Found

Apache/1.3.12 Server at caesar.num.uni-sb.de Port 80 · ') def end_from(self): self.handle_data('

 tags
containing Python examples) where this would foul up the
result beyond the pure aesthetics of xml indentation.

I just wanted to hear if anybody knows of an alternative
solution, or if anybody was willing to fix this. Otherwise
I might try to write a patch myself, or, failing that,
write a separate tool... (This may, of course, be solvable
completely in xslt, though I'm not sure how.)

--

  Magnus Lie Hetland         http://www.hetland.org

 "Reality is that which, when you stop believing in
  it, doesn't go away."           -- Philip K. Dick

From cjensen at bioeng.ucsd.edu  Wed Oct  3 12:24:29 2001
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 03 Oct 2001 09:24:29 -0700
Subject: Tkinter image mapping
Message-ID: <3BBB3BBD.602470C2@bioeng.ucsd.edu>

Is there a convient way to do image mapping in Tkinter?  
ie:  We display an image in a Tkinter frame, and we would like to set a
binding to a mouse click in a certain area of the image.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451

From emile at fenx.com  Tue Oct 16 17:12:13 2001
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 16 Oct 2001 14:12:13 -0700
Subject: breaking a string
References: <9qi6uu$h4c$1@nserve1.acs.ucalgary.ca>
Message-ID: <9qi85i$o2t3m$1@ID-11957.news.dfncis.de>

Depending on the version of python, use
string.split("Hello World!"[:-1])
or
"Hello World!"[:-1].split()

--

Emile van Sebille
emile at fenx.com

---------
"Michael Yuen"  wrote in message
news:9qi6uu$h4c$1 at nserve1.acs.ucalgary.ca...
> I'm pretty new to this Python stuff and i'm trying to figure out how to
> break a string into it's component parts.
>
> For example, if a line of text in a file is:
> "Hello World!"
>
> I want to enter that into my list as just ['Hello', 'World']
> Can someone help me out?
>
> Thanks,
> M
>
> --

From jnlc at bellsouth.net  Wed Oct 31 11:13:44 2001
From: jnlc at bellsouth.net ( 1)
Date: Wed, 31 Oct 2001 11:13:44 -0500
Subject: newbie question
Message-ID: 

I must have missed it in the tutorial or some thing.

I want to know how to open and run a file in python. The tutorial said to
just type " python filename.py" and hit enter. I did this and it comes back
with an error message telling me the the file name is not defined. I wasn't
told that I had to define it first before I can run it. Is there an easier
way to run files otherthan defining them?

thanks
lancer

From MarkH at ActiveState.com  Mon Oct 29 17:16:27 2001
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 29 Oct 2001 22:16:27 GMT
Subject: Windows/COM events/strange behavior
References:  
Message-ID: <3BDDD56E.4040003@ActiveState.com>

Krzysztof Czarnowski wrote:

> Sat, 27 Oct 2001 05:42:26 GMT
> Mark Hammond wrote:
> 
> 
>>As I said, this main thread probably needs to be calling
>>PumpWaitingMessages() each time around the loop.
>>
> Well... This definitely doesn't look very elegant to me!

I don't recall saying it did to me :)  One extra function call in your 
main loop is hardly an imposition though.

> Anyway, if I got it well:
> 
> 1. I should introduce CoInitialize()/CoUninitialize() pair in the main
> thread *and* call PumpWaitingMessages() in its loop? Right? I assume that no
> COM event will be delivered here, or will it?

The main thread does not need a CoInitialize call.  pythoncom 
automatically calls the apartment threaded model version.

> 2. No need to do this in other child threads (like the CORBA thread). Right?

It must be done on every thread that calls any COM function or uses any 
COM objects.

> 3. Will the problem go away if:
> 
>  * the COM event handler is installed in a "free threaded appartment"? (can
> I do this in a threading. thread?)
> 
> or
> 
>  * native Windows threads from ? (now module threading is used)

It depends on the app.  I haven't followed your code/this thread closely 
enough to remember what thread the COM event originates from.

> And generally, what would be *your* favorite solution? (my task is basically
> to receive events from a COM driven subsystem and pass them to a CORBA
> subsystem)

Again, it depends on the app.  However, my favourite solution is 
generally the one that works :)

"Python Programming on Win32" has a reasonably complete discussion on 
threads in COM from Python's POV.

Mark.

From sebrosa at artenumerica.com  Mon Oct 22 05:34:40 2001
From: sebrosa at artenumerica.com (Jose' Sebrosa)
Date: Mon, 22 Oct 2001 10:34:40 +0100
Subject: suggestion for module re
Message-ID: <3BD3E830.3EC94E94@artenumerica.com>

Hi

My first goal is to get all the *named* fields of a regex, for all of the
matches in a source string.  The re.RegexObject.findall method is "almost"
good, but it returns a tuple of matched *groups*, leading to a dumb
pharentesis-counting task and to code hard to maintain.  Just imagine that you
have a regex with several groups and wish to insert a new group in the middle. 
This changes the group counting, and that's no good.

In the beginning I was thinking about making a variant of the findall method
returning a dictionary of named groups instead of a tuple of groups.  But we
have already the re.MatchObject.groupdict method, (along with a bunch of other
useful match objects' methods), so I found it better to make a list of all the
match objects found in a source string.

Here is the method I would like to add to re.RegexObject (sure it could be
better named...)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    def find_all_match_obj(self, source, method = 'match'):
        """Return a list of all non-overlapping match objects in the string.
        """
        pos = 0
        end = len(source)
        results = []
        if method == 'match':
            worker = self.match
        elif method == 'search':
            worker = self.search
        else:
            if type(method) != type(''):
                raise TypeError, ('Invalid type for method: %s'
                                  % `type(method)`)
            raise ValueError, 'Invalid value for method: %s' % `method`
        append = results.append
        while pos <= end:
            m = worker(source, pos, end)
            if m == None:
                break
            append(m)
            pos = max(m.end(), pos+1)
        return results
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Such a method would let users to use the full functionality of match objects in
very simple python constructs.  For example, my first problem (to get all the
named groups of all matches) get solved like this (silly example follows):

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
rx = re.compile(r'((?Pa|A)(?Pb|B))')
match_list = rx.find_all_match_obj('abaBAbAB')
named_group_list = map(lambda m: m.groupdict(),  match_list)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Notes:

 - I'm using Python 1.5.2 (under Linux), but I inspected the docs for Python
2.1 and it seems to me that it too lacks a method like this.

 - I don't know which is better to use, "match" or "search", as the matching
function, so I used both...  But it still remains the question of which default
should we define.  Or if we should define *two* different methods, one with
match and other with search.  What's the usual policy?

 - I like to use the clean syntax obj.name instead of the hairy obj['name']
whenever possible.  Sure it is easy to go from one to another, but it is a pity
to have a groupdict method and not an equivalent groupobject method to return
an object usable with the cleanest syntax.  With it, we could look at a regex
with named groups as a way to define a set of standard names in Python (and to
give them values).  Furthermore, the requirements for named groups in regexs
are that the names must be valid Python names, so it is *really* a pity...! 
Anyway, that's just an aesthetic modification.

Thanks,
Sebrosa

From info at devshare.de  Thu Oct 25 05:21:38 2001
From: info at devshare.de (Johannes Gamperl)
Date: Thu, 25 Oct 2001 11:21:38 +0200
Subject: Learing Python, Newbie question
Message-ID: 

Hello,

i just starting to learn python ... so i've two questions:

1. is there a easy way to test, if a file exists?
2. do you have an example to do a template work for cgi?

   e.g.

   you have this template

   ...

   %name%  

   ... 

   so what i want is, to open this file and replace %name%
   with a variable inside the py-script ...

thx for any help .. :-)
Hannes

From kalle at gnupung.net  Sat Oct 20 05:15:41 2001
From: kalle at gnupung.net (Kalle Svensson)
Date: Sat, 20 Oct 2001 11:15:41 +0200
Subject: struct module
In-Reply-To: <3bd10af8$0$27331$afc38c87@news.optusnet.com.au>
References: <3bd0f13e$0$27332$afc38c87@news.optusnet.com.au>  <3bd10af8$0$27331$afc38c87@news.optusnet.com.au>
Message-ID: <20011020111540.A27786@sandra.lysator.liu.se>

[nobody]
> > H has a two-byte alignment requirement.  There's one byte for the "B",
> then
> > it has to throw in a pad byte to get the correct word alignment for the
> > "H".
> 
> How do I disable that?

You don't.

Peace,
  Kalle
-- 
[ Thought control, brought to you by the WIPO! ]
[ http://anti-dmca.org/ http://eurorights.org/ ]

From db3l at fitlinxx.com  Thu Oct 18 19:49:44 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Oct 2001 19:49:44 -0400
Subject: Is there a small Python runtime environment installer?
References: <3583b331.0110181524.5695bd67@posting.google.com>
Message-ID: 

tex_r at hotmail.com (Tex Riddell) writes:

> More specifically I'm looking for something for the windows
> environment, but one should be able to make something like this
> multi-platform, right?

Well, you'd have a harder time having a portable distribution if you
were looking for a binary, sort of by definition.

But since you're specifically targetted at Windows, why not just use
the standard Windows installer?  It lets you ignore the examples and
documentation and just install the core stuff if you like.  Any
Windows user should be completely at home with the installation
process.

> How can we expect python to become more mainstream if only developers
> can understand the environment that must be set up for scripts to run?

At least under Windows, I don't think installing Python is any harder
than installing any other application.  That's largely true as well on
various Unix platforms, since there are rpms and other install
packages written for Python to conform to local platform expectations.

> I really don't have time to learn distutils (re-inventing the wheel
> that I'm sure so many have done) in order to install a runtime
> environment on someone's machine so they can run a simple 8k script.

Of course, those same people (under Windows) have already installed
MBs of common code just so they can run small VB applications, whether
via their initial Windows installation or updated silently in other
stuff they may have installed. :-)

BTW, there is also the option of constructing a standalone
installation (for Windows, look into the py2exe or installer packages,
and installer also handles Linux) for your script.  For Windows, it
would include the Python DLL and appropriate lib files all packages
into a single file or collection of files that you could then
distribute independently of any other.  On the Windows platform if you
combine that with an installer - like the free Inno Setup - you'd end
up handing your script to your friend as a normal Windows installation
program that took care of installing everything.

Of course, this would be much larger than an 8K script and would thus
be hellishly less efficient if you were sending a lot of scripts this
way (as opposed to just having him install the runtime one time), but
for a one-shot deal it would make the receiver's life simple.

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

From sdm7g at Virginia.EDU  Sat Oct 20 11:17:47 2001
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Sat, 20 Oct 2001 11:17:47 -0400 (EDT)
Subject: silly idea - interesting problem
In-Reply-To: <20011020122801.A3963@cruciatuz.de>
Message-ID: 

On Sat, 20 Oct 2001, Stefan Antoni wrote:

> 
> i got a silly idea about a small script which takes to strings (lets say
> "test" and "dust") and mixes them to "tdeusstt".
> i tried to code it because i thought it would be easy, but now i cannot
> find out _how_ to do it ;)
> 
> i tried to "list()" the string and mix the items. but ...
> 
> Anybody has a hint or even a solution?

>>> import operator
>>> reduce( operator.__add__, map( operator.__add__, 'test', 'dust' ), '' )
'tdeusstt'

-- Steve

From dubal at khimjiramdas.com  Wed Oct 31 02:17:01 2001
From: dubal at khimjiramdas.com (J Dubal)
Date: 30 Oct 2001 23:17:01 -0800
Subject: Help: python-postgres-informix interaction problems
Message-ID: 

Hello good people!

We have developed Informix RDBMS based commercial applications in
Informix-4gl. It is a good procedural language with embeded SQL. We
are trying to rewrite some apps in Python with postgresql backend for
licensing reasons. Python is good too. We are trying to do this with
latest versions of RH linux (7.2), python 2.1 and postgres 7.1.3. We
face following issues:

1. We have to read data from informix db, process it and insert into
postgres. We are using informixdb module (DBAPI 1) which is bit dated.
In Informix we 'select custno, custname, custaddr1, ... into progvar1,
progvar2, ... from custmf where ...' In python, we say fetchone() and
we get a list. We are then moving the elements of the list to named
program variables thus:
        pcustno = list1[0]
        pcustname = list1[1]
        pcustaddr1 = list1[2]
        ... as many as reqd
This way we give meaningful names to data items read from db and
process further.
Is there a better, more elegant/efficient way to to this?

2. To insert this data into postgres, we used to use pygres included
in the postgres distribution (latest). We had to stop using this
because we couldn't find a way to find out how many rows it have
inserted, updated or deleted after such statements are executed ie we
couldn't find a way to check the status or return code. So we switched
over to PoPy which is DBAPI2 and where we could check number of rows
affected. Is this right thing to do?

3. In order to insert data into postgres, we have to compose an insert
statement thus:
        cur.execute("insert into custmf (custno, custname,
custaddr1,...)
            values (%i,'%s','%s'...)" %(pcustno,pcustname,
pcustaddr1....))
Is there a better way of doing this?

4. The above string substitute converts program variables holding null
value into 'None' which goes into target table as string 'None'. To
avoid this we parse all columns to check if they are 'None' and
replace them by ''. Is there a better way to do this?

5. The above check works only on string variables. We have not found
yet a way to handle nulls in numeric columns. Please suggest a
solution.

6. If the character columns happen to contain quotes, the above
composed insert statement breaks. Please suggest a solution.

Any help is greatly appreciated.
Thanks in advance.
Dubal

From lucio at movilogic.com  Tue Oct 23 11:47:16 2001
From: lucio at movilogic.com (Lucio Torre)
Date: Tue, 23 Oct 2001 12:47:16 -0300
Subject: POLL in different OSes
In-Reply-To: <3bd50785@news.actrix.gen.nz>
References: 
Message-ID: <5.1.0.14.0.20011023124514.088ce2e0@imap.hq.movilogic.com>

At 05:54 PM 23/10/2001 +1300, Mark Rowe wrote:
>Hello,
>
> >From the interpreter under Windows:
>
>Python 2.2b1 (#25, Oct 19 2001, 11:44:52) [MSC 32 bit (Intel)] on win32
>Type "help", "copyright", "credits" or "license" for more information.
> >>> import select
> >>> select.poll
>Traceback (most recent call last):
>   File "", line 1, in ?
>AttributeError: 'module' object has no attribute 'poll'
> >>> dir(select)
>['__doc__', '__file__', '__name__', 'error', 'select']
> >>>
>
>I guess its not :)
>
>Mark Rowe

Yes, i did the same in my linux box and got:

Python 2.0.1 (#0, Jun 23 2001, 23:50:30)
[GCC 2.95.4 20010319 (Debian prerelease)] on linux2
Type "copyright", "credits" or "license" for more information.
 >>> import select
 >>> dir(select)
['POLLERR', 'POLLHUP', 'POLLIN', 'POLLMSG', 'POLLNVAL', 'POLLOUT', 
'POLLPRI', 'POLLRDBAND', 'POLLRDNORM', 'POLLWRBAND', 'POLLWRNORM', 
'__doc__', '__file__', '__name__', 'error', 'poll', 'select']
 >>>

so, another question: is bsd sockets the only networking interface Windows 
has? doesnt it have a win32socketlib ( :) ) or something like that?

and deeper again, (for pygame people), does the networking code from 
directx use bsd sockets? or what?

danke.

From matt at mondoinfo.com  Mon Oct 29 19:05:34 2001
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Tue, 30 Oct 2001 00:05:34 GMT
Subject: tkinter - tkFileDialog - how to get dir instead of file
References: 
Message-ID: 

On Mon, 29 Oct 2001 23:14:44 +0100, maxm  wrote:

Max,

>I am writing my first tkinter program and so far it's been pretty painless.

That's great, you're ahead of the game.

>But now I need to open a directory and not a file. I cannot see a
>simple way to get this from the standard module "tkFileDialog".

>Do I really have to write a special class for this?

Alas, you probably do. Tk provides a directory chooser but Python
doesn't (yet) expose it. I filed a feature request on SourceForge for
it but it doesn't look like anything has been done about it yet,
including by me.

Regards,
Matt

From nobody at tek.kom  Tue Oct 16 18:36:38 2001
From: nobody at tek.kom (Kevin)
Date: Tue, 16 Oct 2001 22:36:38 GMT
Subject: Python app. Eroaster and "output = pipe[0].readlines()" trouble
References: 
Message-ID: 

I checked "cdrecord -scanbus" and it writes everything to
stdout.  So, I took out the "2>&1".  I got exactly the same
results, namely:

    Traceback (most recent call last):
      File "/usr/lib/python2.0/site-packages/gtk.py", line 125, in __call__
	ret = apply(self.func, a)
      File "/usr/lib/eroaster/main.py", line 1468, in DetectRecorder
	self.DetectSCSIDevice(deviceRecorder)
      File "/usr/lib/eroaster/main.py", line 1479, in DetectSCSIDevice
	output = pipe[0].readlines()
    IOError: [Errno 4] Interrupted system call

The line I think caused this is the "output = pipe[0].readlines()"
line in the following:

    # Detect SCSI-devices
    def DetectSCSIDevice(self, device):
	pipe = popen2.popen2("cdrecord -scanbus 2>&1")
	output = pipe[0].readlines()
	devicelist = []

Any more ideas?  What is "output = pipe[0].readlines()" supposed
to do?  How can I test that?  Thanks...

In article ,
 Ignacio Vazquez-Abrams  writes:
> The problem is most likely the "2>&1" in the call to
> popen2(). Redirection is a shell function; popen*() just
> execute the process seperately, so cdrecord thinks that "2>&1"
> is a (bad) argument. It should use popen4() instead.

--
Unless otherwise noted, the statements herein reflect my personal
opinions and not those of any organization with which I may be affiliated.

From daves_spam_dodging_account at yahoo.com  Fri Oct 19 17:59:19 2001
From: daves_spam_dodging_account at yahoo.com (David Brady)
Date: Fri, 19 Oct 2001 14:59:19 -0700 (PDT)
Subject: Reverse argument order
Message-ID: <20011019215919.3702.qmail@web21108.mail.yahoo.com>

Thanks, everyone, for your answers.  Also, thanks to
those of you that ways of asking if the question
really needed to be asked.

At the time I had posted, we had considered all of the
non-answer answers that were posted here.  Most were
rejected by the business half of the team, because
they have a goal of reducing the keystroke count as
much as possible.  They're already upset that the new
scripting language has to have parentheses everywhere.
 :-)

Specifically,

We considered using named arguments, but rejected them
because even typing Wait(m=3,s=10) was judged to be
too hard, let alone Wait(minutes=3, seconds=10). 
Personally, I like this approach best, and I want to
just write in the manual, "Look, you're going to have
to learn to type, okay?"  But that's just me.

We also considered using Wait(0,0,5) for a 5-second
wait, but this was also shot down.

One possible compromise was offered by the business
half of the team, that they would be willing to type
Wait(,,5), but *WE* shot that down on the grounds that
it does not work.  (They were thinking it would
evaluate to Wait(None,None,5) which could then have
been turned into 0,0,5).

We considered the string bit, but strings require
quotes....

We even considered modifying the Python source so that
it would accept 00:00:00 as a token identifying a time
type (that we would add).  This was discarded by
everyone as WAY too much effort.  (Imagine getting the
lexer to accept things like "while t < 1:05: pass")

Anyway, thank you all again for your answers and help.
 This project is starting to shape up!

-dB
P.S. We're going to go with the "while len(args)<3:
args.insert(0,0)" model for now.  Much less sucky,
yes, thanks!

=====
--
David Brady
daves_spam_dodging_account at yahoo.com
I'm feeling very surreal today... or *AM* I?

__________________________________________________
Do You Yahoo!?
Make a great connection at Yahoo! Personals.
http://personals.yahoo.com

From phr-n2001d at nightsong.com  Wed Oct 10 03:29:19 2001
From: phr-n2001d at nightsong.com (Paul Rubin)
Date: 10 Oct 2001 00:29:19 -0700
Subject: Curious assignment behaviour
References: 
Message-ID: <7xsncs9cmo.fsf@ruckus.brouhaha.com>

"Tim Peters"  writes:

> [Paul Rubin, on an expression assignment operator]
> > Why on earth does it have to be different from the statement assignment
> > operator?  What's wrong with
> >
> >   while x=get_next(): whatever(x)
> >
> > I just don't buy the rationale that using = instead of == is a big source
> > of bugs.  In 25 years of hacking C code, I think I've seen that error
> > maybe once or twice.
> 
> I suggest you're highly atypical here, Paul.  This was the very first of the
> gotchas listed in Andrew Koenig's classic (late 80s) "C Traps and Pitfalls",
> and has made nearly every list of C "deadly sins" I've seen since then.
> I've wasted weeks of my own life helping people track this error down in C,
> too often under extreme time or customer pressure.  The experience of the
> other folks at PythonLabs is similar, so even if it's a disease you're
> immune to, you can be sure this will never go in.

The thing is, it's just not a matter of MY immunities.  I've spent
plenty of time debugging other people's code too, and have encountered
all kinds of errors in both their code and mine.  The =/== mistake has
just been very rare in my experience.  Can it be that the Unix
compilers I use flag the error a lot of the time, so it gets fixed
quickly, while Windows compilers don't flag it, and you're mostly
using Windows compilers?

Anyway, I can live with "while x:=get_next()" instead of x=get_next.
But I feel seriously cramped in Python when I have to say

  while 1:
     x=get_next()
     if not x: break
     whatever(x)

so I hope something is done about the issue.

> Oops!  The assert turned out to be the perversely self-fulfilling:
> 
>     assert(i = 1);

Ouch!!!! ;-)

From sholden at holdenweb.com  Thu Oct 25 11:22:23 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 25 Oct 2001 11:22:23 -0400
Subject: A small inconsistency in syntax?
References: 
Message-ID: 

"Michael Abbott"  wrote in message
news:Xns91459674D7271michaelrcpcouk at 194.238.50.13...
> It seems to me that there's a little inconsistency in assigning to tuples:
>
>     >>> (x,) = ()
>     Traceback (most recent call last):
>       File "", line 1, in ?
>     ValueError: unpack tuple of wrong size
>
> Yes, that's what I expect.
>
Indeed.

>     >>> () = ()
>     SyntaxError: can't assign to ()
>
> Eh?  Why not?  It's just a matter of checking that the right hand side
> really is an empty tuple...
>
Oh, no it isn't. The best parallel I can think of is writing

>>> 2 = 34
Traceback (SyntaxError: can't assign to literal

although in your case you are trying to assign to a constructor rather than
a literal
.
> I actually hit this in some code, where I wrote:
>
>     time, status, () = myobject.read()
>
> where I happen to know that that in this particular case my object will
> return an empty tuple in its third argument.
>
> Ok, so it's easy enough for me to write
>
>     time, status, _ = myobject.read()
>
> instead, so I'm grumbling about trivia here.

Indeed :-)

Or, if you really don't want to do that, how about

    time, status = myobject.read()[:2]

regards
 Steve
--
http://www.holdenweb.com/

From mmichalski3 at home.com  Tue Oct 16 14:13:47 2001
From: mmichalski3 at home.com (MM)
Date: Tue, 16 Oct 2001 18:13:47 GMT
Subject: mod_python seems to be missing a file
References:  
Message-ID: 

Here is the apache error generated when I try to view a file named
'mptest.py':

######START
Mod_python error: "PythonHandler mptest"

Traceback (most recent call last):

  File "C:\PYTHON21\lib\mod_python\apache.py", line 181, in Dispatch
    module = import_module(module_name, _req)

  File "C:\PYTHON21\lib\mod_python\apache.py", line 335, in import_module
    f, p, d = imp.find_module(parts[i], path)

ImportError: No module named mptest
######END

Here is the configuration directive in the httpd.conf file:

    AddHandler python-program .py
    PythonHandler mptest
    PythonDebug On

I'm not sure what you mean by Python runtime and when the file's loaded.
I've placed the mod_python.dll in the modules directory of Apache and added
the appropriate LoadModule directive into the configuration file.  When
Apache loads it tells me "Apache 1.3.20 (Win32) mod_python/2.7.4
Python/2.1.1 running..."

That's the only configuration I've done with respect to Apache and Python.
Please let me know if I'm missing something.

Jan Dries  wrote in message
news:mailman.1003176151.32729.python-list at python.org...
> MM wrote:
>
>  > I've followed the mod_python documentation to the letter.  It seems
> to be on
>  > the verge of working with the one following problem:
>  >
>  > from mod_python import apache
>  >
>  > this is the first line of the test program you're told to write to see
if
>  > you're up and running.  The problem with this is that the apache.py
> file in
>  > the mod_python package is trying to import a file called _apache.  I
> cannot
>  > find this file anywhere and, apparently, neither can Python, since it
>  > complains about it, as well.  Any help on where this file can be
> found would
>  > be appreciated.
>  >
>
>
> If you check the C source, you will find that the mod_python DLL itself
> export a function init_apache().
> In other words, mod_python.dll is your missing _apache.pyd. When
> mod_python.dll is loaded, the function is registered to the Python
> runtime, and from that moment on the module _apache is known to the
> Python interpreter.
>
> If you could tell what exactly the errors are that you get in trying to
> make mod_python work, maybe I could tell what's going wrong. But it
> certainly isn't the fact that some _apache.py(d) would be missing.
>
> Regards,
> Jan
>
>
>
>

From phr-n2001d at nightsong.com  Sun Oct 28 09:51:19 2001
From: phr-n2001d at nightsong.com (Paul Rubin)
Date: 28 Oct 2001 06:51:19 -0800
Subject: interpreter crashes
References: <7xzo6cd1hw.fsf_-_@ruckus.brouhaha.com> <3BDB8F30.FEE9004C@engcorp.com> <7xd738qr2f.fsf@ruckus.brouhaha.com> <3BDC016C.7E45106@engcorp.com>
Message-ID: <7x4rojvmvs.fsf@ruckus.brouhaha.com>

Peter Hansen  writes:
> All crashes I've ever seen were with other libraries, or when
> I was trying to wrap a DLL with calldll under Windows.  Never 
> seen a crash yet under Linux.  Python is _solid_.

This was Python 2.1.1 under Redhat 7.1.  I am 100% positive I wasn't
using any 3rd party extension modules.  However I had some standard
library modules loaded--I'm not sure which ones.  Also, I had been
repeatedly reloading my own Python modules, which has always seemed
like a flaky process to me.

> Run a separate task which checks the status of the Python app from
> time to time (maybe with cron?).  Kill and restart as necessary.

Good idea, I guess.  Although the app needs to keep some persistent
data structures in ram (for security reasons it can NOT checkpoint
them to disk, so don't bother suggesting that).  Maybe I have to use
the shm module or something like that.

Thanks

From jmcwill at tridog.com  Wed Oct 10 17:16:32 2001
From: jmcwill at tridog.com (Jason McWilliams)
Date: Wed, 10 Oct 2001 15:16:32 -0600
Subject: upload file using cgi on Win2k IIS
References: 
Message-ID: <3BC4BAB0.DB2F605A@tridog.com>

Ignacio Vazquez-Abrams wrote:
> 
> On Wed, 10 Oct 2001, Jason McWilliams wrote:
> 
> > I seem to be doing everything correctly to save a posted
> > multipart/form-data file on IIS server.  Yet, I only
> > seem to be able to save a small portion of the file, not
> > the whole thing.
> >
> > The internet user has permission to the temp directory and
> > I am not getting any errors.  It works properly with the
> > same code on unix, so it must be a IIS thing?
> >
> > I have tried file.read file.value and also tried to just
> > do one line at a time... file.readline .  I can get the
> > file name fine using cgiobj.filename  ... etc.
> >
> > I get about 498 k as the file size on the uploaded file, never
> > more and I am trying to upload a big file.  Maybe there is
> > a post limit in IIS?  I know that you can set a GET limit.
> 
> What is the value of the hidden field named "MAX_FILE_SIZE" set to? Is there
> one at all? If not, then don't expect your file input to work properly.

I did not think this variable was used in cgi.py
Isnt that a php type thing.  Isn't maxlen set to 0 inside cgi.py
to allow for unlimited post length?  That is why I thought
it might be a IIS setting somewhere.

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

From cqx at nefud.org  Mon Oct 22 00:11:19 2001
From: cqx at nefud.org (cqx)
Date: 21 Oct 2001 21:11:19 -0700
Subject: Ops: corrected prime algorithm
References: 
Message-ID: 

John Thingstad  writes:

> Just checked my own algorithm and found several mistakes.
> Here is the correct algorithm:
> 
> 
> def prime(n):
> 	set = [2] + range(3,n, 2)
> 	for n in range(3, int(math.sqrt(n))+1, 2):
            ^                           ^
 which 'n' is this 'n'?                 and this one?

> 		set = [x for x in set if x == n or x % n != 0]
> 	return set

-cqx

From Armin_member at newsguy.com  Thu Oct 25 06:52:06 2001
From: Armin_member at newsguy.com (Armin Steinhoff)
Date: 25 Oct 2001 03:52:06 -0700
Subject: cPickle
References: <9r3ctr02o31@drn.newsguy.com> 
Message-ID: <9r8qsm02d3a@drn.newsguy.com>

Oleg,

thanks a lot for that hint ... it's working now :)

Armin

In article , Oleg says...
>
>On Tue, Oct 23, 2001 at 02:23:07AM -0700, Armin Steinhoff wrote:
>> are there any know problems with cPickle which does prevent reading of a
>> 'pickled' file from a Windows machine on a machine with an other operating
>> system ??
>>  (e.g. QNX6) 
>> 
>> Any ideas about the reasons ??
>
>   On broken Windows OS you must open binary files in binary
>modes (and cPickle fies are binary):
>
>   infile = open(filename, 'rb')
>   outfile = open(filename, 'wb')

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

From sholden at holdenweb.com  Tue Oct 30 15:15:57 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 30 Oct 2001 15:15:57 -0500
Subject: Unpacking extension (Re: A small inconsistency in syntax?)
References: 
Message-ID: 

 wrote in ...
>
> Bernhard Herzog wrote:
> >Personally, I wouldn't mind to see the "[a,b] = sequence" syntax go.
>
> I agree.  I think it's confusing and adds no value (unless I'm missing
> something).
>
But since it's been permissible for a long time, removing it might break
existing code. Would you propose a PEP to go through the "from __future__
..." phase and arrive at "feature removed"? As far as I'm concerned it
remains a quirk, which I could easily live without but equally easily
ignore.

regards
 Steve
--
http://www.holdenweb.com/

From info at mjais.de  Tue Oct  9 15:18:48 2001
From: info at mjais.de (Markus Jais)
Date: Tue, 09 Oct 2001 21:18:48 +0200
Subject: Why not 3.__class__ ?
References: 
Message-ID: <9pvilt$kepge$1@ID-75083.news.dfncis.de>

brueckd at tbye.com wrote:

> On Tue, 9 Oct 2001, Markus Jais wrote:
> 
>> In article , "Guido van
>> Rossum"  wrote:
>>
>> > Marcin, I don't understand why you care so much about being able to
>> > write 3.foo.  It's painful to fix in the lexer, and probably leads to
>> > less useful error messages if someone makes a mistake in a float
>> > literal
>> > (e.g. "3.e 0").  And I see zero use for it: in practice, you will never
>> > ask for 3.foo -- you'll ask for x.foo where x happens to contain the
>> > value 3.
> 
> [snip]
>> but to write 3.foo is something I would like to have in Python
> [snip]
>> It would be great, if this would be possible in Python too
> 
> Care to elaborate on why it would be "great"? Guido's reply to Marcin
> points out that there doesn't seem to be any usefulness to it. So.... what
> problem would it solve or simplifiy?  What exactly would its benefit be,
> anyway?

it would be more consise.
in Ruby, everything is an object and I can invoke methods on everything
with object.method

the result is the same.
I just think, that ruby is "more" object oriented like Python
(another reason for this is, that in Ruby every instance variable
is private!)

but  this is just a matter of taste.
again. both are cool languages and I like both.
markus

From reqhye72zux at mailexpire.com  Sat Oct  6 17:15:38 2001
From: reqhye72zux at mailexpire.com (Robert Amesz)
Date: Sat, 06 Oct 2001 21:15:38 -0000
Subject: Tkinter Html Display
References: <3daa1aa8.0110051100.33815499@posting.google.com>
Message-ID: 

Jeffrey wrote:

> Hi All,
>   I want to display an HTML file on a Tkinter Canvas widget.  Do any
> of you know how to do this?  The principle problem I have is that the
> HTML file can display subscript and symbols and stuff, while
> translated the HTML lines into text, you lose that.  I came up with
> one solution that one could just capture the HTML off the browser and
> turn it into a gif file and then display on the widget.  But I just
> wanted to know if there is anything out there or any solution for
> this....

How desperate are you to be able to display HTML? If it's important 
enough, you might consider switching to wxPython: it has an HTML-
component which can display all common HTML 3.2 elements, like fonts, 
pictures and tables.

Robert Amesz

From tefol at This.Is.Not.An.Address  Sun Oct  7 08:12:46 2001
From: tefol at This.Is.Not.An.Address (tefol)
Date: Sun, 07 Oct 2001 12:12:46 GMT
Subject: newbie form questions...
Message-ID: <9133E264AtefolThisIsNotAnAddr@61.9.128.12>

Hiya everyone....

I am trying to get a form to create a file (or append to a file if the file 
already exists, maybe..  haven't quite decided what to do in case the form 
is processed too foten to quickly yet) on a webserver as a result of a POST 
to a http form.  How I want the file to look is something like this:

field1=stringOfField1
field2=stringOfField2
field3=stringOfField3
field4=stringOfField4
field5=stringOfField5

or alternatively, 

field1 stringOfField1
field2 stringOfField2
field3 stringOfField3
field4 stringOfField4
field5 stringOfField5

But what I get is this:

(i__main__
FeedbackData
p0
(dp1
S'field5'
p2
S'stringOfField5'
p3
sS'field4'
p4
S'stringOfField4'
p5
sS'field1'
p6
S'stringOfField1'
p7
sS'field3'
p8
S'stringOfField3'
p9
sS'field2'
p10
S'stringOfField2'
p11

What the heck is all that stuff?   Is there a way to make the output be as 
I require?  Or should I create some huge regex thing to process this output 
so that it appears as I want??

Or is there an entirely different approach to take?  Basically,  I want a 
script to create a file that a cron script will regularly search for.  If 
it finds it,  it should process the information accordingly.   My script 
and html follow.  

TIA

Tefol

+++++++++++

FORM HTML:

Python Form Test

All fields are required, or else the form funtion cannot proceed.

    Field1

    Field2

    Field3

    Field4

    Field5

++++++++

Python script.

#!/usr/bin/python2

import cgi, os, sys, string, pickle

def bail():
    print "Content-type: text/html\n\n"
    print "You didn't fill something out"
    print "

Double check each field!
"
    print "Press the 'Back' button to do it again."
    sys.exit()

class FormData:
    '''CGI Form information'''
    def __init__(self, form):
        for fieldname in self.fieldnames:
            if not form.has_key(fieldname) or form[fieldname].value == "":
                bail()
            else:
                setattr(self, fieldname, form[fieldname].value)

class FeedbackData(FormData):
    fieldnames = ('field1', 'field2', 'field3', 'field4', 'field5')
    def __repr__(self):
        return '%(name)s' % vars(self)

if __name__ == '__main__':
    sys.stderr = sys.stdout
    form = cgi.FieldStorage()
    data = FeedbackData(form)

pickle.dump(data, open('/home/nemir/output/form.txt', 'w'))

print "Content-type: text/html\n\n"
print "You did it!"
sys.exit()

From ozonehole2k at yahoo.com  Tue Oct 30 01:44:20 2001
From: ozonehole2k at yahoo.com (Ozone Hole near South Pole)
Date: 29 Oct 2001 22:44:20 -0800
Subject: Matlab-style mask in Python??
Message-ID: 

Hi,

I am a long time Matlab user.  I just picked up Python for a short
while.
I wonder what should be the best way to do sth like:

Matlab:
a=[3,4,6,7,2,54,2,1,2]
idx=a>4      ans: [0,0,1,1,0,1,0,0,0]
a(idx)         ans: [6,7,54]

Python:
a=[3,4,6,7,2,54,2,1,2]
idx=map(lambda x: x>4,a) #assume this calculation is lengthy
                         #and we don't want to repeat
last cmd???

While we can always put that into a for loop, then append the ans
element-by-element....  I feel there should be an easier way in
Python.
Any suggestions?

From ransen_spam_me_not at nemo.it  Thu Oct  4 10:21:33 2001
From: ransen_spam_me_not at nemo.it (Owen F. Ransen)
Date: Thu, 04 Oct 2001 14:21:33 GMT
Subject: test - don't read...
References:  
Message-ID: <3bbc2a8b.10142621@news.newsguy.com>

On Thu, 4 Oct 2001 16:50:16 +1000, Richard Jones
 wrote:

>On Thursday 04 October 2001 16:36, Ignacio Vazquez-Abrams wrote:
>> On Thu, 4 Oct 2001, Owen F. Ransen wrote:
>> > On Wed, 03 Oct 2001 22:36:06 -0000, Robert Amesz
>> >
>> >  wrote:
>> > >Owen F. Ransen wrote:
>> > >> don't read
>> > >
>> > >don't write
>> >
>> > don't respond
>>
>> don't care
>
>But you told me you loved me!

That was then, this is now. You see, I
met this snake...

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

From gleki at gol.ge  Tue Oct 23 16:34:17 2001
From: gleki at gol.ge (Giorgi Lekishvili)
Date: Tue, 23 Oct 2001 13:34:17 -0700
Subject: unnecessary change
References: <3BD5BCDF.B40C1422@gol.ge> <6UbB7.7681$Ub2.261294@atlpnn01.usenetserver.com>
Message-ID: <3BD5D449.F7680D7D@gol.ge>

Thank you for the hint!
The bad news is, that the matrix turned to be un-deep-copyable object:

"""
>>> m1=deepcopy(m)
Traceback (innermost last):
  File "", line 1, in ?
    m1=deepcopy(m)
  File "d:\python20\lib\copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "d:\python20\lib\copy.py", line 237, in _deepcopy_inst
    state = deepcopy(state, memo)
  File "d:\python20\lib\copy.py", line 151, in deepcopy
    y = copierfunction(x, memo)
  File "d:\python20\lib\copy.py", line 200, in _deepcopy_dict
    y[deepcopy(key, memo)] = deepcopy(x[key], memo)
  File "d:\python20\lib\copy.py", line 147, in deepcopy
    raise error, \
Error: un-deep-copyable object of type 
"""

however, the copy function did help:

"""

Grtz, Giorgi

Steve Holden wrote:

> "Giorgi Lekishvili"  wrote in message
> news:3BD5BCDF.B40C1422 at gol.ge...
> > Hi folks!
> >
> > My problem is the following:
> > I want to save initial data matrix, as e.g., follows
> >     self.Mdata=Matr
> > In the same time I need a matrix, which is used and processed. The
> > latter is created as follows
> >     self.Mactual=self.Mdata
> >
> > No the problem is, that, when I have finished to use the Mactual matrix,
> > self.Mdata is always changed and equals the final Mactual matrix.
> >
> > What is the systematic solution of the problem?
> >
> > Thanx in advance,
> > Giorgi
> >
> > PS. I use Numeric module.
> >
> This is because Python variable bindings are actually references. You
> probably need to make a copy of the matric. See if the copy module (and
> particularly copy.deepcopy()) will help.
>
> regards
>  Steve
> --
> http://www.holdenweb.com/

From mlh at idi.ntnu.no  Wed Oct 10 14:44:15 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 10 Oct 2001 20:44:15 +0200
Subject: Why not 3.__class__ ?
References: 
Message-ID: <9q24tv$nvh$1@tyfon.itea.ntnu.no>

 wrote in message
news:mailman.1002735730.12645.python-list at python.org...
>
> Yukihiro Matsumoto wrote:
> 
> >I'm not sure Python needs something like 3.method.  Python's
> >integers have far fewer methods than Ruby (because it uses
> >functional style).  And it doesn't have idioms like
> >
> >  100.times {|i|
> >    puts i
> >  }
> >
> >So (3).method is enough for Python.
>
> Agreed.  Invoking methods on literals (like integers) is quite nice when
> you have methods that take unnamed, in-place, code blocks as arguments as
> is possible in Ruby (example above) and Smalltalk.
>
> Obviates the need for unfortunate idioms like "for i in range(100):" -- or
> should we make that "xrange"? ;-).

Well... I'm not sure it's unfortunate. For a beginner, it might be easier to
translate the above into something understandable ("for each value of the
index variable i in the range up to 100, execute the following block") than
it would be for the Ruby version ("call the times method of the number 100
with an unnamed code block, which will then be executed with the numbers
up to 100 supplied for the parameter i").

The Ruby version may be nice, but seems unnecessary to me, as long as you
actually have a looping construct in the language. (And if you haven't, then
methods like .times must have them internally anyway...)

I do, however, think it would be nice to have some syntactic sugar for the
range function, e.g.:

  for i in 1..10: foo()

But I guess that'll never happen.  :)

(And, no, I don't like names like "xrange" either.)

> Jim

--

  Magnus Lie Hetland         http://www.hetland.org

 "Reality is that which, when you stop believing in
  it, doesn't go away."           -- Philip K. Dick

From jwilhelm at outsourcefinancial.com  Wed Oct 31 15:49:18 2001
From: jwilhelm at outsourcefinancial.com (Joseph Wilhelm)
Date: Wed, 31 Oct 2001 13:49:18 -0700
Subject: Is round() broken?
Message-ID: <001201c1624d$82ea6dd0$a905a8c0@JWILHELM>

Hello again everybody!

I'm having some troubles with round() now... either I'm doing it completely
wrong or round() is.  But here's what I'm getting.

>>> round( 43583.010000000002, 2 )
43583.010000000002
>>> round( 43583.010000000002 )
43583.0
>>> round( 43583.010000000002, 1 )
43583.0
>>> round( 43583.010000000002, 2 )
43583.010000000002
>>>

So, what this is saying.. is that rounding to 1 decimal point works... but
anything beyond that is broken?  Rounding negative will round the number
before the decimal point also.

And actually, as a side question.. that number was pulled from a float8
field in a Postgres database, using the 'pg' module. If I look at the field
through pgAdmin, I just see it as "43583.01", instead of that whole big long
decimal. Is it supposed to come out like this?

--Joseph Wilhelm

From office at tangfeng.org  Tue Oct 16 04:44:50 2001
From: office at tangfeng.org (office at tangfeng.org)
Date: Tue, 16 Oct 2001 16:44:50 +0800
Subject: Beijing Tangfeng Culture Exchange Centre
Message-ID: 

An HTML attachment was scrubbed...
URL: 

From mrq at for.mat.bham.ac.uk  Fri Oct 26 09:01:58 2001
From: mrq at for.mat.bham.ac.uk (Martyn Quick)
Date: Fri, 26 Oct 2001 14:01:58 +0100
Subject: Enabling/disabling Pmw menubar items
In-Reply-To: 
References: 
Message-ID: 

Dear All,

I've been creating a menubar for an application using the Pmw MenuBar
class.  For example, I have something like the following:

menubar = self.createcomponent(
    'menubar', (), None, Pmw.MenuBar, (self.interior(),)
menubar.pack()
menubar.addmenu('Menu', 'A dummy menu')
menubar.addmenuitem(
    'Menu', 'command', 'A non-existent command',
    command=dummy_fucntion, label='Something', state='disabled')

This has the effect that my menu item is disabled.  But how do I enable it
again?   (I can't seem to find this out on the Pmw website!)

Thanks,

Martyn

--------------------------------------------------------
Dr. Martyn Quick  (Lecturer in Pure Mathematics)
University of Birmingham, Edgbaston, Birmingham, UK.
http://www.mat.bham.ac.uk/M.R.Quick/

From tim.one at home.com  Wed Oct 31 00:29:47 2001
From: tim.one at home.com (Tim Peters)
Date: Wed, 31 Oct 2001 00:29:47 -0500
Subject: python development practices?
In-Reply-To: <7x3d40sfm4.fsf@ruckus.brouhaha.com>
Message-ID: 

[Paul Rubin]
> ...
> IMO a bunch of the frustration I sometimes feel with Python comes
> from its originally being intended as a "glue" language.  It's too
> good for that, and finds itself used as a work horse or even a race
> horse.  Neither type of horse belongs in the glue factory ;-).

Oooh -- good one!  If Andrew doesn't add it to his Python Quotes collection,
I get first dibs on passing it off as my own .

From l0819m0v0smfm001 at sneakemail.com  Tue Oct  9 18:14:50 2001
From: l0819m0v0smfm001 at sneakemail.com (Joshua Macy)
Date: Tue, 09 Oct 2001 22:14:50 GMT
Subject: Python is better than free (was Re: GNU wars again)
References:  <3BBA56C4.9010904@sneakemail.com> <3BBABE1C.5010108@sneakemail.com> <9pimak$c22$1@tyfon.itea.ntnu.no> <825v7.1025$rq1.37994@atlpnn01.usenetserver.com> <3BBE4E12.16FE036A@earthlink.net> <9pn44q$m1r$1@tyfon.itea.ntnu.no> <3BBF2751.3010309@sneakemail.com> <23891c90.0110090118.7df7917e@posting.google.com>
Message-ID: <3BC34197.2000403@sneakemail.com>

Paul Boddie wrote:

> 
> Who are these Strunk and White people anyway? (Serious question - I
> assume it's an American English style manual people keep referring to,
> or something, but I can't be bothered to search for it.) They must be
> rich by now, though. ;-)

  Strunk & White are William Strunk, Jr. and E.B. White, authors of The 
Elements of Style, which is pretty much *the* American English style 
manual...all the books you've ever seen called The Elements of ____ 
Style (e.g. The Elements of Programming Style by Kernighan and Plauger) 
are patterned after it.  E.B. White was a famous essayist (writing for 
the New Yorker for many years) and author of Charlotte's Web; William 
Strunk, Jr. was a college English professor of his. The Elements of 
Style started as a textbook written and privately printed by Strunk, and 
was revised by White.  It's a slim volume, weighing in at 85 pages, and 
is almost Pythonic in its pithiness.

   Joshua

From thomas at gatsoft.no  Tue Oct 30 07:17:37 2001
From: thomas at gatsoft.no (Thomas Weholt)
Date: Tue, 30 Oct 2001 12:17:37 GMT
Subject: How to read a xml-file sequentially
Message-ID: 

Hi,

I need to read a xml-file sequentially, piece by piece, when a given end-tag
to a given start-tag is found, the xml-data inbetween the start and end are
sent to a different method for processing, then the next piece of data is
read, sort of like old-school record-based files.

Only the needed data is kept in memory.

How can this be done using preferrably expat or sax? No DOM cuz the files
are huge.

Thomas

From mmealman at tarsis.org_NOSPAM  Thu Oct  4 13:35:14 2001
From: mmealman at tarsis.org_NOSPAM (Mark)
Date: 4 Oct 2001 12:35:14 -0500
Subject: Zope vs PHP for the Linux guru
Message-ID: 

I've been doing web app development for the last 3 years using php. I
already plan on learning Python to use as a rapid GUI development
language(maybe even replace some of my perl-use) and was curious about Zope.

I currently develop applications using php to query both remote and local
databases. I use local MySQL databases for fast read-only queries, and use a
remote MS SQL 2000 database for storing mutable data.

I use mod_php compiled into Apache 1.3.xx, and run php apps that range from
a few hundred lines to over 50k. DB access is through php's API using
MySQL's native libs and FreeTDS for connecting to SQL.

On the larger php apps I use an OO design model to keep the project
managable. My smaller apps are pretty much just quickly thrown together
scripts.

Right now I run it all on one Linux box, but have plans on creating farms
of apache servers for future scalability.

Some questions I had about Zope where:

Is developing apps in it a quick process? Php is a very rapid language to
develop in and that has become my most favored trait of the language.

Does Zope have good database support? Is it faster, slower than php with
database queries?

I'm assuming Zope will be better than php for large project development?

Is Zope easy for newbies to learn? I'm not a newbie, but we do have
some clients learning to program on our seperate Cold Fusion server, they
find it easier than php. I'd love to get them off Cold Fusion and onto a
more cost effective open source solution.

Does Zope scale well in large deployments? Any kind of support for web
server farming?

Any other major differences from php? Good things about it, bad things about
it?

Thanks for any input.

-Mark

-----=  Posted via Newsfeeds.Com, Uncensored Usenet News  =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
 Check out our new Unlimited Server. No Download or Time Limits!
-----==  Over 80,000 Newsgroups - 19 Different Servers!  ==-----

From jh at web.de  Tue Oct 30 14:57:20 2001
From: jh at web.de (J�rgen Hermann)
Date: Tue, 30 Oct 2001 20:57:20 +0100
Subject: [ANN] PIRXX 1.2
Message-ID: <9rn0j7$k1b$02$2@news.t-online.com>

PIRXX provides a Python InteRface to Xerces and Xalan, integrating it into
PyXML.

This release features a stable and complete SAX2 binding.

Links:
 * Homepage - http://pirxx.sourceforge.net/
 * Download -
http://sourceforge.net/project/showfiles.php?group_id=25711&release_id=58705

Changes:
 * Added locator support to ContentHandler interface
 * Implemented DTDHandler and EntityResolver
 * Implemented the LexicalHandler interface and SAX2 property
 * Some changes to SAX2 event calls so that PIRXX behaves more like xmlproc,
   especially delivering None instead of empty strings for empty public IDs
   and the like
 * PIRXX passes adapted "test_sax.py"
 * Removed reference leaks (by using PirxxObject nearly everywhere)

From db3l at fitlinxx.com  Fri Oct  5 13:05:05 2001
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Oct 2001 13:05:05 -0400
Subject: another wxPython-question
References: <9pi16t$5pt1t$1@hades.rz.uni-sb.de> <3BBCB53D.407B103B@home.net> <9pk06u$58ljt$1@hades.rz.uni-sb.de>
Message-ID: 

Uwe Schmitt  writes:

> thanks a lot. the import statement above works. i made an error:
> i looked for a file "*wxDC*" which is not on my computer.
> i'll study your example, 

BTW, if you haven't already, be sure and run the demo and then examine
its source as full of example code.  The demo exercises the majority
of the wxPython features, so it's generally pretty easy to find a
portion of the demo at least close to what you're working with and
then look at the code to see how it accomplishes it.

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

From john at yates-sheets.org  Wed Oct 31 16:00:50 2001
From: john at yates-sheets.org (John S. Yates, Jr.)
Date: Wed, 31 Oct 2001 21:00:50 GMT
Subject: MS network provider using Python
Message-ID: 

I am still in my Python learning stage.  (Not yet ready
to roll my own implementation :-)

I am prototyping a network provider (a DLL exporting
an implementation of the MS defined NP API and invoked
by MS's MPR (Multiple Provider Router)).  Since I
anticipate a great deal of experimentation I would
dearly love to do some prototyping in Python.

Can anyone in this news group offer words of wisdom?

What is the best way in this kind of a rapid prototyping
environment to expose my python code as a DLL?

The relevant interfaces are not part of win32 so they
are not supported by Mark Hammond's win32all package.
This suggests that I need some kind of foreign function
interface/package/technology.  Should I try to decipher
CALLDLL?  Is there a better alternative?  (In a past
life I have been alternatively an assembly language coder
and a compiler writer.  So mapping function prototypes
to calling mechanisms and data representations is not
a stumbling block.)

As I mentioned earlier, I would love to pull this off
in Python.  But I am willing to be told by those in the
know that I am really better of sticking to C++...

/john

--
John Yates
40 Pine Street
Needham, MA 02492
781 444-2899

From jmcwill at tridog.com  Wed Oct 10 18:52:03 2001
From: jmcwill at tridog.com (Jason McWilliams)
Date: Wed, 10 Oct 2001 16:52:03 -0600
Subject: upload file using cgi on Win2k IIS
References: 
Message-ID: <3BC4D113.F99FE096@tridog.com>

Ignacio Vazquez-Abrams wrote:
> 
> On Wed, 10 Oct 2001, Jason McWilliams wrote:
> 
> > Ignacio Vazquez-Abrams wrote:
> > >
> > > On Wed, 10 Oct 2001, Jason McWilliams wrote:
> > >
> > > > I seem to be doing everything correctly to save a posted
> > > > multipart/form-data file on IIS server.  Yet, I only
> > > > seem to be able to save a small portion of the file, not
> > > > the whole thing.
> > > >
> > > > The internet user has permission to the temp directory and
> > > > I am not getting any errors.  It works properly with the
> > > > same code on unix, so it must be a IIS thing?
> > > >
> > > > I have tried file.read file.value and also tried to just
> > > > do one line at a time... file.readline .  I can get the
> > > > file name fine using cgiobj.filename  ... etc.
> > > >
> > > > I get about 498 k as the file size on the uploaded file, never
> > > > more and I am trying to upload a big file.  Maybe there is
> > > > a post limit in IIS?  I know that you can set a GET limit.
> > >
> > > What is the value of the hidden field named "MAX_FILE_SIZE" set to? Is there
> > > one at all? If not, then don't expect your file input to work properly.
> >
> > I did not think this variable was used in cgi.py
> > Isnt that a php type thing.  Isn't maxlen set to 0 inside cgi.py
> > to allow for unlimited post length?  That is why I thought
> > it might be a IIS setting somewhere.
> 
> Hmm. I had always thought it was a browser thing, but after doing some
> research I think that you're right. Unfortunately I don't have enough IIS
> knowledge to know exactly which setting it is. It's under the web properties,
> probably, but I have no idea past that.
> 
> FYI, the default value of cgi.maxlen is 0, so that's probably not it.

Here is something interesting.  I can upload .txt and .doc files no
problem,
no matter what the size.  Things die when trying to upload .gif or .zip
or .tgz
or other binaries similar.  Some of it gets uploaded, but only a small
portion?

I am at a loss now.

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

From Virus.Patrol at drsolomon.com  Tue Oct  9 14:36:06 2001
From: Virus.Patrol at drsolomon.com (Dr Solomon's Virus Patrol)
Date: Tue, 09 Oct 2001 18:36:06 GMT
Subject: TROJAN ALERT! (JS/Loop)
References: 
Message-ID: <3bc343a0.fba6e0c5.c0a801.00.R@drsolomon.com>

WARNING! A trojan has been found in an article posted to
the following newsgroup(s):
 comp.lang.javascript, alt.config, comp.lang.perl.misc, comp.lang.python,
 comp.os.linux.misc

Message header follows:

>Message-Id: 
>From: "dvus" 
>Subject: YOU ARE ALL GAY!
>Date: 09 Oct 2001 15:11:40 GMT

Dr Solomon's FindVirus/VirusScan report follows:

 Dr Solomon's FindVirus IN-HOUSE version. Copyright (c) 1999 Network Associates Inc.
 Virus data file v9999 created Oct 08 2001
 Scanning for 58978 viruses, trojans and variants.

 [HTML part] ... Found the JS/Loop trojan !!!

You can download a free evaluation (yet fully functional) copy of
the latest released version of Dr Solomon's FindVirus/VirusScan from
the following locations:

 WWW: http://www.nai.com/naicommon/buy-try/try/products-evals.asp
 CompuServe: GO DRSOLOMON or GO MCAFEE
 AOL: SAFETYONLINE

IMPORTANT! The trojan has been detected by the latest IN-HOUSE version of
Dr Solomon's FindVirus/VirusScan. As new viruses and trojans appear at the rate
of several hundreds a month, the latest RELEASED version of FindVirus/VirusScan
might be unable to detect this particular trojan. If you have downloaded the above
file and your copy of FindVirus/VirusScan does not detect the trojan in it,
contact Network Associates Technical support and ask for the extra driver
to detect the JS/Loop trojan.

--
Dr Solomon's Virus Patrol                UK Support: support at drsolomon.com
NAI Anti-Virus Emergency Response Team         US Support: support at nai.com
WWW: http://www.mcafeeb2b.com                  UK Tel: +44 (0) 1296 318700
CompuServe: GO DRSOLOMON                          USA Tel: +1 408 988-3832

From dash at home.com  Sun Oct 14 18:30:08 2001
From: dash at home.com (David Ashe)
Date: Sun, 14 Oct 2001 22:30:08 GMT
Subject: Tkinter & all widgets
Message-ID: 

Is there a way to get the complete list of widgets in a window as a list,
array, dictionary, or tuple?
I seem to recall this was possible under tcl, but I can't remember how.

D

From brueckd at tbye.com  Thu Oct 11 00:32:07 2001
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Wed, 10 Oct 2001 21:32:07 -0700 (PDT)
Subject: Python questions -> compiler and datatypes etc
In-Reply-To: <7xr8sbc993.fsf@ruckus.brouhaha.com>
Message-ID: 

On 10 Oct 2001, Paul Rubin wrote:

> Richard Jones  writes:
> > Do a search in the python mailing list for "python compiler". The
> > answer is invariably "it's bloody hard". This stems from the
> > fundamental dynamic typing of python. It's very very very hard to
> > write a compiler when you don't know the type of the objects you're
> > dealing with beforehand. Java knows these types, because it's
> > statically typed. Therefore it's relatively trivial (compared to
> > python) to write a compiler for Java.
>
> I think the real answer to why there's no Python compiler is no one
> has really wanted to write one.  Python's runtime semantics are pretty
> similar to Lisp systems, and native-code Lisp compilers have existed
> since the 1960's or maybe even earlier.  It might even be feasible
> to adapt one to compile Python.

That'd be great. How about just creating a Python-to-Lisp translator? Then
you could use whatever Lisp compilation tools are available (assuming, of
course, that the performance of compiled Lisp is good enough to make the
effort worthwhile).

It seems like the really hard part about translating Python to C, for
example, is that you're moving "down" the dynamic language scale. But if
you already have a compiler for a language that is near or higher on that
scale, then somebody has already done a lot of the work for you.

-Dave

From James_Althoff at i2.com  Wed Oct 17 13:35:43 2001
From: James_Althoff at i2.com (James_Althoff at i2.com)
Date: Wed, 17 Oct 2001 10:35:43 -0700
Subject: Conditional Expressions don't solve the problem
Message-ID: 

Fredrik Lundh wrote:
>and the "while 1" pydiom usually doesn't look that way.

>write pydiomatic Python code instead doing things the C/Java

"Pydiom" & "pydiomatic"?  Pretty cool.  Is this new?

Jim

From phr-n2001d at nightsong.com  Sun Oct  7 12:57:20 2001
From: phr-n2001d at nightsong.com (Paul Rubin)
Date: 07 Oct 2001 09:57:20 -0700
Subject: SPAM (Re: ANNOUNCE: "Compiler")
References: 
Message-ID: <7xitdr4ccv.fsf@ruckus.brouhaha.com>

This is a commercial closed-source program that has nothing to do with
Python.  

From sholden at holdenweb.com  Wed Oct 31 08:42:01 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 31 Oct 2001 08:42:01 -0500
Subject: python-postgres-informix interaction problems
References: 
Message-ID: 

"J Dubal"  wrote ...
> Hello good people!
>
Hello yourself. All the *good* people were busy, so they asked *me* to reply
:-)

> We have developed Informix RDBMS based commercial applications in
> Informix-4gl. It is a good procedural language with embeded SQL. We
> are trying to rewrite some apps in Python with postgresql backend for
> licensing reasons. Python is good too. We are trying to do this with
> latest versions of RH linux (7.2), python 2.1 and postgres 7.1.3. We
> face following issues:
>
Excellent. Another company decides to embrace open source products. If only
Microsoft would learn this lesson ...

> 1. We have to read data from informix db, process it and insert into
> postgres. We are using informixdb module (DBAPI 1) which is bit dated.
> In Informix we 'select custno, custname, custaddr1, ... into progvar1,
> progvar2, ... from custmf where ...' In python, we say fetchone() and
> we get a list. We are then moving the elements of the list to named
> program variables thus:
>         pcustno = list1[0]
>         pcustname = list1[1]
>         pcustaddr1 = list1[2]
>         ... as many as reqd
> This way we give meaningful names to data items read from db and
> process further.
> Is there a better, more elegant/efficient way to to this?
>
Yes. Not necessarily more efficient (probably less so), but much more usable
from a programmatic point of view is Greg Stein's dtuple module, whicj
allows you to address the tuple using numerical or field-name subscripts as
well as attributes (t.fieldname). See

    http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/81252

gives you a short example of the advantages of this excellent module.

> 2. To insert this data into postgres, we used to use pygres included
> in the postgres distribution (latest). We had to stop using this
> because we couldn't find a way to find out how many rows it have
> inserted, updated or deleted after such statements are executed ie we
> couldn't find a way to check the status or return code. So we switched
> over to PoPy which is DBAPI2 and where we could check number of rows
> affected. Is this right thing to do?
>
This sounds like an excellent plan.

> 3. In order to insert data into postgres, we have to compose an insert
> statement thus:
>         cur.execute("insert into custmf (custno, custname,
> custaddr1,...)
>             values (%i,'%s','%s'...)" %(pcustno,pcustname,
> pcustaddr1....))
> Is there a better way of doing this?
>
Yes. Use parameterized statements. There are five different schemes, the
most common of which uses question marks in the statement to associate data
values from a tuple provided to the cursor.execute() method as a second
argument.

So you would write:

    cur.execute("""INSERT INTO custmf (custno, custname, pcustaddr1, ...)
                            VALUES (?, ?, ?, ...)""",
                        (pcustno, pcustname, pcustaddr1, ...))

The DB API documentation gives you more detail on this. Note that not only
is this saving you from constructing individual statements, it also gives
the DB API chance to optimize successive uses of the same statement with
different values.

> 4. The above string substitute converts program variables holding null
> value into 'None' which goes into target table as string 'None'. To
> avoid this we parse all columns to check if they are 'None' and
> replace them by ''. Is there a better way to do this?
>
Parameterization should solve this problem too!

> 5. The above check works only on string variables. We have not found
> yet a way to handle nulls in numeric columns. Please suggest a
> solution.
>
And this one!

> 6. If the character columns happen to contain quotes, the above
> composed insert statement breaks. Please suggest a solution.
>
And this one! THe DB module will perform all necessary quoting!

Hope this helps. Good luck with your project.

regards
 Steve
--
http://www.holdenweb.com/

From sandysj at juno.com  Thu Oct  4 15:38:07 2001
From: sandysj at juno.com (Jeff Sandys)
Date: Thu, 4 Oct 2001 19:38:07 GMT
Subject: A little amusing Python program
References: <3bbaf6c9$0$244$edfadb0f@dspool01.news.tele.dk>
Message-ID: <3BBCBA9F.D1777976@juno.com>

What a clever little program.  I have felt for a long 
time that a program that passes the Turing test will 
use the web as a corpus of data.

Microsoft showed a similar program at the IJCAI that 
would answer questions.  It answered some questions 
like "What is A.I." and "Who made the movie A.I.", but 
when asked "Who is Bill Gates married to" it's reply 
was "I'm not sure, either Malinda French or Microsoft" 

Another program shown at an AI conference was a 
document classifier.  To determine which folder to add 
the document to, it simply compare the size of the 
tarred folders before and after adding the document. 
Of all the AI programs used for document sorting it 
came closest to a human sorter.

Thanks,
Jeff Sandys

From gh_pythonlist at gmx.de  Thu Oct 25 16:37:39 2001
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Thu, 25 Oct 2001 22:37:39 +0200
Subject: urllib.urlopen + https = all threads locked?
In-Reply-To: ; from guido@python.org on Thu, Oct 25, 2001 at 04:00:01AM +0000
References:   
Message-ID: <20011025223736.A12493@lilith.hqd-internal>

On Thu, Oct 25, 2001 at 04:00:01AM +0000, Guido van Rossum wrote:
> Gerhard H?ring  writes:
> 
> > In the meantime, this is a quick (untested) patch against current CVS
> > Python:
> [patch snipped]
> > > Is this a known problem?
> > 
> > Unless it's reported as a bug with the Sourceforge bugtracker, it's not
> > a known problem <0.7 wink>.
> 
> Gerhard, would you mind uploading that to SF?

Done (#475045). I hope it's not too late for 2.2.

Jeff, thanks for the test case. It was very useful.

Gerhard
-- 
mail:   gerhard  bigfoot  de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))

From info at mjais.de  Wed Oct 17 17:26:41 2001
From: info at mjais.de (Markus Jais)
Date: Wed, 17 Oct 2001 23:26:41 +0200
Subject: problem with indentation and emacs
Message-ID: <9qkt53$olvlr$1@ID-75083.news.dfncis.de>

hello
I have a problem which drives me crazy!

when I cut and paste some code within a *.py file
then there seems to be problems with identation.

everything looks fine, but the programm crashes with
syntax error or with "IndentationError: expected an indented block"

when I look at the code, everything looks o.k.
when I open the file with vim and reindent everyting,
the programm works.

when I reindent it with Xemacs, I got the same errors

maybe this is a problems with tabs and spaces, but I couldn't find
out anything.

please help. Xemacs is cool and python is cool. but I want to use
them together.

markus

From nomad*** at ***freemail.absa.co.za  Tue Oct  9 08:39:57 2001
From: nomad*** at ***freemail.absa.co.za (Nomad)
Date: Tue, 09 Oct 2001 14:39:57 +0200
Subject: Octonians in Python
References:  <7x3d4wpqi0.fsf@ruckus.brouhaha.com> <_4Lv7.36535$My2.18927726@news1.mntp1.il.home.com>
Message-ID: 

On Sat, 06 Oct 2001 21:55:06 GMT, "David Feustel"
 wrote:

>Lots of interesting stuff on 'onions here.

onions?

No.  No it's too easy, but I can't help it 

-- 
Nomad

Wondering of the vast emptyness of the 'net
in search of something cool.

From grante at visi.com  Wed Oct 24 20:59:49 2001
From: grante at visi.com (Grant Edwards)
Date: Thu, 25 Oct 2001 00:59:49 GMT
Subject: Need help with programming.
References: <20011024194636.07978.00000094@mb-mb.aol.com> 
Message-ID: 

On Thu, 25 Oct 2001 00:50:22 GMT, William Tanksley  wrote:

>>May I remind you that I AM NEW TO PYTHON, and COMPUTER LANGUAGES ALL
>>TOGETHER (Excluding HTML). I'M LEARNING PYTHON FOR SELF INTEREST, SO
>>PLEASE NO FLAMES..
>
>Are you under the impression that screaming at people you've never met is
>socially commendable?

I assume that he is not only new to programming, but is is also
new to Usenet and doesn't know that all caps text is considered
to be shouting.

Except in Zippy the Pinhead quotes where it is an important
part of the literary experience...

-- 
Grant Edwards                   grante             Yow!  .. does your DRESSING
                                  at               ROOM have enough ASPARAGUS?
                               visi.com            

From jim_kerr at agilent.com  Tue Oct  2 20:19:26 2001
From: jim_kerr at agilent.com (Jim Kerr)
Date: 2 Oct 2001 17:19:26 -0700
Subject: problem processing COMException in VB6
Message-ID: <50626d00.0110021619.2000be73@posting.google.com>

I've written a COM server in Python that is accessed by a VB6 client.
The client/server interaction works just fine, but I'm having a
problem displaying error messages in VB when the server throws an
exception. The problem is that VB seems to be displaying a generic
error message based upon the scode for the exception, and ignoring the
error description entirely. For example, if the server executes

 raise COMException(desc="No measurement defined on channel %d" %
channelNum, scode=winerror.DISP_E_BADINDEX)

and the VB client catches the exception with

    MsgBox Err.Number & vbCrLf & Err.Description, vbCritical, "Server
Error"

then the message box displays the text "Subscript out of range"
instead of "No measurement defined on channel 3".

  I must be missing something obvious here...?

  BTW, I'm running ActivePython 2.1.211 on Windows 2000.

-Jim
==========
Jim Kerr
Software Engineer
Agilent Technologies, Inc.

From mlh at idi.ntnu.no  Sat Oct  6 10:02:50 2001
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 6 Oct 2001 16:02:50 +0200
Subject: Generating a random number
References: <9plkkk$98r$1@ins22.netins.net> <4cnsrtsj7ks3jc6emj39ksod2ipmr9f0bb@4ax.com> <3BBE4728.40101@sneakemail.com>
Message-ID: <9pn2ua$lg7$1@tyfon.itea.ntnu.no>

"Joshua Macy"  wrote in message
news:3BBE4728.40101 at sneakemail.com...
>
>
> Dale Strickland-Clark wrote
>
> >
> > from whrandom import randint
> >
> > print randint(1, 10)
>
>
>
>
>    whrandom is deprecated.  Use random instead.

Actually, randint is deprecated as well... Use randrange
instead :)

Note that when using randrange, the endpoints are interpreted
as with the range function, so the statement becomes

  print randrange(1,11)

or
  print randrange(10)+1

>
>   Joshua

--

  Magnus Lie Hetland         http://www.hetland.org

 "Reality is that which, when you stop believing in
  it, doesn't go away."           -- Philip K. Dick

From brueckd at tbye.com  Tue Oct  9 17:56:36 2001
From: brueckd at tbye.com (brueckd at tbye.com)
Date: Tue, 9 Oct 2001 14:56:36 -0700 (PDT)
Subject: Why not 3.__class__ ?
In-Reply-To: <9pvilt$kepge$1@ID-75083.news.dfncis.de>
Message-ID: 

> > [snip]
> >> but to write 3.foo is something I would like to have in Python
> > [snip]
> >> It would be great, if this would be possible in Python too
> >
> > Care to elaborate on why it would be "great"? Guido's reply to Marcin
> > points out that there doesn't seem to be any usefulness to it. So.... what
> > problem would it solve or simplifiy?  What exactly would its benefit be,
> > anyway?
>
> it would be more consise.
> in Ruby, everything is an object and I can invoke methods on everything
> with object.method

Uh... that is not any different than Python, but even if it was, the
conciseness of something useless is irrelevant. :)

From alan.gauld at bt.com  Thu Oct 25 10:54:03 2001
From: alan.gauld at bt.com (Alan Gauld)
Date: Thu, 25 Oct 2001 15:54:03 +0100
Subject: My Web Tutor  has moved
Message-ID: <3BD8278B.9000405@bt.com>

My online beginner's tutorial has moved.
If you currently have a web site linked to:

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

Could you please move the link to:

http://www.freenetpages.co.uk/hp/alan.gauld

The new site has no adverts (hooray!) and
more importantly seems much more stable.

As soon as the crosswinds site becomes stable
again I will put a notice and redirector onto
the welcome page.

Thanks,

Alan Gauld.

From sholden at holdenweb.com  Wed Oct 24 08:20:17 2001
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 24 Oct 2001 08:20:17 -0400
Subject: Executing a string as python code
References:   
Message-ID: 

"Stephen Boulet"  wrote in message
news:ttdbclaabhte92 at corp.supernews.com...
> Yes, but this is no cgi script. It's in my program, which, after all, also
> executes python code.
>
> If I write:
>
> <<
> command = ''
> for i in range(100):
>     command = command + 'list%d = []\n' % i
>
> exec(command)
>
> >>
>
> That's no less secure than typing them out by hand, right?
>
This may be so. But your next problem is then presumably to write a loop to
process each of these different variables. The major point (assuming no user
input is associated with your executed code) is the pollution of your name
space in a way which doesn't help you in your overall programming goal.

You haven't yet demonstrated why you'd rather run your code than something
like

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

If you really need 100 different empty lists, why not create a list, or
dictionary, of lists to hold them? Saving each one as a variable makes them
difficult to deal with, which is quite unnecessary. Your way you have to
create a statement referencing listN and then execute it.

    stmt = "list%d.append(something)" % n
    exec(stmt)

With a list of lists you can just execute

    lst[n].append(something)

Simpler, cleanr, faster. Where are the *dis*advantages?
regards
 Steve

--
http://www.holdenweb.com/

From kragen at dnaco.net  Tue Oct 16 16:34:41 2001
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 16 Oct 2001 20:34:41 GMT
Subject: Directed input
References: <9qi4r2$gpq$1@nserve1.acs.ucalgary.ca>
Message-ID: 

In article <9qi4r2$gpq$1 at nserve1.acs.ucalgary.ca>,
Michael Yuen  wrote:
>I'm just in the process of learning Python right now and i'm wondering if 
>there's a way run a program to take a file via directed input and 
>sending output to another file via the command line.
>
>Here's what I mean:
>
>myprogram.py < my_input > my_output

I assume you're on Unix?  The above should work fine.
-- 
       Kragen Sitaker     
Perilous to all of us are the devices of an art deeper than we possess
ourselves.
       -- Gandalf the White [J.R.R. Tolkien, "The Two Towers", Bk 3, Ch. XI]

From mel.b at mindspring.com  Sun Oct  7 03:45:07 2001
From: mel.b at mindspring.com (Mel Brown)
Date: Sun, 7 Oct 2001 00:45:07 -0700
Subject: Newbie Question
Message-ID: <9pp1ca$ca6$1@slb4.atl.mindspring.net>

I am trying to get Python up and running. Among the modules that I know that
I will need are math, cmath, and string.

When I import sys, then print out the included module names, I get that math
and cmath are included but sting is not. When I import math and then try to
compute something simple like sin(3), I am told that the function sin() is
not available.

My question is how to make the modules math, cmath, and string available for
my use.

Any help, please?

From guido at python.org  Mon Oct 29 02:58:42 2001
From: guido at python.org (Guido van Rossum)
Date: Mon, 29 Oct 2001 02:58:42 -0500
Subject: Dictionary from list?
In-Reply-To: Your message of "Mon, 29 Oct 2001 02:01:28 EST."

References:  
Message-ID: <200110290758.CAA11453@cj20424-a.reston1.va.home.com>

Subclassing list and dictionary etc. should be seen as an experimental
feature; I don't want to fully fix the semantics yet in all cases.

I think of these types as having two sets of interfaces: an internal,
fixed one (e.g. PyDict_SetItem) and an external, overridable one
(e.g. __setitem__).  The external operations are defined in terms of
the internal ones.  When you invoke an operation with a Python
notation (e.g. dict[key]), you invoke the external operation, and
hence overrides will have the expected effect.  But many accesses from
inside the Python VM (e.g. dictionary use by exec) and from other
methods (e.g. dict.update) invoke the internal versions of various
operations, and hence won't see the overrides.

IMO, fixing these internal accesses to always use the overrides would
remove much of the attractiveness of subclassing built-in types; I
believe it would slow things down too much given the current state of
the implementation.

If you want a vaguely dictionary-like object, write a class that
doesn't derive from dictionary but implements the mapping protocol; if
you want a base class that implements most of the operations already,
start with UserDict.

Subclassing a built-in type is appropriate when either (a) you want to
use it in a context where a genuine list/dictionary/file/etc.; or (b)
you want the speed advantage of the built-in type.  In both cases you
have to live with some restrictions.  Remapping the fundamental
accessors (like __getitem__) is probably not a good idea in either
case.  Adding new state and behavior is fine.

We should document this more clearly and in more detail.

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

From unknown at lee-morgan.net  Wed Oct 10 23:15:10 2001
From: unknown at lee-morgan.net (Lee Morgan)
Date: Thu, 11 Oct 2001 03:15:10 GMT
Subject: How to set tab-width to 4 spaces in emacs ??
References: <1J1x7.14385$tu6.623674@news1.oke.nextra.no>
Message-ID: 

"Thomas Weholt"  writes:

> Trying to set the tab-width in emacs to 4 spaces without luck. Using
> Mandrake 8.1 ( great distro btw, python 2.1.1, postgresql 7.1.3 etc. try it
> now !)
> 
> tried :
> 
> # tab-width: 4
>  etc. but no go.
> 
> Any hints ???
> 

I think (setq tab-width 4) is correct, although isnt it both a global and buffer 
variable(or is it a mode var?) - so you may need to add it on to your mode
hook. (Although I've never had reason to change python-mode)

> Thomas
> 
> 

-- 
Lee Morgan

From rjroy at takingcontrol.com  Tue Oct 16 09:39:38 2001
From: rjroy at takingcontrol.com (Robert Roy)
Date: Tue, 16 Oct 2001 13:39:38 GMT
Subject: Link between a C variable and a python object
References: 
Message-ID: <3bcc370d.1705250234@news1.on.sympatico.ca>

On Fri, 12 Oct 2001 13:01:23 +0200 (CEST), Mads Bondo Dydensborg
 wrote:

>On 12 Oct 2001, Armin wrote:
>
>> In article <3BC4366A.DC3F6982 at wanadoo.fr>, guignot says...
>> >
>> >Does Python have a nice feature of Tcl, i.e. linking an external C
>> >variable and an internal variable, with TclLinkVar ?
>> 
>> Yes ... have a look to the C API.
>
>Could you possible be more specific? I have looked at the API, and the
>code generated by SWIG, and it seems that you need to use function calls -
>there seems to be no easy way to share a variable?
>
>Mads

See the sources for fileobject.c. It exposes some C vars to python. 
look for "struct PyMemberDef" and follow the declared variable. Note
that python 2.2 has a new way of doing this. It seems simpler than
before now.

I have not tried this. Your question just got me looking thats all... 

bob

>
>-- 
>Mads Bondo Dydensborg.                               madsdyd at challenge.dk
>It is strange to claim that video games should influence children. If for
>example Pac-Man influenced children born in the 1980'ies, then today we
>would have a lot of young people jumping around in dark rooms, eating pills
>while listening to monotome music
>
>          - Anonym
>
>

From bill-bell at bill-bell.hamilton.on.ca  Mon Oct 15 07:44:34 2001
From: bill-bell at bill-bell.hamilton.on.ca (Bill Bell)
Date: Mon, 15 Oct 2001 07:44:34 -0400
Subject: Python web client anyone?
In-Reply-To: <1003107518.407.27951.m4@yahoogroups.com>
Message-ID: <3BCA93E2.27792.2C9ABBD@localhost>

Paul Rubin  wrote, in part:
> ... I was looking for something that actually parses the HTML on
> the retrieved page like LWP does.  I wonder if there's some way to
> do that with the XML libraries (though HTML is generally not
> well-formed XML ... 

Paul,

If your platform is MSW then you might consider using MSHTML. 
It's the HTML parser+ that's embedded in IE, and it can be 
exercised as a COM object. Clearly a product like IE does an 
excellent job of parsing broken HTML docs and MSHTML is I 
believe freely distributable.

The snag in using MSHTML with Python is that Python is as yet 
unable to process vtable-based interfaces (which is really needed 
to use MSHTML)--ref Mark Hammond's remarks of several weeks 
ago. One way around this problem is to model code on the 'walkall' 
example provided on MSDN and wrap it in some way to make what 
you want accessible in Python.

I have not investigated what's available for parsing HTML on other 
platforms. However, the same general strategy (ie, that of 
exercising one of the best available web clients on the platform) 
might work in those cases too.

Best of luck,

Bill
"It is the time that you have wasted for your rose that makes your rose so important."--St-Exupery

From eyal at hyperroll.com  Sun Oct 28 17:23:02 2001
From: eyal at hyperroll.com (Eyal Lotem)
Date: Mon, 29 Oct 2001 00:23:02 +0200
Subject: Why doesn't this simple Curry-like implemention work?
References: 
Message-ID: <3bdc8494@news.bezeqint.net>

Michael R. Head wrote:

> # for some reason this function doesn't work in python 1.5.2.
> def curry(f, p):
> return lambda x: apply(f, p, x)

apply(f, p, x) is in the context of an internal function.  Without 
nested_scopes (optionally introduced in Python 2.1, and in by default in 
Python 2.2), internal functions cannot use containing-function's scopes.

Python 1.5 code:

def curry(f, p):
        return lambda x,f=f,p=p: apply(f, p, x)

or in Python 2.1:

from __future__ import nested_scopes

def curry(f, p):
        return lambda x: apply(f, p, x)

will work.
I would suggest considering moving to Python 2.1, as nested_scopes seem to 
be safer, at least for this case.  With the default-argument trick, you 
allow more arguments than you really want to allow, which could cause 
trouble.

From jkraska at san.rr.com  Tue Oct  2 20:21:30 2001
From: jkraska at san.rr.com (Courageous)
Date: Wed, 03 Oct 2001 00:21:30 GMT
Subject: How do Java interfaces translate to Python?
References:  
Message-ID: 

>> What exactly is it that you do with interfaces in Java?
>Enforcing a contract between...

_Enforcement_ is antithetical to Python. Python is a
_dynamic_ language. Dynamic languages provided an
environment which is _mutable_ at run-time, by definition.

C//

From john.thingstad at chello.no  Sun Oct 21 09:15:56 2001
From: john.thingstad at chello.no (John Thingstad)
Date: Sun, 21 Oct 2001 15:15:56 +0200
Subject: Array of 2^n
Message-ID: <20011021133449.BLZ12720.mta01@mjolner>

> Okay, python newbie here. I'm liking what I see, but it seems like there
> are holes in the features of python's objects. I remember programming in
> squeak (an extension of smalltalk), and you could do complex operations on
> lists, and it would apply the result to each item. I want to make a n-item
> list such that mylist[i] = 2**i. How can I generate such a list in Python?

The fastest, cleanest way would probably be using list comprehentions:

l = [2**i for i in range(10)]

From tim.golden at iname.com  Wed Oct  3 02:20:34 2001
From: tim.golden at iname.com (Tim Golden)
Date: 2 Oct 2001 23:20:34 -0700
Subject: Problem with COM and Excel
References: <3BBA1BD5.8E833283@vip.fi>
Message-ID: 

Pekka Niiranen  wrote in message news:<3BBA1BD5.8E833283 at vip.fi>...
> How can I run AutoFit to all columns of my Excel sheet from python ?
> 
> I have tried:
> 
> >>> xlApp = win32com.client.dynamic.Dispatch('Excel.Application')
> >>> xlBook = xlApp.Workbooks.Add()
> >>> sht = xlBook.Worksheets("Sheet1")
> >>> xlApp.Visible = 1
> 
> So far so good, but at this stage I added some values to Excel Sheet
> (manually) and then run:
> 
> >>> sht.Columns().AutoFit
> 
> The whole PC is jamming with harddisk running madly. I have to use task
> manager to kill Idle and Excel.
> 
> -pekka-

The following works for me:

def _2xl (headers, rowset, spreadsheet):
  """_2xl - put a cursor output set to a spreadsheet, optionally
applying traffic light formatting

  Parameters:
    headers - sequence of sequences containing column names & types
(typically from cursor.description)
    rowset - sequence containing each row as a sequence
    spreadsheet - full path to a spreadsheet
  """
  n_cols = len (headers)
  n_rows = len (rowset)
  xl = Dispatch ("Excel.Application")
  xl.DisplayAlerts = 0
  try:
    xl.Workbooks.Add ()
    xl.ActiveSheet.Range (xl.Cells (1, 1), xl.Cells (1, n_cols)).Value
= [header[0] for header in headers]
    xl.ActiveSheet.Range (xl.Cells (1, 1), xl.Cells (1,
n_cols)).Font.Bold = 1
    xl.ActiveSheet.Range (xl.Cells (2, 1), xl.Cells (1 + n_rows,
n_cols)).Value = rowset
    xl.ActiveSheet.Range (xl.Cells (1, 1), xl.Cells (1 + n_rows,
n_cols)).EntireColumn.AutoFit ()

    xl.ActiveSheet.SaveAs (Filename=spreadsheet)
    xl.Quit ()
  finally:
    xl = None
    del xl

I suspect the bit you need is the EntireColumn property of the range.
HTH. Tim

From ignacio at openservices.net  Fri Oct  5 04:11:07 2001
From: ignacio at openservices.net (Ignacio Vazquez-Abrams)
Date: Fri, 5 Oct 2001 04:11:07 -0400 (EDT)
Subject: Formatted printing of deep tuples
In-Reply-To: <01100517304104.10490@ike>
Message-ID: 

On Fri, 5 Oct 2001, Richard Jones wrote:

> Bored. Wrote this:

Heh. I write some of my best code when bored.

> def format_tuple(out, tuple, indent=''):
>     out('%s(\n'%indent)
>     new_indent = indent+'  '
>     for element in tuple:
>         if type(element) == type(()):
>             format_tuple(out, element, new_indent)
>         else:
>             out('%s%r,\n'%(new_indent, element))
>     out('%s)\n'%indent)
>
> import sys
> out = sys.stdout.write
> format_tuple(out, (((1, 2),), ((), ('a', (('b', 'c'),)))))
>
>
> Might write another before I go home ;)

Already done. Take a look at this:

---
def format_tuple(input, indent=2, _indent=0, chars='()', first=0):
  ret='%s%s' % (_indent*' ', chars[0])
  for element in range(len(input)):
    if element:
      ret=ret+', '
    if type(input[element])==type(()):
      ret=ret+'\n'+format_tuple(input[element], indent, _indent+indent, '()', 0)
      first=1
    elif type(input[element])==type([]):
      ret=ret+'\n'+format_tuple(input[element], indent, _indent+indent, '[]', 0)
      first=1
    else:
      ret=ret+'%s' % (`input[element]`)
      first=0
  ret=ret+'%s%s%s' % (first*'\n', first*_indent*' ', chars[1])
  return ret

print format_tuple((((1, 2),), ((), ('a', (['b', 'c'],)))))
---

I made the ouput a little clearer than my initial spec (like that's a _bad_
thing...).

-- 
Ignacio Vazquez-Abrams  

From claird at starbase.neosoft.com  Thu Oct 18 13:56:28 2001
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Oct 2001 12:56:28 -0500
Subject: On morgage payments
References: 
Message-ID: 

In article ,
John Thingstad   wrote:
>Isaw earlier a very inefficient way of calculating morgage on a loan.
>The recurence relation:
>
>P = amount
>i = interest
>d = downpayment
>
>P    = i*P    - d
>   n       n-1
>
>was solver by itteration in a loop
>
>I would like to point out:
>
>P     = i * (i*P     - d)  - d
>  i+1              i
>
>P       = i * ( i * (i * P   - d) -d) -d)
>  i + 2                     i       
>
>
>P      = i*i*i*P  -d(i^2 + i +1)
>  i+2            i
>
>by inspecion i*i*i is a exponential and i^2 + i +1 is a geometric series so we have:
>
>P    = i^n * P    - d * (i^n -1 / i -1)
>  n               0
>
>Which can be computed once.
>
>

I can't interpret what appears above in any standard way
to yield internally-consistent results.  Rather than patch
up what I suspect is just mis-parenthesization, I offer

 factor = (1 + i) ** n
 payment = P * i * factor / (factor - 1)

as the most useful pertinent closed-form computation.
-- 

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

From bonzo at vnet.net  Sat Oct 13 14:19:40 2001
From: bonzo at vnet.net (Alex Ravenel)
Date: Sat, 13 Oct 2001 18:19:40 GMT
Subject: Extracting attachments with Python?
Message-ID: <0B%x7.10230$YC3.3926894@typhoon.southeast.rr.com>

Is there a way to have Python extract a series of attachments from a
Microsoft Outlook .dbx file into a file of my choosing? I was sent a series
of documents by my brother, and while they arent really that many (about
100), I think it would be interesting to see if Python could do it for me.
Thanks.

--

Alex Ravenel
bonzo at vnet.net

From gh_pythonlist at gmx.de  Tue Oct 23 12:11:25 2001
From: gh_pythonlist at gmx.de (Gerhard =?iso-8859-1?Q?H=E4ring?=)
Date: Tue, 23 Oct 2001 18:11:25 +0200
Subject: POLL in different OSes
In-Reply-To: <5.1.0.14.0.20011023125024.027cf278@imap.hq.movilogic.com>; from lucio@movilogic.com on Tue, Oct 23, 2001 at 12:52:28PM -0300
References: <5.1.0.14.0.20011022224503.00b4e280@imap.hq.movilogic.com> <5.1.0.14.0.20011022224503.00b4e280@imap.hq.movilogic.com> <20011023101838.A561@lilith.hqd-internal> <5.1.0.14.0.20011023125024.027cf278@imap.hq.movilogic.com>
Message-ID: <20011023181125.A2111@lilith.hqd-internal>

On Tue, Oct 23, 2001 at 12:52:28PM -0300, Lucio Torre wrote:
> At 10:18 AM 23/10/2001 +0200, Gerhard H?ring wrote:
> >On Mon, Oct 22, 2001 at 10:51:02PM -0300, Lucio Torre wrote:
> > >  From the documentation:
> > >
> > > "The poll() system call, supported on most Unix systems"
> > >
> > > and my questions:
> > >
> > > a) is poll supported on Windows?
> > >
> > > my guess is no, but i hoping for a yes.
> >
> >It isn't. Unless, you're using Cygwin. Also, select only works with
> >sockets on Windows.
> >
> >Wouldn't it be possible to emulate poll() with select()? I'd be glad to
> >see a patch appear at Sourceforge if that's possible :-)
> >
> >Gerhard
> 
> i think i could do that. i can make a .c with the code that emulates
> poll from select, if someone else is willing to make it a patch or
> whatever it should be to get into python.

That would be nice. I'd be willing to review, test, document and do the
patch submission at Sourceforge, if you want to avoid that.

> btw: is it worth it? poll is supposed to be faster, so why emulate it?
> it would have no reason to exist!

Consistency across platforms is a good reason to have it. I once wanted
to use a Python module using poll on Windows and had to rewrite it with
select. Had poll() been available on Windows, this wouldn't have been
necessary.

Gerhard
-- 
mail:   gerhard  bigfoot  de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    OpenPGP public key id 86AB43C0
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))

From use-net at schabi.de  Tue Oct 23 17:15:12 2001
From: use-net at schabi.de (Markus Schaber)
Date: Tue, 23 Oct 2001 23:15:12 +0200
Subject: Apologise (was: Re: IF STATEMENTS)
References: 

	<20011018002005.53f58ef2.use-net@schabi.de>
	<20011018102800.11c71435.use-net@schabi.de>

Message-ID: <20011023231512.79e51793.use-net@schabi.de>

Hi,

On Fri, 19 Oct 2001 10:46:27 -0700, "John Roth" 
wrote:

> I saved it and looked at it, using Outlook Express. There is no text
> between the header and the
> first boundary delimiter. It appears that OE may be being overzealous
> in its attempt to protect you from viruses - even though it says txt,
> it still cautions me
> that it may contain harmful stuff, and requires me to validate it.

I'm afraid this is just another strange behaviour of OE (I don't want to call it bug :-)

In fact, it seems a security feature, as OE under some circumstances executes 

Found

Sample Output

You did it!

You did it!

%s

Python Form Test

You didn't fill something out

You did it!