login. This user data must then be
>searchable! Listings on all or some of the data should be able to
>present in different sorting orders, e.g. Lastname, country, etc.
>
>Is Zope suited for this kind of system?

Absolutely.  The big question is whether or not you want 
your system backed by a RDBMS or not.  But zope is happy
either way.

>Maybe someone already has created a system like this?
>

See http://www.zope.org for more details.  You probably
want to get on the zope mailing list. Start with the digest,
this is a high volume list.

-- cary

From wilson at mclachlan.com.au  Thu Mar 16 18:06:25 2000
From: wilson at mclachlan.com.au (Wilson Fletcher)
Date: 16 Mar 2000 23:06:25 GMT
Subject: Assigning output from a simple python statement
References: <01bf8f1f$a611b640$3201a8c0@wraith.mclachlan.com.au> 
Message-ID: <01bf8fa4$9407a240$3201a8c0@wraith.mclachlan.com.au>

Michael Hudson  wrote in article
...
> "Wilson Fletcher"  writes:
> 
[...]
> 
> I leave out my usual "do you really want to `exec' things you don't
> know everything about" comment, I think.
> 

hehehe, I didn't say I didn't know the source, only that I couldn't be sure
of the code.

From eq3pvl at eq.uc.pt  Mon Mar 27 04:37:18 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Mon, 27 Mar 2000 09:37:18 +0000
Subject: Scripting language poll at linux.com
Message-ID: <38DF2BCE.291B7D81@eq.uc.pt>

There is a Scripting language poll at
http://www.linux.com/polls/index.phtml

Go and and show them who rulez.

From tuttledon at hotmail.com  Tue Mar  7 20:27:36 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Wed, 08 Mar 2000 01:27:36 GMT
Subject: Python performance
References:  
Message-ID: 

I've  moved this question to "Is it Python or is it C ?"

From effbot at telia.com  Tue Mar 14 11:02:56 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 14 Mar 2000 16:02:56 GMT
Subject: Problem with tkinter & spawning processes
References: <8ale7h$bl$1@news1.sunrise.ch>
Message-ID: 

nathan  wrote:
> I've programmed a GUI with Tkinter and I've started  it with the
> mainloop() as usual.  Now, I want to start a new process from
> the GUI, by pressing a button.
>
> I've tried with:
> os.system('"C:\Program Files\Python\Python.exe" -i test.py')
>
> The new process starts fine (standard output on the DOS cmmand
> prompt) nd I cannot do anything more.
>
> It seems that the new started process blocks the mainloop() thread

the os.system call doesn't return until the new process
has terminated.

> How can avoid this?

this might work:

    os.system('start "C:\Program Files\Python\Python.exe" -i test.py')

or you could use os.spawn(os.P_NOWAIT, ...).  see
http://www.deja.com/=dnc/getdoc.xp?AN=589728736

(tweaking that script to support a wait/nowait flag is
left as an exercise for the reader)

From quinn at hedono.ugcs.caltech.edu  Tue Mar  7 20:54:50 2000
From: quinn at hedono.ugcs.caltech.edu (Quinn Dunkan)
Date: 8 Mar 2000 01:54:50 GMT
Subject: sigchld vs. os.popen, os.command, etc.
References:  <8a3cj5$15bg$1@nntp6.u.washington.edu> <38C572CC.90C3BA6@endea.demon.nl>
Message-ID: 

On Tue, 07 Mar 2000 22:21:16 +0100, Niels Diepeveen 
wrote:
>Donn Cave schreef:
>
>> It's too bad select() can't wait for processes along with I/O, that
>> would be a nice solution in some cases.
>
>What if you pass the write end of an unused pipe to the child process?
>When the child terminates it will be closed and select() will return
>ready on the read end. At least that is what I would expect.

Hmm, that's a clever idea, but I wound up polling with waitpid + WNOHANG,
which seems simpler and more synchronous anyway (and fits nicely into a 15-sec
event loop).

Hmm, I wonder why we can't just have a waitpid()-like call with a timeout and
multiple pids...  I guess you could set an alarm and waitpid(-1) or
something...

From clopez at abo.fi  Mon Mar 20 13:34:48 2000
From: clopez at abo.fi (Cesar Lopez)
Date: Mon, 20 Mar 2000 20:34:48 +0200
Subject: How do I debug embedded python code?
Message-ID: <38D66F47.E96304D4@abo.fi>

I have to call a python module from a C program, and I don?t know how to
debug this python code.

main
{

......

 co = (PyCodeObject *)v;
 v = PyEval_EvalCode(co, d, d);

}

Where co it is a buffer with the python code of a python program:

test.py

import led
led.Setup(1)

From claird at starbase.neosoft.com  Tue Mar  7 13:14:54 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 7 Mar 2000 12:14:54 -0600
Subject: Python advocacy
References: <8wQv4.33688$pN1.381485@news1.rdc1.nj.home.com> <89p2kd$1arc$1@nntp6.u.washington.edu>  <38C411A1.B552C42@prescod.net>
Message-ID: <01DCAB3E36313BC3.A8B11C96704C6225.0408D4A80634DAF4@lp.airnews.net>

In article <38C411A1.B552C42 at prescod.net>,
Paul Prescod   wrote:
			.
		[Maybe Java can be
		used for everything;
		in any case, Tcl is
		crap]
			.
			.
>Then a language like Python comes along which subsumes the abilities of
>(at least) TCL, Perl, Visual Basic and ten thousand macro languages. It
>is my assertion that there are no large, complex problems that can be
>solved more easily in any of those languages than in Python with a
>suitable library and (in the case of VB) development environment.
			.
			.
			.
>choosing our programming languages for us. Yes, you can use sh and VBA
>for tiny little programs but isn't it just one damn more thing to learn?
>Is it worth learning another language to reduce a 10 line program into
>1?
>
>Given that Python offers us an opportunity to avoid the code
>fragmentation caused by all of these little fragmented language
>communities (Perl for text processing, TCL for Unix GUIs, VB for Windows
>GUIs, other Basic variants for macros), it is only natural to ask if it
>can also swallow the huge category of tasks being done with Java (we'll
>leave C and FORTRAN alone for now...). If Sun succeeds in making Java
			.
			.
			.
> 1. that the one complex feature we still cannot profitably strip out is
>type safety -- if we strip that out we won't be able to build big,
>complex systems (sound familiar?) or at least efficient systems
>
> 2. that there is no good way to introduce this feature into an
>erstwhile dynamically typed language
>
>I am told by those in the know that Common Lisp disproves both of these
>assertions. You can get fast code and complex systems from a language
>with an optional "add-later" type declaration system. In fact, people
>write large, complex systems in competely dynamic Lisps, Smalltalks and
>low and behold, Python. The people writing the largest systems seem the
>least inclined to argue that the lack of a static type system is
>limiting them.
>
>It is debatably the case that static type checking systems help people
>to write code that scales to large systems. It is indisputably clear
>that it can help the readability of the code. It is relatively clear
>that it also helps the performance without massively complicating the
>implementation. That's why Python is likely to eventually grow a static
>type system. Then we can shove another sock in mouth of the apologists
>for languages that are not good at one thing or another.
>
>Can there be one language that does everything? Everything? Possibly
			.
			.
			.
>> Obviously, Paul is entitled to his view that the cost of proficiency in C++
>> is too high. As a proficient C++ programmer, I have to disagree. I  can't
>> imagine writing industrial-strength apps solely in Python (and I do mean
>> industrial -- I've got applications running in steel mills).
>
>To be honest with you, if I was required to write an application that
>was going to run in a steel mill, and I was not constrained by memory or
>CPU (which you often would be) then I would choose a dynamically typed
>language with automatic memory management and pointer safety over a
>statically typed language without, any day. Tbe safe I would probably
>use a language like Java or Eiffel with both, just to be safe.
>Seriously, where reliability is an issue, C++ would be at the bottom of
>my list. Dr. Watson's frequent appearance on Windows NT desktops is a
>testament to C++'s reliability.
			.
			.
			.
Paul and I could happily go off in a corner and split hairs
for--well, for a long time.  I disagree with several claims
about Tcl, Java, ...; those who truly care about the details
are welcome, as usual, to ask.

I'll decorate a few of his more central points, though:
1.  Sun's not going to make Java a success.  I
    think Sun still intends that, but hasn't
    the capacity.

    IBM probably *will* make Java a success.
2.  I have no idea whether Java will make it in
    the embedded world.  My current expectation
    is that the project of porting Java every-
    where's going to collapse in a sorry heap.
    Maybe it'll work.  I'm unconvinced.
3.  Type safety is a tiny and essentially ir-
    relevant part of what it really takes to
    build big, complex systems.  I know you
    know this.  It needs to be repeated.
4.  Do NOT buy into the nonsense, even for a
    moment, that C and C++ are "better supported
    by tools".  They *require* tools to compensate
    for their burdens of memory management, ...
    That's the whole story.
5.  We're not far away from a time when people
    will say, "Controlling a steel mill/submarine/
    power plant/refinery/... with C++ codings?
    Are you serious?  That would be like, well,
    like, driving a car without having the children
    buckled into seat belts."  Yup.  We did it for
    a lot of years.  There's a much better way.
6.  Complex systems never work, anyway.
7.  Complex systems have a shot of working when
    they're constructed from reliable parts.  The
    winning languages of the '90s were those that
    played well with other technologies.
8.  Ada and Eiffel:  they're safe.
9.  As much as Paul wants to do everything with one
    language, I'm probably as strong on the other
    side in looking for a way to accomodate lots of
    different languages.
10.  If I had to choose only one language, it would
    definitely be Python.  It's not really ready,
    though, without Unicode ...
11.  It's the libraries.  Syntax--piffle!

I'm probably still better at C than anything else.  I can
process signals and respond to external ports and maintain
complex data structures and so on without leaking memory or
violating array bounds or confusing argument lists.  I'm an
anachronism, though.  I might as well be knapping flints.
Industrial customers are better off with any of several
other languages, although they don't yet realize it.

My most important point:  yes, we lose some benefits when
moving away from C and C++.  With very few exceptions, those
are costs we can afford, in comparison with the greater
benefits brought by other modern languages.  Don't get stuck
waiting for alternatives which are "perfect".
-- 

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

From see_plus_plus at my-deja.com  Wed Mar  8 00:04:12 2000
From: see_plus_plus at my-deja.com (see_plus_plus at my-deja.com)
Date: Wed, 08 Mar 2000 05:04:12 GMT
Subject: Python performance
References: <200003080330.OAA01737@nexus.csse.monash.edu.au>
Message-ID: <8a4n0a$spu$1@nnrp1.deja.com>

Not only the feathers were ruffled but somebody even fell out of his
tree without any technical & commercial counter pointer.

Whenever somebody writes a spreadsheet program in Python that outsells
MS Excel, C can be considered as obsolete.

cpp

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

From donn at u.washington.edu  Tue Mar  7 19:39:27 2000
From: donn at u.washington.edu (Donn Cave)
Date: 8 Mar 2000 00:39:27 GMT
Subject: sigchld vs. os.popen, os.command, etc.
References: <38C572CC.90C3BA6@endea.demon.nl>  <8a3cj5$15bg$1@nntp6.u.washington.edu>
Message-ID: <8a47fv$9jk$1@nntp6.u.washington.edu>

Quoth Niels Diepeveen :
| Donn Cave schreef:
|> It's too bad select() can't wait for processes along with I/O, that
|> would be a nice solution in some cases.
|
| What if you pass the write end of an unused pipe to the child process?
| When the child terminates it will be closed and select() will return
| ready on the read end. At least that is what I would expect.

Yes, that's worth trying.

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

From greglandrum at earthlink.net  Mon Mar  6 08:47:01 2000
From: greglandrum at earthlink.net (greg Landrum)
Date: Mon, 06 Mar 2000 13:47:01 GMT
Subject: Choice Of Language (was: Python Advocacy)
References: 
Message-ID: <38C3B7AF.FCCD13EA@earthlink.net>

Don Tuttle wrote:
> 
> I have never programmed for a living.  (I enjoy it way too much to ever do
> that! )  So let me ask all you folks that do.  What percentage of
> your paid, professional time do you get to use Python?
> 

I'm (happily) at about 90-95%

-greg

-- 

greg Landrum (greglandrum at earthlink.net)
Software Carpenter/Computational Chemist

From vetler at ifi.uio.no  Sun Mar  5 11:03:04 2000
From: vetler at ifi.uio.no (Vetle Roeim)
Date: 05 Mar 2000 17:03:04 +0100
Subject: Choice Of Language (was: Python Advocacy)
References: 
Message-ID: 

* Don Tuttle
> [Ran]
> > No,  the project manager chooses the language(s) according to the needs
> > of the project and the availability of programmers and other resources.
> 
> I have never programmed for a living.  (I enjoy it way too much to ever do
> that! )  So let me ask all you folks that do.  What percentage of
> your paid, professional time do you get to use Python?

It's possible to get *paid* for programming? 
Where? ;-)

it's-possible-to-do-it-for-a-living
  -without-getting-paid-for-it--ly y'rs, vr

-- 
Vetle Roeim 
Graduate Student 
Departement of Informatics, University of Oslo, Norway

From mbarre at mac.com  Tue Mar  7 21:43:57 2000
From: mbarre at mac.com (Matthew Barre)
Date: Tue, 07 Mar 2000 20:43:57 -0600
Subject: newbie simple exec problems
References: 
Message-ID: 

That works, thank you. What do I need to do to get it to work for cgi? Can I
just put ~user/cgi-bin/.cgi.py?

-Matt
---------
"Ray! When someone asks you if you're a god, you say YES!!" -Winston Zedmore

> From: Michael Hudson 
> Organization: University of Cambridge, England
> Newsgroups: comp.lang.python
> Date: 08 Mar 2000 00:15:52 +0000
> Subject: Re: newbie simple exec problems
> 
> Matthew Barre  writes:
> 
>> I have been learning python on my mac, and I just setup a RH6.1 box on my
>> little LAN to practice Python as a cgi tool etc. Well the cgi thing has me
>> completely baffled and right now I can't even get a Hello World script to
>> run at the prompt ie:
>> 
>> $hello.py
>> bash:unrecognized command
>> 
>> I have tried chmod 0755 hello.py and
>> chmod a+x hello.py
>> 
>> The first line of the file is
>> #!/usr/bin/python
>> 
>> which is where I checked and that's where python is. I'm baffled. Right now
>> if I try to get a cgi the browser just prints the script...ack...what am I
>> doing wrong?
>> 
>> -Matt
>> ---------
>> "Ray! When someone asks you if you're a god, you say YES!!" -Winston Zedmore
>> 
> 
> Try
> 
> $ ./hello.py
> 
> `.' may well not be on your $PATH...
> 
> HTH, HAND
> Michael
> 
> -- 
> very few people approach me in real life and insist on proving they are
> drooling idiots.                         -- Erik Naggum, comp.lang.lisp

From mhammond at skippinet.com.au  Wed Mar 15 16:50:16 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 15 Mar 2000 21:50:16 GMT
Subject: win32ui.GetActiveWindow() problem
References: 
Message-ID: 

"Michal Bozon"  wrote in message
news:Pine.OSF.4.10.10003151652400.19621-100000 at prfdec.natur.cuni.cz...
> On Tue, 14 Mar 2000, Robin Becker wrote:

>  I use Pythonwin build 125 that I think is newest.

128 is the newest public release, but only at
http://starship.python.net/crew/mhammond

(129 just got released to registered users  - source code folding and
multi-views of the same source :-)

Mark.

From mwh21 at cam.ac.uk  Mon Mar  6 12:03:19 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Mar 2000 17:03:19 +0000
Subject: Self Nanny
References: 
Message-ID: 

Gaetan Corneau  writes:

> Would programmers who have a background in a language other than C++ really
> miss "self" in the method declarations? 

As I understand it, it's not there to make your life easier (or
harder), it's there because Python doesn't have declarations.

I tried to explain this in this article:

http://x37.deja.com/getdoc.xp?AN=588006439

and another that doesn't seem to be on deja...

Cheers,
M.

-- 
very few people approach me in real life and insist on proving they are
drooling idiots.                         -- Erik Naggum, comp.lang.lisp

From paulb at infercor.no  Wed Mar  8 06:03:21 2000
From: paulb at infercor.no (Paul Boddie)
Date: Wed, 08 Mar 2000 12:03:21 +0100
Subject: Splitting comp.lang.python
References:  <20000301105629.A4323@nl.linux.org>
Message-ID: <38C63379.423947D9@infercor.no>

gerrit at nl.linux.org wrote:
> 
> 4000 messages in 29 days is a lot. it's 138 messages a day. 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?
> 
>     I propose to split c.l.py in a technical part and a non-technical part.

I think any proposal has to take into account the following factors:

  * How obvious it is to decide where to post messages. If you're a new user,
    is it obvious where to post your message? Consider that new users may do a
    quick, desperate search for a newsgroup and, given that their news feed
    may be unreliable or that they're doing a search for newsgroups on
    deja.com, find only comp.lang.python and not the hypothetical
    comp.lang.python.newuser.

  * How well the traffic is split between the new groups. Many suggestions
    have only shown how good people are at classifying topics rather than the
    suitability of those classifications. It might be possible to create a
    new group for each of the SIG topics, but how popular are each of those
    topics?

  * How encompassing the topic of a particular group is. As I pointed out in
    an earlier message, it is not particularly desirable to have many groups
    where the majority of messages are cross-posted to other groups.
    Discussions tend to appear and disappear as people become tempted to "move"
    threads between groups.

  * Whether a group is really going to help people, or allow meaningful
    discussion to take place. Advocacy groups rarely allow either activity to
    go on.

One thing to remember is that as users become more technically competent, the
more likely it is that they will be able to subscribe to special interest
mailing lists and manage these interests appropriately. The issue here is
whether newsgroups are really necessary for such users or whether mailing lists
are enough.

what-are-we-trying-to-achieve-with-newsgroups-anyway?-ly-y'rs,

Paul

From cpr at emsoftware.com  Wed Mar 15 14:36:05 2000
From: cpr at emsoftware.com (Chris Ryland)
Date: Wed, 15 Mar 2000 19:36:05 GMT
Subject: Loop performance disappearance
References: <38CF938F.D56437EA@helsinki.fet> 
Message-ID: 

I always assumed (perhaps being too much of a newbie) that

for i in range(n):
   :

would automatically be recognized by the compiler and turned into a simple
loop control using i alone (like turning it into an xrange, I guess). Seems
simple enough and common enough that it should be automatic, no? (Forgive
any massive display of ignorance here.)

--
Cheers!
/ Chris Ryland, President / Em Software, Inc. / www.emsoftware.com
"Remco Gerlich"  wrote in message
news:slrn8cva9a.aj8.scarblac-spamtrap at flits104-37.flits.rug.nl...
> Mikael Johansson wrote in comp.lang.python:
> > I was just wondering what the reason for the huge performance decrease
> > in a loop execution when the number of steps exceeds some critical value
> > is. To give an example:
> >
> > for i in range(loops):
> >     pass
> >
> > If loops=500 000 (space added for clarity) the "program" executes in ~2
> > secs on my machine. But if loops is set to 5 000 000, the execution time
> > rises to substantially more than tenfold (I terminated it after one
> > minute). However the CPU-load is quite small, most of the time is spent
> > disk swapping like crazy!
>
> That's because range(loops) is a list. It actually has to construct a list
> that size, and then walks through it. 5 million doesn't fit in your memory
> and it starts swapping.
>
> > Any ideas of the reason for this, work-arounds?
>
> Use xrange instead of range. xrange doesn't return a list, but an object
> that generates values on demand, simulating a list. Slightly slower, but
> doesn't need the memory.
>
> --
> Remco Gerlich,  scarblac at pino.selwerd.nl
>   Murphy's Rules, "Which is why you get 'em so cheap":
>    In SPI's Universe, the sword is prohibited from use at any combat
>    range.

From shichang at icubed.com  Fri Mar 10 17:36:52 2000
From: shichang at icubed.com (Shichang Zhao)
Date: Fri, 10 Mar 2000 22:36:52 -0000
Subject: Anti- Re: De-compiler for *.pyc files
Message-ID: <01BF8AE1.22230BA0.shichang@icubed.com>

If I make a product using Python, that means someone gonna be able to 
decompile it and peek the source? Than how do I protect my private part? 
(no misunderstanding, I am a person supporting Open Source, but you know, 
if you work for other companies, they may not like the idear)

-----Original Message-----
From:	Michael Hudson [SMTP:mwh21 at cam.ac.uk]
Sent:	Friday, March 10, 2000 10:00 PM
To:	python-list at python.org
Subject:	Re: De-compiler for *.pyc files

"Zarkon & Zena"  writes:

> Does anyone have a "decompiler" that I can use to re-create the original
> *.py scripts from the "compiled" *.pyc files?
> I am stuck trying to make changes to some legacy Python code that I do 
not
> have the source for.
> Any help or suggestions would be greatly appreciated.

Do to www.google.com, type "decompyle" into the box, and then hit "I'm
feeling lucky".

Warning: It probably won't work.  But it's worth a try.

Failing that, you can just use the `dis' module on the functions and
work out what they do by hand, which is not very hard, but very
tedious.

Cheers,
M.

--
very few people approach me in real life and insist on proving they are
drooling idiots.                         -- Erik Naggum, comp.lang.lisp
--
http://www.python.org/mailman/listinfo/python-list

From markx.daley at intel.com  Thu Mar 23 10:21:13 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Thu, 23 Mar 2000 07:21:13 -0800
Subject: Jack-pot
Message-ID: 

No, I got a large number of identical listings as well.

- Mark

-----Original Message-----
From: Mikael Olofsson [mailto:mikael at isy.liu.se]
Sent: Thursday, March 23, 2000 1:37 AM
To: python-list at python.org
Subject: Jack-pot

Gosh, what's up? Suddenly I get several copies of each posting to 
c.l.py. I follow the group through its mailing list. Am I the only
one having this problem?

split-ends-ly y'rs

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson 
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    23-Mar-00
Time:    10:33:27

This message was sent by XF-Mail.
-----------------------------------------------------------------------

From nascheme at enme.ucalgary.ca  Wed Mar 29 14:43:06 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Wed, 29 Mar 2000 19:43:06 GMT
Subject: socket.connect() and socket.bind() change
References:  <20000329195448.O3140@coltronix.com>
Message-ID: 

I think this change is going to break a lot of code.  The example
in the documentation used to use:

    s.bind(host, port)

and

    s.connect(host, port)

Unlike the append() change I would consider this to be changing a
documented interface.  All of my socket code will be broken by
this change.  Anyone else?

    Neil

From alex at somewhere.round.here  Mon Mar 27 12:16:17 2000
From: alex at somewhere.round.here (Alex)
Date: 27 Mar 2000 12:16:17 -0500
Subject: Strict Type Checking Objects in Python
References: 
Message-ID: 

> type_string = type("a") # what is the official "name" of the string
>                         # type anyways!?

from types import StringType

assert type ('') == StringType

Alex.

From effbot at telia.com  Wed Mar  1 17:09:05 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 1 Mar 2000 23:09:05 +0100
Subject: will python 3000 break my code?
References:  <040701bf83c5$5b296a20$34aab5d4@hagrid> 
Message-ID: <042801bf83ca$c3a0f000$34aab5d4@hagrid>

Daniel wrote:
>    |     3. be impossible to identify by automatic tools.
>    |     Tim's excellent checkappend.py can spot many
>    |     cases, but not all.
> 
> Actually, this isn't true (that it's impossible).
> If you really like, i'll take a few days of my spring break and write one
> guaranteed to catch every single case. 

how about this one:

    foo = list.append
    for bar in range(100):
        foo(bar, bar, bar)

or this one:

    class Foo:
        def __init__(self):
            self.setup()
            self.add = self.list.append
        def setup(self):
            self.list = []

    foo = Foo()
    foo.add(bar, bar, bar)

From avflinsch at att.net  Tue Mar 21 12:33:19 2000
From: avflinsch at att.net (aflinsch)
Date: Tue, 21 Mar 2000 11:33:19 -0600
Subject: how to 'scanf' in python
References: <006201bf9326$e9cb0520$2801000a@iel.ie>
Message-ID: <38D7B25F.C0579BA5@att.net>

Shaun Hogan wrote:
> 
> ok lads, im trying to figure our how to 'scanf' in python, but the code
> below isnt working properly, can anyone tell me what im doing wrong??
> 
> def rl():
>     line="how you doin"
>     import sys
>     line=sys.stdin.readline()
>     x, y = string.split(line[:-1])
>     x, y = int(x), int(y)
>     print x,y
> 

try 

def rl():
    line="how you doin"
    import sys, string
    line=sys.stdin.readline()
    x, y = string.split(line[:-1])
    x, y = int(x), int(y)
    print x,y

From effbot at telia.com  Wed Mar 29 19:21:49 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 30 Mar 2000 00:21:49 GMT
Subject: Python equivalent of "lynx -dump"?
References: <20000330001154.28633.qmail@web2106.mail.yahoo.com>
Message-ID: 

lewst  wrote:
> > An all Python solution is a little bit more complicated:
> >
> > import htmllib, formatter
> >
> > p =
htmllib.HTMLParser(formatter.AbstractFormatter(formatter.DumbWriter()))
> > f = open('test.html')
> > p.feed(f.read())
> > p.close()
> > f.close()
>
> Yes, but how can I store the output of "p.feed(f.read())" in a
> variable such as `data' like I'm doing above with lynxcmd.  Your
> code writes everything out to the terminal.

did you read the fine manual?

    http://www.python.org/doc/current/lib/writer-impls.html

    DumbWriter ([file[, maxcol = 72]])

    Simple writer class which writes output on the file object
    passed in as file or, if file is omitted, on standard output.

in your case, using a StringIO file object is probably the best
solution:

    import StringIO

    file = StringIO.StringIO()

    # build formatting pipeline
    w = formatter.DumbWriter(file)
    f = formatter.AbstractFormatter(w)
    p = htmllib.HTMLParser(f)

    ...

    data = file.getvalue()

From bjorn at roguewave.com  Sun Mar  5 17:32:55 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Sun, 05 Mar 2000 15:32:55 -0700
Subject: req: OO SQL gen module?
References: 
Message-ID: <38C2E097.BA3C38E8@roguewave.com>

A while ago I was attempting to swig a c++ library that did exactly this, but
ran into problems when I couldn't overload the individual comparison operators
for a python class, e.g. consider (assuming Table.__getitem__() returns a Column
object):

    t = Table('foo')
    s = Select( t['col1'], t['col2'] )
    s.where( t['col3'] > 5 )  # oops, not possible

back when I was doing this, people were saying comparison operator overloads
would be implemented in 1.6, but I don't know if that has happened...

-- bjorn

Greg Wilson wrote:

> [apologies if this is a repeated post --- ISP problems]
>
> I am looking for a module that does for SQL what Perl's CGI.pm, Python's
> HTMLgen, and DOM libraries in various languages do for HTML and XML.  Right
> now, both Perl and Python require me to manipulate SQL at the string level.
> Tools like CGI.pm, HTMLgen, and DOM, on the other hand, let me manipulate
> trees at the semantic level, and then translate the tree structure into text
> when necessary. This seems like the right way to build, store, and
> manipulate SQL queries.
>
> If anyone has done something like this (in any language), I'd be grateful
> for pointers or references. Please mail me directly (gvwilson at nevex dot
> com), as I am not a regular reader of all of these groups.
>
> Thanks,
> Greg
>
> --
> http://www.python.org/mailman/listinfo/python-list

From neelk at brick.cswv.com  Thu Mar  2 20:25:12 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 3 Mar 2000 01:25:12 GMT
Subject: XML DTD for Python source?
References:    
Message-ID: 

Greg Wilson  wrote:
> > "Neel Krishnaswami"  wrote
> > Can I ask what the purpose of this utility would be?

I've rearranged your response a bit, so that I can answer the
interesting part first. :)

> Most of the data I deal with these days is either in XML, or headed
> that way. [...] I'm interested in exploring what would happen if I
> could do with programs what I do with hypertext:
>
> - apply a DTD to switch between Scheme-style parenthesizing, Python-style
>   indentation, or C-style bracing
> 
> - embed arbitrary information (images, optimization hints for the
>   compiler, etc.) in a way that third-party browsers and processors
>   can handle (specially-formatted comments are *not* the answer)
>
> One of the big motivations for my interest is that I don't expect
> kids my niece's age to put up with glass typewriters as a
> programming environment. They're already building web pages with
> images, their choices of color, etc.  Existing IDEs mostly just put
> lipstick on this particular toad...
>
> ...and then define a standard for specially-formatted comments that
> contain URLs and other enriched information, and turn the little
> parser into a browser plug-in so that it can translate .py files on
> the fly, and... That might be the only way forward, but I am
> interested in exploring what happens if we finally do to our
> programs what we're done to everyone else's documents :-).

I have looked into building a rich structuring editor, and let me tell
you, this is a MUCH bigger can of worms than you realize. It's bigger
than I realize, too, even though I think building a high-quality
programming environment could easily generate a dozen PhD theses along
the way. XML might be a useful interchange format, but IMO it doesn't
really help solve the core difficulties in this enterprise.

The two big problems surrounding building a good DE are:

o I have a program, and I want multiple views of and means of changing
  it.

o My program changes over time, with changes made by myself and
  others. I want to be able to understand and manage these changes. 

Simple examples of the first sort of tool are the "tags" program that
Unix programmers use to find function definitions, and Javadoc. A
moderately sophisticated example might be cweb, Knuth's literate
programming tool, or a form-painting GUI builder. Some extremely
sophisticated examples might be the Smalltalk refactoring code
browser, and PLT Scheme's static dataflow analyzer MrSpidey.

A simple example of the second sort of tool would be the "undo"
functionality in editors. A moderately complex version of is CVS. An
extremely sophisticated example of the second is using a database to
get fine-grained version control of the objects in a Smalltalk image.

Some things you should play with/look at are:

o Allegro Common Lisp 
o Visual Age Smalltalk

  These two are worth studying, because they show the level of 
  reflectivity and self-understanding a system needs in order to
  adequately answer the programmer's questions about the system,
  and to learn where the state of the art already is.

o Berkeley's Ensemble project

  This is an attempt to build an advanced multilingual development
  environment, with support for things like flagging the programmer
  if the code s/he's writing has incorrect syntax or doesn't type
  check, showing diagrams of hyperlinked class hierarchies, and so
  on, plus using an ODB to get persistent versioning and ACIDity
  for your code. There are a lot of good papers and theses available
  on the website -- I recommend it highly.

Note: I'm not saying this isn't worth doing -- quite the contrary! I'm
just saying that a) doing the Right Thing is one hell of a lot of
work, and b) your analyzer has to understand both the program and the
programming language to say very much useful about it.

> > what would an XML representation for Scheme programs look like?
>
> I don't know, but I don't think it would look like the token-level
> representation you posted.  I think it would look more like an XML
> encoding of an abstract syntax tree (AST).  

Hah! You fell for my trick question. :)

Seriously, you *can't* store arbitrary Scheme code in semantically
rich XML, because predicting whether you can finish reading a Scheme
program is equivalent to the Halting Problem. Scheme's hygienic macro
system is Turing complete, and you can rebind the symbols bound to the
special forms. It's like a document that can add and delete element
declarations to and from its DTD.

The same is somewhat true of Python, though less forcefully. It's
probably not possible to change they syntactic structure of Python,
but you can't really very infer much semantics from that, short of
actually running the program. For example, consider the __import__
hook. Doing an import can conceivably change nearly anything in the
Python environment.

It's a lot like type-safety: in order to build a language that can be
type-checked, you have to accept that there are good programs that
that your typesystem can't assign a type to.

[I think this analogy can actually be pushed further than I thought,
because if you think of a DTD as defining a datatype, then wanting to
store some semantics of the program in the XML means that you are
trying to guarantee that the program has the "type" defined by the
DTD. You will either have to accept that the XML is sometimes wrong,
or sometimes have to reject vaild programs. Neat!]

Neel

From effbot at telia.com  Mon Mar 13 18:28:39 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Mon, 13 Mar 2000 23:28:39 GMT
Subject: Forth like interpreter
References: <158801bf8bdc$08516040$6401a8c0@dorb>     <38CC03A9.6461C7D6@tismer.com> 
Message-ID: 

Samuel A. Falvo II wrote:
> >It may count for Forth, but for Python this is very fast.
>
> For any language that uses threaded interpretation in this manner, this
type
> of overhead *ALWAYS* counts.

umm.  do you seriously think that the best way to
speed things up is to go find something that takes
less than one percent of the execution time, and
work very hard on making that little part run a few
percent faster?

> What I'd like to know, though, is what kind of performance boost would one
> (theoretically) get if Python switched from its current switch-threading
to
> token, direct threading?

very close to zero, on most modern platforms.

From kragen at dnaco.net  Mon Mar  6 22:23:33 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 07 Mar 2000 03:23:33 GMT
Subject: Python for non-programmers
References:  <3prrbskd8a3tmhmr8a4bnrcf2ft8k52k29@4ax.com>  
Message-ID: 

In article ,
Neel Krishnaswami  wrote:
>Kragen Sitaker  wrote:
>> The Scheme-as-CS-101 crowd seems to think that it's a good idea to
>> put off the whole mutability thing until later --- just have names
>> and values, and don't worry about mutation until later.
>
>My own experience matches that -- when I was introduced to functions,
>they were explained to me as basically substitutions. That is, if I
>had
>
>  (define (square x) (* x x))

I am guessing from this example that you were taught from SICP, at
MIT.  (Your mention of "problem sets" and your email address also
suggest this.)

Obviously 6.001 is at least somewhat successful at teaching MIT
students basic CS --- and, at that, some really heavy stuff considering
that it's just a first-semester CS class.

The question is, I guess, is it equally successful at teaching
programming to people who are too dumb or uneducated to get into MIT?
Indications seem to be "yes", but it's hard for me to say.

>I *hated* mutation, simply because I needed to start thinking about
>the deeply bizarre (to the novice me) and extraneous notion of
>"environments". I didn't see what they had to do with my program *at
>all*, and as far I was concerned they were an ad-hoc invention created
>to make me fail my problem sets.

At least the ad-hoc part is true.  I think they are an ad-hoc invention
created to make computers economically feasible with the technology of
the 1940s.

>I think it's possible for a total novice to handle a language with
>either mutation (eg BASIC), or functions (eg Scheme), but not both at
>once. That will confuse them unless they are sharper than I was when
>starting out -- I honestly have no idea how people were able to learn
>how to program with Pascal. ;)

Very slowly, and they generally remain confused for a long time.

Python is probably not a great language for teaching functional
programming.  Writing numerical functions gets boring in a hurry, and
Python's aggregate data structures are all messy and inefficient to
deal with without mutation.

Python is probably a good language for teaching procedural programming.

I'll post a flaming rant about why CP4E is uber-important some other
time.
-- 
       Kragen Sitaker     
The Internet stock bubble didn't burst on 1999-11-08.  Hurrah!

The power didn't go out on 2000-01-01 either.  :)

From effbot at telia.com  Fri Mar  3 14:39:51 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 03 Mar 2000 19:39:51 GMT
Subject: MIMETOOLS QUESTION
References: 
Message-ID: 

Gregoire Welraeds  wrote:
> According to the Library reference doc, on page 244 (section 12.7) the
> package tool define a Message(fp[,seekable]). Wel, very cool... but what
> are fp and seekable after all ?

the copy I have points out that mimetools.Message is a sub-
class of rfc822.Message.  the arguments are well explained
in that chapter...

> All I want is to parse a MIME header from a string read on a socket...

the Message takes any file object which properly implements
"readline".  you can use StringIO to access your string as if
it were a file -- or in this case, you can use "makefile" on the
socket to wrap the socket itself in a file object.

> I found that, except for the tutorial, there is a lack of example in the
> python documentation. Python is not a strong typed language so it's pretty
> difficult to find the meaning of some arguments used in the API.
>
> Maybe Guido could take some example of the Faqts site to put them in the
> doc... Guido ?

better make that "Fred"

fwiw, there's plenty of cookbooks out there.  here are a few:

+ Python Annotated Archives (by Martin Brown):

    http://www.python.org/psa/bookstore

  50+ sample scripts from various sources; extensive
  annotations by the author.  massive piece of work.

+ Programming with Python (by Tim Altom):

    http://www.python.org/psa/bookstore

  have only browsed it, but it appears to be a budget
  version of the annotated archives.  early reviewers
  seem to think it's no good, so you may wish to check
  it out in a bookstore before buying it.

+ The Python Grimoire (by Andrew Kuchling):

    http://starship.python.net/crew/amk/grimoire/

  work in progress; sample scripts from various (mostly
  unattributed?) sources; sorted by task.  (like martin's
  book, this seems to contain some eff-bot code, so it
  cannot be all bad ;-)

+ (the eff-bot guide to) The Standard Python Library,
  eMatter edition (by Fredrik Lundh):

    http://www.pythonware.com/people/fredrik/librarybook.htm

  320 sample scripts, sorted by standard library module.
  more code and less annotations than the others (after
  all, python code is supposed to be easy to read ;-).

From dworkin at ccs.neu.edu  Sun Mar  5 12:15:28 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 05 Mar 2000 12:15:28 -0500
Subject: my first real program blows up, can someone tell me why?
In-Reply-To: "Jerry F. Davis"'s message of "Sun, 05 Mar 2000 10:26:05 -0600"
References: <38c28b4f$0$40286@news.execpc.com>
Message-ID: 

"Jerry F. Davis"  writes:

> I read the tutorial and tried a variation on what i saw there. and it
> blew up, with a integer overflow. can someone tell me why?

You aren't converting the argument to a string.

The line that reads:

x = sys.argv[1]

should read:

x = int(sys.argv[1])

With that change, your program should only overflow if you try to pass 
an argument that is... hm, one larger than the last fibonacci number
preceding your system's maximum integer.

If you wish to make your program able to handle arbitrarily large
arguments, you should modify it to use Python's long data type.

-Justin

From Peter.Desaulniers at pahv.xerox.com  Wed Mar  8 16:59:03 2000
From: Peter.Desaulniers at pahv.xerox.com (Desaulniers, Peter)
Date: Wed, 8 Mar 2000 13:59:03 -0800 
Subject: HELP: Python.exe Application Error using Extensions on Win2000
Message-ID: <51B8ABCE456FD111899900805F6FD6EE071E90C2@mercury.ADOC.xerox.com>

Hello Fellow Pythoners,

I hope this is the appropriate list for this type of request.

The architecture is a python method that make a call into a C extension
which in turn loads a 3rd party DLL (via LoadLibrary and later
FreeLibrary).  A Py_Object is created in the extension which includes a
member object obtained from a call into the DLL. The Py_Object is returned
to the python method.  The python method then makes calls to the methods on
the 3rd party object via the Py_Object.  The code works fine. Everything
happens as is should then the error occurs at the very end when its cleaning
up. 

I am getting a Python.exe Application error indicating that I am trying to
"write" to an unacceptable memory location on a Win2000 system.  The exact
same code works fine on WinNT 4.0.   I have determined that it is occurring
during the Python cleanup as the Python process is shutting down. This is
confirmed by doing an os._exit(0) as the last line in the top level module
being executed (to prevent the clean up, the problem still occurs with a
sys.exit(0) in the same location).  So, something changed in Win2000 that is
instigating this problem.

If I skip the use of the extension altogether and let the rest of the python
method to work without the feature obtained from the 3rd party DLL then I
get no problem indicating that it is the use of the C extension that is
causing the error.

The extension has a dealloc method  for the Py_Object which gets called by
python.  The dealloc method frees allocated memory and DECREFs objects. The
dealloc is called and does complete before the error occurs.  The problem
occurs whether dealloc is allowed to do its stuff or not (e.g. returning at
the beginning of the dealloc method).  

Has anyone seen a similiar problem with the use of C extensions and / or 3rd
party DLLs on Win2000? And if so, how did you resolve the problem?

Or, in a more general sense, has anyone discovered some less than obvious
things (e.g. freeing of memory, DECREFs, etc) that need to be done for
cleaning up C extensions, particularly those that in turn use a 3rd party
DLL.  I am hunting for any clues here.

Has anyone discovered any errors arising with python applications that works
on NT 4.0 but failed on Win2000?

Any suggestions are greatly appreciated.

Thanks,

-- Peter Desaulniers

From sholden at bellatlantic.net  Mon Mar 27 13:11:10 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Mon, 27 Mar 2000 18:11:10 GMT
Subject: regular expressions questions
References: <11A17AA2B9EAD111BCEA00A0C9B4179305321B0A@molach.origin.ea.com> <38D6DDAA.F69F3BEC@ooc.com.au>  <3d3dpiejnj.fsf@amarok.cnri.reston.va.us> <38DEA7A9.CB1C324F@ooc.com.au> <3d66u8y0rw.fsf@amarok.cnri.reston.va.us>
Message-ID: <38DFA456.5F9A8447@bellatlantic.net>

"Andrew M. Kuchling" wrote:
> 
> Derek Thomson  writes:
> > Really? I couldn't find anything that was even half as good as Perl's
> > Parse::RecDescent, after doing some searching from python.org. Links?
> 
[snip]
>								And
> didn't SNOBOL provide string processing features that were powerful
> but still usable by non-programmers?  (But my memory's fuzzy.)
> 

Well, in theory yes, but in the days of SNOBOL a non-programmer was
someone who was prepared to LEARN programming to get a job done.
That doesn't really apply nowadays, CP4E notwithstanding.

regards
 Steve
--
"If computing ever stops being fun, I'll stop doing it"

From donn at u.washington.edu  Tue Mar  7 12:00:21 2000
From: donn at u.washington.edu (Donn Cave)
Date: 7 Mar 2000 17:00:21 GMT
Subject: sigchld vs. os.popen, os.command, etc.
References: 
Message-ID: <8a3cj5$15bg$1@nntp6.u.washington.edu>

Quoth quinn at mono.ugcs.caltech.edu (Quinn Dunkan):
| Ok, I know this is more of a general unix question, but when I set a handler
| for sigchld, it causes any library function which makes a child and wants to
| wiat() on it throw an IOError because my handler has already wait()ed.  Is
| there any better solution than simply wrapping all such calls in try: ...
| except IOError: pass ?

It's up to you, but in my opinion, where possible it's better to avoid
the problem and not set a handler for SIGCHLD.  The signal handling may
interrupt other things, besides wait();  even if it's not doing that right
now, it may start when you move to another platform (or later release of
the same OS or library), or when your design changes with respect to I/O
or threading.

It's too bad select() can't wait for processes along with I/O, that
would be a nice solution in some cases.  If I remember it right, VMS
has this covered, maybe what we need is GNU VMS.

	Donn Cave, donn at u.washington.edu

From effbot at telia.com  Wed Mar  8 12:41:35 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 08 Mar 2000 17:41:35 GMT
Subject: Python 1.6 and Python 3000's new features?
References: <01BF88F8.148DD690.shichang@icubed.com>
Message-ID: 

Shichang Zhao wrote:
> >spring cleaning.  get rid of most warts, add typing, clean
> >up the (rather messy, by now) standard library.  etc.
>
> typing? does that mean that I have to :
>
>         int x=0
>
> instead of x=0?

not "have to", but "may, if you like, and want
the best possible performance"

or in other worse, change "add typing" to "add
optional static typing".  for more info, see the
type-sig archives.

From milgram at cgpp.com  Mon Mar 20 23:16:22 2000
From: milgram at cgpp.com (milgram at cgpp.com)
Date: 20 Mar 2000 23:16:22 -0500
Subject: installing cephes (or other modules)
Message-ID: 

Hi,

have gotten myself stuck installing cephes (special functions module)
on my system (SW 4.0/kernel 2.2.13(?)/libc5). Compiles a .so OK but
apparently there's something required at installation to allow the
interpreter to (a) find it (b) initialize it. I have never quite
understood how this works.

Can someone point me to general documentation on installing modules? 
This will come up again so I might as well learn to do it right now.

Or: has someone who has installed cephes provide some specific tips?

thanks,

Judah

-- 
Judah Milgram        milgram at cgpp.com
P.O. Box 8376, Langley Park, MD 20787
(301) 422-4626 (-3047 fax)

From junkster at nospam.rochester.rr.com  Thu Mar 30 19:25:46 2000
From: junkster at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Fri, 31 Mar 2000 00:25:46 GMT
Subject: Python w/MS Outlook / MS Exchange
Message-ID: 

Folks,

	I was wondering if anyone has found a way to look up Email address's,
etc, in python from a MS Outlook "global" Address book?

	Basically, at work we are using MS Outlook, and I want to write a 
CGI-BIN
that will take a user name, and list their email information.  
Unfornately, it's a global
address book out on the outlook server....

	Any suggestions on the next step, etc, would be helpful & 
Appreciated.

			- Benjamin

====================================================
          (Remove "NoSpam" to Email me)
====================================================
Please feel free to copy any and or all of this sig.
A little something for spam bots:

root at localhost postmaster at localhost admin at localhost
abuse at localhost postmaster at 127.0.0.1

Chairman William Kennard: bkennard at fcc.gov 
Commissioner Susan Ness: sness at fcc.gov
Commissioner Harold Furchtgott-Roth: hfurchtg at fcc.gov
Commissioner Michael Powell: mpowell at fcc.gov
Commissioner Gloria Tristani: gtristan at fcc.gov
consumerline at ftc.gov
fccinfo at fcc.gov
ssegal at fcc.gov

From milesthompson at sprint.ca  Wed Mar  8 08:29:31 2000
From: milesthompson at sprint.ca (Miles Thompson)
Date: Wed, 08 Mar 2000 09:29:31 -0400
Subject: Python/MySQL Where do "SELECT ..." results go?
References: <38C4F789.71ECCB04@sprint.ca> <8a3jia$pkv$1@news1.xs4all.nl>
Message-ID: <38C655BB.6AD8EE77@sprint.ca>

Where did I get it?  I was either looking through the modules, or picked it
"experitentially" .
Miles

Boudewijn Rempt wrote:

> Miles Thompson  wrote:
>
> >>>> import Mysqldb
>
> Where did you get this? I know of MySQLdb and MySQL (and mySQL)
> as Python modules, but I haven't heard of Mysqldb before. If you
> were using MySQLdb, you would be using the DB-API II api, which
> ought to be portable to other databases (but are there any other
> database modules that conform to this api?)
>
> >>> import MySQLdb
> >>> conn=MySQLdb.connect(host='localhost',user='boud',db='lng',passwd='')
> >>> curs=conn.cursor()
> >>> curs.execute('select * from lng_parameter');
> >>> print curs.fetchall()
> [('testvalue', 'TEST')]
> >>>
>
> So, judging from your output, I'd try a curs.fetchall(), too.
>
> --
>
> Boudewijn Rempt  | http://www.valdyas.org

From gmcm at hypernet.com  Thu Mar 16 10:37:38 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 16 Mar 2000 10:37:38 -0500
Subject: Starting Python from autorun.inf on WinXX?
In-Reply-To: 
Message-ID: <1258901430-9582160@hypernet.com>

Peter Funk wrote:
> Please excuse my ignorance, but I'm a Linux/Unix guy don't know much about
> WinXX.  I've tried to put the following into the autorun.inf
> of a CD-Rom:
> 
> [autorun]
> OPEN=test\dosbin\pythonw.exe test\hello.pyw
> ICON=archiv.ico,0
> 
> That didn't work, because win98 refuses to find python15.dll. 

I know squat about autorun.inf stuff, but if python15.dll is in 
the same directory as pythonw.exe, Windows should find it.

> (Of course, I should have  known this :-( )
> Than I tried to put a .bat file into autorun.inf, that does a
> proper setup of PYTHONPATH, TCL_LIBRARY and so on.  But this also
> didn't work, because WinXX seems to refuse to start a batch file 
> from autorrun.inf?  Is that true?  Is there a workaround so that
> I still can autostart a Python/Tk GUI from CD without having the
> user to install Python on his WinXX box first?

I know people have used my installer for this purpose (having 
made changes to accomodate them).

http://starship.python.net/crew/gmcm/distribute.html

- Gordon

From sparhawk at aon.at  Sun Mar 19 17:07:54 2000
From: sparhawk at aon.at (Gerhard Gruber)
Date: Sun, 19 Mar 2000 23:07:54 +0100
Subject: Conversion to Hex or other bases
Message-ID: 

How can I convert a decimal to another base? i.e. Hex or binary?

-- 
Bye,
   Gerhard
If you are programming dynamic HTML in python you might find 'template.py'
usefull. Look at my homepage: http://members.aon.at/sparhawk

From dirck at pacbell.net  Sun Mar 26 02:08:16 2000
From: dirck at pacbell.net (Dirck Blaskey)
Date: Sat, 25 Mar 2000 23:08:16 -0800
Subject: colostomy
References:    
Message-ID: 

Jason Stokes  wrote in message
news:FliD4.56740$3b6.225857 at ozemail.com.au...
> >> * Dirck Blaskey menulis:
> >>>...
> >>> Please, PLEASE make colon's optional when not needed.
> >>> ...
> ...
> The convention is easy enough to remember: as far as I've been able to
make
> out, any statement that governs other statements (that is to say, indented
> underneath) ends with a colon.
> ...

I guess I can only explain it like this:
My head knows the syntax;  my fingers don't.
My fingers don't seem to learn very fast, either.

From sholden at bellatlantic.net  Mon Mar  6 10:53:32 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Mon, 06 Mar 2000 15:53:32 GMT
Subject: When to use classes?
References: <38C1CFC7.DD9F0918@pitnet.net> 
Message-ID: <38C3D47D.A3FE1C19@bellatlantic.net>

"Daniel T." wrote:
> 
> [Good advice about structuring]
> 
> Just because Python doesn't have any private parts ...
> 
> --
> When a thing ceases to be a subject of controversy,
> it ceases to be a subject of interest.
>                                                    -- William Hazlitt

I hope you're not trying to suggest that Python has
been emasculated (or should that be effeminated?)

regards
 Steve
--
"If computing ever stops being fun, I'll stop doing it"

From kens at sightreader.com  Fri Mar 10 05:40:02 2000
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 10 Mar 2000 02:40:02 -0800
Subject: HELP: restore my faith in Python
References: <38C00D46.8CC5BA67@americasm10.nt.com> <38C373C4.3EBB62D8@sightreader.com> <38C7DAFB.D5FE279D@stud.upb.de>
Message-ID: <38C8D101.CDA3E07B@sightreader.com>

Okay.  You are right.  Better yet let's make sure floats Never work
correctly.  The more they work right, the more people will think they
work right.  We should have a special case for 1/2.  Obviously, 1/2
should be 7.392.

Also, there should be a limit on the size of lists of 100 items (beyond
which
the program should dump core).  If people can have over 100 items in
a list, they might think they can have several trillion, which would
overflow
memory.

Mirko Liss wrote:

> Ken Seehof wrote:
> >
> > Seems to me an improvement would be to have python artificially
> > add epsilon to floats before converting to int.  This would still
> > suck but IMHO it would suck less.
>
> It's not just the conversion to int. As soon as you tell the
> programmer not to expect the worst from floats, he'll invent
> other ways to make a mess with those ugly floating point numbers.
>
> Want an example?
>
> >>> pennies1, pennies2 = 3.0 / 20 , 3 * 0.05
> >>> menu = { pennies1 : "spam" }
> >>> menu[pennies2] =  "lobster"
> >>> menu
> {0.15: 'lobster', 0.15: 'spam'}
>
> Ooops !
> So the division 3.0/20 is just a bit inaccurate, isn't it?
> But who cares, as long as I get at the sea food.
> So I count my pennies and ...
>
> >>> menu[0.15]
> 'spam'
>
> Spicy regards,
>
> Mirko
>
> --
> M Li?, n89553 at squid.upb.de

From tom-main at REMOVEME.home.com  Sat Mar 11 16:03:13 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Sat, 11 Mar 2000 21:03:13 GMT
Subject: What is the Python mainstream for the GUI Lib ..??
References: <8ad724$lpb$1@bw107zhb.bluewin.ch> 
Message-ID: 

> Since I'm also programming in C++ I usually use wxWindows. You can have a
look
> at http://www.wxwindows.org to see for yourself. The interface for the C++
> library is almost exactly the same as for python and there is a big number
of
> supporting classes and it's easy to use.

That sounds like an excellent solution to me.  Thanks for pointing it out.

Tom.

From pf at artcom-gmbh.de  Thu Mar  2 09:15:48 2000
From: pf at artcom-gmbh.de (Peter Funk)
Date: Thu, 2 Mar 2000 15:15:48 +0100 (MET)
Subject: Most common #! format
In-Reply-To: <38BD9017.4A2B5941@mindspring.com> from Chuck Esterbrook at "Mar 1, 2000  4:48: 7 pm"
Message-ID: 

Hi!

Chuck Esterbrook:
> What's the most common #! format?
> 
> #!python
> #!/usr/local/bin/python
> #!/usr/bin/env python

Since MacOS and Windows don't care about this very first line,
this is only interesting to increase portability on unixoid platforms.  

On all flavours of Unix I've ever seen in the last 20 years the utility
'env' used to live in /usr/bin.  This habit seems to be even older than
the invention of the '#!' exec hack (But this is from memory).

So using '#!/usr/bin/env python' as first line is best available solution 
for portable Python programs today.  But:

Today unix like systems can be divided in two major groups:
  - freely available (FreeBSD, NetBSD, Linux, where the several flavours
    of Linux are the most prominent ones)
  - commercial Unices (AIX, HP-Ux, Sun Solaris, SGI IRIX, Unixware)
Today the freely available Unices have about 90% market share (This is 
hard to estimate, since you may copy them over and over without having 
to buy new licenses) and the commercial Unices are slowly dying out 
one after the other.  Most Linux distributions come with a bunch of
useful open source software and so include a ready to run Python system,
where the Python interpreter can usually be found as /usr/bin/python.

However those commercial Unices have some niches, where professionals
may still earn some Money. ;-)  Unfortunately these systems usually
come with the original Unix utilities, which many professionals
(including me) consider simply as broken.  So if you want to make
e.g. a Solaris system usable for many practical tasks you normally end up
first installing gcc, bash, GNU-find and so on and not to forget Python
under /usr/local/bin.  After that you usally extend the default search
path $PATH to also include /usr/local/bin.

However there may be people out there, who are permitted to install
Python under /usr/local/bin but are *NOT* permitted to change the
default search path to include /usr/local/python (e.g. for some
mysterious security policy).  Under this rare circumstances the second
form '#!/usr/local/bin/python' comes into play.  This would allow
to call for example a CGI-script from a web server process, which 
didn't has Python on its search path.  But this situation should 
be considered as very exotic and rare today.

Regards, Peter
-- 
Peter Funk, Oldenburger Str.86, 27777 Ganderkesee, Tel: 04222 9502 70, Fax: -60
The way to do research is to attack the facts at the point of greatest
astonishment.                                                    -- Celia Green

From alex at somewhere.round.here  Mon Mar  6 20:05:02 2000
From: alex at somewhere.round.here (Alex)
Date: 06 Mar 2000 20:05:02 -0500
Subject: Self Nanny
References:  <8a1jlh$l4k$1@nnrp1.deja.com>
Message-ID: 

> Anyway, white space, indentation, self, this, etc. are cosmetics.  A
> real programmer reads and understands a ten thousand lines Fortran
> program without embedded comments or external documents.  If you are
> clueless, a readable program won't make any difference.

Yes.  Personally, I diagnose bugs using hex dumps of core files.  All
the other cosmetics just clutter my view of what's really going
on.  

Alex.

From oscarbp at maptel.es  Wed Mar 22 16:02:02 2000
From: oscarbp at maptel.es (Oscar Bosca)
Date: Wed, 22 Mar 2000 22:02:02 +0100
Subject: Tkinter Documentation
Message-ID: <38D934CA.6AE0292F@maptel.es>

Hi all !

    My question is very simple.
    Where could  I find the most complete documentation about Tkinter? I
have the Fredik Lundh's book but it looks incomplete.

    Thanks in advance.
    Oscar.

From see_plus_plus at my-deja.com  Sat Mar  4 17:41:21 2000
From: see_plus_plus at my-deja.com (see_plus_plus at my-deja.com)
Date: Sat, 04 Mar 2000 22:41:21 GMT
Subject: C++ / Java / Python (was Status of C++)
References: <38bf4389.272576531@news.mindspring.com>  <38C16B0F.2A50@letterror.com>
Message-ID: <89s3eh$thm$1@nnrp1.deja.com>

"Much of the relative simplicity of Java is - like for most new
languages - partly an illusion and partly a function of its
incompleteness. As time passes, Java will grow significantly in size and
complexity. It will double or triple in size and grow
implementation-dependent extensions or libraries. That is the way every
commercially successful language has developed. Just look at any
language you consider successful on a large scale. I know of no
exceptions, and there are good reasons for this phenomenon. -

The same can be said for Perl/Tcl/Python/Rebol/Ruby/etc.

cpp

In article <38C16B0F.2A50 at letterror.com>,
  just at letterror.com wrote:
> dickjoe_nospam_ at mindspring.com:
>
> > C++ is pretty much the default choice for "serious" software, and
> > in that I incude "system" and "application" programming.
>
> "Starting a project with C/C++ is a premature optimization."
>
> (Apparently by someone named Peter Jensen)
>
> Just
>

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

From jcw at equi4.com  Tue Mar 14 12:50:17 2000
From: jcw at equi4.com (Jean-Claude Wippler)
Date: Tue, 14 Mar 2000 18:50:17 +0100
Subject: Can embedded static-linked python still import dll extensions?
References: 
Message-ID: <38CE7BD9.DB05685F@equi4.com>

Warren Postma wrote:
> 
> I recompiled the Python sources for Windows with config.h and put this
> line in:
> 
> #define MS_NO_COREDLL 1  // build a static library not a DLL!
> 
> Now it statically links fine, but it doesn't allow me to import .pyd
> files (Python extension DLLs). [...]
> 
> I want my embedded python deeply embedded but I still want to produce
> .PYD style add ons for the system once it's in the field. Am I stuck?

Eh... yes you are.  This is one of the reasons why Tcl adopted "stubs",
which is simply a vector pointing to all Tcl core routine, handed to the
extension at initialization time so it doesn't need DLL linkage *back*
into the interpreter.  Scriptics has embraced and extended stubs (ahem),
and now calls is TEA, the Tcl Extension Architecture.

No rocket science (MacOS does it, COM does it, and many others), but it
won't be doable for Python without a change: extensions don't get any
info passed into them at initialization time, so there's no way to make
this happen without altering the extension API (slightly).  That could
be done without breaking anything, by adding a second init entry point.

It would be a very useful change, it simplifies versioning, as well as
path search conflicts, and it lets you create standalone executables
which are still dynamically extensible (actually, they are today, on
Solaris/Linux, but not Windows).  It's one of the main reasons why I
keep using Tcl (the other is Tk).

-- Jean-Claude

From dickenson at serviceware.com  Fri Mar 31 14:26:41 2000
From: dickenson at serviceware.com (dickenson at serviceware.com)
Date: Fri, 31 Mar 2000 19:26:41 GMT
Subject: Python Threading and sys.exit() codes
Message-ID: <8c2u4v$cft$1@nnrp1.deja.com>

I'm working my first multithreaded Python program. This program
compares the contents of an ftp site with a local copy. (It is not a
pure mirror site, but there is a 1-1 correspondence of files.) It then
ftp's only files which are newer on the source site than the target
site. There are often many files, so the ftp is set up to use multiple
threads, each of which repeatedly locks a global semaphore, pops a task
off a stack, releases the semaphore, and executes the task (a file to
ftp) ad nauseam.
My problem is this: I want to run this program as a scheduled task
under a job scheduler (currently testing cronDsys from #ifdef). The
single-threaded version of the program posts an appropriate code to the
environment via sys.exit(). The job scheduler traps this and takes
conditional action. So far, so good.
In the multi-threaded version of the program, the exit codes are not
getting posted and the main process is not terminating cleanly, at
least from the perspective of the job scheduler -- it appears to run
until manually terminated from the scheduler control console although
the NT command window disappears from the target desktop as expected.
The code fragments below are the ftp thread class and the __main__
section.
Can anybody help me with this ??
class FTPThread(Threading.Thread):
# def __init__(self,group,target,name,args,kwargs):
# print "in Init"
def run(self):
print self._Thread__name
print self._Thread__args
try:
ftp=ftplib.FTP('ftp.microsoft.com')
ftp.login('anonymous','dickenson at serviceware.com')
ftp.cwd('misc/kb')
except:
sys.exit(3)
subdir=self._Thread__args[0]
# print self,self._Thread__args
b=stacklen(self._Thread__args[1])
print "b at start=",b
while (b > 0):
# print "thread ",self._Thread__name," attempting to get
semaphore"
a=stacksem.acquire()
# print "thread ",self._Thread__name," got semaphore"
msfilepath=getstack(self._Thread__args[1])
newtime=getstack(self._Thread__args[2])
# print "thread ",self._Thread__name," attempting to release
semaphore"
a=stacksem.release()
# print "thread ",self._Thread__name," released semaphore"
#print "msfilepath",msfilepath
local0=string.replace(msfilepath,".","",1)
local1=string.replace(local0,"\\","")
localdrive=Target_Root+"updates\\\\"+subdir+"\\\\"
local=[localdrive,local1]
local2=string.join(local)
localfilepath=string.replace(local2," ","")
file=open(localfilepath,"w")
cmdstrings=["retr ",msfilepath]
cmd=string.join(cmdstrings)
try:
ftp.retrbinary(cmd,file.write)
print b," - file", msfilepath, "retrieved successfully
by thread ",self._Thread__name
if subdir=="revised":
revisecount=revisecount+1
if subdir=="new":
newcount=newcount+1
#print "success by thread ",self._Thread__name
#filecount=filecount+1
#print "filecount=",filecount
except NameError:
pass
else:
print sys.exc_info()
print "file ", msfilepath, " was not retrieved"
#print "failure by
thread ",self._Thread__name
file.close()
os.system("erase "+localfilepath)
sys.exit(1)
file.close()
# print "newtime",newtime
os.utime(localfilepath,(newtime,newtime))
b=stacklen(self._Thread__args[1])
# print "b at exit=",b
ftp.quit
sys.exit(0)
def notify(target,newcount,revisecount):
host="SNERT"
fromstring="Python Auto-FTP"
# tostring=["dickenson at serviceware.com"]
tostring=
["dickenson at serviceware.com","hlewis at serviceware.com","gkindel at servicewa
re.com","mciaramitaro at serviceware.com"]
msg="Subject: AutoFTP results\n\nUpdated files: " + str(revisecount)
+"\nNew files: "+str(newcount)+"\nlocated in subdirs off "+target
+ "\\\\updates"
s=smtplib.SMTP(host)
s.sendmail(fromstring,tostring,msg)
if __name__=='__main__':
Threadpool=12
stacksem=Threading.Semaphore(1)
Month_Num_Dict=
{'Jan':1,'Feb':2,'Mar':3,'Apr':4,'May':5, 'Jun':6,'Jul':7,'Aug':8,'Sep':
9,'Oct':10,'Nov':11,'Dec':12}
Target_Dict={}
Source_Dict={}
Fix_These=Stack(500)
Fix_Time=Stack(500)
New_These=Stack(500)
New_Time=Stack(500)
newcount=0
revisecount=0
Target_Root="S:\\\\Microsoft\\\\"
#Target_Root="M:\\\\msftp\\\\"
print "read source"
a=DateTime.now()
print a.hour, a.minute, a.second
read_source()
print "read target"
a=DateTime.now()
print a.hour, a.minute, a.second
read_target()
print "lookup source"
a=DateTime.now()
print a.hour, a.minute, a.second
newcount,revisecount=lookup_source()
print "start ftp..."
a=DateTime.now()
print a.hour, a.minute,a.second
threads=[]
for i in range(Threadpool):
my_thread=FTPThread(group=None,target=None,name=None,args=
("revised","Fix_These","Fix_Time"),kwargs={})
#my_thread.__init__(group=None,target=None,name="Revised
1",args=("revised","Fix_These","Fix_Time"),kwargs={})
my_thread.start()
threads.append(my_thread)
# for i in range(6):
my_thread=FTPThread(group=None,target=None,name=None,args=
("new","New_These","New_Time"),kwargs={})
# my_thread.__init__(group=None,target=None,name="New 1",args=
("new","New_These","New_Time"),kwargs={})
my_thread.start()
threads.append(my_thread)
for thread in threads:
thread.join()
print "all threads finished"
try:
notify(Target_Root,newcount,revisecount)
print "notify done"
finally:
sys.exit(0)

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

From embed at geocities.com  Fri Mar  3 10:57:33 2000
From: embed at geocities.com (Warren Postma)
Date: Fri, 3 Mar 2000 10:57:33 -0500
Subject: has anyone seen comp.lang.python.announce?
References: 
Message-ID: 

your last post to this group was feb 28th. at least as far as I can see.

> and what's worse, it doesn't show up on deja.com either.

Hmm. Sounds like  a localized propagation problem.

Warren

From mwh21 at cam.ac.uk  Wed Mar 22 09:18:09 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Mar 2000 14:18:09 +0000
Subject: Pascal int()
References:     <868zzfydwr.fsf@g.local> 
Message-ID: 

jstok at bluedog.apana.org.au (Jason Stokes) writes:

> 
> I'm no Lisp expert, but I believe that "set" could be used for this.  Eg:
> 
> (defun increment (symbol)
>     (set symbol 
>        (add1 (eval (symbol)))
>     ) 
> )
> 
> (let ((a 1))
>    (increment 'a)
>    a)

You'd need a dynamically scoped lisp for that to have a chance
(ie. not Common Lisp or scheme).

Cheers,
M.

-- 
very few people approach me in real life and insist on proving they are
drooling idiots.                         -- Erik Naggum, comp.lang.lisp

From gslee111 at daisy.kwangwoon.ac.kr  Sun Mar 19 17:42:54 2000
From: gslee111 at daisy.kwangwoon.ac.kr (???)
Date: Mon, 20 Mar 2000 07:42:54 +0900
Subject: Conversion to Hex or other bases
References: 
Message-ID: 

You can get hex or octal string from a decimal number using

1. hex() function :   hex(23)
2. 'formatted string' : hex = "%x" % 23

both gives you '0x17'

Gang

"Gerhard Gruber"  wrote in message
news:gsjads40dsuo2jbgokhrbktndkdkhk85jg at 4ax.com...
> How can I convert a decimal to another base? i.e. Hex or binary?
>
> --
> Bye,
>    Gerhard
> If you are programming dynamic HTML in python you might find 'template.py'
> usefull. Look at my homepage: http://members.aon.at/sparhawk

From mcalla at home.com  Sat Mar  4 21:15:42 2000
From: mcalla at home.com (Mike Callahan)
Date: Sun, 05 Mar 2000 02:15:42 GMT
Subject: newbe question about lists
References: <0d6dde44.8e4c635f@usw-ex0102-015.remarq.com>
Message-ID: 

Fredrik S  wrote in message
news:0d6dde44.8e4c635f at usw-ex0102-015.remarq.com...
> Sorry for this simple question, but I can't go to sleep before I
> have solved it.
>
> I am running python on Beos for the first
> time.
>
> I'm going through a study course online, and ringt now I
> am trying lists.
>
> I type the following:
> >>>
> l1=["a1","a2","a3"]
> >>> l2=["a4","a5","a6"]
> >>> l3=l1
l3 now references the same object that l1 points to. Therefore, if you
change the object the l1 points to, l3 will also change. What you want to do
is this:
>>> l3 = l1[:]
Which will create a copy of l1 and assign l3 to it.
> >>>
> l3[1:1]=l2
> >>> l3
> ['a1', 'a3', 'a4', 'a2']
Shouldn't this be ['a1', 'a4', 'a5', 'a6', 'a2', 'a3']?
> >>> l1
> ['a1', 'a3',
> 'a4', 'a2']
>
> QUESTION:
> Why is value of l1 changed... The only
> thing I can think of is B U G ...
This is a common newbe mistake. Lists are not the same as simple variables.
Get "Learning Python" and read it carefully.
>
> Can anyone help me out...
>
>
>
> * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network
*
> The fastest and easiest way to search and participate in Usenet - Free!
>

From avv at quasar.ipa.nw.ru  Wed Mar 22 17:00:59 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Wed, 22 Mar 2000 14:00:59 -0800
Subject: Does Guido Have A Will? (The legal kind )
References:  <3110C5495705440B.A7540016610C3F64.6B294A66CE5E80A1@lp.airnews.net>   <7581373A7FF14B66.539C1DD2455262EC.953827F87956C027@lp.airnews.net>
Message-ID: <38D9429B.6DF98AE@quasar.ipa.nw.ru>

Hi,

Cameron Laird wrote:
> He's a celebrity, in the sense that he makes occasional
> appearances in the pages of the *Wall Street Journal*,
> *Forbes*, *New York Times*, and other popular organs
> that are puzzling over Open Source.  Moreover, he's a
> Friend of Python, having delivered the Keynote for the
> IPC7, designed Trove, ...

... and even wrote some modules, like rfc822, found in the standard
Python distribution. Grep gives five, where 'raymond' is mentioned (I
didn't check in what context, though). He also (at least) contributed to
the famous ncurses library. So he contradicts the known saying: "who
cannot do himself, teaches".

Alexander

From gchiaramonte at ibl.bm  Mon Mar  6 03:44:54 2000
From: gchiaramonte at ibl.bm (Gene Chiaramonte)
Date: Mon, 6 Mar 2000 04:44:54 -0400
Subject: Help: sizeof() and reference in python?
In-Reply-To: 
Message-ID: 

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Fredrik Lundh
> Sent: Sunday, March 05, 2000 4:51 PM
> To: python-list at python.org
> Subject: Re: Help: sizeof() and reference in python?
> 
> 
> Gene Chiaramonte  wrote:
> > I guess I should have been more specific.
> 
> unfortunately, that doesn't help in this case ;-)
> 
> > 1 - I am trying to get the size of an object in bytes to 
> monitor how much
> > memory a list or object occupies.
> 
> you can't.
> 
> (well, you can always do it yourself, using a dictionary
> to map type(object) to approximate size of object. see
> the struct definitions in Include/*.h for more info on how
> the basic types are implemented).

Good idea.

> 
> > 2 - I am trying to get the pointer to the memory address of an object.
> Much
> > like the array objects buffer_info(), where the second item in the
> returned
> > tuple is the pointer to the memory address of the first element of the
> > array.
> 
> since you cannot do anything with it anyway, why not
> just use id(object).  it's a pretty good approximation...

Thanks.

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

From tjreedy at udel.edu  Tue Mar 28 17:08:20 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 28 Mar 2000 17:08:20 -0500
Subject: Problem
References: <8bn3gj$7qh$1@the-fly.zip.com.au>
Message-ID: <8brbbv$6uh$1@news.udel.edu>

[posted and emailed]
"InCoGniTo"  wrote in message
news:8bn3gj$7qh$1 at the-fly.zip.com.au...

> I'm trying to write a program which does basic bank functions and I have
a
> problem that I can not get it to read my list to see if there is a match
in
> a name so that the prgram can identify whos account it is. Can someone
> please try and help me.

Let's start with some improvements for your code.

def login():
    #  delete multiple print statements and instead print 1 multiline
string
    print '''\
Welcome to Account System.

You now have to login the user to access the account.

'''
    # use raw_input(prompt) for data
    # note: str(sys.stdin.readline()) is redundant since readline() returns
string
    fn = raw_input( 'Please enter the customers First name')
    ln = raw_input('Please enter the customers Last name')
        for item in customer_list:
            if item.firstname == fn and item.lastname == ln:
                current_person = item
                print 'You have succesfully Logged in, now transfering to
account systems menu\n
                # note that you do not need separate print statement for
extra \n
                time.sleep(3)
                accountmenu()
                break # I added this - once match is found, stop looping
                # if current_person  == None: # broken since indented to
execute within loop
                # instead of after and since current_person not
initialized.  Replace this with below
        else:
            print 'Sorry account records are not present in system records'
            print 'Going back to level 1'
            print 'Loading, please wait ...'
            time.sleep(2)
            print '\n'
            welcome_verfication()
# this sort of construction -- break loop on match, else do non-match
action
# is precisely what else-after-for is meant for.

> There are many other fucntions and classes, for instance customer_list is
my
> list:). I don't get any errors the program just stops and qoutes me the
> module.

Does replacement of broken mismatch test fix problem?
If not, what is last statement executed?
If necessary, add print statements to determine this.

'quotes me the module': what does this mean?

Terry J. Reedy

From warlock at eskimo.com  Thu Mar  9 00:03:23 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Wed, 8 Mar 2000 21:03:23 -0800
Subject: pythonpath
References: <38C3C701.99D7F8AE@quicknet.nl>
Message-ID: 

On Mon, 06 Mar 2000 14:56:01 +0000, 
 Renier Leenders, in the persona of ,
 brought forth the following words...:

>I am using Python 1.5.2 and Redhat 6.0. I am a newbie to Python and I
>want to adjust the PYTHONPATH. Where can I find such a thing?
>

Start up an Xterm, and type

echo $PYTHONPATH

anything it prints out, is the value of PYTHONPATH

It's probably blank. To change it (or set it) Type

export PYTHONPATH=/where/ever/you_want/to/point_to

it's easy. This assumes that you are using Bash as the shell,
which is probably a safe assumption as that is the default for
RH.

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.

From mlauer at trollinger-fe.rz.uni-frankfurt.de  Mon Mar 20 02:40:42 2000
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 20 Mar 2000 08:40:42 +0100
Subject: Which CORBA binding ?
Message-ID: <38d5d5fa@nntp.server.uni-frankfurt.de>

Good Morning... well, for me at least, local time is 8:35...

I want to use ORBit because it has python bindings. So far
so good. But at the moment, there seem to be two different python
bindings for ORBit available. AFAIK there is

pyOrbit, hosted by the theopenlab.uml.edu, where the great
pygnome and pygtk bindings come from,

and there is 

ORBit-python, hosted by http://projects.sault.org/orbit-python/

What's the status of these two projects, has anyone used both and
can recommend using one of them ? Please advice, if you can.

Have a nice day!

--
--
    Regards & Gruesse from Mickey @ http://www.Vanille.de
  ---------------------------------------------------------
  How could anyone know me - when I don't even know myself ?

From mikael at isy.liu.se  Wed Mar  1 10:01:43 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Wed, 01 Mar 2000 16:01:43 +0100 (MET)
Subject: Splitting comp.lang.python
In-Reply-To: <20000301151034.A4873@nl.linux.org>
Message-ID: 

On 01-Mar-00 Gerrit Holl wrote:
 >  
 > > Perhaps the best thing would be to keep c.l.py as it is as the general
 > > place for discussion, and then adding subgroups for specific interests.
 >  
 >  Perl uses:
 >  
 >   comp.lang.perl.announce
 >   comp.lang.perl.misc
 >   comp.lang.perl.tk
 >   comp.lang.perl.modules
 >   comp.lang.perl
 >   comp.lang.perl.moderated
 >  
 >  Maybe it's a good idea too do something like that on c.l.py.

Not necessarily exactly those subgroups, but definitely something in 
that direction, yes. Based on my first post in this thread, I come up 
with the following

  c.l.py             General
  c.l.py.gui         GUI
  c.l.py.os          os/platform dependent issues
  c.l.py.db          DataBase handling
  c.l.py.advocacy    Whitespace and such

The CGI subgroup that I proposed earlier is perhaps not especially
interresting. Many questions and discussions regarding CGI programming 
tend to be originating from poor understanding of Python itself, and 
its modules. I guess that is the way in to the Python world for many
newbies. It sure was for me once.

I think that it is better to have general subgroup names such as gui, 
os, db, and advocacy, rather than more explicit ones like tk, linux, 
sql, and whitespace. Maybe platform is better than os... I leave that
to the os/platform lawyers out there.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson 
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    01-Mar-00
Time:    15:36:08

This message was sent by XF-Mail.
-----------------------------------------------------------------------

From neelk at brick.cswv.com  Thu Mar  9 20:46:53 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 10 Mar 2000 01:46:53 GMT
Subject: What If Python Replaced Elisp?
References:      
Message-ID: 

Fredrik Lundh  wrote:
> Samuel A. Falvo II wrote:
> > >umm.  does using an explicit syntax tree turn Lisp
> > >into a statically typed language?
>
> > No, but I don't see what this has to do with the point I was trying
> > to make either.
> 
> then I don't understand your point -- to compile
> things efficiently, you need to know what types
> your variables have.

There are actually a whole lot of optimizations that could be done,
but aren't, in the current Python interpreter.

For example, in the following code:

import operator

lst = []
for i in range(100):
    lst.append(i**2 + i**2)

result = reduce(operator.add, lst) 

You can statically infer the type of result, the final size of lst,
avoid actually allocating a range() object, eliminate the the
typechecks that make sure that i has a __pow__ method, the bounds
checks on the range object, do available expressions analysis on the
i**2 to avoid calculating it twice, and probably a few more
optimizations that I don't know about.

No, these analyses wouldn't help on super dynamic code, but it doesn't
need to to yield big speedups -- most Python code is in the simple
first-order form that is most amenable to analysis.

This is a lot of work, would greatly complicate the internal structure
of the Python implementation, and is a lot of work, and would somewhat
complicate interfacing with C, and is a lot of work, but it's
definitely possible. (*I* have no interest in doing this, not when it
would involve programming in C rather than Python, anyway.)

> in Python, you don't know that, no matter what tree structure or
> byte code format you use.

You don't have to win every game to make it to the championships. :)

Neel

From nobody at nowhere.nohow  Sun Mar 26 17:57:48 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 26 Mar 2000 22:57:48 GMT
Subject: Language forks (was: Python tutorial/comparison for C++ programmer)
References:     <5FF7ED1C1C902C69.0E5D50066AC4B434.CE45BE37438DE91A@lp.airnews.net>
Message-ID: 

Cameron Laird wrote:
>			.
>>If the language implemented by "MS Visual-Python" isn't 100%
>>compatible with the official, free version of Python, then life
>>gets less pleasant.  If you want to write "portable" python,
>>you had to have some sort pre-processor to sort out the
>>differences between MSVP and real Python.  We could end up with
>>library modules that are available for one language and not the
>>other.
>			.
>			.
>I entirely understand your claims, and feel familiar with much of
>the history that supports them.  We have interesting counter-ex-
>amples at hand, though; how does it happen that CPython and JPython
>happily co-exist?

Because they've decided to?  The goal of one is not to kill the other. That
can rarely be said of MS's attitude towards other entities in the SW world.
MS seems to actively attempt to diverge from standards in order to try to
kill off those standards and establish their own dominance.

-- 
Grant Edwards                   grante             Yow!  I'm reporting for
                                  at               duty as a modern person. I
                               visi.com            want to do the Latin
                                                   Hustle now!

From effbot at telia.com  Fri Mar  3 14:54:28 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 03 Mar 2000 19:54:28 GMT
Subject: HELP: restore my faith in Python
References: <38C00D46.8CC5BA67@americasm10.nt.com>
Message-ID: 

Steven Holton wrote:
> I'm trying to learn Python, and wrote this script last night.  I was not
> expecting to see this behavior, and my faith is severely shaken:
>
> nose-46> 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
> >>> limit = 5
> >>> for target in range(10):
> ...    n=((target/(limit+0.0))-(target/limit))*limit
> ...    print target, n, int(n)
> ...
> 0 0.0 0
> 1 1.0 1
> 2 2.0 2
> 3 3.0 3
> 4 4.0 4
> 5 0.0 0
> 6 1.0 0     <=== huh?
> 7 2.0 1     <=== et tu Brutus?
> 8 3.0 3
> 9 4.0 4

welcome to the wonderful world of floating point
arithmetics:

>>> limit = 5
>>> for target in range(10):
...    n=((target/(limit+0.0))-(target/limit))*limit
...    print target, "%.30g" % n, int(n)
...

0 0 0
1 1 1
2 2 2
3 3 3
4 4 4
5 0 0
6 0.99999999999999978 0
7 1.9999999999999996 1
8 3.0000000000000004 3
9 4 4

I'll leave it to our resident IEEE guru, the timbot,
to explain exactly why this is happening...

From grant at nowhere.  Thu Mar 23 12:41:17 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 23 Mar 2000 17:41:17 GMT
Subject: Python tutorial/comparison for C++ programmer
References:  <8b956c$9tn$1@nnrp1.deja.com> <077C600D9F281A4D.487194D3E94C88DA.D53DE68986E89418@lp.airnews.net>  <8bcr2a$b76$1@nnrp1.deja.com>   
Message-ID: 

In article , Michael Hudson wrote:

>I suppose it could be viewed as competition to VB, but then that's not
>really relavent until Python gets a flashy point n' click GUI, and
>even then, if micorsoft ship Visual Python (tm), how does this impact
>me?

If the language implemented by "MS Visual-Python" isn't 100%
compatible with the official, free version of Python, then life
gets less pleasant.  If you want to write "portable" python,
you had to have some sort pre-processor to sort out the
differences between MSVP and real Python.  We could end up with
library modules that are available for one language and not the
other.

-- 
Grant Edwards                   grante             Yow!  Well, O.K. I'll
                                  at               compromise with my
                               visi.com            principles because of
                                                   EXISTENTIAL DESPAIR!

From bhoel at starship.python.net  Sun Mar 12 11:31:21 2000
From: bhoel at starship.python.net (Berthold H�llmann)
Date: 12 Mar 2000 17:31:21 +0100
Subject: pygtk and GTK+ tutorial
Message-ID: 

Hello,

I try to learn about GTK+ programming using python. I read the tutorial
and port the examples to python. But I can't get "colorsel.py" working
properly. The GtkColorSelectionDialog does not set the drawingarea's
color. Any hint, or shall I send my code?

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 sholton at mindspring.com  Sun Mar 19 19:38:52 2000
From: sholton at mindspring.com (Steve Holton)
Date: Mon, 20 Mar 2000 00:38:52 +0000
Subject: Newbie help with extend()
Message-ID: <38D5731C.BCDF4BE2@mindspring.com>

Greetings-

Can anyone post a short example showing the extend method of list?

Here's what I'm seeing:

[sholton at kest sholton]$ python
Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66
19990314/Li on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> a=[1,2]
>>> type(a)

>>> b=[3,4]
>>> type(b)

>>> a.extend(b)

I'm expecting this to set a to [1, 2, 3, 4]
but instead I get:

Traceback (innermost last):
  File "", line 1, in ?
AttributeError: extend

Is this a 1.5.2 feature?

-- 
sholton at mindspring.com
If the answers are increasing and the questions decreasing,
Then you're on the road to knowledge, but you're headed the wrong
direction.

From tseaver at starbase.neosoft.com  Tue Mar  7 09:41:51 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 7 Mar 2000 08:41:51 -0600
Subject: Self Nanny
References:  
Message-ID: 

In article , Lloyd Zusman   wrote:
>Gaetan Corneau  writes:
>
>> OK, I didn't make myself clear :)
>
>Yes you have, at least for me.
>
>
>> 1) *I* don't mind having to specify "self" in method definitions
>> 
>> 2) *I* don't mind having to write "self.Something"
>> 
>> 3) BUT I think that it would be *nice* not to have to specify "self" in
>> method definitions, and it would certainly help A LOT of newbies
>> 
>> 4) if possible at all, not having to write "self.Something" would be cool,
>> but not as important as #3
>> 
>> And don't tell me #3 can't (technically speaking) be done :)
>
>Of course #3 can be done (technically speaking).  And in return for it
>getting done, would you volunteer to donate your personal time to
>change all the existing lines of Python code in the world that this
>change would break?
>
>Or barring that, would you be willing to pay for the time of all the
>others who would have to do this work?
>
>
>> If I'm mistaken, then OK, only C++ programmers forget to specify
>> "self" as first parameter for a method, I won't insist, you know. If
>> you really don't care about C++ programmers (but remember, there are
>> quite a few), fine.
>
>This is not an issue about "caring" about C++ programmers.  I am a C++
>programmer, and literally hundreds of my colleagues over the past
>decade or so have been C++ programmers.  Many of these people have
>tried Python.  And *none* of them have ever even made a complaint
>about the explicit "self" of Python.  You are the first C++ programmer
>I have encountered who has ever raised this as an issue.  That's one
>out of hundreds of people.

This is moot, given that changes to a feature which *does* have a
demonstrable, vocal minority opposed to it (syntactically-significant
indentation) is no more likely to change than Guido is to spell his last
name with a capital 'V'.

Which-syllable-of-no-isn't-clear?'ly

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com

From godzilla at netmeg.net  Sat Mar 18 21:54:57 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sun, 19 Mar 2000 02:54:57 GMT
Subject: need fast parser for comma/space delimited numbers
References:  
Message-ID: 

> ...
> start_time = time.time ()
> for i in 90000 * [None]:
> ...
> print time.time () - start_time    

(gustav)~/Engineering/dspring/stoplite/matlab/Jue-Data/: python testTime.py 
156.307130933

> Hmm, I must be missing something.  The following code takes about 8s
> on a Sparc 10 and about 5 on a Linux w/ pentium II 400MHz.

you think everyone runs on a 400 MHz machine  ;-)

P90 .... i'm due for an upgrade.... 

From lexberezhny at email.msn.com  Fri Mar 31 23:17:33 2000
From: lexberezhny at email.msn.com (lexberezhny)
Date: Fri, 31 Mar 2000 23:17:33 -0500
Subject: Python/GTK on Windows?
Message-ID: <#XSVXB5m$GA.240@cpmsnbbsa03>

Hi,
  I was wondering if anyone was able to get PyGTK to work on windows, and if
its possible at all. I am trying to port a Linux app to windows that uses
PyGTK. If you know anything about this or can point me in the right
direction, I would really appreciate it. Thanks.

 - Lex Berezhny

From jimn at minusen.force9.co.uk  Sat Mar 25 06:57:10 2000
From: jimn at minusen.force9.co.uk (Jim Nicholson)
Date: Sat, 25 Mar 2000 11:57:10 +0000
Subject: problem w/ Makefile.pre.in
References: <38DC7C13.6606D717@seas.upenn.edu>
Message-ID: 

dave white wrote:
> Hi,
> 
> I'm trying to build the MySQLdb module and I'm having to compile the
> module using Makefile.pre.in for the first time.  I keep getting the
> following error during the build:
> 
> make[1] *** No rule to make target 'usr/lib/python1.5/config/Makefile',
> needed by 'sedscript'.  Stop.
> make: *** [boot] Error 2
> 
> I'm running python 1.5.2, redhat 6.1.  

You need to install the python-devel RPM

Jim

From gchiaramonte at ibl.bm  Sun Mar  5 07:17:43 2000
From: gchiaramonte at ibl.bm (Gene Chiaramonte)
Date: Sun, 5 Mar 2000 08:17:43 -0400
Subject: Help: sizeof() and reference in python?
In-Reply-To: 
Message-ID: 

Is there a way to get the size of an object in python? I'd also like to get
a reference to an object.

example in c:

l = [1, 2, 3]	#python list

t1 = sizeof(l)	// how i'd do it in c
t2 = &l

Thanks,

Gene

From tuttledon at hotmail.com  Sun Mar  5 10:52:07 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Sun, 05 Mar 2000 15:52:07 GMT
Subject: Choice Of Language (was: Python advocacy)
References: <38BEE429.179D4865@prescod.net>  <38BF1990.CAA71DE6@prescod.net> <38BF6169.46FC6A66@netgate.net> <38C0153A.8C277AEA@prescod.net> <38C22B8D.B2D31196@netgate.net>
Message-ID: 

[Ran]
> No,  the project manager chooses the language(s) according to the needs
> of the project and the availability of programmers and other resources.

I have never programmed for a living.  (I enjoy it way too much to ever do
that! )  So let me ask all you folks that do.  What percentage of
your paid, professional time do you get to use Python?

Don

From dworkin at ccs.neu.edu  Fri Mar 17 10:29:59 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 17 Mar 2000 10:29:59 -0500
Subject: Hidden Modules
In-Reply-To: Calishar's message of "Fri, 17 Mar 2000 12:11:25 GMT"
References: <3p74ds4bse2cj7jb7jf3c8c65l8lqoobk4@4ax.com>
Message-ID: 

Calishar  writes:

>   Well, I looked through all my python directories, and there was no
> MD5.py file, but I had faith (think it might have been /f who posted
> the message) so I tried it, and surprisingly enough, it worked.

In a typical UNIX installation of Python, this module is in the file
named 'md5module.so'.

Not all Python modules are implemented in Python.

-Justin

From shogan at iel.ie  Mon Mar  6 04:25:18 2000
From: shogan at iel.ie (Shaun Hogan)
Date: Mon, 6 Mar 2000 09:25:18 -0000
Subject: (no subject)
Message-ID: <000801bf874d$e3a5f900$2801000a@iel.ie>

can anyone tell me where i can get documentation on python/c/c++ integration, either web pages/web tutorials/books anywhere at all.
thanks 
Shaun
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From python at rose164.wuh.wustl.edu  Thu Mar 23 21:00:56 2000
From: python at rose164.wuh.wustl.edu (David Fisher)
Date: Thu, 23 Mar 2000 20:00:56 -0600
Subject: Tkinter (Win98) Bombs but Tcl/Tk is OK
References: <8bcjgg$2m9$1@nnrp1.deja.com>
Message-ID: <01a101bf9534$df0a5ea0$933dfc80@fredicus>

----- Original Message ----- 
From: "Bjorke" 
Newsgroups: comp.lang.python
To: 
Sent: Thursday, March 23, 2000 2:09 AM
Subject: Tkinter (Win98) Bombs but Tcl/Tk is OK

> I've got two Win machines, with identical vrsions of Tcl/Tk and Python.
> The Win98 laptop fails to run tkinter -- i can do this:
> 
>    import _tkinter
>    import Tkinter
> 
> no problem, but when I hit
> 
>    Tkinter._test()
> 
> it fails (as do all Tkinter programs), bombing in _tkinter.create()
> saying:
> 
> TclError: Can't find a usable init.tcl in the following directories:
>     () ./lib/tcl8.0 C:/tcl8.0/library {C:/Program Files/library}
> 
> Where is it getting this? I have tcl installed in C:\Program Files\Tcl
> and it works flawlessly. I've uninstalled, re-installed, changed
> installation ordering, made sure that the autoexec paths are correct...
> ???
> 
> As I say, Tcl/Tk DOES work fine -- I use it for development on that
> laptop regularly, and the Tcl version is a new one

Does it work with version 8.0.5 that comes with the Win installer?

From lss at excosoft.se  Mon Mar 20 14:00:50 2000
From: lss at excosoft.se (lss at excosoft.se)
Date: Mon, 20 Mar 2000 19:00:50 GMT
Subject: Scope issues when embedding/extending
Message-ID: <8b5sgo$os9$1@nnrp1.deja.com>

I am trying to embed Python scripting in a C++ application in a way so
that the script interpreter has direct access to object instances and
variables in the host application. Is this possible by combining
Python's extension and embedding mechanisms, or will the embedded
interpreter not be able to access instances native to the host
application? How do I avoid creating a new object instance in the
interpreter, but rather reference an already existing one in the host
application? Does Python support this form of integration natively or do
I have to go via COM or some other protocol? Suggestions and/or sample
code would be appreciated.

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

From sreiling at elara.tripos.com  Tue Mar 21 10:27:10 2000
From: sreiling at elara.tripos.com (Stephan Reiling)
Date: Tue, 21 Mar 2000 15:27:10 GMT
Subject: Q: organizing classes in modules
Message-ID: <38D794CD.6F2824A9@elara.tripos.com>

I have a question about how to organize classes in modules:
I would like to organize my classes so that every class has its own file, the
file having the same name as the class, e.g.
In file Class1.py
    class Class1:
        ...

I would also like to put these into a module, for example module1. When I now
want to use Class1, I would like to just say:

import module1
c=module1.Class1()

instead I have to say:

import module1
c=module1.Class1.Class1()

I have quite a few of these classes, so this gets annoying after a while.
I looked into what I can do in the __init__.py file of the module. The only
thing I could figure out was to put something like the following there:

from Class1 import *
from Class2 import *
...

Which according to what I have read is not a good idea, and also seems to
increase startup time.
(And what happens if Class1 and Class2 cross-reference each other, or one of my
classes has to be a singleton)

So the question is:
How can I do this? Obviously I am trying to organize my code in a way like C++
and java do it. Right now I have around 30 classes distributed over 6 modules
and sub-modules and I expect the number of classes to grow.

Thanks,
    Stephan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: sreiling.vcf
Type: text/x-vcard
Size: 222 bytes
Desc: Card for Stephan Reiling
URL: 

From wtanksle at dolphin.openprojects.net  Tue Mar 21 13:10:05 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 21 Mar 2000 18:10:05 GMT
Subject: Forth like interpreter
References: <158801bf8bdc$08516040$6401a8c0@dorb>  <53YA4.637$jX3.10074@typhoon.nyroc.rr.com>  <425A5D00B0F916D1.7920112131B1A88A.D3428C8A56515B00@lp.airnews.net>  
Message-ID: 

On 21 Mar 2000 02:53:49 GMT, Samuel A. Falvo II wrote:
>In article , William
>Tanksley wrote:
>>Here, have some cycles of reversed kielbasa.  And ten (10 (0xa (101010b)))

>101010b is 2+8+32, or 42, not 10.

Wow, an EXTRA bonus!  Thanks, Sam.

>Samuel A. Falvo II

-- 
-William "Billy" Tanksley

From effbot at telia.com  Wed Mar  1 04:56:14 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 01 Mar 2000 09:56:14 GMT
Subject: A comp.lang.python code snippet archive?
References:  
Message-ID: <2P5v4.1511$mYj.192557056@newsa.telia.net>

Michael Hudson  wrote:
> The tokenize module could be brought to bear.

1,0-1,3: NAME 'The'
1,4-1,12: NAME 'tokenize'
1,13-1,19: NAME 'module'
1,20-1,25: NAME 'could'
1,26-1,28: NAME 'be'
1,29-1,36: NAME 'brought'
1,37-1,39: NAME 'to'
1,40-1,44: NAME 'bear'
1,44-1,45: OP '.'

From jonp at bhmi.com  Thu Mar  9 17:10:38 2000
From: jonp at bhmi.com (news)
Date: Thu, 9 Mar 2000 16:10:38 -0600
Subject: HELP !!!!!!!!!!!!!!!!!!!!!!!!!
Message-ID: <38c8215f$0$10504@wodc7nh7.news.uu.net>

I am not a Python developer and cannot find an  acceptable answer in any
book that I have looked in.  Simply want to embed a call to a python script
in C++ code.  Do I use PyRun_ ...?

From S.I.Reynolds at cs.bham.ac.uk  Tue Mar 14 05:59:12 2000
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Tue, 14 Mar 2000 10:59:12 +0000
Subject: itoa
References: 
Message-ID: <38CE1B80.59DC@cs.bham.ac.uk>

Moshe Zadka wrote:
> 
> On Mon, 13 Mar 2000, Stuart Reynolds wrote:
> 
> > Is there an integer to ascii (itoa) function in Python? I need to
> > convert integers into various different bases (but to binary, in
> > particular). Most languages come provided with an itoa-like function but
> > I can't seem to find one in the Python API.
> 
> Have a look at the builtin "str" function

Thanks, but this only does base 10.

>>> help(str)
str(object) -> string

Return a nice string representation of the object.
If the argument is a string, the return value is the same object.

Stuart

From martin.franklin at waii.com  Wed Mar 29 02:28:28 2000
From: martin.franklin at waii.com (Martin Franklin)
Date: Wed, 29 Mar 2000 08:28:28 +0100
Subject: processing unix command output lists
References: <8bs7s6$uus$1@nnrp1.deja.com>
Message-ID: <38E1B09B.F40C8F16@bedford.waii.com>

How about.......

import os
command = os.popen('ps -ef | grep whatever','r')
output_list = command.readlines()
return_code = command.close()

> >Hi,
> >In python is there a way to get the output of a unix comamnd
> >in the form of list or something.
> >example:

From robin at alldunn.com  Wed Mar 15 00:00:33 2000
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 14 Mar 2000 21:00:33 -0800
Subject: wxPython for Mac
References: 
Message-ID: 

"Matthew Barre"  wrote in message
news:B4F34378.18B3%mbarre at mac.com...
> I've seen wxPython praised by many here, and as a newbie who felt inundated
> by Tkinter I was looking forward to trying it. I've scoured the wxPython
> site, and I haven't found any predictions on when their Mac port will be
> finished. Does anyone here have any approximations?
>

The Mac version of the wxWindows C++ library has nearly caught up with the
MSW and GTK versions and is stable and robust.  What remains is to put the
Python wrappers on top of that.  Given that I don't have access to Mac
hardware that effort has been going a bit slowly, (okay, it hasn't moved at
all.)

Over the past couple days there has been some activity though.  Chris Barker
 has been drumming up some enthusiasm for doing the work and
trying to get some developers together to work on it.  You maight want to
drop him a line (or ask on the wxPython-users list,) and see if there is some
way that you can help out.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!

From adustman at comstar.net  Thu Mar  2 13:40:29 2000
From: adustman at comstar.net (Andy Dustman)
Date: Thu, 02 Mar 2000 18:40:29 GMT
Subject: MySQL access from Python
In-Reply-To: <38BDDA78.6F8EBA80@mindspring.com>
References: <38BDB625.80A4417E@mindspring.com>  <38BDDA78.6F8EBA80@mindspring.com>
Message-ID: 

On Wed, 1 Mar 2000, Chuck Esterbrook wrote:

> Andy Dustman wrote:
> 
> > 0.1.2 is the current version. I have reports that it dumps core on
> > RedHat/Alpha platforms, but I can't reproduce it on RedHat/Intel; patches
> > welcome. 0.1.1 doesn't have this problem, and it's still available in the
> > same location (no direct link on the page, however). Other people have
> > reported getting it to work on that crackpot OS from out of Redmond. It
> > also includes a patch to make Zope's ZMySQLDA to work with it instead of
> > MySQLmodule. Also check this page:
> 
> Hmm, sounds like yours isn't stable and if MySQLmodule isn't current
> and/or supported that for production purposes, this is kind of scary.

I still consider 0.1.2 to be slightly beta; 0.0.x was alpha, 0.1.x is
beta. The core dumps ONLY occur on Alpha, for unknown reasons; there are
no known problems on Intel. There are apparently a couple of Zope sites
that use MySQLdb in production. I've heard that www.codecatalog.com is one
of these, but can't confirm this.

> Looking at the features, it seemed more "friendly" than Python's DB
> API in that some things were more explicit. For example, you could ask
> for a list of columns for a given table. I'm sure you can do that type
> of thing in SQL, but SQL does vary per server and it was nice to have
> some higher-level functions rather than just basically the infamous
> "execute this SQL" function.
> 
> What do you think?

Hard to say. MySQLdb generally exports most of the MySQL-specific calls
that _mysql has for completeness. So you can do stuff like this (db is a
Connection object):

fields = db.list_fields(table) # returns list of fields as tuple

This is non-portable, of course. Then again, the SQL to do the same thing
is probably also non-portable.

MySQLdb has also been throughly tested against memory leaks when running
multiple threads. If you need threads (for Zope or otherwise), this is
the only real choice.

-- 
andy dustman       |     programmer/analyst     |      comstar.net, inc.
telephone: 770.485.6025 / 706.549.7689 | icq: 32922760 | pgp: 0xc72f3f1d
"Therefore, sweet knights, if you may doubt your strength or courage, 
come no further, for death awaits you all, with nasty, big, pointy teeth!"

From ken at straton.demon.co.uk  Sun Mar  5 10:59:37 2000
From: ken at straton.demon.co.uk (Ken Starks)
Date: Sun, 05 Mar 2000 15:59:37 GMT
Subject: When to use classes?
References: <38C1CFC7.DD9F0918@pitnet.net>
Message-ID: 

Scott Bahling  wrote:

... [ General description of a polygon class ]
>  firstxcoord = poly.points[0].x firstycoord = poly.points[0].y

> The second piece of code is easier to read. But lets say that this Point
> class has no other purpose but to contain two numerical values; is the
> use of a Point class overkill? Is there a 'rule of thumb' for when to
> create a new class?

> Second: What is the general consensus about accessing instance variables
> directly like myobj.x=10 vs. through methods like myobj.setx(10). The
> first method is quicker/easier to code, but is it "bad" coding?  Does
> anyone even care?

When I started Python I also wrote a few 'vector graphics' classes.
I used a class for points and for polygons, and I both set and read
the coordinates using methods.

This was useful for two main reasons:
  1. I wanted the user to use mm or inches or some other proper unit
     It could be floating point.
     I could make it 'snap' to the nearest grid point.

     Internally, the coordinates were fixed point integers related
     to pixel size, postscript, and printer's points; the user
     did not have to know that!

   2. The polygon class had a few derived classes. In the original
     class the shape was held as a sequence of points, but in the
     derived class of 'regular' polygons it was calculated.
     The method xxx.getpoints() returned a sequence in both cases,
     but in the second case it was 'virtual' if you like.

      The same sort of thing applied to some 'curve stitching' shapes
      in which two sequences of points are joined to create an 
      'envelope'. This was for some evolute/involute geometry
      that I never finished.

    3. Actually, I tell a lie. The original class was for a path
    which could contain gaps, could be open or closed, and which
    could include Bezier segments. It had its own local origin and 
    scale factor.

    The 'Polygon' was a derived class of this which was always closed 
    and which had straight segments between all the points.

   But the main point is that the class-based design will allow me
   use all the line-thickness, fill-colour, Etc methods on any
   new objects I come up with. 

Ken,     __O 
       _-\<,_ 
      (_)/ (_)                                                                 
    Virtuale Saluton.

From rmuschall.fih at t-online.de  Wed Mar 29 14:58:24 2000
From: rmuschall.fih at t-online.de (Ralf Muschall)
Date: Wed, 29 Mar 2000 21:58:24 +0200
Subject: Stackless Python 1.1 + Continuations 0.8
References: <38E12457.E719E536@tismer.com>
Message-ID: <38E26060.550DDB5@t-online.de>

Christian Tismer wrote:
> ANNOUNCING:

>               http://www.stackless.com

The text at this URL looks like XML with namespaces and lots
of other stuff, which reproducibly kills Netscape 4.01.
It was readable with IE4, I did not try other browsers.

Might it be possible to change at least the initial page
to something that everybody can read?

Ralf

From joeyGibson at mindspring.com  Fri Mar 31 09:21:21 2000
From: joeyGibson at mindspring.com (Joey Gibson)
Date: Fri, 31 Mar 2000 09:21:21 -0500
Subject: lambda
In-Reply-To: <38e49499.1173061@news.inet.net.nz>
References: <20000329063753.20711.cpmta@c000.sfo.cp.net> <38e1d27a.27069487@news.inet.net.nz> <9sj4esgeh2i9emc4p249ksrj04u1bdm7ur@4ax.com> <8bv7k3$h7e$1@readme.uio.no>  <38e49499.1173061@news.inet.net.nz>
Message-ID: <6vc9essn8as6uaguc6chl1k11v1hm70lpu@4ax.com>

On Fri, 31 Mar 2000 12:06:12 GMT, morbid at doesnotlikespam.inet.net.nz
(Morbid Curiosity) wrote:

||| On Thu, 30 Mar 2000 16:26:43 -0500, Joey Gibson
|||  wrote:
||| >||| > It took me a while to 'get' lambdas at first too. They are most useful
||| >||| > (IMO) when paired with map(), filter() and reject(). These functions take
||| >||| 
||| >||| Where did you find reject()?
||| >
||| >	D'ooh! I work in Smalltalk also, and that is one of its collection
||| >messages. Sorry. Python doesn't have one.
||| 
||| You're not thinking of reduce(), are you?
||| 
||| Morbid

	I could have been. I probably was thinking Smalltalk and Python at the
same time, and transmogrified reduce() to reject: and never noticed the
change. Curious....

Joey

--
-- Sun Certified Programmer for the Java2 Platform
-- My Pocket Smalltalk Stuff: www.mindspring.com/~joeygibson/st/
--
-- "Then slew they the goats, YEA! And placed they the bits,
--  in little pots."

From rasumner at iname.com  Mon Mar 20 04:04:44 2000
From: rasumner at iname.com (Reuben Sumner)
Date: Mon, 20 Mar 2000 11:04:44 +0200
Subject: id() trickery
References: <000301bf9235$1d59a740$c52d153f@tim>
Message-ID: <8b4pj4$qlc$1@news.netvision.net.il>

Tim Peters wrote in message <000301bf9235$1d59a740$c52d153f at tim>...
>but the converse doesn't hold.  For brownie points, figure out the expected
>number of iterations before this program produces some output:

Approximately 82,137.
For brownie points calculate the expected number of iterations before the
program produces some output if the check "if hash2float[h] != x" is
removed.

Reuben

From markx.daley at intel.com  Fri Mar 24 16:47:14 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Fri, 24 Mar 2000 13:47:14 -0800
Subject: Python executables?
Message-ID: 

I read through some of the documentation for Python, and maybe I just
haven't found it yet, but is there a way to make a Python .py file
executable?  I am running Windows NT 4.0, and every time I try that, I get a
DOS window for a couple of seconds, then nothing.  I have my Python
directory in my path.

Regards,

Mark Daley

Return-Path: 
Delivered-To: python-list at python.org
Received: from ebony.nesc.epa.gov (ebony.nesc.epa.gov [134.67.65.43])
	by dinsdale.python.org (Postfix) with ESMTP id 4E0C61CD15
	for ; Wed,  8 Mar 2000 07:07:18 -0500 (EST)
Received: from ralph.rtpnc.epa.gov (ralph.rtpnc.epa.gov [134.67.66.44]) by ebony.nesc.epa.gov (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id HAA17571 for <@ebony.nesc.epa.gov:python-list at python.org>; Wed, 8 Mar 2000 07:09:16 -0500 (EST)
Received: (from rhh at localhost) by ralph.rtpnc.epa.gov (980427.SGI.8.8.8/980728.SGI.AUTOCF) id HAA06136 for python-list at python.org; Wed, 8 Mar 2000 07:09:16 -0500 (EST)
Resent-Message-Id: <200003081209.HAA06136 at ralph.rtpnc.epa.gov>
Received: from relay02.esat.net (192.111.39.21) by mta101.mail.yahoo.com with
 SMTP; 8 Mar 2000 10:33:25 -0000
Received: from (iel.ie) [193.120.149.250] by relay02.esat.net with smtp id
 12SdmL-0008AY-00; Wed, 08 Mar 2000 10:33:24 +0000
Received: from shogan.iel.ie by iel.ie with SMTP (MDaemon.v2.7.SP5.R)
 for ; Wed, 08 Mar 2000 09:55:34 +0000
Date: Wed, 8 Mar 2000 09:54:43 -0000
From: "Shaun Hogan" 
Subject: Re: how does SWIG work? how do i make a python module out of a c
 function?
To: 
Reply-To: shogan at iel.ie
Message-id: <018601bf88e4$54068e20$2801000a at iel.ie>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
X-Mailer: Microsoft Outlook Express 5.00.2314.1300
Content-type: text/plain; charset="iso-8859-1"
Content-transfer-encoding: 7bit
X-Priority: 3
X-MSMail-priority: Normal
X-Apparently-To: aa8vb at yahoo.com via web1004.mail.yahoo.com
X-Track2: 2
X-Track: 1: 40
X-MDaemon-Deliver-To: aa8vb at yahoo.com
X-Return-Path: shogan at iel.ie
References: <012201bf8857$fa33e320$2801000a at iel.ie>
 <20000307141108.A113511 at vislab.epa.gov>
Resent-From: rhh at ebony.nesc.epa.gov
Resent-Date: Wed, 8 Mar 2000 07:09:16 -0500
Resent-To: python-list at python.org
Resent-Sender: rhh at ralph.rtpnc.epa.gov
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta1
Precedence: bulk
List-Id: General discussion list for the Python programming language 

hi again,
i downloaded the 'swig1_3a2_tar' file from
http://download.sourceforge.net/swig/, i was wondering how i make it work or
what do i do with all the files now ive down loaded them. i cant open any of
the files because i have no program associated with them, what program do i
need to view the files? i tried droping them into notepad but that didnt
work?
thanks for your help.
Shaun.

----- Original Message -----
From: Randall Hopper 
To: Shaun Hogan 
Cc: python 
Sent: Tuesday, March 07, 2000 7:11 PM
Subject: Re: how does SWIG work? how do i make a python module out of a c
function?

> Shaun Hogan:
>
>  |
>  |
>  |  |http-equiv=3DContent-Type>
>  |
>  |
>
> Hi.  First, turn off HTML in your mailer (Microsoft Outlook Express I
> gather).  You'll get more replies.  Many folks ignore or filter out posts
> formatted with HTML.
>
>  |   looking at [1]www.swig.org/tutorial.html
>  |
>  |   i can see that this looks like what i need to do, however according
to
>  |   the tutorial "you need to write an "interface file" which is the
input
>  |   to SWIG.", then it gives an example interface file which looks okay
to
>  |   me to do, but my question is.....what application do i use to write
>  |   this"interface file"?,
>
> It's a text file.  So you can use any ASCII text editor you wish.  Since
> you're on an MSWin box, use MSVC++, notepad, edit, whatever (just so long
> as it's a text file when you're done).
>
>  |   and what do i do with it when its wrote?,
>
> You run "swig" on it.  SWIG generates wrapper code (in C, for example) and
> optionally Python wrapper classes (.py) if you tell it to.  E.g. on UNIX:
>
>      swig -shadow -python -I/my/includes -o stuff.i
>
>  |   and what do i do with it when its wrote?
>
> Link the wrapper code with the code it's wrapping, and then load it into
> Python using a standard "import " statement.  I could tell you
> about UNIX, but you'll need to read the docs for MSWin directions.  This
> should all be in the SWIG docs:
>
>      http://sourceforge.net/project/?group_id=1645
>      http://download.sourceforge.net/swig/
>
> See the Doc/ tree in the package.
>
>  |   how do i know if i have written it correctly?
>
> Call your wrapped methods and test them?
>
>  |   further down the page it shows (apparently) how to turn c code into a
>  |   python module for Irix.....i dont know what that is, im using IDLE
and
>  |   windows.
>
> See:
>    http://starship.python.net/crew/mhammond/win32/BuildingExtensions.html
>
>
>  |   can anyone help me, ive been stuck with this problem for a week...and
>  |   im going mad trying to figure it out.
>
> I'm not sure what you're stuck on yet.  Do you have an error message or
> code snippet that's relevent?
>
> --
> Randall Hopper
> aa8vb at yahoo.com
>
>

From: "Zarkon & Zena" 
Newsgroups: comp.lang.python
References:    
Subject: Re: De-compiler for *.pyc files
Lines: 1104
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.00.2919.6600
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600
Message-ID: 
Date: Sun, 12 Mar 2000 08:27:09 -0700
NNTP-Posting-Host: 63.225.7.6
X-Trace: news.uswest.net 952874934 63.225.7.6 (Sun, 12 Mar 2000 09:28:54 CST)
NNTP-Posting-Date: Sun, 12 Mar 2000 09:28:54 CST
Path: news!uunet!ffx.uu.net!news3!logbridge.uoregon.edu!news-out.uswest.net!news.uswest.net.POSTED!not-for-mail
Xref: news comp.lang.python:87846
To: python-list at python.org
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta1
Precedence: bulk
List-Id: General discussion list for the Python programming language 

Ok, I think I am starting to get there... Here is what I get when I try it:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> import doc
>>> import hack
>>> import Command
>>> import ApplyLocation
>>> dir(ApplyLocation)
['ApplyLocation', 'Command', 'DSExceptions', '__builtins__', '__doc__',
'__file__', '__name__']
>>> hack.dis(ApplyLocation.Command)
Traceback (innermost last):
  File "", line 0, in ?
  File "C:\Program Files\Python\Lib\hack.py", line 289, in dis
    dis.disco(c)
  File "dis.py", line 55, in disassemble
    code = co.co_code
AttributeError: co_code
>>>

I am assuming that the function names are being returned by the "dir"
command.
I have attached the files in case you have the time to look at them.

Thanks for your help and ideas.

-Walt

"Michael Hudson"  wrote in message
news:m3n1o5bhe4.fsf at atrus.jesus.cam.ac.uk...
> "Zarkon & Zena"  writes:
>
> > Ok, I have played with "dis" module a little bit, but don't I need to
know
> > the names of the functions inside the *.pyc file before I can get at
them?
>
> Oh that's easy:
>
> Python 1.5.2+ (#13, Jan 14 2000, 09:26:45)
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import string
> >>> dir(string)
> ['_StringType', '__builtins__', '__doc__', '__file__', '__name__',
> '_apply', '_float', '_idmap', '_idmapL', '_int', '_long', 'atof',
> 'atof_error', 'atoi', 'atoi_error', 'atol', 'atol_error',
> 'capitalize', 'capwords', 'center', ...
>
> or
>
> def getfuncs(module):
>     r = []
>     for f in module.__dict__.values():
>         if type(f) is type(getfuncs):
>            r.append( f )
>     return f
>
> or something like that.
>
> I suggest a close reading of the langauge reference if you intend to
> poke around on this level.
>
> HTH,
> M.
>
> --
> very few people approach me in real life and insist on proving they are
> drooling idiots.                         -- Erik Naggum, comp.lang.lisp

begin 666 ApplyLocation.pyc
MF4X-"LE<(SAC``````,```!S,P```'\``'\(`&L``&P```%_"0!K`0!:`0!_
M"P!D``!E``!F`0!D`0"$``"#``!96@(`9 (`4R@#````4QO
M8V%T:6]N8P`````!``(`4QO8V%T:6]N+G!Y7,#`````'P``&D!`&D(`&\:``%_'P!\``!I`0!I
M#0!\``!I!0"#`0`!;A8``7\A`'0.`&D/`'P``&D!`(,!`((!`&0``%,H`P``
M`$YS"@```'9D9E]R97-U;'1S!0```'=R:71E*! ```!S! ```'-E;&9S!P``
M`'-E6QE7,-````07!P;'E,;V-A=&EO
M;@L`7,$````:6-O;G,$````3F]N92@$````7,(````7U]I;FET7U\,`',,`````P$,`0P!#P$<`1T"
M8P$``0`!``,`7,#````7,!````/P@`71H;VX at +4\*"B(B(B *(" @($%U=&]M871I
M8R!0>71H;VX at 1&]C=6UE;G1A=&EO;B!I;B!(5$U,"@H@(" @5&AI2!O;FQY($A4
M34P@:7,@2!M86ME('5S92!O9B!T:&4@' G
M6YA;6EC86QL>0H@(" @9&5F:6YE9"!C;V1E("AL:6ME(&-O9&4@:6X@
M:68N+BYE;'-E+BXN+6-L875S97,I(&ES(&YO="!P87)S960N"@H@(" @5&AE
M(&1O8RUS=')I;F=S(&%R92!P0H@(" @8V]N=&%I;B!S<&5C:6%L(&-H87)A8W1E6]U(&AA=F4@=&\*(" @(" @:6YC;'5D92!S:6YG;&4@;V-C=7)E;F-E2!I;G1O"B @(" @('1H92!(5$U,+69I;&4@>6]U
M(&AA=F4@=&\@8F4 at 8V%R969U;"!A8F]U="!U6]U6)E($D@2!(5$U,"B @(" @("!I;B!M>2!C=7)R96YT(&1O8RUS=')I;F=S(&%N
M>7=A>5T*(" @("T at 8V]D92!C;W5L9"!B92!M861E(&9A2!U71H;VXN;W)G(&9OF5D($A434P@=F5R7EO=7(@4'ET:&]N('-C
M2 H9&]W;FQO
M861A8FQE(&9R;VT@;7D at 4'ET:&]N('!A9V5S(&]N('-T87)S:&EP*0H*(" @
M($AI&5S('1O('1H92!R96=E
M>'!S("AT:&%N:W,@=&\@5&EM(%!E=&5R' @=&\@
M;F]T(&)R96%K(&-O9&4 at 870@8V]M;65N=',*(" @(" @(" @(" H=&AI&5D(&)U9R!T:&%T(&-U="!A=V%Y(&9I2!O9B!+
M96ET:"!$879I9'-O;BD["B @(" @(" @(" @861D960@)V\G(&%S(&QI7, at 879A:6QA
M8FQE(&9R;VT@;7D at 4'ET:&]N('!A9V5S. at H*(" @(" @(" @("!H='1P.B\O
M2!-87)C+4%N9')E($QE;6)U2P@
M86YD(&1I2!I7!O2!A;F0 at 8G5I;&0 at 86X*:6YT
M97)N86P@;V)J96-T('-T2!O;FQY($A434P@:7,@"YC;VUP:6QE*"=<*" J7"EC;&%S%]F=6YC=&EO;B ](')E9V5X+F-O;7!I;&4H)UPH
M("I<*61E9B K7"A;7B@@72M<*2 J)PH@(" @(" @(" @(" @(" @(" @(" @
M(" @(" @)RA<*%PH6UXH*5U%]M971H
M;V0@/2!R>%]F=6YC=&EO;@IR>%]B;V1Y:6YD96YT(#T@(%QN72(%QN72"YC;VUP
M:6QE*"<@*E]?=F5R7"=<(ETK7"E;7"=<
M(ETG*0HC(&-A8VAE('-O;64 at 8V]M;6]N('!A" ](')E9V5X+F-O;7!I;&4H%]P87)TF4@/2 X"@HC('-O
M;64 at 9V]B86P@;W!T:6]N7!O
M3)H=&UL+G!Y+ H)"2,@=7-E('1H:7,@=F5R'0@:&%S('1O(&)E
M(&EN9&5N=&5D(&)Y(&EN9&5N="!S<&%C97,*"2H@=&5X="!M=7-T(&)E(%1!
M0BUF2!D969I;F5D(&]B:F5C=', at +2T@=7-I;F<*"2 @
M:68G" ](')X7W!A)RMS:2LG9G)O;2!)RMS:2LG6UX at 7&Y=)PH)"YC;VUP
M:6QE*')S*0H)'1;7!E
M6RTQ72 ]/2 G("7!E6SHM,5T*"6Q;+3%=6S%=(#T@=" C
M('1H92!L87-T('!A'0I"B @("!L(#T@
M;6%P*'1U<&QE+&PI"B,@(" @<')I;G0@)RHJ(')E='5R;FEN9R!W:71H('!A
M'1;" ](')X7V)O9'EI;F1E;G0*(" @(&EF(')X+G-E87)C:"AT97AT+'-T
M87)T*2 A/2 M,3H*"6$L8B ](')X+G)E9W-;,5T*"7)E='5R;B!B+6$*(" @
M(&5L6)E('1H97)E(&%R92!N;R!N97<@;&EN97,@;&5F=" H
M92YG+B!A="!T:&4 at 96YD(&]F(&$@9FEL92D*"7)X(#T@6EN9&5N
M=%\R"@EI9B!R>"YS96%R8V at H=&5X="QS=&%R="D@(3T at +3$Z"@D@(" @82QB
M(#T@'0I. at H*
M(" @("(B(B!S=6)S=&ET=71E('-U8B!F;W(@979E'0L9FEN9"D*(" @(')E='5R;B!S=')I;F6QE(&YE=VQI;F5S(&5V97)Y=VAE'0I"B @("!R971U'0I"@ID
M968 at 97-C87!E7VQO;F=?'0L)UPB
M7")<(B'1R86-T7V1O8U]S=')I;F71H;VXM
M9F%S:&EO;B!O9B!N86UI;F<@=&AI;F=S"B @(" B(B(*"B @("!T(#T at 9&]C
M+FYA;64*(" @('=H:6QE(&1O8RYO=VYE'0 at 87, at 82!D;V,M'0@/2!T
M97AT"@H@(" @9&5F(%]?&%M<&QE. at H*"2 @(" @(", at 9&5F('5N97-C87!E
M7VQO;F=?'0L)UPB7")<(B'0I. at D)"2,*"2 @(" @(" @;" ]('-T2X*"@D@(" @*B!)9B!Y;W4@<&QA;B!T;R!P=70@=F5R8F%T
M:6T at 2%1-3"UC;V1E(&EN;&EN92!T:&5N('EO=2!C86X*"2 @(" @('5S92!T
M:&ES('-Y;G1A>"!6]U
M)VQL(&AA=F4@=&\@;&]O:PH)(" @(" @870@=&AE('-O=7)C92!C;V1E('1O
M(&9I;F0@;W5T(&AO=R!I="!W;W)KD$M6C M.4 N)5PM72M<*2D$M6C M.4 N)5PM7"\_)BL]?ETK7"DG+ H@(" @(" @(" @(" @
M(" @(" @(" @(" G/$$@2%)%1CTB:'1T<#HO+UQ<,2(^:'1T<#HO+UQ<,3PO
M03XG+'0I"@ET(#T@D$M6C M.4 N
M)5PM7"\_)BL]?ETK7"DG+ H@(" @(" @(" @(" @(" @(" @(" @(" G/$$@
M2%)%1CTB9G1P.B\O7%PQ(CYF=' Z+R]<7#$\+T$^)RQT*0H*"70@/2 G/% ^
M("5S(#PO4#XG("4@= H*"7)E='5R;B!T"@IC;&%S"QS>2QO=VYE"QS>2D*"7-E;&8N;W=N97(@/2!O=VYE<@H)%]C;&%S"YM871C:"AT97AT+'-X*0H):68@'1; H)"YG2UI;F1E;G0*"7-E;&8N
M8F]D>6EN9&5N=" ](&-A;&-?8F]D>6EN9&5N="AT97AT+'-T87)T*0H)(R!B
M5TL'0L"QS=&%R="MY+'-E;&8I*0H)"65L:68@='EP
M92 ]/2 G8VQA'0L"QS=&%R="MY
M+'-E;&8I*0H)"65L:68@='EP92 ]/2 G(R7!E(#T]("=<(EPB7"(G. at H)"2 @
M(" C(&=O="!A(&1O8RUS=')I;F<*"0D@(" @&-E<'0 at 4&%R"QS>2QO=VYE'0@(" @+2!O2P at 8G)O:V5N(&EN=&\@<&%R=',*
M"2(B(@H@(" @(" @('-E;&8N=&5X=" ]('1E>'0*"7-E;&8N'0L'1;"YG'0L2QT>7!E(&EN('-E;&8N<&%R=',Z
M"@D):68@('1Y<&4@/3T@)V1E9B7!E(#T]("=C;&%S
M7!E(#T]("'1;#IS=&%R="MY72DL'0L#IS>5T at 9F]R(&$@;65T:&]D
M(&1E9FEN:71I;VX at 86YD(&%L;"!I=',*"2 @("!M96UB97)S.R!I;G-T86YC
M92UV87)I86)L97,Z"@D@(" @*B!T97AT(" @("T@;W)I9VEN86P at 8V]D90H)
M(" @("H@'0@=VAE2P at 8G)O:V5N(&EN=&\@<&%R=',*"2(B(@H@(" @(" @('-E
M;&8N=&5X=" ]('1E>'0*"7-E;&8N#IS>5TI"@EI9B!S=&%R=" \(# Z"@D@(" @(R!W92=V92!G
M;W0 at 82!P#IS>5T*"2 @("!R86ES92!087)S945R'1;6EN9&5N="D*(PEP2QT>7!E(&EN('-E;&8N<&%R
M=',Z"@D):68@('1Y<&4@/3T@)V1E9B7!E(#T]("=C
M;&%S7!E(#T]("'1;#IS=&%R="MY72DL71H;VX@;6]D=6QE"@D@(" @*B!O=VYE
M2P at 8G)O:V5N(&EN
M=&\@<&%R=',*"2(B(@H)(R!R96%D(&9I;&4L(&5X<&%N9"!T86)S+"!F:7@@
M;&EN92!BF4I"@ES
M96QF+G1E>'0@/2!F:7A?;&EN96)R96%K'0I"@ES96QF+G1E
M>'0@/2!E'1;+3$Z72 A/2 G7&XG. at H)(" @('-E;&8N
M=&5X=" ]('-E;&8N=&5X=" K("=<;B<*"7-E;&8N;F%M92 ](&]S+G!A=&@N
M0H)2QT>7!E(&EN('-E;&8N<&%R=',Z"@D):68@
M('1Y<&4@/3T@)V1E9B2QS96QF*2D*"0EE;&EF('1Y<&4@/3T@)V-L87-S)SH*"0D@(" @(R!G
M;W0 at 82!C;&%S'0L>"QY+'-E;&8I*0H)"65L:68@='EP92 ]/2 G(R7!E(#T]
M("=<(EPB7"(G. at H)"2 @(" C(&=O="!A(&1O8RUS=')I;F<*"0D@(" @#IY72DL'1R86-T7W9E5T*"2 @("!P2" ]('-E;&8N;F%M92LG+FED>"YH=&UL)PH)"MS96QF+FEN9&5X+"!H H);W5T<'5T(#T@(B(B"@D@(" @
M(" @(" @(" \2%1-3#X*"0D@(" @/$A%040^"@D)(" @(#Q4251,13Y$;V-)
M;F1E>#HZ4')O:F5C=#HZ)7,\+U1)5$Q%/@H)"2 @(" \345402!.04U%/2)+
M15E73U)$4R(@0T].5$5.5#TB)7,L(&EN9&5X+"!D;V-U;65N=&%T:6]N+"!P
M>71H;VXB/@H)"2 @(" \+TA%040^"@D)(" @(#Q"3T19($)'0T],3U(](T9&
M1D9&1CX*"0D@(B(B("4@*'-E;&8N;F%M92P@"MS96QF+F-O
M;G1E;G0K7 H)"0D)(" @(" G(R"MS96QF+F-O;G1E;G0K7 H)"0D@
M(" @(""YW
M"YC;&]S92 at I"@D*"2,@;6%K92!C;VYT96YT
M+69I;&4*"7!R:6YT("71H
M;VXL(&1O8RUT;V]L(CX*"0D@(" @/"](14%$/@H)"2 @(" \0D]$62!"1T-/
M3$]2/2-&1D9&1D8^"@D)("(B(B E("AS96QF+FYA;64L H*(" @('!R:6YT("=$;V,M
M5&]O;"!6)RM?7W9E7,N87)G=BD@/" S. at H)<')I;G0@)U-Y;G1A>#HG+'-Y" \<')O:F5C="UN86UE/B!C;VYT86EN:6YG)PH)<')I;G0@)W1H
M92!D;V,M"P@:68G
M"@EP&ET*"D*
M(" @('!R:6YT("=7;W)K:6YG(&]N('!R;VIE8W0Z)RQS>7,N87)G=ELQ72PG
M" ]('-Y" ]("7)I9VAT(&)Y($UA2P@;6]D:69Y+"!A;F0@
M9&ES=')I8G5T92!T:&ES('-O9G1W87)E(&%N9 T*(" @(&ET2!I2!G7)I9VAT(&YO=&EC92!A;F0@
M=&AI7!O
M&%M:6YE#0IC;V1E(&9R;VT@:6YS:61E(&%N(&EN=&5R86-T:79E(&EN=&5R
M<')E=&5R('-E2!D;V,@0T*&-E<'1I;VX@:&5L<&5R3H-"@EP(%TG*0T*#0ID968@7,N;6%X:6YT#0H);" ](&UA<"AE
M>'!A;F1T86)S+&PI#0H)9F]R(&D@:6X@"QI;F1E;G0]:6YD96YT
M.B!X6VEN9&5N=#I=+&Q;,3I=*0T*(" @(&5L&1E<'1H/3(L;&5V96P]
M,"P-"@D at 24Y$14Y4/4E.1$5.5"DZ#0H-"B @(" B(B(@&1E<'1H+&QE=F5L*0T*(" @('-H
M;W=A='1R*&,L)U]?871T&1E<'1H. at T*"7)E='5R;@T*#0H@(" @9F]R
M('@@:6X@:71E;7,Z#0H)=')Y. at T*"2 @("!A(#T at 9V5T871T&-E<'1I;VXJ)PT*"6EF(&QE;BAR*2 ^(#0P. at T*"2 @("!R
M(#T@"QR
M*0T*#0ID968@3H-"@EX
M(#T at 9V5T871T&1E<'1H. at T*"7)E='5R;@T*#0H@(" @&1E<'1H
M+&QE=F5L*0T*#0ID968@&1E<'1H. at T*"7)E='5R;@T*#0H@(" @:68@
M;F]T(&ET96US. at T*"7!R:6YT(""!I;B!I=&5M&-E<'1I;VXJ)PT*"6EF(&QE;BAR*2 ^(#0P. at T*"2 @("!R
M(#T@&1E<'1H. at T*"7)E='5R;@T*#0H@(" @:68@;F]T(&ET96US. at T*"7!R
M:6YT("2D-"@EE>&-E<'0Z#0H)(" @(&L@/2 G*F5X8V5P=&EO;BHG#0H)
M:68@;&5N*&LI(#X at -# Z#0H)(" @(&L@/2!K6SHT,%TK)R N+BXG#0H)=')Y
M. at T*"2 @("!V(#T@&-E<'0Z#0H)(" @('8@/2 G
M*F5X8V5P=&EO;BHG#0H)(" @('9A;'5E(#T at 3F]N90T*"6EF(&QE;BAV*2 ^
M(#0P. at T*"2 @("!V(#T@=ELZ-#!=*R<@+BXN)PT*"6EF(&QE=F5L(#P@;6%X
M9&5P=&@@86YD('9A;'5E(&ES(&YO="!.;VYE. at T*"2 @("!P7!E*&,I(#T]
M('1Y<&5S+E-T7!E2!A8V-U
M&5C(&,@:6X@;F%M97-P86-E#0H@(" @
M96QS93H-"@EI;7!O&5C(&,@:6X at 7U]M86EN7U\N
M7U]D:6-T7U\-"@T*9&5F('=H>2 at I.@T*#0H@(" @(B(B('=H>2 at I("TM('-H
M;W<@;&]C86QS('1H870 at 8V%U'0-"@EF2!P7,N;6]D=6QE7,N;6]D=6QE
Delivered-To: python-list at python.org
Received: from ebony.nesc.epa.gov (ebony.nesc.epa.gov [134.67.65.43])
	by dinsdale.python.org (Postfix) with ESMTP id 8199B1CD0D
	for ; Wed,  8 Mar 2000 07:07:09 -0500 (EST)
Received: from ralph.rtpnc.epa.gov (ralph.rtpnc.epa.gov [134.67.66.44]) by ebony.nesc.epa.gov (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id HAA17562 for <@ebony.nesc.epa.gov:python-list at python.org>; Wed, 8 Mar 2000 07:09:07 -0500 (EST)
Received: (from rhh at localhost) by ralph.rtpnc.epa.gov (980427.SGI.8.8.8/980728.SGI.AUTOCF) id HAA06113 for python-list at python.org; Wed, 8 Mar 2000 07:09:07 -0500 (EST)
Resent-Message-Id: <200003081209.HAA06113 at ralph.rtpnc.epa.gov>
Received: from relay02.esat.net (192.111.39.21) by mta101.mail.yahoo.com with
 SMTP; 8 Mar 2000 10:30:44 -0000
Received: from (iel.ie) [193.120.149.250] by relay02.esat.net with smtp id
 12SdjU-00081s-00; Wed, 08 Mar 2000 10:30:29 +0000
Received: from shogan.iel.ie by iel.ie with SMTP (MDaemon.v2.7.SP5.R)
 for ; Wed, 08 Mar 2000 09:29:46 +0000
Date: Wed, 8 Mar 2000 09:28:55 -0000
From: "Shaun Hogan" 
Subject: Re: how does SWIG work? how do i make a python module out of a c
 function?
To: 
Reply-To: shogan at iel.ie
Message-id: <016a01bf88e0$b9663bc0$2801000a at iel.ie>
MIME-version: 1.0
X-MIMEOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
X-Mailer: Microsoft Outlook Express 5.00.2314.1300
Content-type: text/plain; charset="iso-8859-1"
Content-transfer-encoding: 7bit
X-Priority: 3
X-MSMail-priority: Normal
X-Apparently-To: aa8vb at yahoo.com via web1001.mail.yahoo.com
X-Track2: 2
X-Track: 1: 40
X-MDaemon-Deliver-To: aa8vb at yahoo.com
X-Return-Path: shogan at iel.ie
References: <012201bf8857$fa33e320$2801000a at iel.ie>
 <20000307141108.A113511 at vislab.epa.gov>
Resent-From: rhh at ebony.nesc.epa.gov
Resent-Date: Wed, 8 Mar 2000 07:09:07 -0500
Resent-To: python-list at python.org
Resent-Sender: rhh at ralph.rtpnc.epa.gov
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta1
Precedence: bulk
List-Id: General discussion list for the Python programming language 

hi there,
thanks for your help, the problem i refered to that i was stuck on is:

below is the c file 'tester.c', what im trying to do is:
write python script that does the equivalent to the following : i.e.,
creates the C++ object created in this function (the "writeThisInPython"
function) and then calls the same 2 methods called here, i.e., getName() and
getNumber().

if anyone has any ideas on how i can figure out how TODO these 2 problems,
could they please let me know.
thanks
Shaun

/*----------------------
tester.c
----------------------*/

#include 
#include "Subscriber.h"

void writeThisInPython(void);
void callPythonFromHere(void);

void main(void)
{
 writeThisInPython();
 callPythonFromHere();
}

// TO-DO 1 : write python script that does the equivalent to the following :
// i.e., creates the C++ object created in this function and then calls
// the same 2 methods called here, i.e., getName() and getNumber()

void writeThisInPython(void)
{
 Subscriber subscriber("Mick McCarthy", 99);

 cout << "Subscriber Name   : " << subscriber.getName() << endl;
 cout << "Subscriber Number : " << subscriber.getNumber() << endl;
}

void callPythonFromHere(void)
{
 Subscriber subscriber("Mick McCarthy", 99);

 cout << "Subscriber Name   : " << subscriber.getName() << endl;
 cout << "Subscriber Number : " << subscriber.getNumber() << endl;

 // TO-DO 2 : call python code from here to write the above details
 // to a file (format doesn't matter)
}

----------------------------------------------------------------------------
------------------------
----- Original Message -----
From: Randall Hopper 
To: Shaun Hogan 
Cc: python 
Sent: Tuesday, March 07, 2000 7:11 PM
Subject: Re: how does SWIG work? how do i make a python module out of a c
function?

> Shaun Hogan:
>
>  |
>  |
>  |  |http-equiv=3DContent-Type>
>  |
>  |
>
> Hi.  First, turn off HTML in your mailer (Microsoft Outlook Express I
> gather).  You'll get more replies.  Many folks ignore or filter out posts
> formatted with HTML.
>
>  |   looking at [1]www.swig.org/tutorial.html
>  |
>  |   i can see that this looks like what i need to do, however according
to
>  |   the tutorial "you need to write an "interface file" which is the
input
>  |   to SWIG.", then it gives an example interface file which looks okay
to
>  |   me to do, but my question is.....what application do i use to write
>  |   this"interface file"?,
>
> It's a text file.  So you can use any ASCII text editor you wish.  Since
> you're on an MSWin box, use MSVC++, notepad, edit, whatever (just so long
> as it's a text file when you're done).
>
>  |   and what do i do with it when its wrote?,
>
> You run "swig" on it.  SWIG generates wrapper code (in C, for example) and
> optionally Python wrapper classes (.py) if you tell it to.  E.g. on UNIX:
>
>      swig -shadow -python -I/my/includes -o stuff.i
>
>  |   and what do i do with it when its wrote?
>
> Link the wrapper code with the code it's wrapping, and then load it into
> Python using a standard "import " statement.  I could tell you
> about UNIX, but you'll need to read the docs for MSWin directions.  This
> should all be in the SWIG docs:
>
>      http://sourceforge.net/project/?group_id=1645
>      http://download.sourceforge.net/swig/
>
> See the Doc/ tree in the package.
>
>  |   how do i know if i have written it correctly?
>
> Call your wrapped methods and test them?
>
>  |   further down the page it shows (apparently) how to turn c code into a
>  |   python module for Irix.....i dont know what that is, im using IDLE
and
>  |   windows.
>
> See:
>    http://starship.python.net/crew/mhammond/win32/BuildingExtensions.html
>
>
>  |   can anyone help me, ive been stuck with this problem for a week...and
>  |   im going mad trying to figure it out.
>
> I'm not sure what you're stuck on yet.  Do you have an error message or
> code snippet that's relevent?
>
> --
> Randall Hopper
> aa8vb at yahoo.com
>
>

From effbot at telia.com  Thu Mar 30 15:52:21 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 30 Mar 2000 20:52:21 GMT
Subject: Tk novice question
References: <38E3ACDD.D7D0918B@westlakefinancial.com>
Message-ID: <98PE4.409$n68.188469760@newsb.telia.net>

Joseph Santaniello wrote:
> How do I trap window close events for a given window? I want to (after a
> window is closed by hitting the X or equiv) to cycle through a list of
> spawned windows, but I don't ever know when my window gets closed. Where
> should I look for info on how to do this?

http://www.pythonware.com/library/tkinter/introduction/events-and-bindings.h
tm
=> protocols

From darrell at dorb.com  Sat Mar 11 23:32:53 2000
From: darrell at dorb.com (Darrell)
Date: Sat, 11 Mar 2000 20:32:53 -0800
Subject: Forth like interpreter
Message-ID: <158801bf8bdc$08516040$6401a8c0@dorb>

Wrote a Forth like interpreter in Python.
Probably useless but it gave me something to do today.
http://www.dorb.com/darrell/forth/forth.py

Kragen Sitaker mentioned in "What if Python Replaced Elisp" that Forth
compilers were easy. He's right!

I only did one branching instruction "if" and an assortment of other words.

The compile phase could generate Python byte code or for real crazies
assembler. Real Forth often comes with an assembler built in. So I can see
how some version of Lisp might compile to assembler.

--Darrell

From hamlet at en.com  Tue Mar 28 22:35:53 2000
From: hamlet at en.com (hamlet at en.com)
Date: Wed, 29 Mar 2000 03:35:53 GMT
Subject: US v Microsoft
Message-ID: <38e17b68$0$1374@news.voyager.net>

I think that what many people lose sight of in defending Microsoft is
that any concentration of power of this magnitude has the potential
for being abused. I think that this litigation has been very broadly
based and courageous; I respect the government lawyers for using the
rule of law to attempt to check the tyrannical exercise of power.

I also think that Microsoft has already suffered great reverses in the
court of public opinion, especially expert opinion. It is incumbent on
those of us who have expert knowledge to marshall that knowledge in
the service of fair play.

In one sense, the points raised in this litigation have been made moot
by certain technical advances, most notably the open source movement.
The unavoidable reality is that these developments have come about in
spite of the concerted efforts of Microsoft the thwart them. That, I
believe , gives the lie to any claim of Microsoft to be an
"innovator."

From effbot at telia.com  Sun Mar 12 08:59:19 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 12 Mar 2000 13:59:19 GMT
Subject: Sharing XMLRPC w/ my neighbors
References: <5650A1190E4FD111BC7E0000F8034D26A0F372@huina.oceanic.com>
Message-ID: 

Doug Stanfield  wrote:
> There exists this group of folks who work in ways completely foreign to
me.
> They're known as the IS department and stoke an AS400 with something
called
> RPG-II to do billing .  It would be very useful if they could use
one
> of my interfaces, but I don't know how to tell them what to do.  They want
> to hear something on the order of "Open a socket..." and I jumped at using
> the xmlrpclib precisely to hide those mungy details.

maybe they can run Java on their boxes?  several
Java client libraries are available from:

    http://www.xmlrpc.com/

(alternatively, if RPG-II supports XML and HTTP in some
way, it shouldn't be that hard to implement a small portion
of the XML-RPC spec.  you'll find the specification on the
above site as well).

From hnowak at cuci.nl  Fri Mar 24 01:11:49 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Fri, 24 Mar 2000 07:11:49 +0100
Subject: file / string problem?
In-Reply-To: 
Message-ID: <200003240612.GAA27838@dionysus.fw.cuci.nl>

On 24 Mar 00, at 0:27, Todd Palmer wrote:

> I new to python and I'm having a problem writing a binary string to a
> file.  I'm using the zipfile.py module and getting the data from the zip
> file and writing it to a file object.  The string returned from the
> zipfile.read() function is the correct size, but when I write it to disk,
> the file on the disk is the wrong size and corrupted.  Any ideas?  Do I
> have to do something special with binary data in strings?
> 
> Code:
> 
> import zipfile
> 
> z = zipfile.ZipFile('D:\\tmp\\esbdates.zip')
> try:
>  # read the binary data into a string
>    S = z.read('esbdates.hlp')
>  # note len(S) here returns 362588 which is correct  as per 
>  # zipfile.py AND winzip
>  f = open('D:\\tmp\\esbdates.hlp', 'w')

Try 'wb' rather than 'w'...

Blessed be,

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://www.hvision.nl/~ivnowa/newsite/
Python questions? See http://tor.dhs.org/~zephyrfalcon/snippets
You call me a masterless man. You are wrong. I am my own master.

From shogan at iel.ie  Tue Mar  7 12:10:02 2000
From: shogan at iel.ie (Shaun Hogan)
Date: Tue, 7 Mar 2000 17:10:02 -0000
Subject: how does SWIG work? how do i make a python module out of a c function?
Message-ID: <012201bf8857$fa33e320$2801000a@iel.ie>

as you may have seen from previous postings i need to embed some c code in a python script (make a python module out of a c function) and i dont know how.
looking at www.swig.org/tutorial.html
i can see that this looks like what i need to do, however according to the tutorial  "you need to write an "interface file" which is the input to SWIG.", then it gives an example interface file which looks okay to me to do, but my question is.....what application do i use to write this"interface file"?, and what do i do with it when its wrote?, how do i know if i have written it correctly? how do i get it to work in the script, and get the function to do what its supposed to?

further down the page it shows (apparently) how to turn c code into a python module for Irix.....i dont know what that is, im using IDLE and windows.
can anyone help me, ive been stuck with this problem for a week...and im going mad trying to figure it out.
Thanks 
Shaun   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From claird at starbase.neosoft.com  Thu Mar  2 23:06:41 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 2 Mar 2000 22:06:41 -0600
Subject: Most common #! format
References: 
Message-ID: 

In article ,
Peter Funk  wrote:
>Hi!
>
>Chuck Esterbrook:
>> What's the most common #! format?
>> 
>> #!python
>> #!/usr/local/bin/python
>> #!/usr/bin/env python
>
>Since MacOS and Windows don't care about this very first line,
>this is only interesting to increase portability on unixoid platforms.  
Windows' command.com also supports analogous hackery,
yielding .bat sources which can act either as inde-
pendent executables or as well-formed Pyscripts.  I
have an allergic reaction to the syntactic contortions
involved, so I'll not give an example now.  Maybe we
should include one in the FAQ, though ...
			.
		[sage counsel on
		various more-or-
		less pertinent
		topics]
			.
			.
>mysterious security policy).  Under this rare circumstances the second
>form '#!/usr/local/bin/python' comes into play.  This would allow
>to call for example a CGI-script from a web server process, which 
>didn't has Python on its search path.  But this situation should 
>be considered as very exotic and rare today.
You just don't travel in the right (or wrong) circles
enough.  There are sites where ... well, I'll sum-
marize this way:  #! composition is like GUI toolkit
selection, in that different circumstances lead rea-
sonable people to different choices.  There's no one
right first-line.

Note that 
also explains the 
  #! /bin/sh
  """:"
  exec python $0 ${1+"$@"}
  """                       
alternative which I favor.

Exercise for those designing a c.l.p reorganization:
should this thread take a home in .platform, .question,
.cgi, ...?
			.
			.
			.
-- 

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

From shogan at iel.ie  Fri Mar 10 12:26:32 2000
From: shogan at iel.ie (Shaun Hogan)
Date: Fri, 10 Mar 2000 17:26:32 -0000
Subject: c/python integration
Message-ID: <012c01bf8ab5$c6fdd1e0$2801000a@iel.ie>

right so, heres my problem,
i have this c code (below) and i want to call the functions getname() and
getnumber() from within the python interpreter, can anyone give me the code
i need or tell me how to do it pls
Thanks
Shaun

#define debug
/* #undef debug*/ // Uncomment this to compile with python
#ifndef debug
    #include "Python.h"
#else
    #include 
    #include 
#endif /*ndef debug */

typedef struct
{
    char name [100];
    int number;
}subscriber ;

int get_number (subscriber s);

char * get_name (subscriber s)
{
    char *t = (char *) malloc (sizeof (char) * 100);
    strcpy (t, s.name);
    return t;
}

#ifdef debug
int main (void)
{
    get_sub_details();

}
#endif /*debug*/

int get_number (subscriber s)
{
    return s.number;
}

#ifndef debug
Py_Object * get_sub_details (Py_Object *, Py_Object *)
#else
int get_sub_details(void)
#endif  /*debug */
{
    // this is one of the functions exported
    // to Python
    // For testing , create a sub object here
    subscriber test = {"Shaun", 100};
    printf ("name %s \t age %d\n", get_name (test) , get_number(test));
}

From markx.daley at intel.com  Thu Mar 23 10:58:35 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Thu, 23 Mar 2000 07:58:35 -0800
Subject: List limitations?
Message-ID: 

OK, here's a good one:

I am using this glorious little piece of code, which I will explain in
detail:

def collect():
	familyname = ['name1', 'name2', 'name3', 'name4', 'name5', 'name6',
'name7', 'name8', 'name9', 'name10', 'name11', 'name12']
	servername = ['server1', 'server2']
	sql_date = time.strftime('%Y'+"-"+'%m'+"-"+'%d
%H'+":"+'%M'+":"+'%S', time.localtime(time1[0]))
	current_date = time.strftime('%Y'+"-"+'%m'+"-"+'%d
%H'+":"+'%M'+":"+'%S', time.localtime(time2[0]))
	file_date = time.strftime('%m'+"_"+'%d'+"_"+'%Y',
time.localtime(time.time()))
	for index in servername:
		dbc = odbc.odbc(index+'/username/password')
		crsr = dbc.cursor()
		for name in familyname:
			crsr.execute("""select * from subtestlist\n\
where (errorcode<>'null'\n\
AND familyname='"""+name+"""'\n\
AND Packagestarttime>'"""+sql_date+"""'\n\
AND Packagestarttime<='"""+current_date+"""');""")
			result = crsr.fetchall()
			for item in range(len(result)):
				data1, data2, data3, data4, data5, data6,
data7, data8, data9, data10, data11, data12, data13, data14, data15, data16,
data17 = result[item]
				data10 = time.ctime(data10)
				data11 = time.ctime(data11)
				result[item] = (data1, data2, data3, data4,
data5, data6, data7, data8, data9, data10, data11, data12, data13, data14,
data15, data16, data17)
			mylist = result[:]
			f = open(index+'_'+name+'_'+file_date+'.txt','a')
			for item in mylist:
				f.write(`item`+'\n')
			f.close()

What this does is connect to a server and pull data from it based on each
family name in the familyname list, using an SQL query.  Conceivably, this
code should connect to server1 and pull data for all familynames and write
that data to a file, server_familyname_date.  Then it should connect to
server2 and do the same.  However, when it connects to server 2, it isn't
pulling the familynames from the familyname list correctly.  Instead, it is
pulling single character strings, like a, b, e, g, h, r, S (these are the
actual strings, which may or may not have some bearing on one of my
familyname entries, I haven't been able to verify).  Has anyone else seen
problem remotely related to this?  Any help is appreciated.

Regards,

Mark Daley

From TBryan at UTS.Itron.com  Sun Mar  5 20:19:29 2000
From: TBryan at UTS.Itron.com (Tom Bryan)
Date: Sun, 05 Mar 2000 20:19:29 -0500
Subject: Parsing SQL-Files
References: <38C02CC0.23D2DBA3@weihenstephan.org>
Message-ID: <38C307A1.9B7B546@UTS.Itron.com>

"Dr. Peter Stoehr" wrote:

> I have more than 100.000 lines of SQL code forming a script to create a
> database. Unfortunately, there is no documentation. That's why I'm
> thinking about writing a python based script that parses all the SQL
> statement. 

I did something similar, but it's very limited and special-case.

I had a huge script of "create table" statements but no ER Diagram for 
an Oracle database.  I wrote a script to pull the information about all 
of the tables into Python and to write out (using HTMLgen) a web page 
with all of the tables and columns with many links within the page.

If you want to use it as a starting point for your program, I could 
post it tomorrow.  Let me know if you're interested.

---Tom

From effbot at telia.com  Sun Mar  5 22:24:32 2000
From: effbot at telia.com (Fredrik Lundh)
Date: 5 Mar 2000 21:24:32 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Mar  5)
Message-ID: <06B081C4E2C91FF2.AF83611DE3B513A3.B8442F0447C73E32@lp.airnews.net>

    Major New Internet Portal Python Powered!
        http://www.filibuster.org

    On the edu-sig, David Scherer reminds us what CP4E (and
    Python) is all about:
        http://www.python.org/pipermail/edu-sig/2000-February/000098.html
    "CP4E *is* about computers.  It's about the incredibly
    versatile machines sitting on all our desks, and (soon)
    in our pockets and around our wrists and in our clothing
    and cars and walls.  It's about giving ordinary people
    some real control over those objects, the basic tools
    they will need to learn more about them, and the freedom
    to change or extend software that isn't right for them."

    New guidelines for patch contributions:
        http://www.python.org/patches/

    Python's CVS repository can be accessed via the web
        http://cvs.python.org/

    The edu-sig deals with everything related to using Python
    in education (except, perhaps, whether Python should remain
    case sensitive or not).  Great discussions, this far:
        http://www.deja.com/getdoc.xp?AN=580836413
        http://www.python.org/sigs/edu-sig/ 

    The import-sig will design a new architecture for the import
    facilities in Python (to better support embedding, archives,
    installers, etc):
        http://www.deja.com/getdoc.xp?AN=581571482
        http://www.python.org/sigs/import-sig/

    New Python consortium members that we forgot to mention in
    last week's "Python-URL!"
        ASTi: http://www.asti.com
        BeOpen: http://www.beopen.com

    Danny Yee reviews Dave Beazley's Python Essential Reference:
        http://www.deja.com/getdoc.xp?AN=582232970
        http://www.anatomy.usyd.edu.au/danny/book-reviews/h/Python_Reference.html

    How do I add subject lines to mails sent via smtplib?
        Q: http://www.deja.com/getdoc.xp?AN=582370142
        A: http://www.deja.com/getdoc.xp?AN=582411857

    What's wrong with execfile and local variables?
        Q: http://www.deja.com/getdoc.xp?AN=581374635
        A: http://www.deja.com/getdoc.xp?AN=581374636
        A: http://www.deja.com/getdoc.xp?AN=581843072

    How do I best prepare a Python?
        A: http://www.deja.com/getdoc.xp?AN=581198835

    New releases of plotting software
        http://www.deja.com/getdoc.xp?AN=581191243 (gdchart)
        http://www.deja.com/getdoc.xp?AN=582314930 (pxDislin)

    And to round off this week's URL, here's Python poetry
        http://www.deja.com/getdoc.xp?AN=581006163

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

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

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      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/   

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

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

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

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

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

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

To receive a new issue of this posting in e-mail each Monday morning,
ask  to subscribe.  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 
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html

From sanjay08544 at yahoo.com  Thu Mar  2 19:09:02 2000
From: sanjay08544 at yahoo.com (sanjay pradeep)
Date: Thu, 2 Mar 2000 16:09:02 -0800 (PST)
Subject: printing problem
Message-ID: <20000303000902.27956.qmail@web1702.mail.yahoo.com>

hi i want to grab 3 characters and print
them out as they are entered with no space in
between. the following program waits till
all 3 characters are printed, and then prints,
which is close to what i want but not quite.

#!/usr/local/bin/python
#  import posix
import string
import curses
def reader():
    win = curses.initscr()
    s = win.getch('noecho')

# tty-example-2.py

import sys

try:
    # windows or dos
    import msvcrt
    getkey = msvcrt.getch
except ImportError:
    # assume unix
    import tty, termios, TERMIOS
    def getkey():
        file = sys.stdin.fileno()
        mode = termios.tcgetattr(file)
        try:
            tty.setraw(file, TERMIOS.TCSANOW)
            ch = sys.stdin.read(1)
        finally:
            termios.tcsetattr(file, TERMIOS.TCSANOW,
mode)
        return ch
def main():
    horky=['y','e','s']
    borky=horky
    alpha=range(3)
    for count in alpha:
      jerk=getkey()
      print jerk,;sys.stdout.softspace=0;
      borky[count]=jerk

main()

thanks!
sanjay

__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

From gmcm at hypernet.com  Sun Mar  5 08:27:48 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 05 Mar 2000 13:27:48 GMT
Subject: Calling class objects
References: <38C188AD.1C61F217@flashmail.com>
Message-ID: <38c260d4$0$4306@wodc7nh0.news.uu.net>

Jp Calderone  wrote:

>class Master:
>        def __init__(self):
>                ## some init stuff, like get call the loadClass function
>                ## a few times - this works, printing the results gives
>                ## 
>                playerInitObj = loadClass('Player')
>
>        def connectionMade(self, sock, addr):
>                ## Call the constructor with sock and addr
>                ## This is where it seems to die - CPU usage jumps
>                ## to 100% and the network code stops accepting connections
>                ## - presumably because it is blocking on this call
>                player = playerInitObj(sock, addr)
>
>        def loadClass(name):
>                mod = __import__(name)
>                return eval(name + '.' + name)

>So why doesn't the connectionMade function do what I think it should?

It's difficult to figure out what you think it should do. My guess is this:

import Player
class Master:
  def __init__(self):
    self.playerInitObj = Player.Player()
  def connectionMade(self, sock, addr):
    self.player = self.playerInitObj(sock, addr)

Your Master.__init__ can't possibly work unless you have a global function 
"loadClass" (as written above you'd get a NameError). However, even if it 
did work, it would assign the results of loadClass to a method-local var which 
goes out of scope and gets deleted on the next line. Master.connectionMade 
has the exact same problem.

Because Python does not require variables to be declared, you must explicitly 
use "self" to make a variable belong to the instance.

Since you've hardcoded "Player", I don't see the reason for the dynamic 
import. If you don't want the Player module imported until it's needed, use:

  def __init__(self):
    import Player
    self.playerInitObj = Player.Player()

This has the additional advantage that the second instance of Master won't 
incur the overhead of searching your disk for a module named "Player".

From mhammond at skippinet.com.au  Fri Mar 31 01:41:24 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 31 Mar 2000 06:41:24 GMT
Subject: Python w/MS Outlook / MS Exchange
References: 
Message-ID: 

Your best bet is to look at testExchange.py in the win32com\test
directory.

For the ultimate in performance you may be better off using
win32com.mapi, but there arent too many samples for that (and the
C++ samples you can find should port quite obviously :-)

Mark.

"Benjamin Schollnick"  wrote in
message
news:rNHvjEdhm5Pp-pn2-LE0bjbfqNid6 at d185d18f2.rochester.rr.com...
> Folks,
>
> I was wondering if anyone has found a way to look up Email
address's,
> etc, in python from a MS Outlook "global" Address book?
>
> Basically, at work we are using MS Outlook, and I want to write
a
> CGI-BIN
> that will take a user name, and list their email information.
> Unfornately, it's a global
> address book out on the outlook server....
>
> Any suggestions on the next step, etc, would be helpful &
> Appreciated.
>
> - Benjamin
>
> ====================================================
>           (Remove "NoSpam" to Email me)
> ====================================================
> Please feel free to copy any and or all of this sig.
> A little something for spam bots:
>
> root at localhost postmaster at localhost admin at localhost
> abuse at localhost postmaster at 127.0.0.1
>
> Chairman William Kennard: bkennard at fcc.gov
> Commissioner Susan Ness: sness at fcc.gov
> Commissioner Harold Furchtgott-Roth: hfurchtg at fcc.gov
> Commissioner Michael Powell: mpowell at fcc.gov
> Commissioner Gloria Tristani: gtristan at fcc.gov
> consumerline at ftc.gov
> fccinfo at fcc.gov
> ssegal at fcc.gov
>

From edcjones at erols.com  Wed Mar 15 01:02:42 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Wed, 15 Mar 2000 01:02:42 -0500
Subject: Lisp to Python converter?
Message-ID: <38CF2782.F57144E3@erols.com>

Are there any programs that convert Lisp to Python? Lisp to C or
C++?

Thanks,
  Ed Jones

From wware at world.std.com  Mon Mar 13 13:49:52 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 13 Mar 2000 18:49:52 GMT
Subject: Stackless/microthreads merge news
References:      
Message-ID: 

Konrad Hinsen (hinsen at cnrs-orleans.fr) wrote:
> Indeed. But what then are the applications for which one might
> profitably use microthreads? ...Or is it memory efficiency that you
> are aiming for?

Memory efficiency is a big part of it, and context switching should
presumably be faster (tho I haven't clocked it). What motivated my
interest in microthreading was a wish to run thousands or tens of
thousands of threads simultaneously, which started when I went to a
talk by Gerald Sussman on amorphous computing, described at
http://www.swiss.ai.mit.edu/projects/amorphous/
Sussman and his group are interested in studying computing architectures
where so many processors are networked together that you would never
bother to explicitly address a single processor, but just treat them
as a continuous computing substrate. In simulation, however, you still
need to model each of the processors.

Another application that interests me is trying to simulate
microeconomic behavior. There is some work going on in this area now,
mostly following the "artificial life" tradition, designing
economic agents who learn to transact with one another in an abstract
world (e.g. http://www.brook.edu/sugarscape/). I think it would be
more interesting to base models on empirically observed behavior of
real people and businesses.

Games are another possible class of applications. Microthreads would
allow you to code up the behaviors of game characters as independent,
simultaneous actors. These might be spaceships, or staggering zombie
brain-eaters, or Pacman monsters, or naval ships.

The reasons microthreads would be a win in these situations is that
(1) you can practically run several thousand of them simultaneously,
and (2) syntactically they are fairly painless, and therefore good for
the kind of experimental tinkering somebody might do with simulators or
games where raw performance isn't yet a critical need. The complexities
of OS threads probably mean that you wouldn't casually throw hundreds
of them at a back-of-the-envelope simulation.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com

From owen.corpening at tivoli.com  Thu Mar 23 11:37:05 2000
From: owen.corpening at tivoli.com (Owen Corpening)
Date: Thu, 23 Mar 2000 10:37:05 -0600
Subject: Tangential anti-educationist rant (was: what is easier to learn first?...)
References:   <3E53238E59BADF34.01ECA7E96E32ECCE.EFB19829F3626C71@lp.airnews.net>
Message-ID: <8bdhke$ehl$1@news.tivoli.com>

funny, I thought C was a cinch, very easy to learn and teaches good habits.

owen

Cameron Laird wrote in message
<3E53238E59BADF34.01ECA7E96E32ECCE.EFB19829F3626C71 at lp.airnews.net>...
>In article ,
>Grant Edwards  wrote:
> .
> .
> .
>>more Python.  I'd definitely go with Python.  C is for OS
>>kernels, device drivers, and embedded work.  It really oughtn't
>>be taught to beginners or people who just want to write an
>>application.  C is low-level and unsafe: the programmer's got
>>to do way too much of the housekeeping, and there are countless
>>ways to hang yourself before you've even gotten enough rope out
>>to accomplish anything. -- but at least the semantics are
>>farily simple (assuming you know some assembly language and how
>>compilers work).
> .
> .
> .
>It pretty well captures my impression of
>institutional education that US high
>schools commonly teach C (and C++!) as a
>first programming language.  I find that
>typical of the mental abuse they thought-
>lessly perpetrate.
>--
>
>Cameron Laird 
>Business:  http://www.Phaseit.net
>Personal:  http://starbase.neosoft.com/~claird/home.html

From mhammond at skippinet.com.au  Wed Mar 22 18:44:48 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 22 Mar 2000 23:44:48 GMT
Subject: COM: Passing arrays, SAFEARRAYs, and BSTRs in connection points
References: <38d8e3f2$0$2941@senator-bedfellow.mit.edu>
Message-ID: <8bblte$3h8$1@nnrp1.deja.com>

In article <38d8e3f2$0$2941 at senator-bedfellow.mit.edu>,
"Robert"  wrote:
> The short form of the question:
> pythoncom seems to convert BSTR/unicode strings to normal python
strings
> automatically. Is there any way of recovering the original raw data?
> I am sending raw chars that are converted to BSTRs through COM to
> a python program. For example I am sending 'test' (4 bytes) in a
> SAFEARRAY/vector,
> convert it to a bstr using BstrFromVector, then send it off to
python, which
> somehow converts it to a string '??' (2 bytes). Is there anyway of
> retrieving the
> original 4 bytes?

You should be using a safe array of type VT_UI1 - this is the type that
most objects (eg, MS Database objects, etc) use for binary data.

When Python recieves a variant of this type, the Python program sees
the 4 bytes returned in a "buffer" object.

If you need to pass a 4 byte buffer to an external COM object from
Python, then you use the buffer built-in - eg:

CallSometing(buffer("test"))

Will cause Python to create a VT_UI1 safearray with the 4 bytes.

Looking at your "long story", it appears you are attempting to do this -
 we only added this feature since 125, so if you are not in 128 then
upgrade, and it should work.  If you _are_ on 128, mail me and we can
sort it out.

Mark.

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

From jayfreeman at earthlink.net  Tue Mar 28 14:47:42 2000
From: jayfreeman at earthlink.net (Jay Freeman)
Date: Tue, 28 Mar 2000 13:47:42 -0600
Subject: Checking for data on a serial port
Message-ID: <38E10C5D.895467C1@earthlink.net>

Howdy,

I'm probably missing something obvious in the docs, but I'd like to have
a way to see if there is any data waiting to be read on a serial port.
I'm doing a touch screen driven app on RedHat 6.1. I've tried two
methods of reading the serial port and they both behave the same way.
I've just opened /dev/ttyS0 for read, and used readline(), and I've
redirected stdin and used raw_input(). Both methods just sit on that
statement and wait until some data comes in. What I'd like to do is
check and see if there is any data, if there is then read it, if not
then check the time against a default timeout value so I can kill some
intermediate screen that the user has walked away from and start the
session over (it's a kiosk). Neither termios or tty modules seem to be
able to do what I want (although termios is handy for flushing extra
touches out of the buffer). Any ideas?

Thanks,

Jay

From jjv5 at nih.gov  Tue Mar 21 12:01:04 2000
From: jjv5 at nih.gov (James Vincent)
Date: Tue, 21 Mar 2000 12:01:04 -0500
Subject: PIL text on SGI
Message-ID: <38D7AAD0.5E6343A8@nih.gov>

Hello,

	Does anyone have success using PIL to create text in an image
on an SGI? I am able to do everything correctly under Linux, but under
IRIX 6.5.6
I get no exceptions, but also no text in any image of any type. Lines,
rectangles,
colors all work, just not text. Any help is appreciated.

			Jim

-- 

James Vincent
jjv5 at nih.gov

From reimunds_rei at yahoo.de  Mon Mar  6 10:37:28 2000
From: reimunds_rei at yahoo.de (Rei)
Date: Mon, 06 Mar 2000 15:37:28 +0000
Subject: MySQL module
Message-ID: <38C3D0B8.1B76A9E9@yahoo.de>

hi there!
anyone to help interpret the following errormessage (invoked by 'import
MySQL')?

ImportError: /usr/lib/python1.5/lib-dynload/MySQLmodule.so: undefined
                      symbol: __rawmemchr

best regards, rei

From greg at perceval.be  Wed Mar  1 11:15:55 2000
From: greg at perceval.be (Gregoire Welraeds)
Date: Wed, 1 Mar 2000 17:15:55 +0100 (CET)
Subject: no pyc file
In-Reply-To: <89jdnr$sg8$1@netnews.upenn.edu>
Message-ID: 

In reply to the message of Samuel Nicolary sent on Mar 1 (see below) :

According to the Python tutorial by GVR, in section 6.1.2 :

Normally, you don't need to do anything to create the spam.pyc file.
Whenever spam.py is successfully compiled, an attempt is made to write the
compiled version to spam.pyc. It is not an error if this attempt fails; if
for any reason the file is not written completely, the resulting spam.pyc
file will be recognized as invalid and thus ignored later. The contents of
the spam.pyc file is platform independent, so a Python module directory
can be shared by machines of different architectures. 
[...]
When a script is run by giving its name on the command line, the bytecode
for the script is never written to a .pyc or .pyo file. Thus, the startup
time of a script may be reduced by moving most of its code to a module and
having a small bootstrap script that imports that module. 

--
Life is not fair
But the root password helps
--

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194		
Rue Tenbosch, 9			Fax: +32-2-6403154		
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

On Wed, 1 Mar 2000, Samuel Nicolary wrote:

> Date: Wed, 1 Mar 2000 10:42:04 -0500
> From: Samuel Nicolary 
> To: python-list at python.org
> Newsgroups: comp.lang.python
> Subject: no pyc file
> 
> Is there a reason why a pyc file would not be created upon execution of my
> script?
> 
> It is a simple text parcing application.
> 
> Thanks
> --
> Sam Nicolary
> Systems Admin - SOS - ISC
> University of Pennsylvania - Graduate School of Education
> 3700 Walnut Street
> Philadelphia, PA 19104
> (215) 573-5423 Fax - (215) 573-2119
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
> 

From mfletch at tpresence.com  Mon Mar 13 14:17:00 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 13 Mar 2000 14:17:00 -0500
Subject: Proposal: Weekly FAQ pointer (was Re: Newbie: Is there a faq 
	for this newsgroup?)
Message-ID: 

Writeup is pretty simple...

Subject: Weekly Frequently Asked Questions (FAQ) Pointer
The Python Frequently Asked Questions (FAQ) collection (with searching
capabilities) is available here:
	http://www.python.org/cgi-bin/faqw.py?req=home

The Python FAQ as a single static document is available here...
	http://www.python.org/doc/FAQ.html

Python Software Activity (PSA) members can add new entries to the FAQ (see
the welcome message from the PSA sent when you joined).

Still need someone with presence of mind (or a decent CRON scheduler) to
volunteer to post it every week :) . Enjoy,
Mike

-----Original Message-----
From: aahz at netcom.com [mailto:aahz at netcom.com]
Sent: Monday, March 13, 2000 2:11 PM
To: python-list at python.org
Subject: Proposal: Weekly FAQ pointer (was Re: Newbie: Is there a faq
for this newsgroup?)

In article <38cb9f9a$0$1338 at motown.iinet.net.au>,
Aesop  wrote:
>
>        Been programming in python for, well about 2 days now. Have a whack
>of dumb questions that are probably guaranteed to see me flamed to death.
Is
>there a FAQ anywhere? oe better still anyone got any good links to tutorial
>type sites for people who have not touched a computer language in about six
>or so years?

We've been getting a lot of questions like this lately, despite the
existence of python.org.  I think that a >>very short<< weekly FAQ with
pointers to the five or six most common URLs would be pretty handy.  I'm
willing to write it up if someone else takes the responsibility for
posting it; I'm just not sure whether I should talk to Fred, Barry, or
someone else.  Suggestions?

(Yes, I'm more than aware of the weekly Python-URL!, but that's not
labelled "FAQ" and it's a lot longer than someone who wants a quick
answer to "Where's the FAQ?" will read.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"Along the lines of getting a massage, taking a hot bubble bath, and
swimming in a pool of warm, melted chocolate, can you give me some
innovative ways that you pamper yourself?"
"Spend three hours flaming stupid people on the Net."  --Aahz
-- 
http://www.python.org/mailman/listinfo/python-list

From herzog at online.de  Sat Mar 18 15:14:13 2000
From: herzog at online.de (Bernhard Herzog)
Date: 18 Mar 2000 21:14:13 +0100
Subject: need fast parser for comma/space delimited numbers
References: <1258725464-20166134@hypernet.com> 
Message-ID: 

Les Schaffer  writes:

> >>>>> ">" == Gordon McMillan  writes:
> 
>     >> Second, benchmark against "int" and "float".
> 
> okay. i noticed in the Scientific Python modules K. Hinsen uses

I haven't seen those, but...

> something like this
> 
> numb = exec( str )

..I guess you mean eval()

> with str being things like ' 4.235 ', etc. i wonder which is faster?
> (thinking out loud)

int/float should be faster because eval creates an intermediate code
object which is then executed by the bytecode interpreter. eval has the
advantage that it automatically distinguishes between ints and floats,
though, and other literals and even expressions for that matter.

It is, of yourse a gaping security hole unless you use rexec.

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

From calishar at *NOSPAM*home.com  Wed Mar  1 19:26:37 2000
From: calishar at *NOSPAM*home.com (Calishar)
Date: Thu, 02 Mar 2000 00:26:37 GMT
Subject: Email/Newsgroup client?
References: 
Message-ID: 

On Wed, 1 Mar 2000 17:17:53 -0500, "Joel Lucsy"
 wrote:

>Are there any that work with win32? I found a couple at the Vault, but
>they're based on unix. I'm looking to replace Outlook Express, and no, I
>don't want anything like Eudora,Gravity,Newsxpress,etc. I'd like something
>mostly/completely python based.

I remember a program called Mahogany Mail, it was a Win32 (based or
compatible) mail client that had python scripting. A quick web search
shows the following URL

http://www.phy.hw.ac.uk/~karsten/M/status.html

Calishar

From python-list at teleo.net  Wed Mar  8 11:19:13 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Wed, 8 Mar 2000 08:19:13 -0800
Subject: Is it Python or is it C ?
In-Reply-To: <004401bf8910$a42da4c0$01ffffc0@worldnet.att.net>
References: <0%hx4.3000$N3.36008@typhoon.southeast.rr.com>  <004401bf8910$a42da4c0$01ffffc0@worldnet.att.net>
Message-ID: <00030808323600.01321@quadra.teleo.net>

[Emile van Sebille, on Wed, 08 Mar 2000]

:: In which case, doesn't that transform the question to 'which
:: preferred pydioms yield the best performing bytecode?'? <*1>

:: *1 

If both the main clause of the sentence and the quotation are
interrogative, only one question mark is used:

"""
In which case, doesn't that transform the question to "which preferred
pydioms yield the best performing bytecode."? 
"""
Note that single quotation marks are incorrect. Reserve singles for
quotations nested in other quotations. (Further quotations within the
nested quotation would again use double quotes.)

From dirck at pacbell.net  Sun Mar 26 01:20:20 2000
From: dirck at pacbell.net (Dirck Blaskey)
Date: Sat, 25 Mar 2000 22:20:20 -0800
Subject: colostomy
References:  
Message-ID: 

Cliff Crawford  wrote in message
news:slrn8dqo5c.4sd.cjc26 at ith1-491.twcny.rr.com...
> * Dirck Blaskey menulis:
>>...
>> Please, PLEASE make colon's optional when not needed.
>> ...
> I actually like the colons.  They seem to make code a little easier
> to read.

I don't dispute their aesthetic value,
but for me, in practice, the requirement gets in the way.
I'm not exactly a new Python user either, I've written
a fair amount of Python code.  I just seem to have trouble
sometimes teaching my hindbrain to remember things
that my frontbrain forgets.
(I HATE learning a new text editor.)

I'm willing to bet I'm not the only coder who forgets to put
them in on a regular basis.  And since they're not really necessary,
they should at least be optional.

(The best approach for handling User Input errors
 isn't to provide a good error message,
 but to eliminate the possibility of error.)

>
>> I waste a lot of time going back and filling in the bloody things.
>
> After I learned Python, I had the same problem at first with C and
> semicolons.  I'm not sure why I hate semicolons as EOL markers, but
> like colons as block markers; I guess one is too much redundancy
> while the other is just the right amount. 

After many (10+) years of heavy (Heavy) C coding, I still would
occasionaly forget to type in semi-colons -
and now when typing Python I occasionaly add them in without thinking.
Fortunately, Python is cool enough not to complain.

>> The only other suggestion I have, and this isn't nearly as important:
>> it would be nice to have block comments like c's /* */ ...
>
> You can enclose the comment in """...""".  You can have strings
> anywhere where a statement is expected.

I hadn't known """ was allowed as a block comment -
I had thought doc strings were the only block option.

thanks for the tip,

d
===========================
Dirck Blaskey
Danbala Software
dirck at danbala.com
http://www.danbala.com

From aahz at netcom.com  Wed Mar  8 18:17:11 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Mar 2000 23:17:11 GMT
Subject: FTP hangs with urllib.urlretrieve
References: 
Message-ID: <8a6n1n$vpk$1@nntp6.atl.mindspring.net>

In article ,
Michal Wallace (sabren)  wrote:
>On Tue, 7 Mar 2000, Oleg Broytmann wrote:
>> 
>>    I am developing/running URL checker (sorry, yet another one :). And
>> found that I have consistent hangings on some FTP URLs.
>> 
>>    Here is test program I am using to retest my results - just URLopener
>> and urlretrieve, nothing magical; python is stock 1.5.2, hangs on all
>> platforms I am using - Pentium Linux, Sparc Solaris, Pentium FreeBSD:
>
>This is not limited to ftp urls... urllib just hangs sometimes.

I've submitted a patch that gives httplib.HTTP() a default timeout that
fixes these kinds of problems, but Guido hasn't accepted it yet.  If
someone wants to help me provide the info that Guido wants, I'd really
appreciate it.  Once that's done, it should be straightforward to patch
ftplib to have the same kind of timeout.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"Along the lines of getting a massage, taking a hot bubble bath, and
swimming in a pool of warm, melted chocolate, can you give me some
innovative ways that you pamper yourself?"
"Spend three hours flaming stupid people on the Net."  --Aahz

From darogers at xnet.com  Fri Mar 24 17:56:58 2000
From: darogers at xnet.com (David A. Rogers)
Date: 24 Mar 2000 22:56:58 GMT
Subject: python / COM question
Message-ID: 

I'm learning COM and want to use python to test the objects I'm creating.  I
immediately ran into a problem.

I have an interface with a function like so:

	STDMETHOD(GetDate)(int* year, int* month, int* day);

How do I call this from python when all python functions are pass by value?

Thanks for any help,
dar

From corg at copernic.com  Mon Mar  6 08:54:35 2000
From: corg at copernic.com (Gaetan Corneau)
Date: Mon, 6 Mar 2000 08:54:35 -0500 
Subject: Self Nanny
Message-ID: 

> Yes, I went so far as to define a macro so I could use self.* in C++. 

Ah, c'mon guys: having to declare self may or may not be fun, but the fact
is that it is a gotcha for newbies, so I think self should be passed
implicitly to methods.

Remember: it's stuff like that that turns new users
off.______________________________________________________
   Gaetan Corneau
   Software Developer 
   Copernic Technologies inc. 
   http://www.copernic.com
   E-mail: corg at copernic.com                         
______________________________________________________
"Tu l'as trop ?cras?, C?sar, ce Port Salut"

-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/B/MU d- s+:++ a C++ 
UL+ P--- W+ N- K- W++ t-- 
!5 X- R+ tv-- b++ DI++ 
G e++ h---- r+++ y++++
------END GEEK CODE BLOCK------

From gregm at iname.com  Thu Mar  9 06:17:19 2000
From: gregm at iname.com (Greg McFarlane)
Date: Thu, 9 Mar 2000 22:17:19 +1100
Subject: TK Notebook with Tabs on the bottom
In-Reply-To: <38C4217C.B5CADA8F@exceptionalminds.com>; from Timothy Grant on 06 Mar 2000 at 01:22:04PM
References: <38C4217C.B5CADA8F@exceptionalminds.com>
Message-ID: <20000309221719.58051@nms.otc.telstra.com.au>

Currently Pmw.NoteBook only accepts 'n' and None as values for the
'tabpos' option.  The others ('s', 'e' and 'w') are on the Pmw todo
list.  If anyone would like to contribute ...

On 6 Mar, Timothy Grant wrote:
> Hi all,
> 
> Does anyone have a Tkinter notebook widget with tabs on the bottom? I've
> used the Pmw.Notebook widget successfully, but I haven't been able to
> figure out if those tabs can be moved to the bottom or not.
> 
> I have some rememberance of seeing a widget that would allow you to
> place the tabs wherever you want, but I can't find it, so I'm beginning
> to wonder if I dreamt it!
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Chief Technology Officer              tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.                   (503) 246-3630
> >>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com

From python-list at teleo.net  Mon Mar  6 20:00:56 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Mon, 6 Mar 2000 17:00:56 -0800
Subject: 11-digit integer literal on a 32-bit CPU
In-Reply-To: <0003061647130K.03962@quadra.teleo.net>
References: <0003061647130K.03962@quadra.teleo.net>
Message-ID: <0003061702560L.03962@quadra.teleo.net>

Oh. I should add that the ID needs to be included in a delimited text
data file, so 10000000000L doesn't work.

[Patrick Phalen, on Mon, 06 Mar 2000]
:: I need to assign an 11-digit Transaction ID, according to specs I'm
:: given by a client.
:: 
:: Begin with 10000000000
:: 
:: Increment by 1 for each new transaction -- 10000000001, etc.
:: 
:: (Store as string in a file between transactions)
:: 
:: As an integer literal, this overflows on an Intel CPU.
:: 
:: What's a nice trick for dealing with this?
:: 
:: Total transactions aren't likely to exceed 999,999
:: 
:: -- 
:: http://www.python.org/mailman/listinfo/python-list

From jcw at equi4.com  Thu Mar  2 14:31:29 2000
From: jcw at equi4.com (Jean-Claude Wippler)
Date: Thu, 02 Mar 2000 20:31:29 +0100
Subject: MetaKit for python (Mk4py) on Macintosh ... no way :(
References: <1260348438-26442187@hypernet.com> <87vh36ob49.fsf@den.home.net> <87ln42dzbw.fsf@den.home.net>
Message-ID: <38BEC18E.D2A01466@equi4.com>

Frank,

> Of course, I meant to say 'I have NOT figured out how to do that
> yet with Mk4py'.

Perhaps the following helps:

import Mk4py
db=Mk4py.Storage()
vw=db.getas("people[name:S,age:I,weight:F]")
vw.append(name='Tarzan',age=34,weight=45.6)
vw.append(name='Jane',age=23,weight=34.5)

props = vw.structure()
for r in vw:
	for p in props:
		print getattr(r,p.name),
	print

Please consider posting MetaKit-specific questions on the MK-Scripting
mailing list, see http://sourceforge.net/mail/?group_id=736

-- Jean-Claude

From aahz at netcom.com  Thu Mar  9 10:28:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Mar 2000 15:28:34 GMT
Subject: Duplicates in lists
References: <38C748C7.35162A7B@teleatlas.com>
Message-ID: <8a8fv2$327$1@nntp6.atl.mindspring.net>

In article <38C748C7.35162A7B at teleatlas.com>,
Michael Husmann   wrote:
>
>is there someone who has an efficient function that finds all
>duplicates in a list?  I used a hash which works quite reasonable but
>maybe there is a better way.

If you're already using a dictionary, you'll only be able to do
micro-optimizations; your basic algorithm is the best available in
Python.  If you want us to show you how to optimize Python code, post
what you've got here.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"Along the lines of getting a massage, taking a hot bubble bath, and
swimming in a pool of warm, melted chocolate, can you give me some
innovative ways that you pamper yourself?"
"Spend three hours flaming stupid people on the Net."  --Aahz

From embed at geocities.com  Wed Mar  8 10:47:32 2000
From: embed at geocities.com (Warren Postma)
Date: Wed, 8 Mar 2000 10:47:32 -0500
Subject: What If Python Replaced Elisp?
References: 
Message-ID: 

Why rewrite EMACS? Why not start from scratch with Python and see what
develops?

I never learned EMACS because I hated LISP. I took a LISP course in 2nd year
CompSci.  Blech!

I would love an Editor written in Python, scriptable in Python.  IDLE is
great except for when I need to use a command line machine. In fact, I find
a text screen less distracting when I'm writing reams of code.  I think a
Python+Curses based editor would be great. Does such a beastie exist?

Warren

From landauer at apple.com  Thu Mar  2 17:35:32 2000
From: landauer at apple.com (Doug Landauer)
Date: Thu, 02 Mar 2000 14:35:32 -0800
Subject: Python bindings for Apple's Foundation and AppKit ???
References: <87tkoi$hql$1@nntp1.atl.mindspring.net> <8ED6BCFDChdjsdhdfh75738@news.sshe1.sk.wave.home.com>  <38BDDFFD.F8B83BCE@mindspring.com>
Message-ID: 

echuck at mindspring asks questions unrelated to the posting he's
following up on...

Lemme shake up this 8-ball...

> So when is Apple going to provide Foundation and AppKit bindings for Python

Reply hazy, try again later.

> and support the language in Project Builder? :-)

Reply hazy, try again later.

Dang, it must be broken.

> (Or are you one of those Carbon people...)

Well, when I step on the old spectrometer, it says that there's Carbon
somewhere in there, as well as Nitrogen, Oxygen, Hydrogen, and Chocolate,
but no, I don't work on Carbon.  Or AppKit or Foundation.  Just gcc.  I
barely have time to just compile up a bare python (not even shared-libraries
or a readline library) for my own use, let alone write bindings for something
the size of AppKit.

I haven't even managed to convince them to ship Python on the
system yet :-( -- I just use it to make little tools to help me debug
stuff.
-- 
    Doug Landauer        landauer at apple.com (work)
                         landauer at scruznet.com (not-work)

From scarblac-spamtrap at pino.selwerd.nl  Wed Mar 15 08:37:58 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 15 Mar 2000 13:37:58 GMT
Subject: Should Python documentation for __class__ be improved?
References: 
Message-ID: 

Tom Funk wrote in comp.lang.python:
> While tracking down the UserList problems I've previously discussed I ran 
> across this line of code in UserList.__getslice__():
> 
>     userlist = self.__class__()
> 
> I'd never seen this idiom before and I wasn't sure exactly what was going 
> on.  So I started tearing up the Python docs looking for an explanation.  
> 
> I found a number of references to __class__, all of which made it look as 
> though __class__ was an attribute, rather than a method (though it's not 
> technically a method either).  Nowhere did I find anything that said  
> that __class__ was callable.  Also, there are *no* examples of calling 
> __class__() anywhere in the docs.

Well, it's not really 'callable'. __class__ *is* just an attribute. It's
just a reference to a class. And you can make an instance of a class.

class A:
   pass

a = A()

Now A == a.__class__.

So, obviously,
x = A()
now means the same thing as
x = a.__class__()

It's never explained that you can make an instance of __class__, since you
can make an instance of any class, and this one isn't any different. It's
slightly confusing that making a class instance looks like a function call.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "I sense... golden arches":
   In Original D&D, some swords had the ability to 'detect meal and what
   kind.'

From tismer at tismer.com  Thu Mar  2 15:51:56 2000
From: tismer at tismer.com (Christian Tismer)
Date: Thu, 02 Mar 2000 21:51:56 +0100
Subject: stackless/microthread merge
References:  <38BD197A.4F9E93C1@tismer.com> <9sxv4.28865$Jz3.173654@nnrp1.uunet.ca>
Message-ID: <38BED46C.905A5B9@tismer.com>

Warren Postma wrote:
[skipping nice words - thanks :-)]

> I can see stackless+microthread going on to challenge regular python by very
> careful optimizations at it's core. I'm seriously going to consider using
> this for my embedded control [SCADA, Radio Telemetry] uses, because I think
> thread latency is a pretty key part of any complex control system.

We're doing our best. For my part, I'm writing the fastest possible
but still flexible scheduler kernel that I can consider. Due to
Will's input, it has builtin thread locking already, to provide
atomic operations. But the stuff beyond the absolute basics
(which is about two or three extra support functions), everything
is done by Python, and people can implement their own thread
implementation on top of it.
Will will be the first one, of course. :-)

> I can't wait to see what this looks like when it's finished! :-) Very very
> cool!

I'm finishing the last function, so it will probably be less than
a week, unless I see a hairy bug.

cheers - chris

-- 
Christian Tismer             :^)   
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
     we're tired of banana software - shipped green, ripens at home

From helge.hess at mdlink.de  Thu Mar 30 10:30:15 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Thu, 30 Mar 2000 17:30:15 +0200
Subject: multiple interpreter sessions
Message-ID: <38E37307.18E5CFD0@mdlink.de>

Hi,

as far as I can see Python allows only one interpreter instance to be
embedded, right ? Is that already recognized as a problem or don't
people care ?

I would like to have separate Python runtimes in a single server, so
that they don't affect each other (eg for security and state reasons).
Is that going to be addressed in Python 1.6 or 2.0 ? Maybe someone
already has modified Python to support such stuff ?

Thanks
  Helge Hess

From spooky at dreamfish.org.uk  Fri Mar 31 05:05:09 2000
From: spooky at dreamfish.org.uk (Robert Longstaff)
Date: Fri, 31 Mar 2000 11:05:09 +0100
Subject: CGI support question
In-Reply-To: 
References: 
Message-ID: <00033111134600.00753@fishology>

> Can anyone tell me of an ISP who supports python in the UK? Or even better
> would be a site that would give me free space who supports Python.
>

You could try UK Shells (http://www.ukshells.co.uk). They provide UNIX shell
accounts with webspace and support CGI/SSI. Last time I looked they had Python
v1.5.1 installed so you could use that for your CGI.

An account is about 90 pounds per year (including VAT) for which you get 50MB.

Aside: I've got an account on UK Shells and am currently developing an app
that I'll test (and maybe even run) there, so I might ask them to upgrade to
v1.5.2 ;)

Robert.

From helpdesk at perceval.be  Thu Mar 16 05:29:30 2000
From: helpdesk at perceval.be (Gregoire Welraeds)
Date: Thu, 16 Mar 2000 11:29:30 +0100 (CET)
Subject: CVS acces via Python
Message-ID: 

Hi,

Is there any module that I can use to access a cvs with Python ?

--
Life is not fair
But the root password helps
--

Gregoire Welraeds
greg at perceval.be
Perceval Development team
-------------------------------------------------------------------------------
Perceval Technologies sa/nv	Tel: +32-2-6409194		
Rue Tenbosch, 9			Fax: +32-2-6403154		
B-1000 Brussels			general information:   info at perceval.net
BELGIUM				technical information: helpdesk at perceval.net
URL: http://www.perceval.be/
-------------------------------------------------------------------------------

From r.barendregt at village.uunet.be  Mon Mar 20 17:58:57 2000
From: r.barendregt at village.uunet.be (Robin Barendregt)
Date: Mon, 20 Mar 2000 22:58:57 GMT
Subject: Question about instance creation
References: <1258533308-31724431@hypernet.com>
Message-ID: <38d6aa4b.49104330@news.pandora.be>

"Gordon McMillan"  wrote:

>So, I haven't the foggiest idea *what* you're trying to do. 
>However, this might help clarify things.

My apologies. I'll try to be more precise (the example didn't
illustrate well what I wanted to do):

I have a derived class that I want to make instances of. I want to be
able to change the base class (hence the alias) and I can't just
create an instance of the base class and forget all about the derived
class, because the derived class inherits from another class as well.
So the derived class instance acts like a base class instance with the
extra functionality as defined in the other class. The def looks like
this:
class DerivedClass(BaseClassAlias, SomeOtherClass):

I do not even know if I'm searching in the right direction to get to
above mentioned result. I guess I could change the base class alias
and then directly instantiate the derived class. Any other ideas?

Robin.

From gmcm at hypernet.com  Sat Mar 11 16:58:37 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 11 Mar 2000 16:58:37 -0500
Subject: not at Python Bugs -- __builtins__ madness
In-Reply-To: 
Message-ID: <1259310576-5365836@hypernet.com>

> Could some please explain how Python makes this mistake?
> 
>   jeffp at friday [4:15pm] bin #324> python
>   Python 1.5.2 (#1, Apr 20 1999, 08:57:03)  [GCC 2.7.2.3] on sunos5
>   Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>   >>> __builtins__
>   
> 
> It got the module name wrong.  Unless Python treats __builtins__ THAT
> specially, that's a major goof.  And the 1.5.2 docs say it's
> __builtin__.  This boggles my poor mind.  I'm checking Python Bugs...

No mistake. __builtin__ is a module. __builtins__ is a name 
which normally refers to __builtin__, but may not (eg, in 
restricted execution).

- Gordon

From breiter at usf.Uni-Osnabrueck.DE  Tue Mar  7 09:35:19 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 7 Mar 2000 14:35:19 GMT
Subject: Compile without execute for syntax check?
Message-ID: <8a3437$er1$1@newsserver.rrzn.uni-hannover.de>

The python interpreter in the non inspecting mode(default) finds
some python syntax errors directly before running the code.

Call me old fashioned, but the documentation is not clear about how
to use this build-in syntax check on scripts without executing them.
Nothing in the FAQ ether.

Yes, I should be able to do an
	import py_compile
	py_compile.compile("myscript.py")

which can degenerate into
	python -c "import py_compile; py_compile.compile('myscript.py')"
and might create python bytecode files.
And pythonwin of course has this as a nice button upfront.

So why nothing in the python manpage (where I looked first) in the
FAQ (second) look or in the library documentation?

The answer seems obvious: Just make everything a module then you can
compile it and run the test function in the same moment within
actually doing harm. This is certainly a healthy attitude, but
sometimes I do not want to run a testfunction, I just want the
syntax checked. :)

Just-proposing-another-documentation-improvement-to-learn-the-best-way-ly,
Yours
	Bernhard

ps.: Do I gain speed if I execute script, which I compiled with
py_compile, like "python myscript.pyc"  instead of "python myscript.py"?
-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)

From wware at world.std.com  Mon Mar 13 07:03:53 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 13 Mar 2000 12:03:53 GMT
Subject: Stackless/microthreads merge news
References:  
Message-ID: 

Konrad Hinsen (hinsen at cnrs-orleans.fr) wrote:
> Looks great! One question: are there any provisions to use
> microthreads in C extension modules? If yes, is the Python interpreter
> (the modified one) thread-safe for microthreads, or is there still
> something like the global interpreter lock? Getting rid of that would
> be reason enought to switch to microthreads for me.

Chris wrote all the C code, and he can give better answers. I'm inclined
to say you should be able to do everything in C that the manual shows
being done in Python but of course the syntax would be much uglier. Also,
I'm not 100% certain of that, because running Python code causes a
counter (called uthread_ticker) to count down, and when it hits zero,
a context switch occurs, and to you that appears to be an invisible
background task. So you'd need to explicitly step down that counter and
allow the context switching code to run, and I'm not sure how you'd do
that unless you allowed yourself to return out to the Python world. So
the simplest thing would be to arrange your computation as a bunch of
limited-duration steps, each of which could be a call from Python. That's
probably not what you're looking for, maybe Chris will have a more
satisfactory idea.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com

From nobody at nowhere.nohow  Sun Mar 26 22:51:58 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 27 Mar 2000 03:51:58 GMT
Subject: Newbie - help in opening and reading a file from >>>
References:  
Message-ID: 

On Sun, 26 Mar 2000 21:17:37 GMT, Jerome Chan  wrote:
>
>> I've just installed 1.5 on my Windows 98Se system
>> I'm trying to execute the following-
>> >>>f=open('C:\Autoexec.bat')
>> and I get a no such directory or file error
>> Is there some parameter or path that I must set?
>>  Would apprecaite any clues...
>
>You must escape the \?

Back in the bad old days, I always used forwards slashes when
programming under DOS for just this reason (it also made them
more portable when moving code between DOS, VMS, and Unix).
Does win32 still accept forwards slashes in pathnames the way
DOS used to?

-- 
Grant Edwards                   grante             Yow!  My life is a patio
                                  at               of fun!
                               visi.com            

From quinn at triskaideka.ugcs.caltech.edu  Mon Mar 13 16:05:58 2000
From: quinn at triskaideka.ugcs.caltech.edu (Quinn Dunkan)
Date: 13 Mar 2000 21:05:58 GMT
Subject: Extention to the integer class.
References: 
Message-ID: 

On Mon, 13 Mar 2000 18:25:05 +0100 (CET), Gregoire Welraeds 
wrote:
>In reply to the message of Anthony J Wilkinson sent on Mar 14 (see below) :
>
>> Alternatively you could use a __call__ method:
>>         def __call__(self):
>>             return self.__val
>
>This works great. It also avoid to implement __cmp__() since i can do:
>while i() < 2:
>	...
>
>Thanks too :)

Also note that you can use __coerce__:
class Int:
    def __init__(self, val):
        self.__val = val
    def __int__(self):
        return self.__val
    def inc(self):
        return Int(self + 1)
    def dec(self):
        return Int(self - 1)
    def __repr__(self):
        return repr(self.__val)
    def __coerce__(self, other):
        return (int(self), other)

Your original one mutated itself, but hopefully you didn't really mean that :)
And this one should probably have __add__ etc.

Of course, lists and tuples still won't attempt to coerce their index to an
int, but then you could do:
class CoerceList(UserList.UserList):
    def __getitem__(self, i):
        return self.data[coerce(i, 0)[0]]
    ... etc for setitem, has_key ...

But this seems like a lot of work just to add two methods (and it's gonna be
slow if you use it a lot).  Everyone knows what i = i + 1 means, and it's not
that hard to type.  On the other hand, IMHO it's a problem with python that
you can't subclass built-in types.

But __coerce__ is tons of fun.  If you ever find yourself bored, go through
all your code and make it use __coerce__ extensively.  After that, you'll
never have a dull moment again.

From dan at cgsoftware.com  Wed Mar  1 15:40:06 2000
From: dan at cgsoftware.com (Daniel Berlin)
Date: Wed, 1 Mar 2000 12:40:06 -0800 (PST)
Subject: will python 3000 break my code?
In-Reply-To: <38BD6C19.CB058A2@roguewave.com>
Message-ID: 

> 
> No, it's a design change by a standards committe, that has evaluated all the
> options, giving individual vendors the chance to veto any changes that would be
> too radical for their customers (I know, I've been there).  It's an enormous leap
> from that to deciding that list.append(1,2,3) might be so confusing to some
> newcomers that people who know what they were doing have to change their code
> (ask Frederik if he doesn't have more interesting things to do than fixing
> multiargument appends ;-)
Actually, it's not a design change in about 95% of the incompatible
changes in g++. It's just the implementation happened to be allowing
illegal (by the language standard) things because nobody had gotten around
to saying "that's illegal". It's not a world away from deciding that
something that was never legal in the first place, and was accepted, is
now 
not accepted.

> > Should people stop fixing bugs in the python implementation cause some major
> > customer has programmers who wrote buggy code?
> > Nope.
> 
> Was this a bug?  Not really.  Was it a source incompatible design change? You
> bet.

Not a bug?
Can you point out where it says  in the documentation, that this is
allowed?
If you can point out where it says this was supposed to be allowed, i'll
happily agree that it's a source incompatible design change, and thus, is
a bad thing (TM).
 > 
> >
> > God damn it. It's not a feature. It's a bug. A bug. It should have been an
> > error nit he first place. It's only by the grace of Guido and a slip of the
> > eye that it wasn't.
> 
> And nobody would have complained if it _was_ an error in the first place.
I'll give you that.
But sometimes things slip through.

> 
> >    | the-days-of-legacy-code-are-upon-us'ly y'rs
> > No they aren't.
> 
> If you're going to have any luck getting a company (vs. individual dedicated
> programmers) to use your tools, is if the tools are stable enough that a company
> doesn't have to spend money every time you do a new release.  If you do think a
> change is neccessary, and they often are although I'm not sure this particular
> one rises to that level, you should at least warn customers by upping a major rev
> number (or waiting until it's time to do so).
> 
I have companies that use my tools.
They are stable.
When people report bugs,I fix them.
I don't retroactively decide that things that were features are now bugs.
Which is what you seem to be trying to say is happening here.
Look. I'll hapilly agree that if that were the case, it'd be wrong to just
change it without warning.
That would be legacy code.
If it's not the case, then the cod eis not legacy code, it's buggy code.
Regardless of how long it worked for, it's still buggy.

> The-price-of-purity-is-obscurity'ly y'rs
> -- bjorn

> 
> ps: since Guido hath spoken, there really is no reason to discuss this further,
> so this'll have to be my last words on the issue -- I know we're all heartbroken
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

From tuttledon at hotmail.com  Wed Mar 22 18:03:07 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Wed, 22 Mar 2000 23:03:07 GMT
Subject: Does Guido Have A Will? (The legal kind )
References:  <3110C5495705440B.A7540016610C3F64.6B294A66CE5E80A1@lp.airnews.net>  
Message-ID: 

Eh...... I own his 1st edition of The New Hackers Dictionary.  I
bought it back in '92 but had forgotten his name.

Don

From wware at world.std.com  Thu Mar 16 10:00:23 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 16 Mar 2000 15:00:23 GMT
Subject: Chat system
References: 
Message-ID: 

Chris Armstrong (punck at SpamIsBadPenguinPowered.com) wrote:
> Another thing about the chat system: I need for it to be able to constantly
> stream in new content while still being able to type in your message and
> send it whenever you want. I know some programs like IRC clients do
> this without having two processes (one for the streaming content and
> one for the input). Oh, and one more thing, this'll all be using curses,
> if it matters.

You might be able to use microthreads, which should be available within
a week or two. The URL is
http://world.std.com/~wware/uthread.html
I'm not familiar enough with curses to know if there'd be any
incompatibility. I'd be interested in how things turn out; a few weeks
ago I emailed an economics professor, the author an educational
microeconomics simulator currently running in Vax Basic, who would like
to make his simulation web-based and run interactive sessions across the
country. The URL for his simulator is
http://ecedweb.unomaha.edu/ecosim.htm
I myself haven't done anything like an interactive web-based game and it
would be silly to try to reinvent the whole thing myself. Is there a URL
where I can read more about your project? Thanks.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com

From phd at phd.russ.ru  Tue Mar 14 11:00:33 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 14 Mar 2000 16:00:33 +0000 (GMT)
Subject: Simple string formatting question
In-Reply-To: 
Message-ID: 

On Tue, 14 Mar 2000, Matthew Hirsch wrote:
> print '%5.3f %5.3f %5.3f 1i'% 
> (temp_list[0],temp_list[1],temp_list[2],temp_list.index(min(temp_list)) 
> + 1)
> 
> but it won't fit on one line.  It keeps thinking that I have two 
> separate incomplete statements.  What is the correct way to get around 
> this problem?  Thanks for your help.

print '%5.3f %5.3f %5.3f 1i'%  \
(temp_list[0],temp_list[1],temp_list[2],temp_list.index(min(temp_list)) + 1)

   (backslash is escape character; escaping end-of-line means continue)

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

From robinhat at rhk.dk  Sat Mar 11 10:08:44 2000
From: robinhat at rhk.dk (Lars Hoeyrup Jensen)
Date: Sat, 11 Mar 2000 16:08:44 +0100
Subject: How to handle timed out connections
Message-ID: <38CA617C.37948FB9@rhk.dk>

I have a program that tries to connect to a certain web page with
urlopen(). Sometimes the page doesn't respond: if I try to open it in
Netscape, after a couple of minutes I get a message like this:

"ERROR

The requested URL could not be retrieved

.....

(110) Connection timed out

...

Generated by .... (Squid/2.2.STABLE4)"

How do I handle this in Python? I would like to generate a message
like "The page is currently down" or something if the page doesn't
respond, but I don't know how to "detect" this timeout in Python.

Thanks in advance,
Lars

From boud at rempt.xs4all.nl  Thu Mar 30 10:01:05 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 30 Mar 2000 15:01:05 GMT
Subject: Rational-Rose and Python ...
References: 
Message-ID: <8bvq7h$gvn$1@news1.xs4all.nl>

Kiss, Arpad  wrote:

> http://boa-constructor.sourceforge.net/

Dash it! I was looking for boaconstructor, without the dashed dash...

-- 

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

From hathawa2 at marshall.edu  Tue Mar  7 13:09:03 2000
From: hathawa2 at marshall.edu (Mark Hathaway)
Date: Tue, 7 Mar 2000 13:09:03 -0500
Subject: Self Nanny
References: 

Message-ID: <38C5709D.59E8@marshall.edu>

>> Gaetan Corneau  writes:
>>
>> OK, I didn't make myself clear :)

> Lloyd Zusman wrote:
>
> Yes you have, at least for me.

me too

>> GC:
>>
>> 1) *I* don't mind having to specify "self" in method definitions
>>
>> 2) *I* don't mind having to write "self.Something"
>>
>> 3) BUT I think that it would be *nice* not to have to specify "self" in
>> method definitions, and it would certainly help A LOT of newbies
>>
>> 4) if possible at all, not having to write "self.Something" would be cool,
>> but not as important as #3
>>
>> And don't tell me #3 can't (technically speaking) be done :)

Personally, I'd rather not ever have to write 'self' or 'self.anything'.
It contributes nothing to the code and doesn't describe what I'm trying
to write. It's extra verbosity without any payoff.

Was there a problem with doing it the other way around...

    class aClass:
        def __init__(self,number):
            self.num = number
            n = self.num

    class aClass:
        def __init__(number):
            local.n = number

where the roles are reversed. All object attributes are
referred to by their names and any class (global) attributes
are specified by "local.". This would decrease tremendously
the need to write 'self.' and would still easily allow the
differentiation of class-wide attributes from object attributes.

Which do you write more of? Object attribute names.
Why make that the more verbose version?

> LZ:
>
> Of course #3 can be done (technically speaking).  And in return for it
> getting done, would you volunteer to donate your personal time to
> change all the existing lines of Python code in the world that this
> change would break?

This is both true and annoying. In our search for better ways to do
things we keep running into this barrier that languages are written,
tried and much time & effort put into that trial. Then when someone
says, "Couldn't we do it better this way?", the answer is, "But, we've
put so much work into this..."

Somehow there has to be a willingness to put the effort in and then,
if it's not as perfect as we'd like, throw it away and start with
another
piece of whole cloth. Well, perhaps we could write a translator to
convert
the old to the new. That wouldn't take quite as much man-power, would
it?

> Or barring that, would you be willing to pay for the time of all the
> others who would have to do this work?

Pay? Who paid the first time? Didn't most of the Python code come about
because people liked the language and just wrote stuff to be distributed
via the internet for free? O.K. here's a penny. Ya satisfied?

I know you're saying people's time & effort is considerable. All I'm
saying
is there should be ways to move to new better ways of doing things
without
having to face such stiff resistance.

> This is not an issue about "caring" about C++ programmers.  I am a C++
> programmer, and literally hundreds of my colleagues over the past
> decade or so have been C++ programmers.  Many of these people have
> tried Python.  And *none* of them have ever even made a complaint
> about the explicit "self" of Python.  You are the first C++ programmer
> I have encountered who has ever raised this as an issue.  That's one
> out of hundreds of people.

Perhaps they are simply more accepting of what is, and don't challenge
it.
That doesn't mean they haven't thought about it. I'm not a C++
programmer,
so I don't know either.

> And as I have mentioned in another post, I challenge you to come up
> any signficant number of people who would be glad to use Python if
> only it didn't have an explicit "self".

It's not just about that issue. It's about an effort to move toward
something better. You're not against that altogether, are you?

> Your arguments have always been clear.  And they are still rather
> unconvincing.
>  Lloyd Zusman
>  ljz at asfast.com

A closed mind can't even be convinced by the truth or a stick up side
the head.

P.S. I like Python. I'm still learning it (a real newbie). I even accept
     the indentation rule.  :-)

Mark Hathaway
e-mail: hathawa2 at marshall.edu

From kojo at tamu.edu  Mon Mar 27 16:27:16 2000
From: kojo at tamu.edu (Kojo Idrissa)
Date: 27 Mar 2000 21:27:16 GMT
Subject: JPython vs. Javascript?
References: <38DF007A.64AFBE6E@tamu.edu> 
Message-ID: <8bojnk$3g8$1@news.tamu.edu>

Thanks for the response.  I'm aware that Java !=Javascript, but I guess my 
general question was more of a "Java Applet vs. Javascript" question, not a 
"JPython vs Javascript" question.  I thought that might be the case.  It's just 
that since I don't really know Java, the way I was going to make applets was 
through JPython.  From your response, it looks like the functionality of the 
two is very different.  You're right, I basically want to add some functions to 
static web pages...mostly the ability to have people be able to "respond" to a 
page or set of pages and have their repsonses sent to me for analysis.  Ah, the 
life of a researcher.

Thanks again, and if there's anything I've missed or am overlooking, please let 
me know.

In article , cpr at emsoftware.com says...
>
>I think you're conflating Javascript with Java. They're completely different
>beasts.
>
>Javascript is most often used embedded in web pages to do client-related
>(DHTML) work (what it sounds like you want to do), usually mucking with the
>HTML of a given page dynamically. Javascript (in this case) is built in to
>the browser, and (in modern browsers, e.g., IE 4/5, NS 4) can be used to do
>some fairly complex things.
>
>Java applets (and thus JPython applets) usually run in a little "visual
>sandbox" (to say nothing of their file system sandbox), i.e., a rectangular
>portion of a web page. They can't really do the kinds of things that
>Javascript scripts can do, in terms of augmenting the overall UI of a web
>page.
>
>--
>Cheers!
>/ Chris Ryland, President / Em Software, Inc. / www.emsoftware.com
>
>"Kojo Idrissa"  wrote in message
>news:38DF007A.64AFBE6E at tamu.edu...
>> Hello all.  I am new to Python and I've got a question (surprise).  I'm
>> teaching myself Python, but I'm also thinking about doing some stuff
>> with Javascript to add some functionality to some web-pages (surveys,
>> on-line tests, allowing people to interact through the web and "respond"
>> to a site).
>>
>> My question is this:  How do JPython/Java applets compare with
>> Javascript from a functionality standpoint?  I'm thinking it would be
>> nice to not worry about learning another language and get more mileage
>> out of my knowledge of Python by using JPython to turn Python code into
>> Java applets.  I'm just not sure of the pros and cons.  My guess is that
>> it's probably more an Applet vs. Javascript question, but I'm not sure.
>> Any insight would be appreciated.
>
>
>

-- 
****************************
Kojo Idrissa
KPMG Scholar
Accounting Doctoral Student
Texas A&M University

kojo at tamu.edu
401M Wehner Bldg.
409-862-2726
****************************

From thomas.heller at ion-tof.com  Mon Mar 27 15:02:52 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Mon, 27 Mar 2000 22:02:52 +0200
Subject: tp_itemsize
Message-ID: <030a01bf9827$6f8002b0$4500a8c0@thomasnotebook>

Well,

after browsing through the source I have a vague impression
what tp_itemsize is for.
Is somewhere documentation for this?

Thomas Heller

From drugar at my-deja.com  Thu Mar 16 18:49:52 2000
From: drugar at my-deja.com (drugar at my-deja.com)
Date: Thu, 16 Mar 2000 23:49:52 GMT
Subject: How can I get a PyObject reference to an object already created?
Message-ID: <8arruu$dpm$1@nnrp1.deja.com>

In particular this is  Win32 project that embeds the python
interpretor.  Can I get a PyObject reference to an object that is
created by a script?
For example:

PyRun_SimpleString("import CustomScript\n");
PyRun_SimpleString("t=CustomClass()\n");

Is there anyway to get a reference to t? or any other object that is
created by importing CustomScript?  Assuming I know it's "name".

Thanks

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

From hartmut at oberon.noris.de  Tue Mar 28 15:27:35 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Tue, 28 Mar 2000 21:27:35 +0100
Subject: Bug in module time: localization?
References: <16600602@oberon.noris.de>  <16600605@oberon.noris.de> 
Message-ID: <16600611@oberon.noris.de>

Berthold H?llmann (bhoel at starship.python.net) schrieb:

 >It's no workaround. Programs only honour the locale settings if you
 >tell them so. An when you write Python Programs, and want to use the
 >locale settings, you have to tell it. An you do this with the locale
 >module.

IC! Thanks for the help!

 >Berthold

+++hartmut

From gchiaramonte at ibl.bm  Sun Mar  5 10:22:30 2000
From: gchiaramonte at ibl.bm (Gene Chiaramonte)
Date: Sun, 5 Mar 2000 11:22:30 -0400
Subject: Help: sizeof() and reference in python?
In-Reply-To: <20000305073630.C7327@quark.emich.edu>
Message-ID: 

I guess I should have been more specific.

1 - I am trying to get the size of an object in bytes to monitor how much
memory a list or object occupies.
2 - I am trying to get the pointer to the memory address of an object. Much
like the array objects buffer_info(), where the second item in the returned
tuple is the pointer to the memory address of the first element of the
array.

Thanks for the help though.

Gene

> -----Original Message-----
> From: Jeff [mailto:jam at quark.emich.edu]
> Sent: Sunday, March 05, 2000 8:37 AM
> To: Gene Chiaramonte
> Subject: Re: Help: sizeof() and reference in python?
>
>
> On Sun, Mar 05, 2000 at 08:17:43AM -0400, Gene Chiaramonte wrote:
> > Is there a way to get the size of an object in python? I'd also
> like to get
> > a reference to an object.
> >
> > example in c:
> >
> > l = [1, 2, 3]	#python list
> >
> > t1 = sizeof(l)	// how i'd do it in c
> > t2 = &l
> >
> > Thanks,
> >
> > Gene
> >
>
> you can get the *length* of an object in python, which I suspect is the
> closest you can come to sizeof()..  what are you trying to do
> with sizeof()
> that len() can't already handle?
>
> the reference thing is easy ;).. everything is a reference in
> python. so if you
> want 't2' to be a reference to 'l', you assign it as normal 't2 = l'.
>
> hope that helps..
>
> regards,
> J
> --
> || visit gfd 
> || psa member -- 
> || New Image Systems & Services, Inc. 

From grant at nowhere.  Thu Mar 30 11:14:28 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 30 Mar 2000 16:14:28 GMT
Subject: Philosophical long integer question...
References: <38E23160.B16860C2@bioreason.com>
Message-ID: 

In article <38E23160.B16860C2 at bioreason.com>, Brian Kelley wrote:

>In the same vein as should 1/2 equal 0.5:
>
>should (1 << 1000) produce a long int?  

My vote would be no, but I write embedded, real-time software
most of the time and am used to being constantly aware of the
machine representation of what I'm writing.

-- 
Grant Edwards                   grante             Yow!  My forehead feels
                                  at               like a PACKAGE of moist
                               visi.com            CRANBERRIES in a remote
                                                   FRENCH OUTPOST!!

From jstok at bluedog.apana.org.au  Sat Mar  4 20:22:47 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Sun, 5 Mar 2000 12:22:47 +1100
Subject: newbe question about lists
References: <0d6dde44.8e4c635f@usw-ex0102-015.remarq.com>
Message-ID: 

Fredrik S wrote in message <0d6dde44.8e4c635f at usw-ex0102-015.remarq.com>...
>Sorry for this simple question, but I can't go to sleep before I
>have solved it.
>
>I am running python on Beos for the first
>time.
>
>I'm going through a study course online, and ringt now I
>am trying lists.
>
>I type the following:
>>>>
>l1=["a1","a2","a3"]
>>>> l2=["a4","a5","a6"]
>>>> l3=l1
>>>>
>l3[1:1]=l2
>>>> l3
>['a1', 'a3', 'a4', 'a2']
>>>> l1
>['a1', 'a3',
>'a4', 'a2']
>
>QUESTION:
>Why is value of l1 changed... The only
>thing I can think of is B U G ...

This is just standard reference semantics.  Lists are mutable structures,
first of all.  Assigning the first list to l1 creates the list object and
binds it to the name "l1".  Assigning l1 to l3 binds "l3" to the *same*
object.  Therefore, any changes in the object referred to by the "l1" name
are now reflected in the object referred to by the "l3" name, because they
are one and the same.

To prevent this behaviour bind l3 to a *copy* of the object referred to by
l1, like this:

>>>l3 = l1[:]

From m.faassen at vet.uu.nl  Thu Mar 16 12:38:52 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 16 Mar 2000 17:38:52 GMT
Subject: SPAM
References: 
Message-ID: <8ar67d$f2e$1@newshost.accu.uu.nl>

Michal Bozon  wrote:
> __hi__

> What is it "spam" ?

Besides the other answers, the Python conferences are also sometimes referred
to as SPAM (i.e. SPAM7). I'm not sure why. :)

Regards,

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

From python-list at teleo.net  Tue Mar  7 21:41:41 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Tue, 7 Mar 2000 18:41:41 -0800
Subject: A Question
In-Reply-To: <8a4dgt$naa$1@nntp.itservices.ubc.ca>
References: <8a4dgt$naa$1@nntp.itservices.ubc.ca>
Message-ID: <0003071846440C.00825@quadra.teleo.net>

[Purang Abolmaesumi, on Tue, 07 Mar 2000]

:: So, here is the problem: We would like to have a GUI which gets images from
:: a frame-grabber installed on the PC and show them on the screen (we have the
:: C functions right now which are running perfectly), send some data in
:: real-time to the VXWORKS machine (again, we have the C functions written)
:: and at the same time, interacts with user.
:: I have the following questions:
:: 1. Can we use Python for this?
:: 2. Can we call C functions in Python and vice versa easily?
:: 3. How fast is a Python program? Can we use it for a real-time application
:: like the one we have?
:: 4. If we use TKinter as the GUI, can we use our C functions for the
:: frame-grabber to show the image on the screen?

Since you're running under Linux, you might want to have a look at
SVGAlib (www.svgalib.org). Since it's a low-level library, it might meet
your real time needs better?

From donn at u.washington.edu  Mon Mar 13 12:29:51 2000
From: donn at u.washington.edu (Donn Cave)
Date: 13 Mar 2000 17:29:51 GMT
Subject: event loops and Python?
References:  <200003120619.QAA11915@piglet.dstc.edu.au>
Message-ID: <8aj8if$dk2$1@nntp6.u.washington.edu>

Quoth kragen at dnaco.net (Kragen Sitaker):
...
| There are a few things that were bothering me.
|
| One is that you can't have two event loops in the same thread.  Event
| loops are one of the few things you can only have one of.  Accordingly,
| if you have two pieces of code designed to run from two different event
| loops, and you want to build a program containing both, your choices
| are somewhat restricted --- modify one to work with the other's event
| loop, run them in different threads, or give up.
|
| So a standard event loop, or at least a standard event-loop API,
| for all event-loop-based Python applications would be very useful.

Sounds like asyncore is running unopposed for that, so far.

| Another is that handling signals safely in Perl is tough, and I
| assume the same thing is true of Python.

True, and I say good riddance.  Signals are easier to use in C, but
are really appropriate for only low level stuff, in my opinion -
for example from a software engineering point of view, try to get your
two pieces of code to cooperate over signals, or even protect themselves
from each other's use of signals.

| Another is that, while I'm comfy with writing select()-based
| event loops in C, I'm not particularly comfy with writing C code that
| interfaces with the Python interpreter, particularly at such an
| intimate level.  I was hoping I could put that off until later.
|
| Enough English.  Now it is time for me to write code before posting more.

Eek, use Python's select() function, if you don't use asyncore!  The
Python select() implementation is much more convenient

	Donn Cave, donn at u.washington.edu

From nnever50 at yahoo.com  Wed Mar 15 08:56:28 2000
From: nnever50 at yahoo.com (nathan)
Date: Wed, 15 Mar 2000 14:56:28 +0100
Subject: sorry..the program runs!
References: <8ale7h$bl$1@news1.sunrise.ch> 
Message-ID: <8ao4o6$5nh$1@news1.sunrise.ch>

..but anyway
I cannot do anything with the GUI until the process is ended.
As I already wrote, I have a process that runs during all the time and I
wont to have the possibility to interact with the GUI.
There's other ways?
Thank you!
N.

From embed at geocities.com  Mon Mar 27 12:03:42 2000
From: embed at geocities.com (Warren Postma)
Date: Mon, 27 Mar 2000 12:03:42 -0500
Subject: __getattr__ / __setattr__ - infinite recursion!?
Message-ID: <0yMD4.624$HG1.20784@nnrp1.uunet.ca>

Okay my first attempt at a "static type checking object" for Python blows
up, probably
a stack overflow, probably due to infinite recursion or some such thing....

Python blows up on the line that instantiates this object! I suspect I'm too
clueless to know how to do a proper __getattr__ / __setattr__
implementation. A better solution to this would be a built in dictionary
type with type-information as part of it's C code.  Has anyone built such a
beast directly in C?

Warren

#---------------------------------------------------------------------
# Why does this blow up Python?
#---------------------------------------------------------------------

import os, sys, string, pprint

# "type objects"
int_type       = type(0)
string_type    = type("X")
float_type     = type(0.1)

# required field flags?
required = 1
optional = 0

# Class with strict type checking and a single key:
class zdb_rec:
    def __init__(self, pattern_dict=None, keyname=None):
        "create an object with a strict pattern of required values to make
up a type"
        self.set_valid(pattern_dict)
        self.keyname = keyname
        self.reset_to_defaults(self)

    def set_valid(self,pattern_dict):
        self.attrs = pattern_dict

    def set_keyname(self,keyname):
        self.keyname = keyname

    def reset_to_defaults(self):
        # fill in default values for each attribute
        self.values = {}
        for x in self.attrs.keys():
            self.values[x] = self.attrs[x][2]

    def keyname(self):
        return self.keyname

    def key(self):
        return self.values[self.keyname]

      # setattr does typesafety check:
    def __setattr__(self,name,value):
        "custom __setattr_ with typesafety check"
        try:
            if (value == None) and (not self.attrs[name][1]):
                self.values[name] = None
            else:
                if (type(value) == self.attrs[name][1]):
                    self.values[name] = value
                else:
                    raise Error, "Invalid type in assignment"+`type(value)`
        except:
            raise Error, "Invalid attribute "+`name`
    def __getattr__(self,name):
        try:
            return self.values[name]
        except:
            raise Error, "Invalid attribute "+`name`

    def attr_count(self):
        return len(self.attrs)

    def validate(self):
        "raises an exception if validation fails, ie requird field not
provided"
        for x in self.attrs.keys():
            if self.attrs[x][1]:
                if self.values[x] == None:
                    raise Error, "validate() failed - required field not
set"

# "Person Database Schema":
person_attributes = {
 # attribute    type          required?   initialvalue
     "name":    (string_type,    required,  ""),
     "address": (string_type,    required,  ""),
     "age":     (int_type,       required,  0 ),
}

x = zdb_rec(person_attributes,"name")

print "done"

From aahz at netcom.com  Thu Mar 23 11:43:47 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Mar 2000 16:43:47 GMT
Subject: List limitations?
References: 
Message-ID: <8bdhk3$42d$1@nntp6.atl.mindspring.net>

In article ,
Daley, MarkX  wrote:
>
>What this does is connect to a server and pull data from it based on each
>family name in the familyname list, using an SQL query.  Conceivably, this
>code should connect to server1 and pull data for all familynames and write
>that data to a file, server_familyname_date.  Then it should connect to
>server2 and do the same.  However, when it connects to server 2, it isn't
>pulling the familynames from the familyname list correctly.  Instead, it is
>pulling single character strings, like a, b, e, g, h, r, S (these are the
>actual strings, which may or may not have some bearing on one of my
>familyname entries, I haven't been able to verify).  Has anyone else seen
>problem remotely related to this?  Any help is appreciated.

Questions/suggestions:

Have you tried connecting just to server2?
What happens if you switch server2 and server1?
Have you considered using mxODBC?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.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 Lisowski.Tomasz at sssa.nospam.pl  Sat Mar 25 03:10:33 2000
From: Lisowski.Tomasz at sssa.nospam.pl (Tomek Lisowski)
Date: Sat, 25 Mar 2000 08:10:33 GMT
Subject: else clauses in while and for loops
Message-ID: 

Hi!

Which is the reason for else clauses in while and for loops? At the first
look it seems, that the statements in the else suit is executed always after
processing the loop. Wouldn't it be simpler, if I just put those statements
after the loop, and omit the else clause completely?

Please give some examples showing the real need for else clauses in this
context.

Thanks in advance

Tomasz Lisowski

From Marc_Battyani at csi.com  Tue Mar 14 16:07:47 2000
From: Marc_Battyani at csi.com (Marc Battyani)
Date: Tue, 14 Mar 2000 22:07:47 +0100
Subject: Can embedded static-linked python still import dll extensions?
References: 
Message-ID: 

Warren Postma  wrote in message
news:cItz4.33201$Jz3.306758 at nnrp1.uunet.ca...
> I recompiled the Python sources for Windows with config.h and put this
line
> in:
>
> #define MS_NO_COREDLL 1  // build a static library not a DLL!
>
>
> Now it statically links fine, but it doesn't allow me to import .pyd files
> (Python extension DLLs). Do you have to use the external Python DLL to
allow
> importing modules from DLLs, or can you statically link python but still
> allow external DLL add ons somehow?
>
> I want my embedded python deeply embedded but I still want to produce .PYD
> style add ons for the system once it's in the field. Am I stuck?

No. Nothing prevents you to exports functions from an exe.
Here is a small C program

#include 
__declspec( dllexport ) void foo(){printf("Foo\n");};
__declspec( dllexport ) void bar(){printf("Bar\n");};
main()
{
  foo();
  bar();
}

Let's compile it:

D:\temp>cl export.c
Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 12.00.8168 for 80x86
Copyright (C) Microsoft Corp 1984-1998. All rights reserved.

export.c
Microsoft (R) Incremental Linker Version 6.00.8447
Copyright (C) Microsoft Corp 1992-1998. All rights reserved.

/out:export.exe
export.obj
   Creating library export.lib and object export.exp

****** You can see it creates a .lib and a .exp **********

D:\temp>dumpbin /exports export.exe
Microsoft (R) COFF Binary File Dumper Version 6.00.8447
Copyright (C) Microsoft Corp 1992-1998. All rights reserved.

Dump of file export.exe

File Type: EXECUTABLE IMAGE

  Section contains the following exports for export.exe

           0 characteristics
    38CEA936 time date stamp Tue Mar 14 22:03:50 2000
        0.00 version
           1 ordinal base
           2 number of functions
           2 number of names

    ordinal hint RVA      name

          1    0 00001012 bar
          2    1 00001000 foo

******* Here you can see the exported functions are... exported. **********

Good luck
Tell us informed.

Marc Battyani

From gmcm at hypernet.com  Wed Mar 15 09:51:38 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 15 Mar 2000 09:51:38 -0500
Subject: For loop
In-Reply-To: <38cf6e09$0$6366@businessnews.de.uu.net>
Message-ID: <1258990590-4218974@hypernet.com>

Boris Ottlewski wrote:
> 
> I got this code snipet: (See the colons!)
> 
> for a in i.something():
>      for cf in a.somethingelse():
>           print cf.property
> 
> It works in python.exe (shell). How do I use this in embedded python? When a
> build a string with this code and do a Py_RunSimpleString() it raises a
> syntax error. Is there an alternative syntax? Is the colon someting just
> known to the python shell?

Make sure the lines are delimited properly, and the parser can 
detect a dendent and end-of-file:

"for i in a:\n for j in b: \n  something\n\n\0"

- Gordon

From prodos8 at yahoo.com  Sun Mar 26 19:26:04 2000
From: prodos8 at yahoo.com (carbon)
Date: Mon, 27 Mar 2000 00:26:04 GMT
Subject: Calculating Derivatives?
Message-ID: <260320001926276702%prodos8@yahoo.com>

Is there any python source that relates to calculating derivatives? 
I'd like to try writing software that does this (for polynomials), but
I don't know where to start.  any resources or help would be greatly
appreciated.  thanks.

From x at x.x  Mon Mar  6 16:50:11 2000
From: x at x.x (x at x.x)
Date: Mon, 06 Mar 2000 16:50:11 -0500
Subject: Problems with uploading binary files using cgi module
References: <01BF86EC.F2D432E0.shichang@icubed.com>
Message-ID: 

In article <01BF86EC.F2D432E0.shichang at icubed.com>,
Shichang Zhao  wrote:
> I am not sure there is a bug in Python cgi module or I did somethin wrong. 
> Here is what I am trying to achieve:
> 
> Goal:
>      Uploading a binary file, say a Microsoft Word file name test.doc, to 
> Web server's cgi-bin directory with a name
>      tmp.doc.
:
> 
> After the file is uploaded, I compared the original file's size (19,456 
> bytes) with the file saved onto the server's cgi-bin directory (tmp.doc, 
> size: 13,073 bytes). Some of the bytes are lost after the file is
uploaded.

Are you by any chance running this off a Windows server? I'm thinking
because you mentioned Microsoft Word; if so, I think I know what the
problem is: Standard input is opened by default in "text" mode, and
you're probably running into an end-of-file character (ascii 26) and
that's causing the file to truncate.

If that's the case, you're going to need to dig into the msvcrt module 
and reset sys.stdin at the beginning of your cgi script. If I recall,
you need the "OSF handle" functions. I had to do that once, I can
dig up the code if you need it.

I do remember that doing an os.fdopen(0, "rb") wouldn't do it in
Win32. It was necessary to get into the msvcrt.dll gunk.

From a.eyre at optichrome.com  Tue Mar 21 06:40:35 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 21 Mar 2000 11:40:35 -0000
Subject: How to retrieve an objects content
In-Reply-To: <38d6278b$0$29742@businessnews.de.uu.net>
Message-ID: <000101bf932a$45ab7fc0$3acbd9c2@optichrome.com>

> I would like to get an objects content into a string. (Like the 
> python print command) The print command uses PyFile_WriteObject
> which is quite unhandy. Any ideas?

PyObject_Str(), or PyObject_Repr() should do the trick. See:

http://www.python.org/doc/api/object.html

You'll also need to convert from a Python string to a C string.
For that you'll need PyString_AsString(). See:

http://www.python.org/doc/api/stringObjects.html

-----------------------------------------------------------------
Adrian Eyre  - http://www.optichrome.com 

From Vladimir.Marangozov at inrialpes.fr  Tue Mar 14 09:24:51 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Tue, 14 Mar 2000 15:24:51 +0100
Subject: Dictionary in Python
References:   <9t9pusynelo.fsf@mraz.iskon.hr>
Message-ID: <38CE4BB3.D81D2275@inrialpes.fr>

Hrvoje Niksic, replying:
>
> AFAIK Python dictionaries increase dynamically.

They also shrink dynamically. For those who are interested, there was
a good discussion on the dict implementation in the august 1999 archives
of python-dev. Look for a thread "shrinking dicts". Tim Peters explains
some subtleties of the implementation, especially in this msg:

http://www.python.org/pipermail/python-dev/1999-August/002162.html

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

From calishar at *NOSPAM*home.com  Tue Mar 14 23:11:24 2000
From: calishar at *NOSPAM*home.com (Calishar)
Date: Wed, 15 Mar 2000 04:11:24 GMT
Subject: Stuck at CH9 of  PythonW32 book, VBA
References: <2KCz4.33305$bm.298215@news1.alsv1.occa.home.com>
Message-ID: 

On Wed, 15 Mar 2000 02:39:58 GMT, "Adolfo Aguirre" 
wrote:

>Hi.
>
>I just bought the new book "Python Programming on W32". CH9 starts with an
>excel exercise that requires the VBA editor. Instructions are given for
>Office 97.
>
>I have Office2000. And I don't see a VBA editor in it. When I tried to
>reinstall Office 2000 I did not see VBA as amodule option either.

The visual basic editor is under the Macros option under the Tools
menu. It's possible the tools menu won't display the Macros if you
havent used it before, in which case, just hold the mouse cursor (for
about a second) over the double down arrow at the bottom of the menu,
or select the double down arrow like any other menu option.

thanking-Microsoft-for-changing-how-menus-work-ly yrs,
Calishar

(neat, I finally get to use a -ly-yrs)

From wware at world.std.com  Thu Mar 23 14:48:58 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 23 Mar 2000 19:48:58 GMT
Subject: Off Topic: looking for a paper
References: <38DA69F0.336A7729@sage.att.com>
Message-ID: 

Garry Hodgson (garry at sage.att.com) wrote:
> i'm looking for a paper i believe i've seen referenced here,
> but can't recall.  it describes a new jersey style of
> design, as exemplified by unix...

Go to www.google.com and search on "new jersey mit style unix". (I
conveniently remember that the competing style group was MIT.) In
the first ten hits, you'll see it appear five times.

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com

From tismer at tismer.com  Thu Mar  9 17:07:27 2000
From: tismer at tismer.com (Christian Tismer)
Date: Thu, 09 Mar 2000 23:07:27 +0100
Subject: Making sense of Stackless
References: <146101bf8164$fc19df90$6401a8c0@dorb> <38B98A01.9DC6C456@tismer.com> <148f01bf8185$fd355eb0$6401a8c0@dorb>
Message-ID: <38C8209F.E95D4A44@tismer.com>

Darrell wrote:
...
> It would be nice if I could do this.
> 
> def classIterator(seq):
>     offset= -1
>     return continuation.current(1)
>     offset=offset+1
>     return seq[offset]
> 
> The parm into current indicates I want the continuation to point just past
> this point.

This is included into continuations 0.8 which are about to be
published. It is even easier: Whatever I can express as a
function avoids to have to juggle with opcodes.
Here we go:

continuation.return_current()

Does exactly what we need: Return ourself and be in position
to be called again.

ciao - chris

-- 
Christian Tismer             :^)   
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
     we're tired of banana software - shipped green, ripens at home

From ljz at asfast.com  Tue Mar  7 22:37:51 2000
From: ljz at asfast.com (Lloyd Zusman)
Date: 07 Mar 2000 22:37:51 -0500
Subject: Self Nanny
References:   <38C5709D.59E8@marshall.edu>
Message-ID: 

Mark Hathaway  writes:

> >> Gaetan Corneau  writes:
> >>
> >> [ ... ]
> 
> > Lloyd Zusman wrote:
> >
> > [ ... ]
> 
> Personally, I'd rather not ever have to write 'self' or 'self.anything'.
> It contributes nothing to the code and doesn't describe what I'm trying
> to write. It's extra verbosity without any payoff.
> 
> Was there a problem with doing it the other way around...
> 
>   [ ... examples snipped ... ]
> 
> where the roles are reversed. All object attributes are
> referred to by their names and any class (global) attributes
> are specified by "local.". This would decrease tremendously
> the need to write 'self.' and would still easily allow the
> differentiation of class-wide attributes from object attributes.
>
> Which do you write more of? Object attribute names.
> Why make that the more verbose version?

In Section 9.4 of the official Python tutorial (did you read it?),
Guido says this about his reasons for the explicit 'self':

  There is no shorthand for referencing data attributes (or other
  methods!) from within methods. I find that this actually increases
  the readability of methods: there is no chance of confusing local
  variables and instance variables when glancing through a method.

  Conventionally, the first argument of methods is often called
  self. This is nothing more than a convention: the name self has
  absolutely no special meaning to Python. (Note, however, that by not
  following the convention your code may be less readable by other
  Python programmers, and it is also conceivable that a class browser
  program be written which relies upon such a convention.)

You may disagree with this.  Nonetheless, this expresses Guido's
reasoning (or at least a part of it).  If you haven't done so already,
I suggest you read all the existing Python documentation.

And did you know that Python was first created some 10 years ago?  And
did you know that in the early stages of its development, many people
such as yourself raised the issues that you are now raising?  And did
you know that now, Python is a mature language and that there are
millions of lines of Python code in use in production applications all
over the world?

... and even if somehow this 'self' change ever were to happen in
Python, all the existing production applications would be very
expensive to change in terms of time and money.

> > LZ:
> >
> > Of course #3 can be done (technically speaking).  And in return for it
> > getting done, would you volunteer to donate your personal time to
> > change all the existing lines of Python code in the world that this
> > change would break?
> 
> This is both true and annoying. In our search for better ways to do
> things we keep running into this barrier that languages are written,
> tried and much time & effort put into that trial. Then when someone
> says, "Couldn't we do it better this way?", the answer is, "But, we've
> put so much work into this..."

So tell the truth: do you really think that this 'self' change is so
much of an improvement over the way that Python works today that it
would justify changing 100 percent of all existing Python programs
that contain method definitions?

Also, keep in mind that there indeed have been many many useful and
meaningful "Couldn't we do it better this way?"  discussions about the
implictness or explicitness of 'self' ... they took place maybe 5-10
years ago when Python was still in the design and debate stage of
development.

But since at least a few years ago, you can no longer think of Python
as a "trial" language.  As I mentioned, there are now millions of
lines of Python being used in real, production-grade applications
today.

If there was a new or young language being considered, discussion of a
behavior change like this would be quite useful and quite appropriate.
It's just that you're way too late to expect a such a major,
legacy-code-breaking change in a language which already has so much
existing production-grade code out there.

There is an undocumented and probably unintended behavior of the
Python 'append' method that people have been exploiting in certain
applications.  Recently, it was decided that in the next major release
of Python, this undocumented behavior will no longer work, and that
the 'append' method will then behave only as documented.  This change
will affect some existing Python applications, but many fewer than
would be affected by this suggested 'self' alteration.

And even in the case of this very small change which removes an
*undocumented* behavior of a method, Guido has recently given us a 3-4
month warning so that the small percentage of code which currently
uses this undocumented behavior could be altered.  And even with this,
there are costs involved in existing, production applications, and
some people are requesting that the change to 'append' not be made.

Given the impact of this tiny change to remove an undocumented
behavior of one method, can you imagine how much of a cost (money and
time) there would be in changing to an implicit 'self'?  As I
mentioned above, this change would require 100 percent of all method
definitions to be changed in all existing Python code.

> Somehow there has to be a willingness to put the effort in and then,
> if it's not as perfect as we'd like, throw it away and start with
> another piece of whole cloth.  [ ... ]

That's quite appropriate for a new or very young computer language.
But you're several years and a few millions of lines of production
code too late for Python.

It's very appropriate to "start over" when creating a new language.
For example, Stroustrup did that to C when inventing C++.  Larry Wall
did that when creating Perl.  Guido did that when creating Python.

And I invite you to do that with the new language that *you* could
create to supersede Python.  What's stopping you from doing that?

> [ ... ] Well, perhaps we could write a translator to convert the old
> to the new. That wouldn't take quite as much man-power, would it?

Even the best translator cannot be counted on to work in all cases.
Therefore, even after running an existing program through this
hypothetical translator, significant testing would have to be
performed to make sure that the converted program still works as it's
supposed to.  And for production applications where time and money is
on the line, this testing would be essential.  Hence, even with a code
translator, the greatest amount of expensive testing would have to be
done in those places where the downtime of an application would cost
the most.

> > Or barring that, would you be willing to pay for the time of all the
> > others who would have to do this work?
> 
> Pay? Who paid the first time? Didn't most of the Python code come
> about because people liked the language and just wrote stuff to be
> distributed via the internet for free? O.K. here's a penny. Ya
> satisfied?

The time and money I'm talking about wouldn't be for development
... it would be to compensate all the people who would have to expend
their own time and money to convert current, production Python
applications so that they would once again work once this hypothetical
change would go into effect.  And remember, this change would break
nearly 100 percent of all existing Python programs.

Would you be willing to explain what advantage this proposed change
would give to people already running these existing applications to
justify the large time and money cost of changing all of them?

The only potential advantages that anyone ever mentioned about having
an implicit 'self' are (1) that a certain small subset of new Python
users might find Python a little easier to learn if it handled 'self'
closer to the way that they are used to C++ handling 'this'; and (2)
that a small minority of existing Python coders find the explicit
'self' to be annoying.

Are there other advantages you see to this proposed change that would
justify the huge time and money cost of converting all the existing
Python programs?

> I know you're saying people's time & effort is considerable. All I'm
> saying is there should be ways to move to new better ways of doing
> things without having to face such stiff resistance.

I wholehearedly agree.  We should always move towards better ways of
doing things.  That's one of the main reasons that people invent new
computer languages.  Therefore, as I've suggested before: given that
it's years too late to make such a drastic change in Python, why not
propose a new language that has the better features that you are
suggesting?  I consider that to be a very worthwhile endeavor.

> > This is not an issue about "caring" about C++ programmers.  I am a C++
> > programmer, and literally hundreds of my colleagues over the past
> > decade or so have been C++ programmers.  Many of these people have
> > tried Python.  And *none* of them have ever even made a complaint
> > about the explicit "self" of Python.  You are the first C++ programmer
> > I have encountered who has ever raised this as an issue.  That's one
> > out of hundreds of people.
> 
> Perhaps they are simply more accepting of what is, and don't
> challenge it.  That doesn't mean they haven't thought about it. I'm
> not a C++ programmer, so I don't know either.

Well, perhaps there is a Great Silent Majority of Python users out
there who have been seething with resentment about 'self' having to be
explicit, and who will rise up and revolt to rally behind a leader who
finally raises his or her voice to fight the Python Establishment.

But I doubt it.

... especially given how much this has already been discussed over the
past 5-10 years with the end result being the Python language we see
today.

What is much more likely is that there are some people who would be
willing to participate in the design and development of a new O-O
scripting language which differs from Python and other existing
languages in ways that these people feel are important.  Would you be
willing to be part of that effort?

> > And as I have mentioned in another post, I challenge you to come up
> > any signficant number of people who would be glad to use Python if
> > only it didn't have an explicit "self".
> 
> It's not just about that issue. It's about an effort to move toward
> something better. You're not against that altogether, are you?

As you can tell from reading what I have already written above and
elsewhere, I am not against that at all, and in fact I'm for it.

> > Your arguments have always been clear.  And they are still rather
> > unconvincing.
> >  Lloyd Zusman
> >  ljz at asfast.com
> 
> 
> A closed mind can't even be convinced by the truth or a stick up side
> the head.

Please re-read this post and my others.

I have said this before and I'll say it again: this sort of feature
would be perfectly valid to discuss for a new language, or one that is
still in the design and debate stage of development ... and I
encourage you and others to do just that.  It's just too late to
effect the change in Python, given the amount of expensive work it
would take to re-tool the existing applications that are out there to
accomodate this change, and given the low amount of desire for such a
change among most Python users.

Also, if you somehow think that this is a new and undebated
suggestion, you might want to go back and try to find some of those
earlier debates about the way that Python handles 'self' that took
place 5-10 years ago when at that time it *was* a young language.

-- 
 Lloyd Zusman
 ljz at asfast.com

From alex at somewhere.round.here  Tue Mar 14 14:37:49 2000
From: alex at somewhere.round.here (Alex)
Date: 14 Mar 2000 14:37:49 -0500
Subject: Is UserList.py "strange" or is it cmp()?
References: 
Message-ID: 

Hi, Tom.

> >>> ul == aList
> in UserList.__cmp__(self=[0, 1, 2, 3], other=[0, 1, 2, 3]):
>    false leg: cmp(self.data, other)=0
> 1
> >>> cmp(ul,aList)
> in UsreList.__cmp__(self=[0, 1, 2, 3], other=[0, 1, 2, 3]):
>    false leg: cmp(self.data, other)=0
> 0

cmp should actually return 0 when it thinks its arguments are equal.  It
confused me for a while, too, but it makes sense in the context of an
ordering.  Usually, the way it's supposed to work is

cmp (a, b) => -1  if a < b
cmp (a, b) =>  0  if a == b
cmp (a, b) =>  1  if a > b

which is consistent with the results you got.  I guess that 'a == b' is
actually returning (not cmp (a, b))

Alex.

From grant at nowhere.  Mon Mar 13 14:34:39 2000
From: grant at nowhere. (Grant Edwards)
Date: Mon, 13 Mar 2000 19:34:39 GMT
Subject: How do I ...
References: <38CD2368.9D29554F@abo.fi>
Message-ID: 

In article <38CD2368.9D29554F at abo.fi>, Cesar Lopez wrote:

>How do I acces to the low-level of my system? I mean, for example, I
>want to access and configure the speed of my serial port using a Python
>program. I've tried building an extension C function and the call it
>from Python, but doesn?t works.

1) You really need to do a better job on your "Subject:" line.
   The bandwidth in this group is high enough that a lot of
   people aren't going to read all of the articles.  I read
   yours by accident -- I generally only read articles when I
   either know the answer to the question or am otherwise
   interested in the particular subject.

2) Read the FAQ

      http://www.python.org/doc/FAQ.html#4.68

-- 
Grant Edwards                   grante             Yow!  Now KEN is having
                                  at               a MENTAL CRISIS beacuse
                               visi.com            his "R.V." PAYMENTS are
                                                   OVER-DUE!!

From tuttledon at hotmail.com  Sun Mar 26 19:46:53 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Mon, 27 Mar 2000 00:46:53 GMT
Subject: Language forks (was: Python tutorial/comparison for C++ programmer)
References:     <5FF7ED1C1C902C69.0E5D50066AC4B434.CE45BE37438DE91A@lp.airnews.net> 
Message-ID: <1cyD4.5258$8E5.54893@typhoon.southeast.rr.com>

[Grant Edwards]
> MS seems to actively attempt to diverge from standards in order to try to
> kill off those standards and establish their own dominance.

I think it's more accurate to say that Microsoft, like all commercial
companies, seeks a competitive advantage.  Just look at the Sun vs.
Microsoft battle.

Sun has visions of Java running everywhere on everything.  Java rules!
Microsoft is history! Microsoft is scared and decides if it's going to keep
its client base, they have to make Windows the most attractive OS to run
Java on.  So Microsoft 'optimizes' Java to run faster on Windows then any
other OS.  Then they create the most advanced Java development system
available (at the time) Visual J++.

Time move on.  Sun sues Microsoft and wins.  But Java fails to deliver on
the grand vision of write once, run anywhere.  Sun seeks support from a
"standards" committee.  But after 2 tries gives up because they can't find a
standards committee willing to  rubber stamp 'their' plan without having to
reveal all of their Java code, which would be the loss of a competitive
advantage to them.

Meanwhile Microsoft developed the open source protocol SOAP, which uses XML
to make remote procedure calls (and more)in a heterogeneous environment.
SOAP is OS and application neutral.  A PERL app on Linux can interact with a
Netware box running a COBRA app.  It just doesn't matter!

And now Sun is scared because SOAP could be the final straw to it's vision
of Java everywhere. Did Microsoft create a proposed open source standard
simply to make the world a better place? Of course not! Microsoft believes
it can leverage its development strengths to create the best applications at
each end of the SOAP pipeline.  And that, of course, translates to a
competitive edge. ;-)

Don

From jcrowe at io.com  Wed Mar  1 14:52:07 2000
From: jcrowe at io.com (Joseph Crowe)
Date: Wed, 1 Mar 2000 13:52:07 -0600
Subject: Python Rocks!
In-Reply-To: <38854F3A.4FBA74F9@exceptionalminds.com>
References: <860ftk$bas@news.or.intel.com>  <862jqd$l9s@news.or.intel.com>  <38854F3A.4FBA74F9@exceptionalminds.com>
Message-ID: 

On Tue, 18 Jan 2000, Timothy Grant wrote:
> You know Francois, this was my first reaction to this also. A good
> editor makes such a huge difference that I can barely work in one that
> doesn't have a "python" mode and syntax highlighting. However, it does
> take a little work to configure and editor to make it work.
> 
> Tye, have you taken a look at an editor with a Python mode? or used Idle
> or Pythonwin?
Hi All,

    Actually, vim...or more precisely, gvim seems to work very well in
Python mode as far as syntax highlighting goes.  Also, it's freely
available for most any platform and I think delivers with just about
any distro of Linux.
> Stand Fast,
>     tjg.

Joseph Crowe
http://www.io.com/~jcrowe
email jcrowe at io.com

From jwbnews at scandaroon.com  Mon Mar 20 14:57:01 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 20 Mar 2000 19:57:01 GMT
Subject: Pascal int()
References: <1258726940-20077342@hypernet.com> <38D66680.316F21D1@roguewave.com>
Message-ID: 

In article <38D66680.316F21D1 at roguewave.com>, Bjorn Pettersen 
 wrote:

> Gordon McMillan wrote:
> > [...]
> > I can assure you that no Python god, archangel, angel,
> > seraphim or cherubim would ever write anything but
> >  a = a + 1
> > 
> > - Gordon
> 
> ... of course, if P3K changed the meaning of 5 from "the object 5" to
> "a constructor for an Integer class, setting the value to 5" you could
> do all kinds of fun things, including:
> 
> 	a = 5
> 	a.inc()
> 
> And even though it would be backwards compatible, I don't really have
> high hopes for it being accepted (change? we don't need no stinking
> change ;-)
> 
> -- bjorn
> 

Nothing would prevent building an int wrapper similar to UserDict 
(except, perhaps, a slavish desire for performance of one's integers).  
The hypothetical UserInt's hypothetical subclass could certainly define 
the method inc().

Personally, I'd rather write a = a + 1.  I seldom used inc() whist 
writing Pascal, either, except for incrementing variables of types like 
(red, blue, green, khaki, oliveDrab).

(Various language implementations--some LISPs, some SmallTalks--show 
that by sacrificing some of the range of int, one can make int a class 
quite nicely.  I doubt that there would be many votes in favor of 
reducing int to 29 bits (or whatever).)

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com

From jfarrell at mincom.com  Sun Mar  5 21:02:12 2000
From: jfarrell at mincom.com (John Farrell)
Date: Mon, 06 Mar 2000 12:02:12 +1000
Subject: Splitting comp.lang.python
References:  <019601bf843c$b216b500$0201a8c0@spkydomain> <38BF1E16.7AD5FE93@mincom.com> <3572DB0C6C3EE456.0F7DD462C78FAF91.B661EE7244D9906C@lp.airnews.net>
Message-ID: <38C311A4.5F9E79C9@mincom.com>

Cameron Laird wrote:
> In article <38BF1E16.7AD5FE93 at mincom.com>,
> John Farrell   wrote:
> >I would also like to see a newsgroup for people who are working on the
> >Python implementation, but I can't think of a good name for it.
> "[T]he Python implementation"?!?  My rough extrapolations lead
> me to conclude that somewhere near the year 2037 there'll be
> more Python implmentations than there are Python programmers
> 

Except that some of the Unix ones will break due to the Unix Y2K+37 bug,
and it might take till 2039 to get the numbers back up :-).

John
-- 
Dr John Farrell - Research Architect - Mincom Limited

I don't suffer from stress.  I am a carrier.
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS d- s++:+ a C+++ U+ 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++++(*)
------END GEEK CODE BLOCK------
This transmission is for the intended addressee only and is confidential
information. If you have received this transmission in error, please delete
it and notify the sender. The contents of this E-mail are the opinion of the
writer only and are not endorsed by Mincom Limited unless expressly stated
otherwise.

From Johannes.Nix at mail.uni-oldenburg.de  Thu Mar 16 13:56:13 2000
From: Johannes.Nix at mail.uni-oldenburg.de (Johannes Nix)
Date: 16 Mar 2000 19:56:13 +0100
Subject: A python program as a Win "Service" or Unix "Daemon"?
References: <6D8A17398E28D3119F860090274DD7DB4985A7@pces.cadlab.it>
Message-ID: <8araoe$d2k@news.Informatik.Uni-Oldenburg.DE>

Alessandro Bottoni  writes:

> Is there any way to install a Python program as a Unix Daemon or a Windows
> Service? (Together with an instance of the interpreter, of course..)

Speaking for Unix, a daemon program is not very special.

It has normally the following features:

- It runs in the background, without attached terminal (standard
input). That is in C usually achieved by closing the standard input
and doing a fork() (or vice versa?). This is almost equivalent to
creating a new thread, so the threads interface should be enough.

- It sometimes reacts specially to interrupts, for example it may
read-read its configuration file if it receives a SIGHUP

- It uses frequently protected services and ports and thus runs with
root or other special privileges.

- An advanced daemon may send log messages to the syslog daemon, which
will write it to a file as /var/log/messages, or send it to another
computer.

- It starts normally automatically by a call from the system startup
scripts.

As can be seen, this should be achieved easily with standard commands,
except perhaps the interrupt protection.

Johannes

From michael.stroeder at inka.de  Fri Mar  3 09:52:47 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 03 Mar 2000 15:52:47 +0100
Subject: cgi, python and forms
References: <2WPv4.367$H95.27613@wagner.videotron.net>
Message-ID: <38BFD1BF.83C6D4C5@inka.de>

Akhar wrote:
> 
> I woud like to know how can I tell my python cgi script to differenciate
> these two:
> 
> 
> I am using the cgi module but can't find any info on these.

By using  and
reading the value of field SUBMITFIELD.

Ciao, Michael.

From pinard at IRO.UMontreal.CA  Thu Mar 30 10:31:09 2000
From: pinard at IRO.UMontreal.CA (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 30 Mar 2000 10:31:09 -0500
Subject: Incompatibility between socket and httplib in python CVS build.
In-Reply-To: "Fredrik Lundh"'s message of "Wed, 29 Mar 2000 19:23:04 GMT"
References:    
Message-ID: 

"Fredrik Lundh"  ?crit:

> [...] anyone insisting on using the bleeding edge development CVS version
> (instead of a stable version) can save a lot of time by sub-scribing to
> the python-cvs mailing list.  in that way, you'll *know* what's changed
> in the bleeding edge version.

Just to put things straight, I'm not "insisting" to use it.  Whatever
in production for me still uses 1.5.2.  But there is no chance that I
contribute to the development effort (by reporting bugs, essentially)
if I do not try using 1.5.2+ in some limited environments, where I know
it is bearable that things can break.

But your suggestion is a good idea.  I'm not much of a CVS user, that's
why I did not think of subscribing to anything having `-cvs' in it :-).
But if modifications are described in there, it could be useful indeed.

Thanks for your feedback.

Oh, by the way, is there a specific mailing list to use (and to be) for
discussing findings if any, or is `python-list' just acceptable?

I took a slightly better look at Unicode implementation in 1.5.2+, and feel
quite satisfied by it.  Some of the things are left to users to handle
(like dynamic combining, as per one of my previous messages), but Python
is such that this can be worked out without too much pain.  I now have
the feeling it would be easy to bring my Unicode applications to Python.

Looking at Lib/encodings/, I see that a few are missing from what the
official `recode' has, and would feel like providing now, or soon, directly
in Python, all those encodings which interface naturally with Unicode
(that is most of them, by the way).  In fact, if I go further and consider
the `recode' in pretest, all Asian encodings are missing, which are many.
But since `recode' relies on `libiconv' for those, I believe it would be
much more natural to directly interface Python and `libiconv'.

I'm also seriously pondering the idea of developing a Python implementation
of `recode' that I could use for faster prototyping, and which I would
even include within the `recode' distribution, next to all the C code.
Prototyping, while using C uniquely, requires a lot of energy, and there
are many new ideas to explore.

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

From reic0024 at bulldog5.d.umn.edu  Wed Mar 15 13:51:53 2000
From: reic0024 at bulldog5.d.umn.edu (Aaron J Reichow)
Date: Wed, 15 Mar 2000 12:51:53 -0600
Subject: wxPython for Mac
In-Reply-To: 
References:   
Message-ID: 

On Wed, 15 Mar 2000, Brett Tribble wrote:

> While I assume that this work is directed at Mac OS 8.x 9.x, what happens
> with OS X and its BSD kernal? Will we see a port of the *nix variant? I
> suppose this depends on the implementation of wxWindows itself, but has
> there been any movement on this?

I personally have no clue, but getting it going under Carbon shouldn't be
a godlike feat.

Aaron

From loewis at informatik.hu-berlin.de  Tue Mar 21 14:19:22 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 21 Mar 2000 20:19:22 +0100
Subject: "builtin" methods
References: <200003152307.AAA09960@dionysus.fw.cuci.nl>
Message-ID: 

"Hans Nowak"  writes:

> Paragraphs 3.3.1 to 3.3.6 have all these methods. (Well, I think it's all 
> of them... :)

I noticed the "special" names used by pickle are not mentioned
(__getstate__, __setstate__, __getinitargs__) are not mentioned. But
then, they aren't really that special, either :-)

Regards,
Martin

From aahz at netcom.com  Wed Mar 15 12:05:08 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Mar 2000 17:05:08 GMT
Subject: Type testing
References: <6D8A17398E28D3119F860090274DD7DB498597@pces.cadlab.it>
Message-ID: <8aofs4$io3$1@nntp6.atl.mindspring.net>

In article <6D8A17398E28D3119F860090274DD7DB498597 at pces.cadlab.it>,
Alessandro Bottoni   wrote:
>
>Is it possible to test for the type of an object (I mean, a built-in object:
>a digit, a string, etc...)?

Use the built-in type() function and the types module.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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

From nospam at noreply.com  Fri Mar 31 01:13:21 2000
From: nospam at noreply.com (Brian)
Date: Fri, 31 Mar 2000 06:13:21 GMT
Subject: Advice requested: GUI project beginning
References: <8c0dpv$djc$1@nntp6.u.washington.edu> <38e3fa10.298739985@news.mr.net>
Message-ID: <5mXE4.9512$lX4.216839@paloalto-snr1.gtei.net>

> >- Tkinter is flaky on the Mac. For instance it doesn't run in the IDL
> >and it doesn't fully quit.
>
>  Ain't cross-platform GUI a bitch?

So when is Python going to have a GUI library as tightly bound with it as Tk
is to Tcl and Swing/AWT is to Java?  Is wxWindows heading in that direction?
Seems like having to distribute Tcl/Tk with your Python program to have a
GUI is not the favorite thing for most Python developers.  Unfortunately Tk
is one of the most portable toolkits ... anyone know if Guido has expressed
any interest in wxWindows being the standard Python GUI and distribute
wxWindows with Python?

 wrote in message news:38e3fa10.298739985 at news.mr.net...
> On Thu, 30 Mar 2000 12:35:09 -0800, "Russell E. Owen"
>  wrote:
>
> >I am starting work on a telescope remote control user interface. It's a
> >big program with lots of windows and controls and display graphics
> >supporting the telescope and numerous instruments.
>
> Sounds like a fun project!
>
> >Python is a leading contender because I'd like a system that offers:
> >- rapid development/change/test
> >- robust
> >- safe (automatic garbage collection and bounds checking)
> >- runs under both unix and MacOS, if possible (windows wouldn't hurt,
> >either)
> >- will probably be around in 10 years
>
> Yup...Python is guilty on all counts.
>
> >
> >My development platform is Mac. I'd like to stick to that, but fear it's
> >not reasonable with Python because:
> >- No threads on the Mac. (yet)
> >- Tkinter is flaky on the Mac. For instance it doesn't run in the IDL
> >and it doesn't fully quit.
>
>  Ain't cross-platform GUI a bitch?
>
> >
> >Any comments? If I do switch, is Python development or support notably
> >superior under unix or windows -- aside from OS preferences? (Please no
> >OS flame wars.)
>
> If switching is an option it would certainly give you better Python
> support _now_. As for your platform question, various flavors of Unix
> seem to be the best platform, but that distinction is getting to be
> less and less relevant. The win32 & com extensions by Mark Hammond, et
> al, work wonderfully.
>
> *nix support still has the edge, though. Hell, it's a standard part of
> many Linux distros, and easilly added to FreeBSD via the ports
> collectioin.
>
> In addition, Linux might possibly give you the option of staying on
> your current machine(s) with the use of Yellow Dog Linux or other
> LinuxPPC ports. I believe NetBSD has a PPC port as well,  and both
> have 68K ports if you have an older Mac.
>
> >What Python windowing environment do you recommend for a new project?
> >wxPython? Tkinter? Java (via JPython)? (Is there anything else?). I'd
> >prefer something that is easy to use and scalable.
>
> I've mostly used Tkinter, because there is a decent knowledge base out
> there, given its de facto standard status. The main platforms I use
> (SUN, Win32 & Linux) all either come with tcl/tk as an (optional) part
> of the OS distribution or, in the case of Win32, as part of the Python
> install. Makes for one less thing I have to get the user to install to
> use my code.
>
> I've grabbed copies of wxPython/wxWindows and FLTK with its Python
> interface. You might want to check them out as well.
>
> As you've noticed, cross platform GUI isn't very smooth today. I'd
> submit that this is true using *any* language system today, thought.
> Java/Swing is often touted as the 'silver bullet', but experience here
> at work tells me that it ain't. It does have the advantage of using
> the vast Java hype flow as lubricant for getting the infrastructure
> for your program installed on your customer's computers, tho.
>
>
>
> >-- Russell
> >
> >P.S. I am also considering Smalltalk (superb language, but ugly GUI and
> >hard to sell to folks used to C/tcl/perl) and Java (reasonable Mac
> >support and long-term prospects, but not very malleable). Are there any
> >other languages I should consider?
>
> No opinion on that here. Of late the only languages I tend to consider
> for use are Perl if I need a quick one-off, C if I need to hit the
> metal or for performance reasons, Java if I need to fix or modify a
> project that was already written in Java and Python for everything
> else.
>
> --
> "If your map and the terrain differ...trust the terrain." -- unknown
>

From python-list at teleo.net  Sat Mar  4 17:54:01 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Sat, 4 Mar 2000 14:54:01 -0800
Subject: HELP: restore my faith in Python
In-Reply-To: 
References: 
Message-ID: <00030414581106.02808@quadra.teleo.net>

[Moshe Zadka, on Sat, 04 Mar 2000]

:: until-the-timbot-implements-slash-arithmetic-for-intel-cpus-ly y'rs, Z.

That'll never happen ... doesn't fit the bot's oft-expressed Criterion
of Uselessness. 

From tim_one at email.msn.com  Thu Mar  2 00:06:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 2 Mar 2000 00:06:10 -0500
Subject: Splitting comp.lang.python
In-Reply-To: <836DAFF48BB74487.F181DAFF1B4CD5BF.DCA44F8A445A3160@lp.airnews.net>
Message-ID: <000f01bf8405$07e1ba80$732d153f@tim>

[Manuel Gutierrez Algaba]
> comp.lang.python.tim.peters  where Tim Peters would fight
> against ruby, viper, Haskell ... whatever, where most syntax
			.
[Cameron Laird]
> That clever rascal!  Here I thought he was
> fighting *for* all those things.

It does rather dull the point of Manuel's post:  since he has clearly
already learned to filter out what I post, it wouldn't do him much actual
good to banish me to e.g. comp.lang.python.bots (I see no reason to banish
Tim Peters to his own little corner!  Uncle Timmy is a generally amiable and
helpful fellow.  Not a bad programmer, either, to judge from the flawless
reliablity of my self-satisfaction module.).

if-it-weren't-for-me-instead-of-ints-python-would-still-be-using-
    guido's-original-dutch-roman-numerals-ly y'rs  - tim

From fredrik at pythonware.com  Tue Mar 14 08:10:16 2000
From: fredrik at pythonware.com (fredrik at pythonware.com)
Date: Tue, 14 Mar 2000 13:10:16 GMT
Subject: Q. on Introduction to Tkinter
References: <38CE331C.F5AD9029@pivot.net>
Message-ID: <8aldnm$lcs$1@nnrp1.deja.com>

Wayne  wrote:
> "Note that the 'name' option can only be set when
> the widget is created."  Does "name" refer to the
> 'option = value" pair?

"the 'name' option" refers to a special option called
"name", which is described back in chapter three (see
"more on widget names").

> Second,  If one does not define an "option = value"
> at creation of the widget, does that mean you can
> not use config () to set the value later?

nope.  the "name" option is special; it can *only*
be specified when you create the widget.  all other
options can be modified at any time.

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

From gerrit at nl.linux.org  Wed Mar  1 11:48:08 2000
From: gerrit at nl.linux.org (Gerrit Holl)
Date: Wed, 1 Mar 2000 17:48:08 +0100
Subject: Splitting comp.lang.python
In-Reply-To: <09ED98078750F0D8.61776C14C63DBD6A.80A949C9746AAA0B@lp.airnews.net>; from claird@starbase.neosoft.com on Wed, Mar 01, 2000 at 09:51:17AM -0600
References:  <09ED98078750F0D8.61776C14C63DBD6A.80A949C9746AAA0B@lp.airnews.net>
Message-ID: <20000301174808.A10405@nl.linux.org>

> I'm not so sanguine about c.l.p.platform.  What are
> examples of postings you anticipate there?

COM, Windows programming using win32api, Unic programming such as
processes and sockets, Mac programming with the Macintosh
extension modules.

regards,
Gerrit.

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

From phd at phd.russ.ru  Tue Mar  7 05:15:43 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 7 Mar 2000 10:15:43 +0000 (GMT)
Subject: FTP hangs with urllib.urlretrieve
Message-ID: 

Hello!

   I am developing/running URL checker (sorry, yet another one :). And
found that I have consistent hangings on some FTP URLs.

   Here is test program I am using to retest my results - just URLopener
and urlretrieve, nothing magical; python is stock 1.5.2, hangs on all
platforms I am using - Pentium Linux, Sparc Solaris, Pentium FreeBSD:

----------
#! /usr/local/bin/python -O

import sys, urllib
urllib._urlopener = urllib.URLopener()

# Some sites allow only Mozilla-compatible browsers; way to stop robots?
server_version = "Mozilla/3.0 (compatible; Python-urllib/%s)" % urllib.__version__
urllib._urlopener.addheaders[0] = ('User-agent', server_version)

url = sys.argv[1]
print "Testing", url

try:
   fname, headers = urllib.urlretrieve(url)
   print fname
   print headers

except Exception, msg:
   print msg
   import traceback; traceback.print_exc()
----------

   The program always hangs on some (but not all) FTP URLs. One is
well-known for Python community: ftp://starship.python.net/pub/crew/jam/ :)
   Others are:

ftp://ftp.sai.msu.su/
ftp://ftp.radio-msu.net/
ftp://ftp.relcom.ru/pub/
ftp://ftp.sunet.se/pub/
ftp://ftp.cs.wisc.edu/
ftp://ftp.cert.org/pub/

   I've tested these sites with FTP clients (Midnight Commander, Netscape
Navigator, ncftp) - all are accessible. It seems like a bug or bugs in
ftplib.
   The first two are very near to me in terms of Internet distance (hop
counts), so timeouts should not be a problem. These are sites near my ISP
(Radio MSU, in Moscow State University).

   Can anyone with better knowledge of FTP protocol look and help? Does
latest python (from CVS) perform better (if anyone willing to test)?

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

From corg at copernic.com  Mon Mar  6 17:03:38 2000
From: corg at copernic.com (Gaetan Corneau)
Date: Mon, 6 Mar 2000 17:03:38 -0500 
Subject: Self Nanny
Message-ID: 

> still-no-chance-of-changing-ly y'rs

Prrrrrrrllllzzztbll!  :P
______________________________________________________
   Gaetan Corneau
   Software Developer 
   Copernic Technologies inc. 
   http://www.copernic.com
   E-mail: corg at copernic.com                         
______________________________________________________
"Tu l'as trop ?cras?, C?sar, ce Port Salut"

-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/B/MU d- s+:++ a C++ 
UL+ P--- W+ N- K- W++ t-- 
!5 X- R+ tv-- b++ DI++ 
G e++ h---- r+++ y++++
------END GEEK CODE BLOCK------

From eq3pvl at eq.uc.pt  Tue Mar 21 12:31:23 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Tue, 21 Mar 2000 17:31:23 +0000
Subject: Simulating a Web Browser in Python
References: 
Message-ID: <38D7B1EB.627FD865@eq.uc.pt>

Brian Fritz wrote:

> After checking through the standard library and a couple of web sites I've
> not found (or maybe I didn't notice) a decent library that would allow me
> to write a simulated web browser.
>
> I need to be able to read and write into edit controls and accept and
> return cookies.
>

Probably you know it, but just in case you didn't noticed, GvR wrote
a web browser in Python. It's called Grail and you can get it from
www.python.org

From python at rose164.wuh.wustl.edu  Fri Mar 31 23:52:07 2000
From: python at rose164.wuh.wustl.edu (David Fisher)
Date: Fri, 31 Mar 2000 22:52:07 -0600
Subject: Python Threading and sys.exit() codes
References: <8c2u4v$cft$1@nnrp1.deja.com>
Message-ID: <00b501bf9b98$73298b60$cbe9093f@spkydomain>

----- Original Message -----
From: 
Newsgroups: comp.lang.python
To: 
Sent: Friday, March 31, 2000 1:26 PM
Subject: Python Threading and sys.exit() codes

> My problem is this: I want to run this program as a scheduled task
> under a job scheduler (currently testing cronDsys from #ifdef). The
> single-threaded version of the program posts an appropriate code to the
> environment via sys.exit(). The job scheduler traps this and takes
> conditional action. So far, so good.
> In the multi-threaded version of the program, the exit codes are not
> getting posted and the main process is not terminating cleanly, at
> least from the perspective of the job scheduler -- it appears to run
> until manually terminated from the scheduler control console although
> the NT command window disappears from the target desktop as expected.

Well for one thing, sys.exit() doesn't do what you think it does.
Sys.exit() raises an exception SystemExit which if not caught will cause the
thread to exit silently.  Read the docs for thread which threading is built
on.  I don't know of a way to catch an exception that another thread raises.
Doesn't mean there isn't one, just that I don't know one.

Another thing you look at is the Queue module.  That's what got my attention
because you are setting a global lock and poping stuff off of a fifo stack,
and Queue does all that for you. Here's some code:

import Queue
import thread    #yes, I know I'm supposed to use threading

def worker(jobqueue):
    while 1:
        job = jobqueue.get()    # this blocks until a job is available
        print job    # or whatever

def main(joblist):
    jq = Queue.Queue(0)
    for i in range(10):
        thread.start_new_thread(worker,(jq,))
    for job in joblist:
        jq.put(job)

There are lots of things that could be prettified in this code, but I hope
it gives you some ideas.  You might try a second Queue to gather return
information.

Good luck,
David

From greglandrum at earthlink.net  Sun Mar 12 16:31:48 2000
From: greglandrum at earthlink.net (greg Landrum)
Date: Sun, 12 Mar 2000 21:31:48 GMT
Subject: PIL and Windows Installations
Message-ID: <38CC0DA9.5EF572BD@earthlink.net>

Well, I've just gotten bit (again) by the "installing PIL under Windows"
problem.  I did a bit of Deja-News searching and discovered a couple
threads from late last year which provided all the answers I needed.

In the hopes of saving others from the frustration of trying to get PIL
(which is so great that it's a MUST HAVE) installed on Windows, I'm posting
my solution.  This is based upon the most recent binary PIL distribution:
http://www.pythonware.com/downloads/pil-win32-991101.zip
and *requires* that Tkinter is installed on your machine.  It comes by
default with the windows python distributions, so this shouldn't be too
much of a problem.  The binary distribution of PIL appears to require
Tcl/Tk anyway.

Step 1:
Unzip the archive (it doesn't matter where), this will create a directory
py152.
Copy the files/folders from py152 to your Python installation directory
(c:\program files\python by default).

That handles the installation.  Now try running one of the PIL sample
programs (I recommend Scripts/viewer.py).  If you get an error, then you'll
need to move to Step 2:
Edit the file c:\program files\python\pil\Image.py
Move down to around line 47, where you'll find:
    import _imaging
change this to:
    import sys
    if sys.platform == "win32":
	import FixTk # Attempt to configure Tcl/Tk without requiring PATH
    import _imaging
Save the file and try running Scripts/viewer.py again.  It ought to work.

I hope this is helpful.
-greg

-- 

greg Landrum (greglandrum at earthlink.net)
Software Carpenter/Computational Chemist

From alex at somewhere.round.here  Wed Mar  1 12:31:10 2000
From: alex at somewhere.round.here (Alex)
Date: 01 Mar 2000 12:31:10 -0500
Subject: I wasn't complaining! (Was: will python 3000 break my code?)
References: <87tkoi$hql$1@nntp1.atl.mindspring.net> <08Zu4.6976$3z.60274@newscontent-01.sprint.ca> 
Message-ID: 

Fredrik and Gerrit, thanks for the info.  I didn't realize it would be a
type error.  I wasn't complaining, though, just giving an example of why
I think some rewriting will be necessary.  And I'm not suprised that it
turns out to be bad form.  I'm not a professional coder, and I'm
learning those sorts of aesthetics as I go along.

Alex.

From markx.daley at intel.com  Tue Mar 21 18:47:31 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Tue, 21 Mar 2000 15:47:31 -0800
Subject: Re-inserting data
Message-ID: 

The only problem with that is I don't have any control over whether or not
tuples are used.  I am using SQL to query data from a server and this is the
format it uses to return the data.  Worse yet, the tuples are not strictly
strings, but dates and numbers, also.  I am trying to pull a date from each
tuple, convert it, and return it to the tuple before it gets written to a
file.  Any ideas?

- Mark

-----Original Message-----
From: Alex [mailto:alex at somewhere.round.here]
Sent: Tuesday, March 21, 2000 1:46 PM
To: python-list at python.org
Subject: Re: Re-inserting data

> Some clarification is necessary as to my earlier message.  Here is what
the
> nesting is like.
> 
> 	  result = [(a, b, c, d, e), (a1, b1, c1, d1, e1),...]

Oh, I guess you are having trouble changing the elements because they
are tuples.  Try using lists instead.  Then you can do things like

for item in times:
    time2 = item[-1]
    time2 = munged_time (time2)
    item[-1] = time2

Alex.

From jstok at bluedog.apana.org.au  Sun Mar 26 08:41:46 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Sun, 26 Mar 2000 23:41:46 +1000
Subject: Internals of interning strings
References:   
Message-ID: 

Michael Hudson wrote in message ...
>Bernhard Herzog  writes:
>
>> That's something I've also wondered about. How likely is it that one and
>> the same string object will be interned several times? That seems to be
>> the only thing that's really optimized.
>
>If you have this in one module:
>
>import sys,os
>
>and in another
>
>import sys,string
>
>then the string "sys" will get interned twice during compilation.  So
>to answer your question, pretty likely.

But those are *different* string objects.  The optimization we're discussing
affects only *the same* string object being interned several times.

From claird at starbase.neosoft.com  Tue Mar  7 12:25:34 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 7 Mar 2000 11:25:34 -0600
Subject: Python performance
References: 
Message-ID: <67980F7CE7203C29.8576A630FEE7042D.1147C70DD44393FD@lp.airnews.net>

In article , Chris Ryland  wrote:
>The recent Python-for-Lisp-ers note and other recent notes have brought up
>Python performance.
>
>I'm just curious (being an old language hacker for 30+ years but new to
>Python): why is Python performance ultimately any worse than Lisp?
>
>Is it because there's only been one major implementation effort so far, and
>so people haven't had a chance to learn the techniques required to make it
>run fast, while Lisp has had 30+ years to mature?
>
>What's so different about compiling Python vs. Lisp? Some Lisp compilers
>have gotten quite good over the years.
			.
			.
			.
There have, in fact, been several Python implementations already .
They look "major" to me, although I can imagine you might be setting the
boundaries in a different place from me.

Performance is an explicit motivation for most of the alternative
Pythons.

Guido regularly professes a lack of interest in performance.  At
least part of the time, he's using that to attract attention to his
profound belief in the importance of correctness; he works very, very
hard to get Python "right", and, from what I can see, has faith that
performance will follow and/or can be left to others.

One difference between Python and LISP is that with Python, as with
most its contemporaries, one has more of a feeling that the performance-
pertinent action is happening in the libraries rather than the language.

My conclusion:  Python performance can be (roughly) as good as LISP's.
It is, already, for the domains that matter most to the big players of
the Python world.
-- 

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

From tom at parlant.com  Tue Mar 14 17:25:03 2000
From: tom at parlant.com (Thomas Lane)
Date: Tue, 14 Mar 2000 15:25:03 -0700
Subject: Return statement question
References: 
Message-ID: <38CEBC3F.9CAC0A31@parlant.com>

Are you sure func3 was never executed? I cut your code (modified to print stuff
to the screen) and it definitely ran func3.

Here's the code I ran:

def func1():
    print 'func1'

def func2():
    print 'func2'
    return 1

def func3(sv):
    print 'func3'

func1()
val= func2()
func3(val)

The output (as expected) was:

func1
func2
func3

-Tom

Gregoire Welraeds wrote:
> 
> 
> In this cas func3 is never executed...
> 
> why ?
>

From cgw at fnal.gov  Mon Mar 27 12:34:20 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: Mon, 27 Mar 2000 11:34:20 -0600 (CST)
Subject: copy directory from a to b, how?
Message-ID: <14559.39836.789275.35413@buffalo.fnal.gov>

Use the "copytree" function in the (oft-overlooked) "shutil" module.

No reason to resort to os.system'ing some non-portable command.

From tiddlerdeja at my-deja.com  Tue Mar 28 06:28:08 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Tue, 28 Mar 2000 11:28:08 GMT
Subject: copy directory from a to b, how?
References: <8bne0u$jlu$1@nnrp1.deja.com> <8bnfof$le9$1@nnrp1.deja.com> <38DFA2DD.4C2EE42C@bellatlantic.net>
Message-ID: <8bq4vu$jk6$1@nnrp1.deja.com>

Why use os.path.walk() to delete files? Am I missing something here?

Would it not just be easier to use shutl.rmtree() ?

Also, I hadn't thought that there may be a permission/sharing issue
with c:\inetpub\wwwroot. I just didn't want to remove it as it may take
some time if wwwroot stores a large number of files. I also felt it to
be a bit unknerving that I would be remving the root directory of my
web server.

In article <38DFA2DD.4C2EE42C at bellatlantic.net>,
sholden at BellAtlantic.net wrote:
> davec999 at my-deja.com wrote:
> >
> > In article <8bne0u$jlu$1 at nnrp1.deja.com>,
> > tiddlerdeja at my-deja.com wrote:
> > > I'd like to copy a directory from c:/Inetpub/staging to
> > > c:/Inetpub/wwwroot.
> > >
> > > How do I do this in Python? I know that I have shutil.copytree()
at my
> > > disposal, but this requires that the destination directory doesn't
> > > exist. I don't want to remove c:/Inetpub/wwwroot so that I can
perform
> > > the copy.
> > >
> > > What is the best way to do this? Any help appreciated.
> > Is is essential to do it in Python? I would use the line:
> > os.system("xcopy c:\\inetpub\\staging c:\\inetpub\\wwwroot\\ /E")
> > Remember the backslashes in the correct place (at the end of the
second
> > argument and not at the end of the first) and you should be fine.
> > Of course, if you want it to work on multiple OSs this will
obviously
> > not work. :)
> > Dave
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
> Also be aware of the various nastinesses that NT (this *is* an NT
problem,
> right?) will put in your way: the C:\Inetpub\wwwroot directory may
well
> be impossible to rename or delete, which I suspect is the motivation
> behind the original question.
>
> If you were to stop the WWW publishing service you should find that
you
> can delete all files and subdirectories (use os.path.walk to find
them)
> from the existing web, and replace the old web with a new one either
by\
> using walk again or by the effective method suggested above.
>
> regards
> Steve
> --
> "If computing ever stops being fun, I'll stop doing it"
>

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

From akuchlin at mems-exchange.org  Tue Mar  7 13:57:02 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 07 Mar 2000 13:57:02 -0500
Subject: delete comments with re
References: <8a33u1$mir$1@nnrp1.deja.com>
Message-ID: <3dya7uwqv5.fsf@amarok.cnri.reston.va.us>

laurent8 at sxb.bsf.alcatel.fr writes:
> with this regular expression I've got
> >>> r="/\*.*\*/"
> >>> re.sub(r,'',t)
> 'a=b;  class foobar {}'

Instead of .* in the middle, try a non-greedy match with .*? :

pat = re.compile('/[*] .*? [*]/', re.VERBOSE)
t='a=b;  ...'
print repr(pat.sub('', t))

This outputs 'a=b; o=1; j=1; class foobar {}'.  

Comments in C don't nest, luckily, so the comment will end at the
first */ encountered after the comment starts.  (The comment in
'/*a/*b*/c*/' is '/*a/*b*/', that is.)  

	.* will run to the end of the string and then backtrack until
	it finds a */; it matches as *much* as possible.

	.*? will repeatedly advance by one character and try to
	match */ at every point, so it matches as *little* as possible.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
The criterion of simplicity is not necessarily based on the speed of the
algorithm or in its complexity in serial computers.
    -- Armand De Callatay, _Natural and Artificial Intelligence_

From embed at geocities.com  Tue Mar  7 10:35:18 2000
From: embed at geocities.com (Warren Postma)
Date: Tue, 7 Mar 2000 10:35:18 -0500
Subject: Python/MySQL Where do "SELECT ..." results go?
References: <38C4F789.71ECCB04@sprint.ca>
Message-ID: 

> >>>curs.execute("Select * from test;")

Try dir(curs) and see if there are next, prev, first, last, etc methods.
Always a good idea with an object to ask it "what can you do for me"...
dir(obj) is how you ask.

Warren

From culliton at clark.net  Tue Mar 14 09:44:13 2000
From: culliton at clark.net (Tom Culliton)
Date: Tue, 14 Mar 2000 14:44:13 GMT
Subject: [] vs. ()
References:  
Message-ID: <1fsz4.793$u%1.20182@iad-read.news.verio.net>

You're missing the critical mutable/immutable distinction.  Tuples can
be used as dictionary keys since they are immutable.  They can also be
used to get pass by value semantics (even though values are actually
always passed by reference) when calling functions since you know that
they can't be modified by the called function.  Not to mention that
they're somewhat more efficient.  It seems like I'm forgetting another
use for their immutablility, but thats already more than enough
rationale for their existance.

In article ,
Jason Stokes  wrote:
>
>Michal Bozon wrote in message ...
>>__Hi__
>>Could anyone tell me the difference between tuples and lists? (except from
>>that a list has few methods (append, extend, remove...) and tuple
>>doesn't.)
>
>Well, that's the difference.  In an object oriented language, objects differ
>by their interfaces.  Tuples are fixed size ordered collections of objects,
>and lists are variable sized ordered collections.

From sbahling at pitnet.net  Sat Mar  4 22:08:55 2000
From: sbahling at pitnet.net (Scott Bahling)
Date: Sat, 04 Mar 2000 21:08:55 -0600
Subject: When to use classes?
Message-ID: <38C1CFC7.DD9F0918@pitnet.net>

I am just starting to use python and I love it so far.

I have a couple of philosophical questions about OOP programming with
python. 

First: When and when not to create a new class? An example would be if I
have a class for a Polygon, I need to store a set of points defining the
polygon object. One method would be to store a list of tuples containing
the polygon's points. If I want to access the individual x and y coords
of each point, I end up with code that looks like:

		firstxcoord = poly.points[0][0]
		firstycoord = poly.points[0][1]

Or I could create a Point class and use a list of them in the Polygon
class. Then I would have code that looks like:

		firstxcoord = poly.points[0].x
		firstycoord = poly.points[0].y

The second piece of code is easier to read. But lets say that this Point
class has no other purpose but to contain two numerical values; is the
use of a Point class overkill? Is there a 'rule of thumb' for when to
create a new class?

Second: What is the general consensus about accessing instance variables
directly like myobj.x=10 vs. through methods like myobj.setx(10). The
first method is quicker/easier to code, but is it "bad" coding?  Does
anyone even care?

I know these are general OOP questions. I am asking in case there are
aspects of the python language that would help to determine when and
when not to do these things.

-thanks

scott

From tuttledon at hotmail.com  Wed Mar  8 22:42:02 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 09 Mar 2000 03:42:02 GMT
Subject: Is it Python or is it C ? (long)
References: <0%hx4.3000$N3.36008@typhoon.southeast.rr.com> <38C5C0D5.6ED417DE@houston.rr.com>  
Message-ID: 

[Fredrik Lundh]
> summing up, the more you do in C, the faster your
> program runs.  but C doesn't help if you use a lousy
> algorithm... [2]
>
> or in other words, if your Python program doesn't
> run  fast enough, use the profiler.  or ask for help
> on comp.lang.python!

Whew!  I hope you get to use all this good info in your next version of (the
eff-bot guide to) the standard python library. (Which I own, btw. Free and
clear with no lean against it!  ;-)

I did not know about the profiler. Until tonight I just thought it was a bad
TV show .

Thanks to everyone,
Don

From punck at SpamIsBadPenguinPowered.com  Sun Mar 12 17:28:32 2000
From: punck at SpamIsBadPenguinPowered.com (Chris Armstrong)
Date: Sun, 12 Mar 2000 17:28:32 -0500
Subject: import & __import__
Message-ID: <2MUy4.5284$bt3.17551@newsfeed.slurp.net>

Hi all. I'm pretty much a newbie to Python, and I have a question about 
import vs __import__. I'm implementing dynamically imported modules in
a game I'm writing, and I need to use __import__ to do this. One thing I
noticed is that __import__ doesn't seem to execute the modules that it
imports, while import does.. The docs for __import__ don't mention anything
about this, and I'm just curious about it. I tested this by making a simple
module that just has `` print "Hi." '' in it. `` import modulename '' caused
it to print out "Hi." while `` foobar = __import__("modulename") '' didn't.
This isn't really important to me, I was just curious about it.

-- 
Chris Armstrong
http://drol.sourceforge.net

Geek code version 3.12
GCS d? s:-- a--- C+++ UL++++>++++ P++ L+++ E--- W++ N+ o? K- w-- 
O-- !M V PS+(+++) PE? Y+(--) PGP(--) t+ 5 X+ R tv+ b+>++ DI+ D+++ 
G e- h!(--) r- y?
Go to www.geekcode.com to find out what on earth this is.

From Alessandro.Bottoni at think3.com  Wed Mar  8 04:43:02 2000
From: Alessandro.Bottoni at think3.com (Alessandro Bottoni)
Date: Wed, 8 Mar 2000 10:43:02 +0100 
Subject: Python 1.6 and Python 3000's new features?
Message-ID: <6D8A17398E28D3119F860090274DD7DB498538@pces.cadlab.it>

I just read the "Frankly Speaking" column at O'Reilly's site (
http://www.oreilly.com/frank/ ), and I was quite chocked by two of the news:

[Python 3000] The contents of this version are still under discussion, but
will almost certainly be backward-incompatible. 
[Python 1.6] In addition, Guido will include some existing bug fixes and
probably Fredrik Lundh's new regular expressions engine

Why Python 3000 should be backward-incompatible? Is it possible to have a
"big-picture" of this major version? 

Why new regular expressions? What was wrong with the "old" re module?

(I'm NOT criticizing anybody. I'm just curious...)

Thanks

--------------------------------
Alessandro Bottoni
(Alessandro.Bottoni at Think3.com)
(alessandro.bottoni at libero.it)
Web Programmer
Think3 inc.
(www.think3.com)

From mindlace at imeme.net  Thu Mar  2 18:06:10 2000
From: mindlace at imeme.net (ethan mindlace fremen)
Date: Thu, 02 Mar 2000 23:06:10 +0000
Subject: compiling python w/large files
Message-ID: <38BEF3E2.F3782546@imeme.net>

I'm trying to compile python with large files on Linux, using kernel
2.3.47 and Libc6 2.1.13-5.

http://www.python.org/doc/lib/posix-large-files.html

tells me how to do it for , but not for Linux.

If i just ./configure, large file support doesn't happen.

Thanks in advance,

Ethan Fremen
http://imeme.net

From thantos at chancel.org  Wed Mar  1 18:11:28 2000
From: thantos at chancel.org (Alexander Williams)
Date: Wed, 01 Mar 2000 23:11:28 GMT
Subject: Python for non-programmers
References:   
Message-ID: 

Gentlemen, there is a /reason/ that Introduction to Computing 101 and
Basics of Programming 102 take half a year to introduce together, and
its not just because of the quarter system in the local college.  Many
people need that much time to pick up the lexicon and assumptive web
of associations that we, as programmers, make use of every single
day.  Like any other field with its own symbology and culture, it
takes time to get the grounding in the true basics.

Think of the field of computing as a culture unto itself with many
different regional languages and you'll be far closer to understanding
/why/ this difficulty is as it is.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins

From hernanf at my-deja.com  Fri Mar  3 11:41:48 2000
From: hernanf at my-deja.com (Hernan M. Foffani)
Date: Fri, 03 Mar 2000 16:41:48 GMT
Subject: has anyone seen comp.lang.python.announce?
References: 
Message-ID: <89oq0c$m5l$1@nnrp1.deja.com>

In article ,
  "Fredrik Lundh"  wrote:
> on all newsservers I have access to, c.l.py.announce has
> been empty for about as long as I remember,
> ....
> and what's worse, it doesn't show up on deja.com either.
>
In my My-Deja the last one is your Dr. Dobbs dated March 1st.
Something happened before though: On Feb 25th there were 3 different
Dr. Dobbs articles and one on Feb 6.

> can this be fixed, or should the newsgroup be scrapped?
> I find it hard to believe that the python universe doesn't
> produce enough new stuff to keep the newsgroup alive...
Warm winter up there? Too many news to read at c.l.p now? :-)

Regards,
-H.

--
Hern?n Mart?nez Foffani
hernanf at my-deja.com

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

From markx.daley at intel.com  Thu Mar 23 12:22:09 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Thu, 23 Mar 2000 09:22:09 -0800
Subject: List limitations?
Message-ID: 

If I connect to server2 alone, it works properly.  If I connect to server2
first, then server 1, server2 works and server1 returns strange characters
for familynames.  I am using mxODBC, at least, as far as  understand it from
the notes I downloaded.
-Mark

-----Original Message-----
From: aahz at netcom.com [mailto:aahz at netcom.com]
Sent: Thursday, March 23, 2000 8:44 AM
To: python-list at python.org
Subject: Re: List limitations?

In article ,
Daley, MarkX  wrote:
>
>What this does is connect to a server and pull data from it based on each
>family name in the familyname list, using an SQL query.  Conceivably, this
>code should connect to server1 and pull data for all familynames and write
>that data to a file, server_familyname_date.  Then it should connect to
>server2 and do the same.  However, when it connects to server 2, it isn't
>pulling the familynames from the familyname list correctly.  Instead, it is
>pulling single character strings, like a, b, e, g, h, r, S (these are the
>actual strings, which may or may not have some bearing on one of my
>familyname entries, I haven't been able to verify).  Has anyone else seen
>problem remotely related to this?  Any help is appreciated.

Questions/suggestions:

Have you tried connecting just to server2?
What happens if you switch server2 and server1?
Have you considered using mxODBC?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.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 tismer at tismer.com  Tue Mar 28 15:33:07 2000
From: tismer at tismer.com (Christian Tismer)
Date: Tue, 28 Mar 2000 22:33:07 +0200
Subject: Stackless 1.1 Homepage: www.stackless.com
Message-ID: <38E11703.C98FA418@tismer.com>

    *** ANNOUNCING *** ANNOUNCING *** ANNOUNCING ***

The Stackless Python Homepage has its own website:

             http://www.stackless.com

Stackless Python 1.1 is about to be published there, soon.

Version 1.1 is a major overhaul of speed and versatility.

Stackless
Python 1.1 + Continuations 0.8 - a version of Python 1.5.2 that
does not need space on the C stack, and first-class callable
continuation objects for Python.  (28-Mar-2000)

Christian Tismer
Mission Impossible 5oftware Team

From nospam.newton at gmx.li  Mon Mar 13 03:09:41 2000
From: nospam.newton at gmx.li (Philip 'Yes, that's my address' Newton)
Date: Mon, 13 Mar 2000 08:09:41 GMT
Subject: Is it Python or is it C ?
References: <0%hx4.3000$N3.36008@typhoon.southeast.rr.com> <38C5C0D5.6ED417DE@houston.rr.com>  <004401bf8910$a42da4c0$01ffffc0@worldnet.att.net> <38C69E6D.CD268578@bellatlantic.net> <38C706CE.66B7C6AA@cosc.canterbury.ac.nz> <38C8309B.9CE1D7F6@earthlink.net> <38cb370e.471286558@news.nikoma.de> <38CBCCA4.50D9C2@bellatlantic.net>
Message-ID: <38cc9f3e.62191398@news.nikoma.de>

On Sun, 12 Mar 2000 16:58:10 GMT, Steve Holden
 wrote:

>All of which goes to show that there will bever be an adequate
>substitute for the pain of intensive proofreading.  And even
>*then* bugs will slip in to documentation.

Proofreading by someone who is competent technically, too -- just
catching spelling errors won't help you if it turns >>"dd".<< into
>>"dd."<< .

Cheers,
Philip
-- 
Philip Newton 

From Jim.Tittsler at tokyopc.org  Thu Mar  9 02:23:23 2000
From: Jim.Tittsler at tokyopc.org (Jim Tittsler)
Date: 9 Mar 2000 07:23:23 GMT
Subject: Python and MS Excel
References: 
Message-ID: 

On Sat, 4 Mar 2000 19:24:01 GMT, Scott Barron  wrote:
>Does anyone have a link to a tutorial on creating/using Excel spreadsheets
>with Python?

An article by Andrea Provaglio in Software Development Online uses driving
Excel as a way to introduce Python.

http://www.sdmagazine.com/features/fr_4.htm

From gmiller at classic-games.com  Thu Mar 23 16:00:11 2000
From: gmiller at classic-games.com (Greg Miller)
Date: Thu, 23 Mar 2000 15:00:11 -0600
Subject: Embedding Python in a MUD
References: <38D7EAA8.6863DEE3@classic-games.com>  
Message-ID: <38DA85DB.33E02193@classic-games.com>

Phillip Lenhardt wrote:
> Might I suggest checking out Moebius?
> 
> Features:
> 
>         * Written in Python: almost everything is runtime-modifiable.

And it provides appropriate multiuser security?

>         * SQL database backend.

Yuck. Historically, people who've tried this with muds have reported
poor performance in exchange for almost no benefit. SQL is nice for its
intended purposes, but not so well suited for muds.

>         * Very dynamic: both data and code come from the database.
>         * No save/load issues.

I take it you mean its fully persistent across reboots and crashes?

>         * Threaded.
>         * Player scripting in Python.

Does it expose synchronization primitives to users?
-- 
http://www.classic-games.com/
History does not repeat itself, but it does rhyme.
*** Please limit .sigs to four lines and avoid HTML mail or posts. ***

From python at rose164.wuh.wustl.edu  Tue Mar 21 16:10:12 2000
From: python at rose164.wuh.wustl.edu (David Fisher)
Date: Tue, 21 Mar 2000 15:10:12 -0600
Subject: Simulating a Web Browser in Python
References:   
Message-ID: <007701bf93af$0a4de460$7a3dfc80@spkydomain>

----- Original Message -----
From: "Brian Fritz" 
Newsgroups: comp.lang.python
To: 
Sent: Tuesday, March 21, 2000 1:37 PM
Subject: Re: Simulating a Web Browser in Python

>
> Well to make my intent a bit more clear... What I'm writing is going to
> be a Web Robot.
>
> However, I'd heard of grail and it may be a good source of example and
> howto code.
>

Well then, in the Tools directory of the Python source is webchecker.  It
will traverse a web tree looking for bad links.  In the same directory is
websucker, which will mirror a remote web site locally when pointed at the
root url.  I use it all the time when I want to pull content from the web
and maintain it's structure.

dnf

From haskin at ptway.com  Fri Mar 17 01:44:10 2000
From: haskin at ptway.com (Brian Haskin)
Date: Fri, 17 Mar 2000 01:44:10 -0500
Subject: SHA Hash in python?
Message-ID: <38D1D43A.6EF351BE@ptway.com>

Does anyone know of a python module for the SHA cryptographic hash?

Thanks,
Brian Haskin
haskin at ptway.com

From claird at starbase.neosoft.com  Thu Mar  9 08:53:08 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 9 Mar 2000 07:53:08 -0600
Subject: What If Python Replaced Elisp?
References:  
Message-ID: <86BBD5F8F3E4194C.5E28784F01135B4A.5AB14C055AD33D4B@lp.airnews.net>

In article ,
Matt Curtin   wrote:
			.
			.
			.
>  Tad> Would it be easier to add a COM (automation) interface to
>  Tad> XEmacs if Python were used instead of Elisp?
>
>I do not see why.
			.
			.
			.
Because Mark Hammond has already worked miracles
in educating Python about COM.

Your general principle that elisp is already
quite refined, and more capable and mature than
naive observers realize, is unarguable.
-- 

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

From S.I.Reynolds at cs.bham.ac.uk  Tue Mar 14 06:31:55 2000
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Tue, 14 Mar 2000 11:31:55 +0000
Subject: itoa -- oops.
References: <38CD3583.2694@cs.bham.ac.uk>  
Message-ID: <38CE232B.3B86@cs.bham.ac.uk>

Alex wrote:
> 
> > I'm not aware of one, but it's easy to roll your own...
> 
> That wasn't really safe.  You could get into infinite loops with
> non-integer or negative inputs.  This version should cope with that:

Great. I've included a quick fix to also let it work up to base 36 (10
digits + 26 letters).

> import string, types
> 
> def itoa (n, base = 2):
>     if type (n) != types.IntType:
>         raise TypeError, 'First arg should be an integer'
    if (type (base) != types.IntType) or not (2<=base<=36):
        raise TypeError, 'Second arg should be an integer between 2 and
36'

>     output = []
>     pos_n = abs (n)
>     while pos_n:
>         lowest_digit = pos_n % base

        if lowest_digit>=10:
            output.append( chr( ord('a')+lowest_digit-10 ) )
        else:
            output.append( str(lowest_digit) )

>         pos_n = (pos_n - lowest_digit) / base
>     output.reverse ()
>     if n < 0:
>         output.insert (0, '-')
>     return string.join (output, '')
> 
> if __name__ == '__main__':
>     assert itoa (16) == '10000'
>     assert itoa (17) == '10001'
>     assert itoa (-17) == '-' + itoa (17)

    assert itoa(10, 16)  == 'a'
    assert itoa(15, 16)  == 'f'
    assert itoa(16, 16)  == '10'

    assert itoa(255, 16)  == 'ff'
    assert itoa(256, 16)  == '100'

    assert itoa(35, 36)  == 'z'
    assert itoa(36, 36)  == '10'
    assert itoa(-35, 36)  == '-z'
    assert itoa(-36, 36)  == '-10'

Stuart

From akira.kiyomiya at autodesk.com  Tue Mar 14 22:06:30 2000
From: akira.kiyomiya at autodesk.com (Akira Kiyomiya)
Date: Tue, 14 Mar 2000 19:06:30 -0800
Subject: Don't Laugh....
Message-ID: <8amunr$qe8@autodesk.autodesk.com>

I am very new to Python stuff, so I am going to ask you a really dummy
question.

I am trying to run Python, but it seems like I cannot.

I am using Windows 98 and installed Python on my machine.   I am simply
using Python command line that tells me "Python 1.5.2 Copyright 1991-1995
Stichting Mathematish Centrum, Amsterdam".

I created one document called "module.py" by using my text editor  The
"module.py" only has a sentense " print 'Hello World!'"  and that's it.

I saved this file on my desktop and c drive, so I simply typed >>>module.py
but I keep getting a message "Traceback : File "",
line1 , in ? NameError: module"

I even tried >>>import module but it is same thing.   I don't think I can go
to any specifc path that I saved my python file.

In case you are wondering if I installed Python correctly, my Python
interpretor works since I can type "print 'Hello World!"

Okay, what am I doing wrong??

From kc5tja at garnet.armored.net  Mon Mar 27 08:14:28 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 27 Mar 2000 13:14:28 GMT
Subject: expanding persistant objects
References: 
Message-ID: 

In article , Moshe Zadka
wrote:
>> i'm worried if i start saving the data one way, when i go
>> to add extra fields i wont be able to load the files anymore.
>
>Pickle (and hence shelve) were designed to deal with this.

Pickle works _perfectly_ for this.  Shelve, in my experience, blows chunks.

The reason is that shelve uses a dbm-type database to store pickled objects
in.  The file it produces is, consequently, a BINARY file, and is therefore
proprietary to the database manager used.

I have had to use the shelve module on several occasions, particularly for
in-house development projects where I work.  Moving databases between
equivalently configured machines (e.g., both running Debian and configured
using apt-get the same way) proved to be a very expensive operation. I found
cases where different versions of same-family dbm libraries couldn't read
older files, and at least two machines couldn't even *FIND* the dbm
libraries, despite having them properly installed.  These were problems of
Windows proportions -- something I'd never seen before on Linux.

I have since switched from using shelves to just plain Pickle files, and
have never, ever had this problem since.

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

From shogan at iel.ie  Fri Mar 24 04:58:49 2000
From: shogan at iel.ie (Shaun Hogan)
Date: Fri, 24 Mar 2000 09:58:49 -0000
Subject: test exercises???
Message-ID: <002b01bf9577$8d557ca0$2801000a@iel.ie>

hi ive been learning python and c/python integration for the last number of weeks.
can anyone tell me if there are any websites that will give me problems to solve so i can put what i think i know to the test.
thanks.
=====================
Shaun Hogan
Interactive Enterprise Ltd.
alt. E-mail : shaun_hogan at yahoo.com
+353 86 8342529

-------------- next part --------------
An HTML attachment was scrubbed...
URL: 

From milesthompson at sprint.ca  Tue Mar  7 07:35:22 2000
From: milesthompson at sprint.ca (Miles Thompson)
Date: Tue, 07 Mar 2000 08:35:22 -0400
Subject: Python/MySQL Where do "SELECT ..." results go?
Message-ID: <38C4F789.71ECCB04@sprint.ca>

Dear All,

Please be patient with me, if I've done something incredibly dumb. I've
workekd with FoxPro, Access, SQl Server for years, but the Python/MySQL
combination is new to me. My query results disappear and all I get is a
count of the records retrieved.

Here's what I've done. (Explore is the database, and table Test has 5
records.)

>>> import Mysqldb
>>>conn = Mysqldb.mysqldb('explore')
>>>conn

#
# here's my first attempt at a select
#
>>>conn.execute("Select * from test;")
5
#
# wasn't what I wanted, so then I did this
#
>>>curs = conn.cursor()
>>>curs.execute("Select * from test;")
5
#
# Nope. Try assigning to a variable?
#
>>>data = curs.execute("Select * from test;")
>>>data
5
#
# and at this point it was 11:30, and over the past two days I'd
# hammered myself through the Python tutorial, various Python docs,
# set up MySQL and played with it interactively, reinstalled a new
version
# of Python (1.5.2) and the Python module for MySQL.
# I used python-MySQLmodule-1.4.0-7.i386.rpm
# So  I went to bed.

I must be close, but the documentation is "scanty" and short on
examples. Suggestions, or a pointer to a web site that answers the
question, has examples, will be welcome.

TIA - Miles Thompson
milesthompson at sprint.ca

From S.I.Reynolds at cs.bham.ac.uk  Mon Mar 13 13:37:55 2000
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Mon, 13 Mar 2000 18:37:55 +0000
Subject: itoa
Message-ID: <38CD3583.2694@cs.bham.ac.uk>

Is there an integer to ascii (itoa) function in Python? I need to
convert integers into various different bases (but to binary, in
particular). Most languages come provided with an itoa-like function but
I can't seem to find one in the Python API.

Stuart

From jwbnews at scandaroon.com  Fri Mar 17 10:25:39 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 17 Mar 2000 15:25:39 GMT
Subject: Pascal int()
References:  
Message-ID: 

In article 
, 
japhy at pobox.com wrote:

> First, you'd better not call it int(), because that'd cause some sillyness
> -- there's a builtin int().  For sake of argument, I'll call it inc().

inc() is a likely name:  it's what N. Wirth called the thing.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com

From rurban at x-ray.at  Fri Mar  3 13:18:36 2000
From: rurban at x-ray.at (Reini Urban)
Date: Fri, 03 Mar 2000 18:18:36 GMT
Subject: win32com.client.dynamic: VARIANT type not supported for SAFEARRAYS
References:  <5OBv4.2211$vW.5056@news-server.bigpond.net.au>
Message-ID: <38c0020a.22381332@judy>

Mark Hammond wrote:
>"Olaf Trygve Berglihn"  wrote in message
>news:dn1n1oi7mw4.fsf at kikp0524.chembio.ntnu.no...
>> TypeError: The VARIANT type is not supported for SAFEARRAYS
>>
>> Any explanations and/or suggestions?
>
>The Python COM framework doesnt have support for that variant type in
>safe array :-(  I have no idea what the variant type is, but it must
>be one of the more obscure ones.  What is the property defined as
>returning?  Presumably it is an array, but of what?

Not a obscure one.
A safearray of variants is a generic array which holds the union type
(aka Variant, i.e. any type) instead of immediate (aka typed) data.

The Variant is the basic IDispatch (late bound OLE) type, used to pass
any types to/from ActiveX objects.
The Safearray is an extension of a variant, holding a range-checked and
type-checked array.
A Safearray of variants is the most generic array type.
--
Reini Urban
http://xarch.tu-graz.ac.at/autocad/lisp/ffis.html

From jajs at NOSPAMretemail.es  Sun Mar 26 14:31:40 2000
From: jajs at NOSPAMretemail.es (*puntero_loko)
Date: Sun, 26 Mar 2000 19:31:40 GMT
Subject: Python tutorial/comparison for C++ programmer
References:  <8b956c$9tn$1@nnrp1.deja.com> <077C600D9F281A4D.487194D3E94C88DA.D53DE68986E89418@lp.airnews.net>  <8bcr2a$b76$1@nnrp1.deja.com>   
Message-ID: 

Michael Hudson dijo sobre Re: Python tutorial/comparison for C++ programmer,
en fecha: 23 Mar 2000 13:35:57 +0000

|I suppose it could be viewed as competition to VB, but then that's not
|really relavent until Python gets a flashy point n' click GUI, and
|even then, if micorsoft ship Visual Python (tm), how does this impact
|me?

Then we have to hide Glade...
- 

From erno at iki.fi  Tue Mar 14 02:26:02 2000
From: erno at iki.fi (Erno Kuusela)
Date: 14 Mar 2000 09:26:02 +0200
Subject: How do I ...
References: <38CD2368.9D29554F@abo.fi>
Message-ID: 

>>>>> "Cesar" == Cesar Lopez  writes:

    Cesar> How do I acces to the low-level of my system? I mean, for
    Cesar> example, I want to access and configure the speed of my
    Cesar> serial port using a Python program.

It depends on your platform, as the std python distribution doesn't
have platform specific serial io modules and accessing serial ports is
not defined by C.

If you happen to be using a unixish system, just opening
the serial port os.open() and using the termios module or stty(1)
should work.

    Cesar> I've tried building an extension C function and the call it
    Cesar> from Python, but doesn?t works.

You are doing it wrong then :) If you tell us how it fails to
work, people may be able to help.

 -- erno

PS. This is in the FAQ. You should look in there before posting,
so the newsgroup/mailing list doesn't get fooded with
repeated questions on the same topic.

From tgdeveloper at my-deja.com  Fri Mar 10 16:10:42 2000
From: tgdeveloper at my-deja.com (tgdeveloper at my-deja.com)
Date: Fri, 10 Mar 2000 21:10:42 GMT
Subject: python-mode with GNU Emacs?
Message-ID: <8aboce$23t$1@nnrp1.deja.com>

Does python-mode.el (from www.python.org) work
with GNU Emacs (20.5.1)? When I try to
byte-compile the file, I get the following:

---
Compiling file
c:/Emacs/emacs-20.5/lisp/textmodes/python-mode.el
at Fri Mar 10 10:50:51 2000
** assignment to free variable zmacs-region-stays
While compiling the end of the data:
** The following functions are not known to be
defined:
py-beginning-of-def-or-class,
py-end-of-def-or-class,
make-extent, set-extent-property,
buffer-syntactic-context,
py-mark-block, py-goto-block-up
---

Then when I try to load-file "python-mode.elc" I
get a message stating "too many arguments."

Any help would be appreciated.

Tom

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

From mhammond at skippinet.com.au  Tue Mar 28 06:30:53 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 28 Mar 2000 11:30:53 GMT
Subject: Determining Ethernet HW Address
References:  
Message-ID: 

"Darrell"  wrote in message
news:dhSD4.7202$JE5.97355 at typhoon.nyroc.rr.com...
> "Jerome Chan"  wrote in message
> > Is there a portable way of determining the Ethernet HW
Address?
> >
> For windows without using Win32 try parsing the output of
ipconfig /all

I couldnt resist :-)  In win32all-130 you will be able to use the
Netbios function:

    from netbios import *
    # code ported from "HOWTO: Get the MAC Address for an
Ethernet Adapter"
    # MS KB ID: Q118623
    ncb = NCB()
    ncb.Command = NCBENUM
    la_enum = LANA_ENUM()
    ncb.Buffer = la_enum
    rc = Netbios(ncb)
    if rc != 0: raise RuntimeError, "Unexpected result %d" %
(rc,)
    for i in range(la_enum.length):
        ncb.Reset()
        ncb.Command = NCBRESET
        ncb.Lana_num = ord(la_enum.lana[i])
        rc = Netbios(ncb)
        if rc != 0: raise RuntimeError, "Unexpected result %d" %
(rc,)
        ncb.Reset()
        ncb.Command = NCBASTAT
        ncb.Lana_num = ord(la_enum.lana[i])
        ncb.Callname = "*               "
        adapter = ADAPTER_STATUS()
        ncb.Buffer = adapter
        Netbios(ncb)
        print "Adapter address:",
        for ch in adapter.adapter_address:
            print "%02x" % (ord(ch),) ,
        print

Yields on my dual-NIC server:
Adapter address: 00 a0 cc 54 22 2d
Adapter address: 00 40 05 6b fc ca

(both of which are correct :-)

Mark.

From dma at cyrus.andrew.cmu.edu  Tue Mar 21 13:07:39 2000
From: dma at cyrus.andrew.cmu.edu (David Andersen)
Date: Tue, 21 Mar 2000 13:07:39 -0500
Subject: Building Mac Python 1.5.2
Message-ID: <875417.3162632859@inuvik.hcii.cs.cmu.edu>

I'm trying to build the Macintosh PythonCorePPC from the Python 1.5.2c1
source, using CWGUSI from Jack Jansen's site (built for PPC only) and
CodeWarrior Pro 5.3.

Compiling is no problem, but I then get the following link errors: Does
anyone know what is going on?

Link Error   : export symbol 'fwalk' is undefined

Link Error   : export symbol '_prealloc_newpool__3stdFUl' is undefined

Link Error   : export symbol '_set_newnonptrmax__3stdFUl' is undefined

Link Error   : export symbol '_set_newpoolsize__3stdFUl' is undefined

Link Error   : export symbol '__malloc_pool' is undefined

Link Error   : export symbol '__num_to_file' is undefined

Link Error   : export symbol '__file_to_num' is undefined

Link Error   : export symbol '__pool_free' is undefined

Link Error   : export symbol '__pool_realloc' is undefined

Link Error   : export symbol '__pool_alloc_clear' is undefined

Link Error   : export symbol '__pool_alloc' is undefined

Link Error   : export symbol '__pool_preassign' is undefined

Link Error   : export symbol '__pool_preallocate' is undefined

Link Error   : export symbol '__init_pool_obj' is undefined

Link Error   : export symbol '__vfscanf' is undefined

Link Error   : export symbol '__end_critical_region' is undefined

Link Error   : export symbol '__begin_critical_region' is undefined

Link Error   : export symbol '__vfwscanf' is undefined

Link Error   : export symbol 'SIOUXclrscr' is undefined

From effbot at telia.com  Sun Mar 12 09:12:38 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 12 Mar 2000 14:12:38 GMT
Subject: Compiling a Python script
References: <_wwy4.14142$ql2.184042@typhoon.we.rr.com>
Message-ID: 

LineNoise  wrote:
> I wrote a smal script for a friend to use, but he doesn't have the Python
> interpreter, and he isn't going to have it anytime soon.  I downloaded the
> source distribution and tried to use the freeze tool, but it didn't work.
I
> think I have to compile/install the source distr to get it to work, but
how
> do I do that?  My computer is running win98, and I have a copy of Visual
C++
> that I can use, but I have no experience with it.

start here:
http://www.mcmillan-inc.com/install1.html

From alex at somewhere.round.here  Wed Mar 29 12:09:49 2000
From: alex at somewhere.round.here (Alex)
Date: 29 Mar 2000 12:09:49 -0500
Subject: Incompatibility between socket and httplib in python CVS build.
References:  
Message-ID: 

> iow, update your CVS copy and reinstall the standard library, and this
> problem will go away.

Oh, right.  Sorry.

Alex.

From jpc at informatics.jax.org  Wed Mar  8 15:25:18 2000
From: jpc at informatics.jax.org (John Corradi)
Date: Wed, 08 Mar 2000 15:25:18 -0500
Subject: emacs can't find python-mode
Message-ID: <080320001525180664%jpc@informatics.jax.org>

Hi All,

I've tried to install the python-mode.el for Emacs 20.3.1, but it tells
me it can't find python-mode.  I followed all of the instructions on
the Python site.  I've never customized Emacs before and the only lines
in my new .emacs file are the ones for using the python-mode library. 
When I try "M-x locate-library RET python-mode RET" it tells me that it
can't find python-mode in the search path despite the fact that I've
added the appropriate line to my .emacs file.  Am I missing something? 
How do I make sure python-mode is in the load-path?  TIA.

John 

-- 

John Corradi
jpc at informatics.jax.org

From mwh21 at cam.ac.uk  Thu Mar 16 04:56:58 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 16 Mar 2000 09:56:58 +0000
Subject: Assigning output from a simple python statement
References: <01bf8f1f$a611b640$3201a8c0@wraith.mclachlan.com.au>
Message-ID: 

"Wilson Fletcher"  writes:

> I would like to execute an unknown python statement and store the output.
> 
> It will typically be a print statement. eg. "print 'Hello World'"
> 
> I won't know the statement until runtime becuase it will come from a
> database. BUT I want to capture anything that goes to stdout and process
> it. 

try this:

import sys,StringIO

def capture_output(code):
    so = sys.stdout
    sio = StringIO.StringIO()
    sys.stdout = sio
    try:
        exec code
    finally:
        sys.stdout = so
    sio.seek(0)
    return sio.read()

not tested, but should work.

I leave out my usual "do you really want to `exec' things you don't
know everything about" comment, I think.

Cheers,
M.

-- 
very few people approach me in real life and insist on proving they are
drooling idiots.                         -- Erik Naggum, comp.lang.lisp

From wlfraed at ix.netcom.com  Sat Mar 25 00:00:06 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Fri, 24 Mar 2000 21:00:06 -0800
Subject: HELP Newbie solve this Problem
References: <8beo97$fm7$1@nnrp1.deja.com> <2atlds0ordevolp8ubbfghdi75ih0bdvst@4ax.com> <38DC2E16.6096@marshall.edu>
Message-ID: 

On Fri, 24 Mar 2000 19:06:25 -0500, Mark Hathaway
 declaimed the following in comp.lang.python:

> I can't really believe the Fartran code takes that much space.
> I suspect it's given as a set-up, so the Python code to follow
> will be all the more impressive. Well, here's a Python version.
> 
> import string
> lines = open('file.txt').readlines()
> for each in lines:
>     eachlist = string.split(each)
>     str = eachlist[1] + " " + eachlist[0]
>     print str
> 
> If you'd wanted the output printed back to a file then you could
> add something other than the final 'print str' statement.
> 
> Amazing isn't it? When a language doesn't get in your way the code
> is brief and more readable. It's all the niggling little details
> that a language like Fortran or C require that makes them very ugly.
>
	I think either you missed a requirement, or I went too far...
>From the original poster:

history> The ?data?? file is the preamble to the constitution.
history> The objective is to write a command that will give a list of
how many
history> times the letters appear in the preamble. Try not to use awk.

 The original program input is not that list of numbers and letters, but
plain text to be counted (ignoring case and punctuation/numerics). I was
just surprised the example output implied that the 
format was more difficult to produce than the 
ordering...

--
 > ============================================================== <
 >   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 moshez at math.huji.ac.il  Mon Mar 13 20:07:02 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 14 Mar 2000 03:07:02 +0200 (IST)
Subject: itoa
In-Reply-To: <38CD3583.2694@cs.bham.ac.uk>
Message-ID: 

On Mon, 13 Mar 2000, Stuart Reynolds wrote:

> Is there an integer to ascii (itoa) function in Python? I need to
> convert integers into various different bases (but to binary, in
> particular). Most languages come provided with an itoa-like function but
> I can't seem to find one in the Python API.

Have a look at the builtin "str" function
--
Moshe Zadka . 
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com

From harold.taylor.remove-this at delete-this.uni-tuebingen.de  Thu Mar  9 07:07:13 2000
From: harold.taylor.remove-this at delete-this.uni-tuebingen.de (H.V.Taylor)
Date: Thu, 09 Mar 2000 13:07:13 +0100
Subject: how to catch signals
References: 
Message-ID: <55HHOFE191QSaSQR5a2xg24K1Jpq@4ax.com>

quinn at mono.ugcs.caltech.edu (Quinn Dunkan) wrote:

>Hello!  I have a process that starts children, and perhaps kills them at some
>time.  The children need to be unregistered when they die, so I set up a ...
>.. snip ...
>... after I kill (I can't just wait() afterwards, because it's
>possible someone else killed it).

Gosh, you sound blood-thirsty ;-)) I guess I'll have to read Kragen
Sitaker's "Crime in Python" as well...

Harold

From nick at spam_me_notvideosystem.co.uk  Mon Mar  6 05:26:27 2000
From: nick at spam_me_notvideosystem.co.uk (Nick Trout)
Date: Mon, 6 Mar 2000 10:26:27 -0000
Subject: DirectX 7 and python
References: <38c012da$0$2943@senator-bedfellow.mit.edu>
Message-ID: <38c3854d@news.xtml.co.uk>

"Robert"  wrote in message
news:38c012da$0$2943 at senator-bedfellow.mit.edu...
> Hi,
>      With the release of DirectX 7, MS included a type library for Visual
> Basic users
> to access DX functionality. Apparently (from using the makepy and win) it
> does not
> provide automation/dispatch interfaces, so python can't access it. My
> question is two-fold:
> 1)Has anyone written pythoncom extensions to provide the necessary
> interfaces?
> 2)How does one go about writing a custom interface in pythoncom? Does anyone
> have some sample code?

I've been working on a wrapper for DX7 RM. ie. not COM/direct access. Pet
project so slow progress!

I did look for other projects doing this but found none, which surprised me.
Have you come across anyone else working on DX Python stuff?

From alex at somewhere.round.here  Wed Mar 29 19:50:06 2000
From: alex at somewhere.round.here (Alex)
Date: 29 Mar 2000 19:50:06 -0500
Subject: How to tell in site.py if python is being invoked interactively?
References:  <20000329194024.A1006339@vislab.epa.gov>
Message-ID: 

Hi, Randall.

No, I just have control over the python program they're going to use.

Sorry about the bogus address.  If I set the Mail-Followup-To header to
the list address, won't replies to my messages that are CC'd to the
newsgroup show up twice?  On the other hand, a null address means mail
people send me just disappears...

Alex.

From scherbi at bam.com  Wed Mar 29 13:00:36 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 29 Mar 2000 13:00:36 -0500
Subject: Rational-Rose and Python ...
References: <8bt5fh$p2t$1@news.urz.uni-heidelberg.de>  <38E22E94.CC6E66E@bam.com> 
Message-ID: <38E244C4.D69D3B4A@bam.com>

Warren Postma wrote:

> > It sounds like the modelling wasn't your primary problem on that project.
>
> Yes, probably. However, I find most new technologies, in attempting to
> overcome a developer's resistance to trying something new, end up oversold.
> A nice thing about Python, IMHO, is that because it's a language to which I
> have the source code, it's not oversold. If it can't do something, I can
> make it do something [like Christian Tismer's Stackless Python].
>
> I think the same would be true for UML.  The Tigris stuff looks exciting for
> the same reason. If there's something my Case tool won't do, I can go in and
> add it, send a patch, and it might even get put into the program.  In other
> words, the latest ROSE doesn't do much for me, even with it's much enhanced
> two-way tools, but an open source UML tool might end up being just as handy
> as Python itself. Even better if it worked from inside
> Python/PythonWin/IDLE/etc, and was truly "compact" in much the same way as
> Python is.
>
> Warren
>

agreed.

From Richard.Jones at fulcrum.com.au  Wed Mar 15 17:44:45 2000
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Thu, 16 Mar 2000 09:44:45 +1100
Subject: Pointers 
In-Reply-To: Message from Curtis Jensen of 2000-Mar-15 14:25:18,
            <38D00DCE.9740380F@be-research.ucsd.edu> 
Message-ID: <200003152244.JAA27546@envy.fulcrum.com.au>

[Curtis Jensen]
> I haven't found anything about pointers in python.  Is there a pointer
> type?  If so, what is the syntax?

   In Python, the things you have come to know as variables are best thought of 
as "labels" for objects. In old-fashioned pointer-language-speak, they're about 
as equivalent to references or pointers as you're going to get.

>>> a = 'hello world'
>>> id(a)
135105304

   In this case, 'a' is a label for the string object 'hello world'. We could 
assign another "pointer" to the same string by:

>>> b = a
>>> id(b)
135105304

   So 'b' is now a label on the same object as 'a'. In pointer-language-speak, 
it's a pointer referencing the same data. Note that Python special-cases some 
very common objects (only immutable ones) so that you're using a reference to 
the same object:

>>> c = 1
>>> id(c)
134955600
>>> d = 1
>>> id(d)
134955600

    Neat huh? Common strings that you use in your programs may be intern()'ed 
... but I'll leave that as an exercise for you to find out about (see the 
Library Reference for details about intern() and id()).

        Richard

From gmcm at hypernet.com  Thu Mar  9 23:29:45 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 9 Mar 2000 23:29:45 -0500
Subject: Python -> Win32 EXE
In-Reply-To: 
Message-ID: <1259459908-1440638@hypernet.com>

A[r]TA wrote:
> 
> > not really -- but what you really want is the "installer"
> > from the nice people at McMillan Enterprises, Inc:
> >
> > http://www.mcmillan-inc.com/install1.html
> >
> 
> The program gives a message:
> Can't find DLL's, copy them to the right dir?
> 
> Can somebody help me? I just trie to compile something with simple.py and it
> doesn't work!

The message most likely to match your vague description is 
produced when the installer can't figure out what your system 
directory is (because you don't have the Win32 extensions 
installed). In that case, before starting installer, do

c:\>PATH=%PATH%;C:\Windows\System

or whatever your system directory is.

- Gordon

From craigf at ilid.com.au  Mon Mar 20 22:06:42 2000
From: craigf at ilid.com.au (Craig Findlay)
Date: Tue, 21 Mar 2000 14:06:42 +1100
Subject: stdout and redirection (win NT)
Message-ID: <3npddssd9fksp73pkor1tl3m0mqvg672tm@4ax.com>

What am I doing wrong here?

My sample python program test.py is:

import sys
sys,stdout.write('Hello World')

>From a command prompt, I type:
test.py
and I get the expected Hello World output on the console.

Next, I type:
test.py > c:\myfile.txt

I expect the file myfile.txt to contain the redirected output from the
python script. Instead, myfile.txt exists but is empty.

What is happening?

Craig Findlay

From mikael at isy.liu.se  Tue Mar 28 09:48:07 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 28 Mar 2000 16:48:07 +0200 (MET DST)
Subject: No Vault my Fault?
In-Reply-To: <8bqfrr$q14$1@weber.a2000.nl>
Message-ID: 

On 28-Mar-00 Ilja Heitlager wrote:
 >  Don't know where else to send this to, but is the Vault of Parnasses closed
 >  or something?
 >  Everytime I would like to enter www.vex.net/parnassus it asks me for a
 >  password Could it be because I am on a Window$ machine with (UPC) cable
 >  connection?  (You Dutch readers know what I mean ;-)

No problem for me...

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson 
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    28-Mar-00
Time:    16:47:54

This message was sent by XF-Mail.
-----------------------------------------------------------------------

From tim_one at email.msn.com  Thu Mar  9 21:23:18 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 9 Mar 2000 21:23:18 -0500
Subject: What If Python Replaced Elisp?
In-Reply-To: 
Message-ID: <000701bf8a37$992c12a0$58a2143f@tim>

[Neel Krishnaswami]
> There are actually a whole lot of optimizations that could be done,
> but aren't, in the current Python interpreter.
>
> For example, in the following code:
>
> import operator
>
> lst = []
> for i in range(100):
>     lst.append(i**2 + i**2)
>
> result = reduce(operator.add, lst)
>
> You can statically infer the type of result, the final size of lst,
> avoid actually allocating a range() object, eliminate the the
> typechecks that make sure that i has a __pow__ method, the bounds
> checks on the range object, do available expressions analysis on the
> i**2 to avoid calculating it twice, and probably a few more
> optimizations that I don't know about.

Actually not, you can't do any of that now!  You currently have no way to
know that, e.g., some *other* module didn't import this one (let's call this
one "neel"), and redirect neel.range and neel.reduce to some arbitrary
functions of its own devious choosing.  Heck, the call to neel.range may
even reach back in and change neel.lst to anything at all before the loop
body is entered.  "i" may end up referring to an object whose __pow__ method
sends email to Afghanistan; etc.

This is why the Types-SIG is so keen to impose restrictions on what modules
can do to each others' namespaces (in the presence of optional static
typing).

> ...
> You don't have to win every game to make it to the championships. :)

you-don't-even-need-to-win-one-but-it-sure-can't-hurt-ly y'rs  - tim

From wware at world.std.com  Fri Mar 17 16:41:11 2000
From: wware at world.std.com (Will Ware)
Date: Fri, 17 Mar 2000 21:41:11 GMT
Subject: system calls to sh
References: <38D295F8.C88CD9E4@stsci.edu>  
Message-ID: 

Alex (alex at somewhere.round.here) wrote:
> >>> del os.environ['MY_VAR']

That doesn't work, but this will:
>>> os.environ['MY_VAR'] = ''

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com

From jeffp at crusoe.net  Fri Mar 10 07:38:14 2000
From: jeffp at crusoe.net (Jeff Pinyan)
Date: Fri, 10 Mar 2000 12:38:14 GMT
Subject: Duplicates in lists
In-Reply-To: 
References: <38C748C7.35162A7B@teleatlas.com> <38C81A12.6B4E85EE@exceptionalminds.com> 
Message-ID: 

[posted & mailed]

On Mar 10, Quinn Dunkan said:

>>> is there someone who has an efficient function that finds
>>> all duplicates in a list?

>def uniq(a):
>    r = {}
>    for i in a:
>        r[i] = None
>    return r.keys()

Your approach loses the order of the keys.  If this is not acceptable, I
offer:

  def ord_uniq(a):
    r = {}
    ret = []
    for i in a:
      if not r.has_key(i):
        ret.append(i)
        r[i] = 1
    return ret

Or, using OrderedDict.py:

  def ord_uniq(a):
    r = OrderedDict.OrderedDict()
    for i in a: r[i] = 1
    return r.keys()

-- 
MIDN 4/C PINYAN, NROTCURPI, US Naval Reserve             japhy at pobox.com
http://www.pobox.com/~japhy/                  http://pinyaj.stu.rpi.edu/
PerlMonth - An Online Perl Magazine            http://www.perlmonth.com/
The Perl Archive - Articles, Forums, etc.    http://www.perlarchive.com/

From python at rose164.wuh.wustl.edu  Fri Mar  3 05:55:00 2000
From: python at rose164.wuh.wustl.edu (David Fisher)
Date: Fri, 3 Mar 2000 04:55:00 -0600
Subject: Access to _inp()/_outp() on Win95?
References: <38bf3add@news.gvsu.edu>
Message-ID: <002801bf84fe$ec5ada80$0201a8c0@spkydomain>

After I wrote the first message I got to thinking, if you have SWIG, this is
even easier.  So, version 2.
This implements the exact same module as the first one, but as you can see,
you don't need to do anything with the Python/C api.  Two steps (after SWIG
is installed):

swig coniomodule.i

will make the c file coniomodule_wrap.c (and coniomodule_wrap.doc but that's
not relevant).  Now your Setup.in file should be one line:

conio coniomodule_wrap.c

Then run the compile.py script (by David Ascher, typo in the first message)
to create the project files (this will also include the swig step in the
project file, so you can rebuild in one step, cool huh?).

If you knew all of this already, feel free to ignore me (and use the code
).  I needed to brush up on my extension coding, and somebody else
might find this interesting.

David

----- Original Message -----
From: "Andrew Sterian" 
Newsgroups: comp.lang.python
To: 
Sent: Thursday, March 02, 2000 10:07 PM
Subject: Access to _inp()/_outp() on Win95?

> Has anyone exposed the _inp() and _outp() functions under Win95/98
> in an extension module? I need to access some I/O ports directly and
> was hoping to not reinvent the wheel.
>
> Thanks,
>
> --
> Andrew Sterian                 | 
> Assistant Professor            |

> Padnos School of Engineering   | (616) 771-6756
> Grand Valley State University  |
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: coniomodule.i
Type: application/octet-stream
Size: 131 bytes
Desc: not available
URL: 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: coniomodule.c
Type: application/octet-stream
Size: 192 bytes
Desc: not available
URL: 

From effbot at telia.com  Thu Mar 30 12:29:04 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 30 Mar 2000 17:29:04 GMT
Subject: parsing CSV files with quotes
References:  
Message-ID: 

Warren Postma  wrote:
> Another is that any code with eval() can do "way more than I intended" by
> calling eval, since eval() is basically opening up the entire parser.
>
> Is there some code to do eval_literals_only() that knows about parsing
> literal expressions but which can't be made to call any functions?

value = eval(expression, {"__builtins__": {}})

for more info, see the eff-bot guide.

From aa8vb at yahoo.com  Tue Mar 28 07:47:38 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 28 Mar 2000 07:47:38 -0500
Subject: Can Python handel bignum or big floating point data type please?
In-Reply-To: <0OqOOIs29gof883lmOsGthhxEC6Q@4ax.com>
References: <0OqOOIs29gof883lmOsGthhxEC6Q@4ax.com>
Message-ID: <20000328074738.B786492@vislab.epa.gov>

http://www.python.org/topics/scicomp/numbercrunching.html

-- 
Randall Hopper
aa8vb at yahoo.com

From eric.hopper at ebenx.com  Thu Mar  9 19:36:37 2000
From: eric.hopper at ebenx.com (Eric Hopper)
Date: Thu, 09 Mar 2000 18:36:37 -0600
Subject: How do I...
References: 
Message-ID: <38C84395.1F428E0F@ebenx.com>

Brett Tribble wrote:

> How do I find the name of the script that is currently running from within
> the script?
>
> I.E. If I'm running C:\python\myscript.py  how do I determine this?
>
> Brett

import sys
print sys.argv[0]

From bboard at parc.xerox.com  Tue Mar 28 15:21:52 2000
From: bboard at parc.xerox.com (Mail Archiving Daemon)
Date: Tue, 28 Mar 2000 12:21:52 PST
Subject: changing LD_LIBRARY_PATH for Python script
Message-ID: <00Mar28.122133pst."3426"@watson.parc.xerox.com>

Hi.  I'm in a situation where I need to set the LD_LIBRARY_PATH before
loading a Python module.  But I can't find a way of doing this from
Python.  By the time I'm actually in the Python script, the
interpreter is running, and changing the value of os.environ has no
effect on that process (or on the dl mechanism).  I could use a trick
like Martin's

#!/bin/sh
true=0;
then=1;
fi=1;
if true :
	then
	exec python "$0" "$@"
fi

del true, then, fi

If I could only figure out the right strings to put in there :-).

The code that is affected is a shared library loaded from *another*
shared library that is in turn loaded from Python as a Python module.

Bill

From bjorn at roguewave.com  Wed Mar 15 19:28:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 15 Mar 2000 17:28:31 -0700
Subject: Pointers
References: <200003152259.JAA27601@envy.fulcrum.com.au> <38D01D46.28817769@be-research.ucsd.edu>
Message-ID: <38D02AAF.78B3D200@roguewave.com>

Curtis Jensen wrote:
> 
> Richard Jones wrote:
> >
> > [Curtis Jensen]
> > > Pointers are sometimes usefull.  I have a module that creates a
> > > dictionary.  I want to include some elements from the distionary in one
> > > class and other elements in another class.  If I had pointers, I could
> > > do that.  Is there another way?
> >
> >    As I explained in my response to your post (unlike the unhelpful Bjorn ;)
> > you're already dealing with a reference there. The dictionary is an object and
> > your only control over it is via your reference to it. You can make labels
> > (variables) for that reference, or pass that reference around to functions or
> > methods. I think you'll find that you can already do what you want:
> >
> > >>> class A:
> > ...  def foo(self, a):
> > ...   a['A'] = 1
> > ...
> > >>> class B:
> > ...  def foo(self, b):
> > ...   b['B'] = 1
> > ...
> > >>> a=A()
> > >>> b=B()
> > >>> d={}
> > >>> a.foo(d)
> > >>> b.foo(d)
> > >>> d
> > {'B': 1, 'A': 1}
> >
> >         Richard
> 
> Maybe I'm missunderstanding, but it appears that this is just a copy
> routine.  I want something like this:
> 
> >>> d={'A':0,'B':0}
> >>> class A:
> ...   def __init__(self):
> ...     self.a -> (Pointer to) d['A']
> ...
> >>> class B:
> ...   def __init__(self):
> ...     self.b -> (Pointer to) d['B']
> >>> a=A()
> >>> b=B()
> >>> print a.a
> 0
> >>> print b.b
> 0
> >>> d['A'] = 1
> >>> d['B'] = 2
> >>> print a.a
> 1
> >>> print b.b
> 2
> 
> If the data in the dictionary changes, I want the data in the class to
> automaticaly change also.  Is this possible?  Or at least something
> close?

yes. If you read through what Richard wrote above a couple of times,
while remembering that ints are immutable and lists aren't, you should
be able to understand the following:

d = {'A':[0], 'B':[0]} # use a list as a box to put ints in
class A:
	def __init__(self):
		self.a = d['A']
class B:
	def __init__(self):
		self.b = d['B']
a = A()
b = B()
print a.a[0] # should be 0
print b.b[0] # should be 0
d['A'][0] = 1
d['B'][0] = 2
print a.a[0] # should be 1
print b.b[0] # should be 2

Please don't do this   There is almost always a more
straightforward way of doing things like this.

-- bjorn

ps: in Richard's example the dictionary is _not_ passed by value, so
there is no copying going on. The technical term is
pass-by-object-reference, but you can think of it as pass-by-pointer in
e.g. C++/C but with automatic dereferencing.

From nielsenjf at my-deja.com  Fri Mar 10 20:04:19 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Sat, 11 Mar 2000 01:04:19 GMT
Subject: Python and Active Directory
References:  
Message-ID: <8ac62h$c7m$1@nnrp1.deja.com>

Using the constant 1 seemed to work (throwing in erroneous constants
does make errors), but I didn't sniff packets to make sure.

Btw, the pythonldap module (works fine w/exchange), looks to support
kerberos(I haven't tested it), which may be useful for those playing
with win2000.

It is nicer than ADSI in some ways since it leverages python
dictionaries. And, of course, pythonldap allows a unix box to talk to an
MS box.

john

In article ,
  "Mark Hammond"  wrote:
>
> "Augustine David"  wrote in message
> news:t8Yx4.1012$Qf.532452 at news-west.usenetserver.com...
> > -----BEGIN PGP SIGNED MESSAGE-----
>
> > I am attempting to reproduce the SDK examples below in Python.  I
> have ran
> > "makepy" on the type library,
> > but I am unsure how to call any of its classes using
> > "win32com.client.Dispatch".  Any input you can give me
> > would be appreciated.
>
> This is a little painful unfortunately.  Note the second line - this
> is necessary to force the constant ADS_SECURE_AUTHENTICATION into the
> COM constants namespace.   This second line was generated by running
> "makepy.py -i".  Otherwise you could just use the literal constant 1
> :-)  Normally the constants are "magically" loaded by the time you
> need them, but not in this case.
>
> This requires build 128 of win32all.  Also I cant make it work, but I
> think I simply can't get the strings correct (Ive struggled with this
> before :-(

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

From joacim at home.se  Fri Mar 17 10:30:02 2000
From: joacim at home.se (JJ)
Date: Fri, 17 Mar 2000 16:30:02 +0100
Subject: Java editor for Python
References: <8atgsd$ava$1@vg170.it.volvo.se> 
Message-ID: <8atj21$k2m$1@vg170.it.volvo.se>

So what!
I want to use a Java editor even if you don't like Java or not.

Java IS a great programming language but there is several more. I can't say
that Python is one because I can to little about it (for now). I use JPython
in my Java application and it seems to work just fine.

// Joacim

"Darrell"  wrote in message
news:pXrA4.24285$W5.355862 at typhoon.nyroc.rr.com...
> "JJ"  wrote in message
> news:8atgsd$ava$1 at vg170.it.volvo.se...
> > Hi!
> >
> > Anyone who knows where to find an editor written in Java with syntax
color
> > highlighting?
> >
> > It's a VERY big plus if there is a java bean and/or source code
available.
> >
>
> Java ? Java !
> We don't do no stinking JAVA!
>
> Be there is a colonizing editor that comes with Python for Python.
> Checkout Idle in the tools directory or Pythonwin.
>
>
> --Darrell
>
>
>
>
>

From ehagemann at home.com  Mon Mar 20 21:38:14 2000
From: ehagemann at home.com (Eric Hagemann)
Date: Mon, 20 Mar 2000 21:38:14 -0500
Subject: Building and Transvering multi-level dictionaries
Message-ID: <01BF92B4.99B88AC0@cc264973-a.hwrd1.md.home.com>

Nathan,
	Thanks for the correction. I guess my many years of C programming shows through !

Cheers
Eric

-----Original Message-----
From:	Nathan Clegg [SMTP:nathan at islanddata.com]
Sent:	Monday, March 20, 2000 9:05 PM
To:	Eric Hagemann
Cc:	python-list at python.org
Subject:	Re: Building and Transvering multi-level dictionaries

On 21-Mar-2000 Eric Hagemann wrote:
>#add a key and setup for a sub-dictionary
> if(D.has_key('a') == 0)
>     D['a'] = {}

In my opinion, this is a very dangerous habit to get into.  Unlike C,
perl and python have more than one false value.  I don't know that
you can be sure that a built-in such as this is guaranteed to be
consistent between versions of python.  Even more dangerous is that
modules and functions written by others could use any of the false
values.  I personally always return None for false.  Seems to me much
safer to do:

if not D.has_key('a'):
        D['a'] = {}

----------------------------------
Nathan Clegg
 nathan at islanddata.com

From tim_one at email.msn.com  Tue Mar  7 00:32:56 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 7 Mar 2000 00:32:56 -0500
Subject: Python performance
In-Reply-To: 
Message-ID: <000d01bf87f6$97ffede0$2b2d153f@tim>

[Chris Ryland]
> ...
> I'm just curious (being an old language hacker for 30+ years but new to
> Python): why is Python performance ultimately any worse than Lisp?
>
> Is it because there's only been one major implementation effort
> so far, and so people haven't had a chance to learn the techniques
> required to make it run fast, while Lisp has had 30+ years to mature?
> ...

Python was implemented by essentially one guy in his "spare time", outside
of academia, and raw speed was never one of his primary goals.
Portabililty, robustness, maintainability, and ease of extending with C
*were* primary goals, and Python meets or beats mosts Lisps on all of those
counts despite having enjoyed less than a fraction of a percent of a percent
of the total development effort that's been poured into Lisps (or Javas or
Smalltalks, for that matter).

nothing-a-few-million-dollars-and/or-legions-of-grad-student-slave-
   labor-couldn't-repair-but-the-mere-passage-of-time-won't-change-
   it-ly y'rs  - tim

From Lisowski.Tomasz at sssa.nospam.pl  Sat Mar 25 02:41:19 2000
From: Lisowski.Tomasz at sssa.nospam.pl (Tomek Lisowski)
Date: Sat, 25 Mar 2000 07:41:19 GMT
Subject: Python on Alpha OpenVMS
Message-ID: 

Hi!

I have tried to compile Python sources on my VAX OpenVMS system, but soon I
realized, that instead of my trusty DEC C++ compiler I need a native C
compiler. Unfortunately my C++ compiler cannot be reverted to a native C
language mode. Has anyone succeeded to create VAX OpenVMS Python binaries?

Note: Apart from VAX OpenVMS I have also an Alpha OpenVMS server, but here I
don't have any C/C++ compiler available. If anyone has created Alpha OpenVMS
Python binaries, please contact me too.

Thanks in advance

Tomasz Lisowski

From clopez at abo.fi  Wed Mar 15 10:47:49 2000
From: clopez at abo.fi (Cesar Lopez)
Date: Wed, 15 Mar 2000 17:47:49 +0200
Subject: wrapping drivers for Python
Message-ID: <38CFB0A5.1636CAA8@abo.fi>

I?m want to wrap some drivers that I have running in my system, but I
want use this drivers with pythno programs.

I?ve build and linked statically an external module for my Linux machine
that emulates the behaviour of one of my devices, a green led. But I
don?t know how to access to the low-leve system I only print on/off on
my screen.

If you know how to wrap or access to the low-leve or your machine with
Python tell me.

Here is my driver code:

//led.h

int Setup(int state); // state = 1 --> ON ; state = 0 --> OFF
int static _state;

// Defining communication with the device, Port, Adress, configuration
registers
 and so  on ...
#define PBIOR (*(volatile short int *) (0x5ffffc6))
#define PBCR1 (*(volatile short int *) (0x5ffffcc))
#define PBCR2 (*(volatile short int *) (0x5ffffce))
#define PBDR (*(volatile short int *) (0x5ffffc2))

//led.c
#include "led.h"

int
Setup(int state)
{
PBIOR |= 0x8000;      // Set the bit for output
PBCR1 &= ~0xc000;     // Make it a normal I/0

        switch (state){
            case 1:{
              PBDR |= 0x8000; // Set the LED on
             _state=1; // to read the state of the led
                return 0;
                }
            case 0:{
              PBDR &= 0x0000; // Set the LED off
                            _state=0;
                return 0;
                }
            default: return -1;  // Error: Invalid code
            } //switch
}

From tuttledon at hotmail.com  Thu Mar  2 17:39:07 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Thu, 02 Mar 2000 22:39:07 GMT
Subject: Python and Windows Scripting Host
References:  <89jfkk$qcp$1@nnrp1.deja.com> <38BE7D61.E17FEBF2@gssec.bt.co.uk>
Message-ID: 

[Alan G.]
> Nope, WSH doesn't do any GUI stuff - its designed
> to replace BAT files agfter all!
>
???   Try running this script.

import win32com.client
shell = win32com.client.Dispatch("WScript.Shell")
rslt=shell.Popup("Would you like to run Notepad?", 0, "Python WSH Test!",
35)
if rslt == 6:
    shell.run("notepad")
else:
    shell.Popup("Maybe next time.", 4, "Bye!")

 Using Python to access WSH objects, like Mark Hammond suggested,  makes
alot more sense than using WSH as the controlling environment.  (providing
you not forced into a mixed environment like Gaetan has mentioned).
Although I still prefer WinBatch for Admin stuff.

Don

From tiddlerdeja at my-deja.com  Wed Mar 15 09:44:15 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Wed, 15 Mar 2000 14:44:15 GMT
Subject: read NT registry with python
References: <8alo1n$tdc$1@nnrp1.deja.com>  <8ansn6$ece$1@nnrp1.deja.com> <8anvhr$1756@newton.cc.rl.ac.uk>
Message-ID: <8ao7jt$m6t$1@nnrp1.deja.com>

Thanks for that, very much appreciated.

However, the second line:
    win32api.RegQueryValueEx('AdminPassword')

I get an error with this:

>>> win32api.RegQueryValueEx('AdminPassword')
Traceback (innermost last):
  File "", line 1, in ?
TypeError: RegQueryValueEx requires exactly 2 arguments; 1 given

Any help with this is greatly appreciated.

p.s. I realise this is lame me asking this, but I'm new to registry
terms and I don't know what the second var should be.

In article <8anvhr$1756 at newton.cc.rl.ac.uk>,
"Richard Brodie"  wrote:
>
>  wrote in message news:8ansn6
$ece$1 at nnrp1.deja.com...
>
> > [HKEY_LOCAL_MACHINE\SOFTWARE\WOF]
> >
> > [HKEY_LOCAL_MACHINE\SOFTWARE\ABC\PlusSoftware]
> > "AdminPassword"="Admin"
>
> import win32api
> from win32con import *
>
> k=win32api.RegOpenKeyEx(HKEY_LOCAL_MACHINE,
> 'SOFTWARE\\ABC\\PlusSoftware',0,KEY_QUERY_VALUE)
>
> win32api.RegQueryValueEx('AdminPassword')
>
>

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

From tim_one at email.msn.com  Tue Mar  7 00:19:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 7 Mar 2000 00:19:17 -0500
Subject: HELP: restore my faith in Python
In-Reply-To: 
Message-ID: <000b01bf87f4$b0444f60$2b2d153f@tim>

[bunching together several replies here]

[Tim]
>     int(0.999999999999999999993427) == 1
>     int(0.99999999999999999993427) == 1
>     int(0.9999999999999999993427) == 1
>     int(0.999999999999999993427) == 1
>
> and so on?  So long as you draw the line *somewhere*, you haven't solved
> anything, you've merely moved the surprises from where they're
> predictable across *all* IEEE-754 platforms to something unique to Python.
>
> The problems with floating-point are deep and can't be patched
> over.  "Quick hacks" always backfire.  If somebody wants to indulge in a
> notion of "safe epsilon", there's nothing to stop them from writing
> that explicitly.

[Konrad Hinsen]
> APL has a "comparison tolerance", implemented as a global (but
> localizable) variable that can be changed by the user, and which
> specifies the maximum relative error that is considered zero for the
> purpose of comparisons. It's a handy feature for quick calculator-type
> calculations, but it can get in the way when implementing more complex
> algorithms.

Note that the original poster was aiming at "fuzzing" float->int, though.
Fudging fp comparisons can almost be made principled, but e.g. there's no
reasonable way to do it without making it possible to find distinct x, y and
z s.t. x==y and y==z but NOT x==z.  Note too that APL is rife with implicit
"under the covers" float->int conversions, while Python has none (i.e., the
Python programmer must always say explicitly how they want float->int to get
done).

[John W. Baxter]
> ...
> 4.  A decimal floating point class could certainly be written for
> Python.

Such have been; AFAIK almost nobody uses them, though.  There's more call
for it in the business world, where a penny rounded away is a billion-dollar
lawsuit .

[Neel Krishnaswami]
> ...
> Is this possible? I've forgotten most of my analysis, but I was under
> the impression that you could divide the continuum as follows:
>
>   o Rationals -- eg 2/3. Countably infinite.
>   o Algebraic irrationals -- eg sqrt(2). Countably infinite.
>   o Transcendental numbers -- everything else. Uncountably infinite.
>
> So presumably you could write numeric classes that properly (if you
> don't care about speed or memory usage) handle rationals and algebraic
> irrationals, but the full transcendentals are just plain impossible.
> Maybe some interesting subset of the transcendentals are possible;
> if you throw in e and pi most people will be happy most of the time?
> Though on reflection this is getting pretty close to a full symbolic
> computation package.

What you're groping for is called the "constructive" or "computable" reals.
There's a well-developed theory, and a few scattered implementation pkgs.
The most recent influential text is by Pour-El and Richards -- curiously
enough, I studied mathematical logic under the former and real analysis
under the latter, so can vouch for their sterling characters .  A
sketchy intro to the main results can be found in Jean Vuillemin's "Exact
Real Computer Arithmetic with Continued Fractions", IEEE Transactions on
Computers, Vol 39 No 8, August 1990.  H-J Boehm did a widely used
implementation in the now-defunct Russell language, still available (last I
checked) from Rice University.  Just last year he released a more accessible
(but plainer) implementation in Java, available from somewhere or other on
the SGI web site.

As I said at the start, though, *all* approaches to real arithmetic suck in
deep ways.  For example, while exp(log(pi + 1e-5000)) is exactly equal to pi
+ 1e-5000 under the computable reals, equality is in general undecidable,
float->int also, and even figuring out the first digit of a constructive
real's decimal expansion is in general undecidable too.

> I guess Mathematica is that expensive for a reason -- arithmetic is
> harder than I thought. :)

Check out Macsyma (http://www.macsyma.com/).  A sixth the price and more
than adequate for most anything you'd like to explore.  It does not
implement the constructive reals, though.

[Michael Hudson]
> Food for thought: the computable numbers are countable.
>
> Proof left as an exercise for the reader (it's not hard!).
>
> (Is it even meaningful to ask whether they are [recursively]
> enumerable?)

Sure!  That's meaningful for any countable set.  It turns out that while
they're countable, they're not recursively enumerable.  If they were, you'd
have an effective algorithm for computing one not in the enumeration, via
the usual diagonalization trick.  Since you *can* enumerate all possible
programs, this may appear paradoxical.  The "out" is that deciding whether a
program produces a computable real is undecidable.  And if your programming
language is so weak that that question *is* decidable, then the same
diagonalization argument proves that it's not strong enough to capture the
notion of computable reals.

almost-as-much-fun-as-continuations-ly y'rs  - tim

From effbot at telia.com  Thu Mar 30 12:14:26 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 30 Mar 2000 17:14:26 GMT
Subject: extending python: c++ problem
References: <8bvmko$gki1@mailgate1b.telekom.de>
Message-ID: 

Torsten H?rmann  wrote:
> Does anyone know about the following problem and have a solution?
> I've written a python extension using c++. No Problems with msvc, but on
> HPUX with aCC. When importing the , the python interpreter can
> not find the "initmod" function in the created lib. Error message:
>
> >>> import mod
> Traceback (innermost last):
>   File "", line 1, in ?
> ImportError: dynamic module does not define init function (initmod)
> >>>

name mangling problems?

using "nm" to figure out what happens to you "initmod" function
might give you some clues...  (maybe iostream.h

the README and Misc/HPUX-Notes contain additional information
on building shared libraries under HPUX:

README:

    HP-UX: Please read the file Misc/HPUX-NOTES for shared libraries.
    When using threading, you may have to add -D_REENTRANT to the
    OPT variable in the top-level Makefile; reported by Pat Knight
    this seems to make a difference (at least for HP-UX 10.20)
    even though config.h defines it.

Misc/HPUX-NOTES:

    There are two important points.  First, the python executable must be
    linked with the -E option to explicitly export all symbols.  This
    works with the vanilla interpreter, but I am not sure how friendly it
    will be when I try to embed the interpreter in a larger application.
    It may be necessary to hand tune the exports using the -e option.
    Anyway, the additional flag to $(CC) is "-Wl,-E", which passes the -E
    flag to the compiler.  My link line (from an actual run) looks like
    this:

        cc config.o -Wl,-E libModules.a  ../Python/libPython.a
        ../Objects/libObjects.a ../Parser/libParser.a   -lm  -ldld
        -o python

    [Guido's note: as of Python 1.5, replace the four libraries with
    ../libpython$(VERSION).a]

    Second, the dynamic module must be compiled with the +z option to make
    it position independent and then linked into a shared library:

        ld -b -o module.sl 