From m.faassen at vet.uu.nl  Thu Jun 15 08:07:31 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 12:07:31 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com>
Message-ID: <8iagu3$ici$1@newshost.accu.uu.nl>

Garry Hodgson <garry at sage.att.com> wrote:
> Aahz Maruch wrote:

>> Like Grant, I've been using Unix about fifteen years, and I find Perl
>> bizarre and difficult.

> after reading several responses like this, i wonder if perl
> is particularly irritating to old timers.

I'm definitely not an old timer and it's frustrating to me too.
(Gordon McMillan makes comments like "Even Martijn knows it, and he's
young" :)

Regards,

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


From sjuranic at condor.ee.washington.edu  Wed Jun 28 14:02:37 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Wed, 28 Jun 2000 11:02:37 -0700
Subject: Python design philosophy
Message-ID: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>

I am *brand* new to python (as of Sunday), and I just got to the section
in the tutorial about classes.  I was wondering why there really isn't
such an idea as a "private" member of classes?  I understand that members
can be hidden, but not really protected from the user.  I was wondering
what the benefit of this would be?  Wouldn't it open the possibility of
security holes?  Also, what benefit is gained by allowing a user to
add/delete members from existing classes?  Wouldn't it be more desirable
to create his/her own derived class?

Please understand, I am not attempting to start a flame war.  These are
all questions asked in sincerity.

Thanks in advance. 

---------------------------------------------------------------------------
Stephen W. Juranich				sjuranic at ee.washington.edu
Electrical Engineering  	   http://students.washington.edu/sjuranic
University of Washington		http://truk.ee.washington.edu/ssli




From thomas at xs4all.net  Fri Jun 16 07:33:55 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Jun 2000 13:33:55 +0200
Subject: Python segfault
In-Reply-To: <3949C958.D3AA1F4@stud.ntnu.no>; from petersc@stud.ntnu.no on Fri, Jun 16, 2000 at 08:29:44AM +0200
References: <slrn8kits2.ggk.perly@typhoon.xnet.com> <3949C958.D3AA1F4@stud.ntnu.no>
Message-ID: <20000616133354.N26436@xs4all.nl>

On Fri, Jun 16, 2000 at 08:29:44AM +0200, Peter Schneider-Kamp wrote:

> > I don't know if anyone has seen this,  but python segfaults when given a
> > rather large input.

> > [pat at perly pat]$ perl -e "print '2+2+' x 10000" > input
> > [pat at perly pat]$ python < input

> Strange. Looks like the limit is exactly at 8192. That makes 16384
> operations the parser (or whatever crashes) can hold.

> Not that the problem is not the length of the input itself:

> $ perl -e "print '22222222 ' x 200000 > input
> $ python < input

> works just fine. I think the problem is the add (or mul) operation.
> So the problem might be in the expression parser.

> Is there anyone with some more insight around?

The problem is not the length of the expression, nor the 'depth' of the
expression (as someone else suggested), but the 'width' of the expression. 

Each expression (a 'node') is parsed into X different expressions (also
nodes) which are the 'children' of the current node. And each of those nodes
is parsed just like the first one, and so on.


So, the expression '2 + 2 + 2 + 2' is parsed somewhat like this:

            '2+2+2+2'
               /|\
              / | \
             / /|\ \
            / / | \ \
           / / /|\ \ \
          / / / | \ \ \
          2 + 2 + 2 + 2

(actual operation to perform ommited, obviously)
Whereas '(2 + 2) + 2 + 2' is parsed like this:

         '(2+2)+2+2'
               |\
              /| \
             / | |\
            /  | | \
         (2+2) 2 + 2
          /|\
         / | \
        2  +  2

The problem is not the 'height' of this graph, or the total number of
'nodes' (which is limited only by memory), but the 'width' of the first
tree. The 'node' struct is defined like this:

typedef struct _node {
        short           n_type;
        char            *n_str;
        short           n_lineno;
        short           n_nchildren;
        struct _node    *n_child;
} node;

and the 'n_nchildren' member is the number of immediate subexpressions (not
grandchildren, only children) -- and it is limited to a short, which is 16
bits on most systems. And the short is signed, which means the maximum
number of children is 32767. This includes the operators itself, so you can
operate on just 16384 items in a single expression. If you try to operate on
more, the short wraps, and the parser tries to index node->n_child[-32768],
which usually segfaults.

If you really hit this limit, you can add a couple of parentheses to solve
the problem ;) But I agree the segfault is not very nice. There should
probably be an overflow detection somewhere in node.c, but I'm not sure what
kind of error it should generate.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From drew.csillag at starmedia.net  Fri Jun  9 14:23:42 2000
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Fri, 9 Jun 2000 14:23:42 -0400 (EDT)
Subject: Python Productivity over C++
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F339E@VRTMAIL>
Message-ID: <Pine.LNX.4.10.10006091418020.10231-100000@naxos.skunk.org>


On Fri, 9 Jun 2000, Mike  Fletcher wrote:

> Just a minor note, having spent all of last night (unsuccessfully) trying to
> track down a memory leak.  Yes, memory leak tools would be nice to have in
> Python.  Hidden memory leaks (cycles) are rare, but when they happen in a
> complex project they are often a nightmare to track down.  An optional
> garbage collector with a debug mode saying "hey, I found a cycle that wasn't
> cleaned up, the objects were: object types, object cycle description" would
> be really cool.  Yes, I'd probably remove it for the final proggy, but as a
> development tool it would be very, very, useful, and is currently (to my
> memory), not available.

Good point.  But there are actually two cycle detectors (one may or may
not be derived from the other) that I know of: plumbo and Cyclops (I've
used both).  I believe (I may be totally wrong on this) that one of them
will be included with 1.6.

> 
> 0.02
> Mike(y)
> 
> -----Original Message-----
> From: Drew Csillag [mailto:drew.csillag at starmedia.net]
> Sent: Friday, June 09, 2000 1:07 PM
> To: Steve Mullarkey
> Cc: python-list at python.org
> Subject: Re: Python Productivity over C++
> 
> 
> 
> 
> On Fri, 9 Jun 2000, Steve Mullarkey wrote:
> ...
> 
> Python's tools are very good (and don't need them a lot of the time
> anyhow).  The main thing missing is a GUI builder (but glade and PyGlade
> are getting there quickly).  Memory leak tools just don't apply (Python
> has reference counted garbage collection).  Programmers editors are
> numerous and one is included (IDLE).
> 
> ...
> 

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






From jkraska1 at san.rr.com  Sat Jun 10 17:14:49 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 10 Jun 2000 21:14:49 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net> <39420BAF.D86DA2BF@stud.ntnu.no>
Message-ID: <3942B05F.47FDBC0A@san.rr.com>

> if you write in lower-/mixed-case you see some letters sticking
> out of the top or the bottom. Studies have shown that words are
> not recognized letter by letter but as a complete pattern.

Studies have furthermore shown that you are correct: the variety
in size of the symbols eases the task of reading.



C/

(who once majored in cognitive science, which studies such things)


From gmcm at hypernet.com  Wed Jun  7 15:58:48 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 07 Jun 2000 19:58:48 GMT
Subject: Gordon M's Installer- Missing DLL
References: <01bfd08e$d36f8da0$0100007f@Cyberus>
Message-ID: <8F4CA44E5gmcmhypernetcom@199.171.54.195>

g_will at cyberus.ca (Gordon Williams) wrote: 

>I am trying to distribute a program that I have packaged into a
>standalone. 
> The package is installing ok but I am getting a report back that a dll
> is 
>not loading when I run the standalone.
>
>C:\Scopal>scopal03b.exe
>Traceback (innermost last):
>  File "<string>", line 1, in ?
[snip]
>  File "E:\Program Files\Python\wxPython\__init__.py", line 20, in ?
>    import wxc
>  File "E:\Program
>  Files\Python\Installer_3f\Installer\support\imputil.py", 
>line
> 54, in _import_hook
>    return self.__chain_import(name, globals, locals, fromlist)
>ImportError: DLL load failed: One of the library files needed to run
>this applic
>ation cannot be found.

Python is reporting an error that Windows gave it when it tried to load 
wxc.pyd. The missing dll is almost certainly wx21_13.dll (maybe a different 
verion embedded in the name) which is in your system directory.

- Gordon


From thomas at cintra.no  Wed Jun 14 03:11:15 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 14 Jun 2000 07:11:15 GMT
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com>
Message-ID: <394a2fed.280565932@news.online.no>

Mandrake 7.0 and 7.1 has it installed, and 7.1 has PIL included too. A
very good distro.

Thomas

On Tue, 13 Jun 2000 15:49:46 GMT, ge at nowhere.none (Grant Edwards)
wrote:

>I've got a Tkinter-based GUI interface used to do
>administrative stuff for a device driver under Linux.  I know
>RedHat installs Python and Tkinter by default.  Is there a
>handy list somewhere of which Linux distributions can be
>expected to have Tkinter installed?
>
>I know it's available on almost all of the distributions, but
>I'd like to know which ones install it in the "normal"
>configuration that joe-windows-user is going to get if he does
>a default workstation or server install.
>
>Had I the spare time and disk space I'd love to install a half
>dozen distributions, but resources don't permit that at the
>moment.



From gansevle at cs.utwente.nl  Wed Jun 28 07:07:57 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Wed, 28 Jun 2000 11:07:57 GMT
Subject: SystemError: bad argument to internal function
References: <8jcltq$1li$1@nnrp1.deja.com>
Message-ID: <8jcma6$1v2$1@nnrp1.deja.com>

I forgot to mention:

python.exe:
  sys.version: '1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)]'
  sys.platform: 'win32'
  os.name: 'nt'

also installed:
  win32all-129.exe
  win32all-129-update.exe

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From eviltofu at rocketmail.com  Sat Jun 10 12:07:41 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 10 Jun 2000 16:07:41 GMT
Subject: namespace question
Message-ID: <eviltofu-8E2F45.12073510062000@news.earthlink.net>

If I do the following:

import socket

socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)

how do I call a function from the socket module later on?


From wlav at hpatl26.cern.ch  Wed Jun 14 04:23:08 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 14 Jun 2000 08:23:08 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB334@ALTNET>
Message-ID: <8i7fdc$nti$1@sunnews.cern.ch>

Mike Steed <MSteed at altiris.com> writes:
>> From: wlav at atlas03.cern.ch [mailto:wlav at atlas03.cern.ch]

>> All that STL containers require is that your dtor and copy assignment
>> don't throw. The compiler generated members fullfill that requirement.

>True.  And although it's not required, it *is* a good idea to make sure that
>each class's copy constructor and assignment operator behave properly.

For that matter, it *is* a good idea to make sure that all your code
behaves properly. So?

>The default copy ctor often does not do what you want, e.g., if your class
>contains pointers to other objects.

If generated members are 'often' not good enough, then all red lights
should start flashing, as you have a design flaw. It is extremely hard to
make such code exception safe. Instead, you should encapsulate the code
that needs a specialized cctor and use it as member or private base.

I've done a quick scan through some of my code base and I have found 2
classes that need an explicit cctor.

>> Please inform yourself better before making such statements. Blindlessy
>> drawing conclusions based on the C++ implementation of one particular
>> vendor from Redmond isn't good for your credability.

>And assuming that I'm blindlessly drawing conclusions based on Microsoft's
>compiler is not good for your credibility.

It is the only compiler that comes with an STL that beeps at you when you
don't provide said members. If I may ask then, where did you find your
enligntment?

>The default copy ctor (regardless of the compiler) simply does a
>member-by-member copy, which is often the wrong thing to do.

Again, I strongly disagree with your 'often'.

>The point is that, even if the default is
>acceptable, the C++ programmer has to (or should) think about it.

There is such a thing as 'second nature'.

>This discussion illustrates the original point: these are details that the
>Python programmer does not need to deal with.

I fully agree that Python is far simpler, but STL is not so complex as you
seem to want it to be.

Best regards,
         Wim Lavrijsen


From moshez at math.huji.ac.il  Sat Jun 10 01:33:33 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 08:33:33 +0300 (IDT)
Subject: Gnome Applets and Python
In-Reply-To: <39416F2B.96AD2322@oreilly.com>
Message-ID: <Pine.GSO.4.10.10006100832420.1511-100000@sundial>

On Fri, 9 Jun 2000, Stephen R. Figgins wrote:

> Any good suggestions on Gnome Applets and Python information?

There's a clock-applet.py example in the gnome-python module on the GNOME
cvs archive. Very cut-n-pastable <wink>

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From mfletch at tpresence.com  Thu Jun  8 14:37:01 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 8 Jun 2000 14:37:01 -0400 
Subject: setblocking in Win32
Message-ID: <CC59868A0A76D311B1F50090277B321C1F339B@VRTMAIL>

Non-blocking ports most definitely work on win32 (almost everything I write
uses them).  The EWOULDBLOCK (10035) error is thrown only when you're in
non-blocking mode (otherwise the call would just wait until it could accept
before returning).  See asyncore.py for examples of how to work with
non-blocking sockets (mostly just catch the socket.error and see if the
first argument is 10035, if so, go about your other business, else re-raise
the error).

Congrats, you're non-blocking :) .
Mike

-----Original Message-----
From: jkingry at my-deja.com [mailto:jkingry at my-deja.com]
Sent: Thursday, June 08, 2000 1:57 PM
To: python-list at python.org
Subject: setblocking in Win32


Hi, I think I'm going to be told what I don't what to hear, but from
the error below, I think I have surmised that trying to set non-
blocking on a socket in a Win32 environment doesn't work.

Are there any work arounds?

Joe

Traceback (innermost last):
  File "D:\Prog\Python\Lib\threading.py", line 376, in __bootstrap
    self.run()
  File "D:\Prog\Python\Lib\threading.py", line 364, in run
    apply(self.__target, self.__args, self.__kwargs)
  File "io.py", line 80, in run
    sock,(address,port) = server.accept()
  File "D:\Prog\Python\Lib\plat-win\socket.py", line 36, in accept
    sock, addr = self._sock.accept()
error: (10035, 'winsock error')


Sent via Deja.com http://www.deja.com/
Before you buy.
-- 
http://www.python.org/mailman/listinfo/python-list



From gregm at iname.com  Mon Jun 12 09:36:45 2000
From: gregm at iname.com (Greg McFarlane)
Date: Mon, 12 Jun 2000 23:36:45 +1000
Subject: Tkinter: disable a item in a Menubar
In-Reply-To: <393E114E.1F59AD44@teleatlas.com>; from Klaus Bruns on 07 Jun 2000 at 11:09:34AM
References: <393E114E.1F59AD44@teleatlas.com>
Message-ID: <20000612233645.02575@nms.otc.telstra.com.au>

It is possible to do what you are trying to do.  To work out how to do
it, you need to carefully study the Tkinter and Pmw reference manuals. 
There you will discover that:

    1) The Pmw.MenuBar addmenu() method creates a menu component.
       In this case, the name of the menu component is 'FILE-menu'.
    2) Pmw allows you to access the components of a megawidget, using
       the component() method.
    3) To configure a Tk menu item, use the entryconfigure() method.
    4) To disable a Tk menu item, set the 'state' option to 'disabled'
       and to enable it, set it to 'normal'.

Putting this all together:

   def open_file(self):
      ## change the state of 'open' to DISABLED
      fileMenu = self.menuBar.component('FILE-menu')
      fileMenu.entryconfigure('open', state = 'disabled')

   def save_file(self):
      ## change the state of 'save' to DISABLED and 'open' to Normal
      fileMenu = self.menuBar.component('FILE-menu')
      fileMenu.entryconfigure('open', state = 'normal')
      fileMenu.entryconfigure('save', state = 'disabled')

Greg

On 7 Jun, Klaus Bruns wrote:
> Hi all,
> 
> I want to change the state of an item in a Menubar, but I don't now how
> to do.
> For example: in the Menu 'File' there is a funktion to open and save a
> file. After using the 'open' funktion this funktion is disable till
> using the 'save' funktion.
> 
> I use Pmw for creating the Menubar:
> 
> import Pmw
> from Tkinter import *
> 
> class mymenu:
>    def __init__(root):
> 	Pmw.initialise(root)
> 	self.balloon=Pmw.Balloon(root)
> 	self.menuBar=Pmw.MenuBar(root,hull_relief=RAISED,hull_borderwidth=1,
>                                  balloon=self.balloon)
>         self.menuBar.pack(fill=X)
>         self.menuBar.addmenu('FILE','open and save files')
> 	self.menuBar.addmenuitem('FILE','command','open a file',label='open',
> 				  command=self.open_file)
> 	self.menuBar.addmenuitem('FILE','command',save a file',label='save',
> 				  command=self.save_file)
> 	
>    def open_file(self):
> 	file  = open(filename)
> 	## here I want to change the state of 'open' to DISABLED
> 
>    def save_file(self):
> 	## do something to save the file
> 	## and change the state of 'save' to DISABLED and 'open' to Normal
> 
> root = TK()
> mymenu(root)
> root.mainloop()
> 
> Can anyone help me?
> 
> 	Klaus
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From nascheme at enme.ucalgary.ca  Wed Jun 21 14:27:14 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Wed, 21 Jun 2000 18:27:14 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au> <39505457.5223ADC5@san.rr.com> <20000621155338.C11308@fangorn.csse.monash.edu.au>
Message-ID: <slrn8l22bv.ma2.nascheme@cranky.arctrix.com>

Toby J Sargeant <tjs at cs.monash.edu.au> wrote:
>:set lcs=tab:?\ 
>:set list

Unfortunately both those settings are global instead of per
buffer.  I guess you could hack some macro to change them when
you enter a .py buffer.

  Neil

-- 
"Reliability means never having to say you're sorry." -- D. J. Bernstein


From herzog at online.de  Thu Jun 29 08:02:40 2000
From: herzog at online.de (Bernhard Herzog)
Date: 29 Jun 2000 14:02:40 +0200
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <8F61BBD91gmcmhypernetcom@199.171.54.194> <slrn8ll4ak.25h.hzhu@rocket.knowledgetrack.com> <8F61DA247gmcmhypernetcom@199.171.54.194>
Message-ID: <m3n1k4k84f.fsf@greebo.nodomain.de>

gmcm at hypernet.com (Gordon McMillan) writes:

> Lets say you have a (gag) 2,000 line module: foo.py. To speed things up 
> (avoid the compile), you write a 2nd script:
> ------
> import foo
> foo.main()
> ------
> 
> You shouldn't have to add the directory this is in to PYTHONPATH for this 
> to work. (This being why "." *should* be on the path.)

That has nothing to do with '.' being in sys.path. The directory that
has to be in the path is not usually the cwd, it's the directory where
the script is. Python automatically prepends that directory to the path,
and it even follows symlinks to find it.

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


From jerome.quelin at insalien.org  Thu Jun 15 18:11:52 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 16 Jun 2000 00:11:52 +0200
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com> <memo.20000615202836.52185E@digitig.compulink.co.uk>
Message-ID: <961107112.124995831@news.libertysurf.fr>

Hello,

>> I wonder if Python runs faster than Perl.
>Check out:
> http://wwwipd.ira.uka.de/%7Eprechelt/documents/jccpp_tr.pdf
>Which somebody posted here recently. Where the best Python was shown to 
>run faster than the best Perl, and the worst Python was shown to run a 
>/lot/ faster than the worst Perl. 

In this report, it is stated as one final conclusion that Python and
Perl are equivalent, with maybe an advantage for Perl :
[snip]
Within the script langages, Python and in particular Perl are faster than Rexx
and Tcl
[snip]


But the report also states that you can't issue conclusions on language speed,
advantages and drawbacks :
[snip]
For all program aspects investigated, the performance variability due to
different programmers (as described by the bad/good ratios) is about as large
or even larger than the variability due to different languages on average.
[snip]

So you can't tell which is faster between Perl or Python. Let's just stop those
flamewars.


I just would like say that I'm fond of Perl since a long long time, and if I
went to Python, it's mainly because I got frustrated with it when dealing with
larger applications.

I currently consider myself in a learning phase (even if I've already
read some books and tutorials on Python, and experienced a little with the
interpreter and made some very little scripts) since I have not applied it on
medium-size examples or on a real-world problem (in fact, I'm looking for an
idea of program I could work on, if you have any idea/suggestions...). This to
say that my opinion on Python may be a little inaccurate, but I hope it's not
too far from reality.

The point I wanted to underline is that different languages have different
scopes. And even if Perl and Python are really close, this doesn't mean they are
interchangeable.

I won't use Python for some tasks where Perl is (in my mind) the best suited.
For example, I think text processing is by far best handled by Perl with all
its text-oriented features (its native regular expression integration, the
facilities to handle files opening, reading and parsing, etc...) and the
short-cuts it provides (the Perl motto TIMTOWTDI).

And despite the fact I'm Perl addicted, I consider Python a lot better
for larger applications (and especially with GUI, since Perl/Tk isn't really
easy to handle - believe me ;-) ), due to its syntax, its OO model and the
readibility it procures.

But not for any kind of applications : you are to be clever enough to turn
yourself to C/C++ (or even ASM) if you need to perform low-levels operations on
your system, or to speed up your code.

Unless you are to develop a 1'000'000+ lines applications ? Then choose ADA or
another language that will allow you to manage your code efficiently...

Do not blindly choose Python just because you're Python addicted. Make your
choice with clever information. And if finally it appears that Python is the
best suited for the problem you want to solve, you will have the satisfaction to
know why (Q: "Why are you coding in Python ?" A: "Because it's the best."
Q: "And why is it the best ?" A: "Hermm....").

Sincerely,

Jerome
--
jerome.quelin at insalien.org





From gmcm at hypernet.com  Fri Jun 30 08:59:36 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 30 Jun 2000 12:59:36 GMT
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org> <8F62AF73Egmcmhypernetcom@199.171.54.194> <36D5F4DFCA9BCA46.B5AE15F2F19F393C.28F9B766F10B5E23@lp.airnews.net> <8F62C20E2gmcmhypernetcom@199.171.54.154> <C311C39CF376029F.651A9619D7C71B72.66E09700B142DAB0@lp.airnews.net>
Message-ID: <8F6354CB4gmcmhypernetcom@199.171.54.154>

claird at starbase.neosoft.com (Cameron Laird) wrote in
<C311C39CF376029F.651A9619D7C71B72.66E09700B142DAB0 at lp.airnews.net>: 

>In article <8F62C20E2gmcmhypernetcom at 199.171.54.154>,
>Gordon McMillan <gmcm at hypernet.com> wrote:
>               .
>               .
>               .
>>The whole inode thing in that code is because it is a common Linux
>>idiom to rely on the fact that once a process has opened a file, said
>>file can be deleted by another process without the reading process
>>noticing. 
>>
>>This idiom is completely impossible on Windows, so there's no sense in 
>>checking whether a new file with the same name has now been created.
>>
>>So what'll it be - a portable version that's useless on Linux, or one
>>that spends half it's time checking for something that can't occur on
>>Windows 
>><wink>?
>>
>>on-Windows-once-you've-grabbed-the-tail-
>> you've-got-the-whole-damn-dog-ly y'rs
>>
>>- Gordon (who never uses popen on Windows anyway)
>
>It's a good question.
>
>Is tail(1) in POSIX?  I'm out of copies of POSIX documentation
>again ...

Apparently not...

>It's never been a difficult question for me.  I've ended up
>coding different tail functions (methods) for different pro-
>jects.  Am I committing redundancy in my coding?  Barely,
>and at a level with which I'm comfortable.  The general cat-
>egory here is, "ambiguously defined stuff where the ideas
>are so clear it's easier recoding them than trying to merge
>conflicting requirements/definitions."

I'll agree with that. And I'll point out that I wrote the [long ago 
snipped] code *instead* of using os.popen("tail -f...."), because I really 
didn't want yet another shell running, and because I would still have had 
to deal with the writing process moving on to a new inode.

>I most often need such a tail for log-scraping, and part of
>my tail implementations generally is *not* to do what you
>describe, but to stop reading the mv-ed i-node and pick up
>with a newly-created file (a log, in my case).  This illu-
>strates the ambiguity in tail's semantics.  I repeat:  I
>resolve the ambiguity by my knowledge of requirements spe-
>cific to a project.

Hmmm. We definitely are having a problem in communicating. That was the 
whole reason for the inode checking - so the "tail"ing process could move 
on to the new file.


>You and I might have a slight vocabulary clash.  My code
>often involves tests of os.name.  I think of the results
>as "a portable version", without embarrassment.  Perhaps
>your habit is to regard the same as *not* "pure" Python,
>and correspondingly "importable".

Oh no, I have no problem with that. I was just pointing out that besides 
the platform syntactic differences, there are also platform semantic 
differences (i.e., yet more ambiguity).

C'mon, Cameron. You *know* if we disagree, at least one of us has 
misunderstood ;-).

- Gordon


From jvickroy at sec.noaa.gov  Mon Jun 12 12:29:16 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 12 Jun 2000 10:29:16 -0600
Subject: enabling threads on HP-UX
Message-ID: <39450FDB.F5998449@sec.noaa.gov>

Hello,

Our systems administrator has installed Python 1.5.2 (downloaded from
www.python.org) on an HP-UX (v 11) OS with threads enabled.  File
./configure was run with the '--with-thread' option.

The following is a sample session illustrating the problem:

$ python
Python 1.5.2 (#1, Jun  9 2000, 14:42:40)  [GCC 2.95.2 19991024
(release)] on hp-uxB
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> def count():
...     for i in range(1,10): print i
...
>>> count()
1
2
3
4
5
6
7
8
9
>>> import thread
>>> thread .start_new_thread (count, () )
Traceback (innermost last):
  File "<stdin>", line 1, in ?
thread.error: can't start new thread



Any ideas about what we have done incorrectly?

Thanks for your time.



From invalid at dev.null  Fri Jun 30 01:35:54 2000
From: invalid at dev.null (Alexander K)
Date: Fri, 30 Jun 2000 05:35:54 GMT
Subject: command-line history in win
Message-ID: <395c304e.1224185@news.student.lu.se>

aloha!

i did a backsearch for this on deja. found a question but no answers.
and the question is: isnt there a historyfunction in the python
command-line interpreter for win?.

you know, the bash-kinda thing.


  thanks in adv. / alex k


From sdhyok at email.unc.edu  Sat Jun 17 16:53:38 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Sat, 17 Jun 2000 13:53:38 -0700
Subject: Python vs. Matlab
References: <724s6sa0k5.fsf@rogn.ii.uib.no>
Message-ID: <8ige1a$5gp$1@slb2.atl.mindspring.net>

One problem of Matlab I found is that there is no explicit control
in reference copying.
As soon as you change an element of a huge array in a function,
Matlab copies all the elements into a new array.
This has a significant effect on performance
in s/w of large scale.
In Numpy, you have the perfect control in copying array
, as well as its source.

Peter

Helge Avlesen <avle at rogn.ii.uib.no> wrote in message
news:724s6sa0k5.fsf at rogn.ii.uib.no...
>
> I have become almost addicted to Matlab after using it extensively for
> some years, however, the licencing feels very strict after getting
> used to free and very good quality software like Emacs, Linux, etc.
>
> My main use of Matlab is reading in large binary files into
> arrays(output from simulations), do some processing with various
> matrix routines, then extract sections of the 3D arrays for plotting.
>
> Spotted Python some months ago, and liked it!  Can Python, Numpy and
> the many plotting extensions be used as a replacement for Matlab, is
> Python slower/faster etc.?
>
> It would be nice to hear some experiences!
>
> --
> Helge Avlesen




From r2d2 at mao.acc.umu.se  Thu Jun 15 03:13:47 2000
From: r2d2 at mao.acc.umu.se (Niklas Frykholm)
Date: 15 Jun 2000 07:13:47 GMT
Subject: Piping output to a web browser...
References: <8i8moh0q8l@enews3.newsguy.com>
Message-ID: <slrn8kh0ha.q2p.r2d2@mao.acc.umu.se>

>A while back I saw a thread about keeping a web coonection alive (too bad I
>saw it 4 hours AFTER I came up with the same general plan).  But I am still
>struggling with how I can pipe the output to the screen on a streaming
>basis.  The code I have seems to cache it all up and pushes it to the screen
>all at once once I return out of the function.  Any ideas?
>
>    print "Content-type: text/html"
>    print
>    sys.stdout.flush()

Flushing stdout may not help in a CGI program, because your output is not
sent directly to the user, it is sent to the web server and from there
it is relayed to the user. The web server keeps its own buffers and your
output might get stuck there.

What you need to do is tell the web server to use unbuffered output,
or that you want to send your output directly to the user. How you do
this differs from web server to web server.

In Apache you specify that you want a direct communication line with
the user by making sure that your script name starts with "nph-" for
non-parsed headers (it is a strange solution, I know). You must also
change your program so that it writes a complete HTTP header starting
with

HTTP/1.1 200 OK

(this is normally added by the server, since you bypass the server
you have to write it yourself)

// Niklas


From kens at sightreader.com  Fri Jun 16 01:23:36 2000
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Jun 2000 01:23:36 -0400
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E27@mail.sauder.com>

It's happened to me many times.  Usually when cutting and pasting from other
peoples code (the other code having tabs, where my code does not).  Over the
past two years this has cost me at least 80 to 90 seconds of development time.
The solution is simple: make tabs illegal.  The biggest time saver, of course,
is that it would stop this utterly rediculous whitespace debate.

In any case, indentation/block rules don't seem to be the most important issue
when selecting a language for a task.

Mark Jackson wrote:

> "Warren Postma" <embed at NOSPAM.geocities.com> writes:
>
> > Having something
> > look correct on the screen, but not work because of misuse of tabs versus
> > spaces, or incorrect indentation, is something that still continues to bug
> > me.
>
> Indulge my curiosity - how many times has this actually happened in
> your six months of using Python?
>
> --
> Mark Jackson - http://www.alumni.caltech.edu/~mjackson
>         You can never solve all difficulties at once.
>                                 - Paul Dirac

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000616/6161ee69/attachment.vcf>

From db3l at fitlinxx.com  Wed Jun 21 18:00:59 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Jun 2000 18:00:59 -0400
Subject: Frozen socket.
References: <Pine.GHP.4.21.0006161522100.3877-100000@don.et.byu.edu> <uln05qioe.fsf@ctwd0143.fitlinxx.com> <Pine.GHP.4.21.0006211035450.5957-100000@don.et.byu.edu>
Message-ID: <ur99qhexw.fsf@ctwd0143.fitlinxx.com>

"C. Porter Bassett" <porter at et.byu.edu> writes:

> I made two small programs, a server and a client, and I got the client to
> "freeze" in the exact same way.  My test server accepts three messages,
> but after that, it freezes.  For some reason that I don't understand, the
> client keeps sending messages (about 1000 of them, just like before), even
> though the server is not receiving any.  After about 1000 messages, it
> freezes, just like the program I am debugging.

Um, right - this pretty much maps to the behavior I mentioned in the
last response.  If your server is not actively receiving data, then it
will begin to back up within the queues within the system.  There is
some amount of data that may be held on the server's receiving end
(within the networking kernel buffers), and data in networking queues
on the sending end.  There's also some data in transit in the network
- while that's not normally additive to the overall queue size as
perceived by the application.

> Why is the client able to keep sending messages if the server is not
> receiving them?  

Because the messages are being queued up (either on the client or
server end) until you max out the internal network queues provided by
the systems involved.  Queue sizes may also be configured for a given
socket or default values via network kernel configuration.

When you use send() you are handing off your data to the network
kernel, and in particular the TCP protocol.  That protocol queues up
the data to be transmitted as soon as possible, but there might be
some earlier data that TCP is still working on, so it understands how
to hold onto your data.  It's these internal buffers that provide the
"elasticity" you are seeing.  You aren't actually transmitting data
with send() - you are just passing it off to be transmitted by the
underlying network layers.

> In short, what is really happening, and how do I safeguard against it?

Well, it's working as designed, so I'm not entirely sure what you are
trying to safeguard against.  TCP works as a stream protocol - you
send it any amount of data and it'll consider it a stream of bytes
that need to get to the other side eventually.  Since TCP is in
complete control of the transmission, eventually might take a long
time - in this case your eventual blocked send() could wait for a long
time unless the server shuts down the connection - but if the server
starts reading again, everything will immediately pick up again.  So
in one respect, this is the simplest behavior for the application.
You just keep using send() and you know you're data will get to the
other end, or eventually you'll get an error from send().  You don't
really care if it blocks because it'll eventually start up again when
the server is ready for more data.

If you need different behavior, then you have to design it into your
application protocol, and TCP may or may not be the appropriate
underlying network protocol to use.

For example, if you want to be sure that your information is getting
through, you could do a simple ACK/NAK sort of setup where the server
sends a response to the client for each transmission.  This would let
the client know for sure the information is received before trying to
send again.

But without knowing the semantics you are looking for at the
application level, it's hard to say what else you might want to do.
But in this particular scenario you are testing here, things are
pretty much working as designed, so it may be a mismatch between your
expectations and the design, or just that you need to consider the
ramifications of a real, variable network in the middle of your two
apps :-)

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


From thomas at xs4all.nl  Mon Jun 26 07:09:04 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 11:09:04 GMT
Subject: SuSE Linux, python and setgid
References: <8j72mh$7q0$1@sun27.hrz.tu-darmstadt.de>
Message-ID: <slrn8leee3.2v5.thomas@centurion.xs4all.nl>

On Mon, 26 Jun 2000 10:04:01 +0200, Thomas Leonhardt
<leonhardt at informatik.tu-darmstadt.de> wrote:
>I have got the problem that I should install a script on our server
>which was written in python. The problem is that the script checks
>wether the group id was set as specified by the file permissions.

>But somehow the gid is not set when I execute the script. Is there
>something like suidperl for python or smoethingelse special with
>SuSE linux which causes the gid not beeing set??

It isn't SuSE specific, nor Python specific: scripts (interpreted files)
can't be setuid or setgid, under most UNIX systems, for reasons of security.
You'll have to create a wrapper (in C or another non-interpreted language)
that takes the proper gid, and then executes the python interpreter for you.
Take a look at Misc/setuid-prog.c, in the Python source tree.

Secure-ly y'rs,
	Thomas.



From hungjunglu at hotmail.com  Fri Jun  9 14:54:50 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Fri, 09 Jun 2000 11:54:50 PDT
Subject: Python Productivity over C++
Message-ID: <20000609185450.91797.qmail@hotmail.com>

--- In python-list at egroups.com, "M.-A. Lemburg" <mal at l...> wrote:
>Steve Mullarkey wrote:
> > 3. I use incremental compiling and linking. This reduces wait times to
> > very small amounts.
> > > 4. I can also use background compiling and linking.
>
>Python compiles on-the-fly at run-time and only if needed.
>You can even program interactively and have the program
>compile its own code.

I really have to raise my voice here: in Python, running means also 
compiling. Steve: sorry to point this out to you, but you really ought to 
take a look at dynamical reloading of modern script languages like Python. 
To me, that's the biggest advantage of modern script languages. In Python, 
there are no separate commands for compiling and running. And you can reload 
a Python module on the fly. Do you understand that? For many applications 
(like webservers), it is absolutely crucial that the main application keeps 
running while you are adding/modifying part of the program. That's an area 
where C++ can not do well. In Python, I often have programs running, and 
then I need to modify part of the program. I go ahead and modify the module, 
and since I have designed the main program to reload the submodules, I never 
have to stop the main application, this even if there are compiling/runtime 
errors from the submodule. Doing that in C++ is an absolute nightmare.

Python not only does incremental compiling and linking, it does it in REAL 
TIME. You can modify your program while it is running. This is absolutely 
critical for applications like webservers: you don't want to stop the main 
application because you want to change one line of code. This is a 
tremendous time saving over the traditional compile-run-debug,
compile-run-debug, compile-run-debug,... cycles. In Python, it was more 
like: debug-click, debug-click, debug-click,... cycle. I just can't imagine 
how I was going to do the same thing in C++ within the same time frame. I 
would say that 10 times shorter development time is probably an 
understatement in that particular case. In my case, it was between a 
possible project and an impossible project.

The biggest advatages of script languages are that they make dynamic 
reloading easier and they offer reflection (or introspection): the language 
is capable of looking into itself, and even perform meta-programming. I know 
you can make your C++ program to do this kinds of things, but you have to 
spend a tremendous amount of development time before you can reach some 
minimum features of dynamic reloading and reflection. (I have also used Java 
reflection features, but Python's reflection is way simpler and superior.)

When you have dynamic reloading, designing GUI in Python is still not as 
comfortable as in Visual Basic, but darn close.

I worked for a multimedia CD-ROM game company before, and it really sucked 
to use C++ or Java because of the lack of dynamic reloading. Programmers 
waste too much time in the compiling process. With Python, you can have the 
application running, find out the bug, fix it, and all while the game is 
still running. Too bad I already changed job.

Hung Jung

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From rod at inf.bme.hu  Fri Jun  9 03:38:52 2000
From: rod at inf.bme.hu (Gabor Papp)
Date: Fri, 09 Jun 2000 09:38:52 +0200
Subject: where is PyOpenGL?
Message-ID: <39409F0C.8935C54F@inf.bme.hu>

Where can I download PyOpenGL for win32 from? I've tried the
following:
http://starship.python.net/~da		- the link seems to be broken
http://PyOpenGL.sourceforge.net/	- it's empty
thanks in advance,
Gabor Papp


From long_way_home at my-deja.com  Wed Jun  7 19:42:19 2000
From: long_way_home at my-deja.com (long_way_home at my-deja.com)
Date: Wed, 07 Jun 2000 23:42:19 GMT
Subject: numpy installation woes
Message-ID: <8hmmko$qd3$1@nnrp1.deja.com>

Hi.

I am new to pyhton and have tried to install numpy, but the darn thing
doesn't recognise it's existence.  I have also installed MatPy and that
works fine.  When I try to import it I get the message
>>> from Numeric import *
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: No module named Numeric

although the directory is there.  I am using an SGI O2 and have compiled
with the SGI compilers, will python complain about the .so suffix.  I'm
buggered if I know how to fix this and would really appreciate some
help, for the NumPy manual has no more useful info.

Cheers
Geoff
--
Geoff Low
"I am the king of all I see, my kingdom for a
voice"


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


From scherbi at bam.com  Thu Jun 29 12:40:41 2000
From: scherbi at bam.com (Bill Scherer)
Date: Thu, 29 Jun 2000 16:40:41 +0000
Subject: Text in Tables
References: <IBEALDELHCKACLINDGEKCECLCAAA.mwelch@incigna.com>
Message-ID: <395B7C09.EB1EBB9A@bam.com>

Sorry I can't really answer your question, but are you using Report
Lab's pdflib?
If not you might want to look into it:

    http://www.reportlab.com

Mike Welch wrote:

> Hello, I'm a software developer trying to learn Python to generate
> PDF files from ASP pages. One problem I'm having with tables in my
> PDF files are that cells do not word wrap. This would not be a
> problem if I knew how to put a line break (or new line) in my text.
> In the docs it says \n works for new lines, but this does not work in
> tables. Can someone please tell me how this is accomplished in
> tables?
>
> Thanks!
>
> _________________________________________________
> Michael B. Welch
> Software Developer
> Incigna, Inc. <http://www.incigna.com/>
> 424 East 4th Street
> Cincinnati, Ohio 45202
> E-mail: mwelch at incigna.com
> Office: (513) 333-0313 Ext. 14
> Fax: (513) 333-0413
>
> "Your eSolutions Company"
>
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From dimus63 at my-deja.com  Thu Jun 22 19:55:15 2000
From: dimus63 at my-deja.com (dimus63 at my-deja.com)
Date: Thu, 22 Jun 2000 23:55:15 GMT
Subject: Using anydbm.
References: <m3n1l980k3.fsf@underlevel.underlevel.net> <63c9jscp2qeratnaa1r3rgivh36l92g186@4ax.com>
Message-ID: <8iu90t$urg$1@nnrp1.deja.com>

It is fixed already. Take whichdb.py from
http://www.musi-cal.com/~skip/python/Python/Lib/whichdb.py
Everything should work after that

In article <63c9jscp2qeratnaa1r3rgivh36l92g186 at 4ax.com>,
  Tim Roberts <timr at probo.com> wrote:
> Jordan Katz <webmaster at underlevel.net> wrote:
> >
> >I'm having trouble getting anydbm to work correctly.  For example,
> >
> >  >>> import anydbm
> >  >>> db = anydbm.open('foo', 'c')
> >  >>> db['bar'] = 'baz'
> >  >>> db['bar']
> >  'baz'
> >
> >Works fine, but if I try to execute it again after restarting python,
> >I get the error:
> >
> >  Traceback (innermost last):
> >    File "<stdin>", line 1, in ?
> >    File "/usr/lib/python1.5/anydbm.py", line 83, in open
> >      raise error, "db type could not be determined"
> >  anydbm.error: db type could not be determined
>
> Good question; I have exactly the same problem.  I replaced "anydbm"
with
> "dumbdbm" and it worked fine, but I expected "anydbm" to work.  If you
get
> an answer privately, please post it for me.
> --
> - Tim Roberts, timr at probo.com
>   Providenza & Boekelheide, Inc.
>


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


From none at none.com  Mon Jun 19 14:28:10 2000
From: none at none.com (none)
Date: Mon, 19 Jun 2000 20:28:10 +0200
Subject: bl
Message-ID: <6E528A6CBCFED311A7C1009027B721F92B89F9@hermes.demogr.mpg.de>





From PClaerhout at CREO.BE  Thu Jun 22 08:28:04 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 14:28:04 +0200
Subject: CurrentUser for remote machine under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301217@msgeuro1.creo.be>

> Hello,
> 
> how can you get the current logged in user for a remote nt system using Python?
> 
> Pieter



From robin at reportlab.com  Tue Jun 20 11:40:43 2000
From: robin at reportlab.com (Robin Becker)
Date: Tue, 20 Jun 2000 16:40:43 +0100
Subject: Roman Number Representation
Message-ID: <S9Q1YMA7B5T5Ew1U@jessikat.demon.co.uk>

I could have sworn I had seen a discussion of Roman Numbers somewhere on
clp, but can't find it. Anyone got any clues?
-- 
Robin Becker


From JoeSmith at bogusaddress.com  Sat Jun  3 16:18:24 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Sat, 03 Jun 2000 20:18:24 GMT
Subject: Python GUI requires tcl?
References: <8hbeol$obo$1@samba.rahul.net> <8hbjl0$611$1@news1.xs4all.nl>
Message-ID: <393968C8.CBFAAF1@bogusaddress.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000603/4f166d7b/attachment.html>

From bragib at my-deja.com  Mon Jun 19 21:01:48 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Tue, 20 Jun 2000 01:01:48 GMT
Subject: Help with telnet...
Message-ID: <8imfpo$bmj$1@nnrp1.deja.com>

This is a repost as I am still clueless!

Bragi


I am trying to telnet from a UNIX box to an NT to get the doskeys. The
NT that I login to runs a couple of .bat
 scripts which clear the screen set up some stuff etc.  When I try to do
a read_all() it fails the first time with the
 exception:
 Traceback (innermost last):
   File "telnet.py", line 17, in ?
     tn.read_all()
   File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
229, in read_all
     self.fill_rawq()
   File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
367, in fill_rawq
     buf = self.sock.recv(50)
 socket.error: (131, 'Connection reset by peer')


 That is why I put the first try in a try: except: clause.  The second
time I try to do a read_all() it works fine but I
 don't get all the keys I am missing some of the first keys.  I assume
that the first read_all() actually did read
 some stuff and then the second read got the rest.

 Can anyone see anything obviously wrong with this.


 Thanks, Bragi

 import sys
 import telnetlib
 HOST = 'shadow'
 USER = 'bragi'
 tn = telnetlib.Telnet()
 tn.open(HOST)
 tn.read_until("login: ")
 tn.write(USER+'\r')
 tn.read_until("password: ")
 tn.write('my_password\r')
 tn.write('doskey /macros\r')
 tn.write('exit\r')
 print 'done'
 try:
     all1 = tn.read_all()
 except:
     pass
 all = tn.read_all()
 print all



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


From kens at sightreader.com  Tue Jun 13 08:03:59 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 13 Jun 2000 05:03:59 -0700
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no>
Message-ID: <3946232F.61B28793@sightreader.com>



Thomas Weholt wrote:

> On Tue, 13 Jun 2000 16:39:23 +1000, "Jason" <it at onestopent.com.au>
> wrote:
>
> >> import string
> >> file = open ("file")
> >> lines = file.readlines()
> >> for line in lines:
> >>     pair = string.split(line,":")
> >>     print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
> >> file.close()
>
> The line :
>
> file = open("file")
>
> change "file" to the complete path of your text-file, like if your
> file resides in c:\temp\music.txt, change it to :
>
> file = open("c:\\temp\\music.txt")
>
> And YES, the double backslashes are needed, or the \t in \temp is
> interpreted as TAB. Just change that and run the script from PythonWin
> or IDLE. They both come with Python, or eh ... at least IDLE does.
>
> Thomas

Or put an r before the string to neutralize the backslashes:

  file = open(r"c:\temp\music.txt")

BTW, you can use forward slashes instead of backslashes if you want.

  file = open("c:/temp/music.txt")

This is a portability feature.  I'm pretty sure the converse, backslashes
in unix, works too.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000613/a4ff3b49/attachment.vcf>

From bgrigsds at khsdfklsf.com  Thu Jun 15 19:24:56 2000
From: bgrigsds at khsdfklsf.com (news-server.austin.rr.com)
Date: Thu, 15 Jun 2000 23:24:56 GMT
Subject: Python in home automation
References: <fsteele-E5789E.13533715062000@news.inet-systems.net>
Message-ID: <cBd25.38645$cH1.355549@typhoon.austin.rr.com>

Sounds like an opportunity to me.  A good start would be to develop some
low -level classes to control the more common devices.  That would simple
enough... A MisterHouse type project would require substantial effort
though.

"Frank Steele" <fsteele at mindspring.com> wrote in message
news:fsteele-E5789E.13533715062000 at news.inet-systems.net...
> I'm beginning to think about extending the home network to include some
> simple home automation/X10 functions. I've started looking around at
> products and software, and even went so far as to pick up the latest
> copy of <A HREF="http://www.tpj.com">The P*&l Journal</A>, which has a
> lengthy story on <A HREF="http://www.misterhouse.net">MisterHouse</A>.
>
> MisterHouse is a cross-platform system, all in P*&l, that allows voice
> and web control of X10 components, and uses text-to-speech to
> communicate alerts to the home occupants.
>
> Here's the problem/question -- if it's my system, I don't want it in
> P*&l. I would prefer to stick to the Python world. Unfortunately, in a
> quick sweep through the Vaults of Parnassus, The Daily-URL,
> Python.FAQts, and the O'Reilly Python dev-center, I don't see any
> modules, extensions, or projects on this.
>
> So, is anyone aware of Python modules or extensions to do X10 home
> automation? I'm interested hearing about them on _any_ platform, but I
> intend to implement on Linux and MacOS....
>
> Thanks,
>   Frank
>   fsteele at mindspring.com




From bjorn at roguewave.com  Wed Jun 28 20:57:34 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 28 Jun 2000 18:57:34 -0600
Subject: ?Does Python have sorting via "Dictionary Order"?
References: <395A92D0.65A205B1@email.sps.mot.com>
Message-ID: <395A9EFE.5422BA4F@roguewave.com>

Norman Shelley wrote:
> 
> Does Python have sorting via "Dictionary Order"?  This would be a very
> handy addition as an option to list's sort() function.
> 
> This is defined by D. Richard Hipp as:
> "The case of alphabetic characters is ignored, except to break ties.
> Thus "B"
> comes before "b" but after "a".  Also, integers embedded in the strings
> compare
> in numerical order.  In other words, "x10y" comes after "x9y", not
> before it as
> it would when using strcmp().

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> print [].sort.__doc__
L.sort([cmpfunc]) -- sort *IN PLACE*; if given, cmpfunc(x, y) -> -1, 0,
1
>>>

i.e. for a given list L, you can say "L.sort(f)" where f is a function
taking two arguments (a,b) and returns -1, 0, 1 if a<b, a==b, a>b
respectively. So all you have to do is implement f to compare two items
by Dictionary Order and you're done.

-- bjorn



From e.smith at cs.bham.ac.uk  Wed Jun 28 12:23:28 2000
From: e.smith at cs.bham.ac.uk (Elliot Smith)
Date: Wed, 28 Jun 2000 17:23:28 +0100
Subject: Tkinter widgets problem
Message-ID: <395A2680.3E33@cs.bham.ac.uk>

Hello all,

I've been having slight problems with Entry widgets in Tkinter, running
under sunos5. For some reason, when I double click on them, Python
closes the whole application and reports 'Bus error'. Any ideas? I've
tried overriding the binding for double mouse clicks to no avail.

Thanks in advance,
Elliot
-- 
email: e.smith at cs.bham.ac.uk
homepage: http://www.cs.bham.ac.uk/~ezs/
"A scholar is just a library's way of making another library" (Richard
Dawkins)


From MarkH at ActiveState.com  Mon Jun 19 20:08:12 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 20 Jun 2000 00:08:12 GMT
Subject: Python Programmer wanted. Sydney Australia
References: <8ik45k$rd0$1@nnrp1.deja.com> <dQq35.29$pa6.2496@news.clarkson.edu>
Message-ID: <MBy35.3132$qz4.24172@news-server.bigpond.net.au>

"Brad Clements" <bkc at Murkworks.com> wrote in message
news:dQq35.29$pa6.2496 at news.clarkson.edu...

> > considered. Come to the city of which Lonely Planet says ".you would
> > have to die and go to heaven before you see a more spectacular setting
> > for a city." Please send r?sum?s to jobs at memetrics.com.

> That sounds like something Ian Wright would say.

Its certainly not something that a Melbournite would say (even if it _is_
true :-)

Mark (in Melbourne :-)





From pehr at pehr.net  Tue Jun 27 23:38:25 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 28 Jun 2000 03:38:25 GMT
Subject: Need help: Python and Oracle on Linux x86
References: <B57EB243.44D%Dan@Grassi.org>
Message-ID: <39597329.CE225BD2@pehr.net>

Dear Dan, 

We got this to work here at Eink.  
The DCOracle database driver for python is
very good and does exacly what we need.

My understanding is that every version of 
Oracle has libraries that behave somewhat 
differently so sometimes you need to fix
little things like file order to make it all work.

Could you post your exact compliation trace so that
I and others can look it over?  
Also let us know which version of Oracle you are using.

Thanks,
	pehr anderson


Dan Grassi wrote:
> 
> I am trying to use Python on a Linux system but need Oracle DB access.  I
> have tried compiling the DCOracle interface with no success, there are
> compile errors.
> 
> I am trying to replace the Perl that is used with python but the DB access
> issue is a killer.  Can anyone help?
> 
> Thanks,
> 
> Dan


From rmartin at objectmentor.com  Sun Jun 18 21:07:21 2000
From: rmartin at objectmentor.com (Robert C. Martin)
Date: Sun, 18 Jun 2000 20:07:21 -0500
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net>
Message-ID: <skqskfte7f111@news.supernews.com>

Jerome Chan <eviltofu at rocketmail.com> wrote in message
news:eviltofu-A13457.22304614062000 at news.earthlink.net...
> Has anyone done any extreme programming in python? Care to share your
> experiences?
>
> <http://www.extremeprogramming.org/index.html>
   http://www.egroups.com/group/extremeprogramming/


We've been doing XP in Python for the last couple of months.  It's been
working very well.  We use a wiki to post user stories; and do iteration
plans on special wiki pages.  It all works like a charm.


--

Robert C. Martin    | "Uncle Bob"              | Training Courses:
Object Mentor Inc.  | rmartin at objectmentor.com | OOD, Patterns, C++, Java,
PO Box 85           | Tel: (800) 338-6716      | Extreme Programming.
Grayslake IL 60030  | Fax: (847) 548-6853      | http://www.objectmentor.com

"One of the great commandments of science is:
    'Mistrust arguments from authority.'" -- Carl Sagan






From shan_mu  Tue Jun 20 21:18:51 2000
From: shan_mu (Aeneas)
Date: Tue, 20 Jun 2000 21:18:51 -0400
Subject: Current date and time
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com> <20000620010356.A892@better.net>
Message-ID: <395017fb@newsprime.tidalwave.net>

How do you format the result as--
"Tuesday  20 June 2000  21:18hr" ?
Thanks.

"William Park" <parkw at better.net> wrote in message
news:20000620010356.A892 at better.net...
> On Tue, Jun 20, 2000 at 04:54:35AM +0000, dsavitsk wrote:
> > what is the easiest way to get the current date and or time?
> > something like
> > >>> print Now()
> > is what i am hoping to find.
> > thanks
> > doug
>
> >>> import time
> >>> time.ctime(time.time())
> 'Tue Jun 20 01:02:03 2000'
>
> --William
>




From 12klat at sightreader.com  Tue Jun  6 18:32:29 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 15:32:29 -0700
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and 
 others" is this possible?
References: <393692FB.8610EBCA@dallas.net> <007901bfcbf2$c3b4e680$1906a8c0@fc.fenx.com> <3936FEFB.BA248CA3@sightreader.com> <003301bfcc30$57851820$1906a8c0@fc.fenx.com>
Message-ID: <393D7BFD.F50F9932@sightreader.com>

I'm running PythonWin
I tried python from the command line.  The process hangs.

>>> import os
>>> os.popen('dir').readlines()
 (hangs . . .)

IDLE gives the same results as PythonWin:

>>> import os
>>> os.popen('dir').readlines()
Traceback (innermost last):
  File "<pyshell#1>", line 1, in ?
    os.popen('dir').readlines()
OSError: (0, 'Error')

Apparently you are right about Window pipes.  <MSJAB> Not surprising.
Microsoft would probably get rid of the command line completely if not
for its legacy value.  It's hard to control the world when the prolls
have access to a usable command shell. </MSJAB> :-)

-- Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

Emile van Sebille wrote:

>  Ken, Are you trying this in IDLE?  I get the same error there.If so,
> try starting python from within a dos window andsee if that makes a
> difference.  Pipes in winxx are somewhatflakey as I understand.
> Emile van Sebille
> emile at fenx.com
> -------------------
> ----- Original Message -----
> From: Ken Seehof
> To: Emile van SebilleCc: python-list at python.orgSent: Thursday, June
> 01, 2000 5:25 PMSubject: Re: Python to call commands, e.g. "find ,
> perl scripts , nmap , and others" is this possible?
>  On NT I get:
>
> >>> mylist = os.popen('dir').readlines()
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
> OSError: (0, 'Error')
>
> (same thing with any command)
>
> popen (command[, mode[, bufsize]])
>     Open a pipe to or from command. The return value is an open
>     file object connected to the pipe, which can be read or written
>     depending on whether mode is 'r' (default) or 'w'. The bufsize
>     argument has the same meaning as the corresponding argument
>     to the built-in open() function. The exit status of the command
>     (encoded in the format specified for wait()) is available as the
>     return value of the close() method of the file object, except that
>
>     when the exit status is zero (termination without errors), None is
>
>     returned. Availability: Unix, Windows.
>
> - Ken Seehof
>
> Emile van Sebille wrote:
>
>> Jim, Take a look at os.popen, as in: import osmylist =
>> os.popen('locate myname').readlines()for eachfile in mylist: do
>> something with eachfileHTH,
>> Emile van Sebille
>> emile at fenx.com
>> -------------------
>> ----- Original Message -----
>> From: Jim
>> To: python-list at python.orgSent: Thursday, June 01, 2000 9:44
>> AMSubject: Python to call commands, e.g. "find , perl scripts , nmap
>> , and others" is this possible?
>> Dear list,
>>
>> I'm new to Python, and I'm sorry if I'm asking about anything that
>> has been
>> already discussed.
>>
>> I would like to use Python to call commands, e.g. "find , perl
>> scripts  , nmap , and others" is this possible?
>>
>> I do not need GUI. All I need is to link the out put and work with
>> it. Is perl or shell scripts my only
>> answer?
>>
>> jim
>>
>> --
>> Proprietary & Confidential.  The information transmitted is intended only for the person or entity to whom it is addressed and may
>> contain confidential and/or privileged material.  Any review, retransmission, dissemination, or other use of, or taking of any
>> action in reliance upon, this information by persons or entities other than the intended recipient is prohibited.  If you received this in error, please
>> contact the sender and delete the material from any computer.
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/75764b0a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/75764b0a/attachment.vcf>

From root at rainerdeyke.com  Mon Jun 19 19:36:47 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 19 Jun 2000 23:36:47 GMT
Subject: {Q} OOP Animal class.
References: <8iluf0$le$1@nnrp1.deja.com>
Message-ID: <j8y35.1072$ul.67488@news1.rdc2.tx.home.com>

<johnvert at my-deja.com> wrote in message news:8iluf0$le$1 at nnrp1.deja.com...
> I want to make sure I understand the sequence of events right:
>
> 1. I call the method reply()
> 2. The interperter sees reply() does not exist in class Hacker
> 3. It goes one class up the hierarchy to class Primate
> 4. reply() doesn't exist in class Primate
> 5. It goes one class up the hierarchy to class Mammal
> 6. reply() doesn't exist in class Mammal either
> 7. It goes one class up the hierarchy to class Animal
> 8. reply() exists in class Animal, it calls it, and self.reply() gets
> translated to `Primate.reply()' since Hacker is of type Primate
> 9. Primate.reply() is called
> 10. `Hello world!' is printed

You missed a couple of steps.  Here's my interpretation of what happened.
Note that this is fairly specific to Python.  Most OO languages work
differently.

1. You try to access data.reply.
2. data.reply does not exist as an instance variable, so look for
Hacker.reply.
3. Hacker.reply does not exist, so look for Primate.reply.
4. Primate.reply does not exist, so look for Mammal.reply.
5. Mammal.reply does not exist, so look for Animal.reply.
6. Animal.reply exists, and is a function, so a bound method is created.
7. You call the bound method with no argument.
8. The call to to the bound method is translated into a call to Animal.reply
with the argument data.
9. Within Animal.reply, you try to access self.speak.
10. self.speak does not exist as an instance variable, so look for
Hacker.speak.
11. Hacker.speak exists, and is a function, so a bound method is created.
12. You call the bound method with no argument.
13. The call to the boundmethod is translated into a call to Hacker.speak
with argument self.
14. Within Hacker.speak, 'Hello World!' is printed.


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




From srn at fn.com.au  Sun Jun 11 19:39:07 2000
From: srn at fn.com.au (Stephen Robert Norris)
Date: Sun, 11 Jun 2000 23:39:07 GMT
Subject: time.clock() always returns 0?
References: <GEL05.4911$HD6.127120@iad-read.news.verio.net>
Message-ID: <vqV05.2397$cQ.20001@news-server.bigpond.net.au>

In article <GEL05.4911$HD6.127120 at iad-read.news.verio.net>, "Doug Fort"
<dougfort at downright.com> wrote:
> On my system (RedHat Linux 6.2 Python 1.5.2 as shipped with RedHat)
> time.clock() always returns 0.  time.time() returns a good value.  Is
> this known behavior?  Am I doing something wrong?

time.clock() returns the amount of CPU time consumed since startup or the
last time.clock() call, not anything to do with the current time.

>>> print time.clock.__doc__
clock() -> floating point number

Return the CPU time or real time since the start of the process or since
the first call to clock().  This has as much precision as the system
records.

However, on my RH6.2 machine, the first call gave me:

>>> import time 
>>> time.clock()
0.05

This is on a PIII 533, so unless your CPU is _really_ fast, you should get
some CPU usage!

	Stephen



From thomas at cintra.no  Wed Jun 28 05:01:00 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 28 Jun 2000 09:01:00 GMT
Subject: XML-code examples - mini-review of Python XML
References: <394f7bc2.365514882@news.online.no> <00062201025104.08285@quadra.teleo.net> <8jbteo$q3v$1@netnews.upenn.edu>
Message-ID: <395ebd37.180924415@news.online.no>

Hmmm .. I just ordered it. It should give me something to get started
with and I`ll take it from there. At least I`m giving something back
to a  writer of Python literature, by buying his book. Perhaps his
next project will be more usefull if he gets some support or feedback.


I`ll get back with some comments when I`ve read thru it.


On Wed, 28 Jun 2000 00:03:27 -0400, "Andrew Diller"
<dillera at isc.upenn.edu> wrote:

>His book is out there! We got ours last week.
>
>I like the book, but I also found it dissapointing. Most of it is an intro
>into the Python language itself. What's left was mostly about his Pyxie and
>PYX. I'm not sure why I'd want to learn PYX, when what I want is a solid
>book on manipulating XML with python.
>
>I didn't think I got that. I'm interested in learning about DOMs, SAX and
>things that I'm reading and hearing about. PYX seems like a bizzare
>off-shoot.
>
>As for the into to Python part of the book (about 3/4 of it) that was a good
>intro if you've never used python.
>
>-andy diller



From garry at sage.att.com  Thu Jun  1 10:40:03 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Thu, 1 Jun 2000 14:40:03 GMT
Subject: Using python (or Tcl) on the web
References: <MPG.1365adcfa8470eef9896aa@news> <BC4B61746187C623.3FA0E7D58DF4EA7E.ECC1D97F577DEDE5@lp.airnews.net> <slrn8g28ah.cc.amphgobb@eccie.blackflag.net> <L3iZ4.154057$681.2782248@news-east.usenetserver.com> <8h4eeu$bpg$1@saltmine.radix.net>
Message-ID: <393675C3.88BB0DBB@sage.att.com>

Cary O'Brien wrote:

> Python has got mod_python and pyapache.

does anyone have any insight about which of these to use?


-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From digitig at cix.co.uk  Thu Jun 15 15:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Jun 2000 20:28 +0100 (BST)
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com>
Message-ID: <memo.20000615202836.52185E@digitig.compulink.co.uk>

In article <8ib1bs$1pv at autodesk.autodesk.com>, akira.kiyomiya at autodesk.com 
(Akira Kiyomiya) wrote:

> I wonder if Python runs faster than Perl.
> 
> Could someone give me some comparison and contrast with some languages 
> such
> as Perl, Tcl, JavaScript, etc?
> 
> I am just curious.

Check out:

 http://wwwipd.ira.uka.de/%7Eprechelt/documents/jccpp_tr.pdf
 
Which somebody posted here recently. Where the best Python was shown to 
run faster than the best Perl, and the worst Python was shown to run a 
/lot/ faster than the worst Perl. Only on one task, admittedly, but one 
that Perl should have been good at...


From aahz at netcom.com  Fri Jun 16 15:23:45 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jun 2000 19:23:45 GMT
Subject: Lot's of one-time-pads available for $15 each! [was: Nth digit of PI]
References: <Pine.A32.3.90.1000616142212.16829D-100000@elvis.med.Virginia.EDU>
Message-ID: <8idus1$enb$1@slb3.atl.mindspring.net>

In article <Pine.A32.3.90.1000616142212.16829D-100000 at elvis.med.Virginia.EDU>,
Steven D. Majewski <sdm7g at virginia.edu> wrote:
>
>Only problems is that encryption based on a physical artifact is 
>vulnerable to physical snooping: someone dressed in black breaks into
>your house in the middle of the night and catalogs your CD collection 
>( or if you're not too smart, finds the one left sitting next to the
>computer! ) Books used to be used as one time pads, and they had the
>same vulnerability -- sometimes it was possible, by close inspection, 
>to tell where the book had been frequently left open. ( The Bible 
>was a popular choice, as there were reasons to make carrying one
>around constantly not be considered suspicious behaviour. ) 

Close, but no cigar.  What you're describing that's vulnerable is the
use of a book as an encryption key rather than a one-time pad.  If it
were literally a one-time pad, there'd be no particular physical marks.
Of course, I don't think the concept of a one-time pad existed as such
back then.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From emile at fenx.com  Mon Jun  5 19:38:36 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 5 Jun 2000 16:38:36 -0700
Subject: operator overloading
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com> <393C2386.307A160D@sightreader.com>
Message-ID: <0c2a01bfcf47$32b1bd40$1906a8c0@fc.fenx.com>

Ken,

You missed my point, while your point is taken.

My point, if I understood the question, was how could
one overload the less than operator.  In absence of an
example, I took that to mean how could one change
the behavior of the less than symbol.  Which needs
to happen in a class definition, which then allows
the class to decide what less than means.  So I 
provided an example that showed changed behavior
for less than.  ;-)  Of course, only the class
can know how two instances should compare, and it
should then return an appropriate value.

That the example should have fully shown that 
return values can be negative, zero, or positive,
indicating less than, equals, and greater than
respectively, is of course my fault.

What-if-a-six-turned-out-to-be-nine-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message ----- 
From: Ken Seehof 
Newsgroups: comp.lang.python
To: python-list at python.org 
Sent: Monday, June 05, 2000 3:02 PM
Subject: Re: operator overloading


Emile van Sebille wrote: 
  Here's a quick example of using __cmp__ 
  >>> class Test: 
   def __init__(self,a): 
    self.a = a 
   def __cmp__(self,other): 
    return self.a < other.a 

  >>> a = Test(3) 
  >>> b = Test(4) 
  >>> print a<b 
  0 
  >>> print a>b 
  1 

  HTH, 

  Emile van Sebille 
  emile at fenx.com 
  -------------------

Um.  No.  Not exactly.  Usually 1 designates "true" and 0 designates "false" unless I'm mistaken.  Also IMHO, > means "greater than" and < means "less than".  BTW there is yet another comparison operator == than means "equal to".  :-) 
The return value of __cmp__ should be -1, 0, 1 for <, ==, >, respectively. 

The following more closely conforms to these conventions: 

>>> class Test: 
...  def __init__(self,a): 
...   self.a = a 
...  def __cmp__(self,other): 
...   if self.a == other.a: 
...    return 0 
...   elif self.a < other.a: 
...    return -1 
...   else: 
...    return 1 
... 
>>> a = Test(1) 
>>> b = Test(2) 
>>> c = Test(2) 
>>> a<b, a>b, a==b, a!=b 
(1, 0, 0, 1) 
>>> b==c, b<c, b!=c 
(1, 0, 0) 

-- Ken Seehof 
kens at sightreader.com 

  ----- Original Message ----- 
  From: <kazan at kazan.fenx.com> 
  Newsgroups: comp.lang.python 
  To: <python-list at python.org> 
  Sent: Monday, June 05, 2000 1:01 PM 
  Subject: operator overloading 
  > Hello 
  > 
  > How can I overload the less than operator? 
  > learning python did mention something about __cmp__ but 
  did 
  > not provide any examples. 
  > 
  > 
  > Thanks in advance 
  > 
  > Erling 
  > --

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000605/789d3c55/attachment.html>

From sabren at manifestation.com  Sat Jun 17 18:09:13 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 17 Jun 2000 18:09:13 -0400 (EDT)
Subject: Why whitespace denotation of blocks doesn't work.
In-Reply-To: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <Pine.BSF.4.10.10006171806020.6221-100000@chi.pair.com>

On Sat, 17 Jun 2000, Thaddeus L. Olczyk wrote:

> Emacs also doesn't reindent correctly. So I have to go through the

Huh? Well, not if you use the tab key, I guess.. But "C-c >" and
"C-c <" seem to work just fine for me, using python-mode.

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------





From hanche at math.ntnu.no  Fri Jun  2 09:59:12 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 02 Jun 2000 09:59:12 -0400
Subject: Bad Math Module Notation re:  atan2
References: <010620001459577031%pecora@anvil.nrl.navy.mil>
Message-ID: <pcovgzstc9b.fsf@math.ntnu.no>

+ Lou Pecora <pecora at anvil.nrl.navy.mil>:

| Maybe you've all kicked this around before, but I've just picked up
| Python in the last month or so.  The notation for atan2(x,y) is 
| 
|  atan2(x,y)= arctangent( x/y)
| 
| Given the usual association of x with the abscissa and y with the
| ordinate in mathematics, science and engineering it is easy to mistake
| atan2(x,y) as arctangent (y/x) (x and y reversed!).  Or in another
| view, according to the majority of mathematically trained people
| atan2(x,y) as defined appears to deliver the arccotangent.  
| 
| Am I mising something?

Yes.  The documentation (meaning the library reference) is wrong, but
the implementation is right:

>>> math.atan2(0,1)
0.0
>>> math.atan2(1,0)
1.57079632679

In fact, even the function's doc string is right (Py 1.5.2):

>>> print math.atan2.__doc__
atan2(y, x)

Return atan(y/x).

you-should-listen-to-what-they-do-but-not-what-they-say-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From millja at wwc.edu  Tue Jun 13 17:06:18 2000
From: millja at wwc.edu (millja)
Date: Tue, 13 Jun 2000 14:06:18 -0700
Subject: ASP, Python, VBScript and sharing like friends
Message-ID: <0a9f5150.911771a5@usw-ex0109-068.remarq.com>

I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m having a bit of an issue.  My current situation is that
I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m able to use Python with ASP and I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m able to use
VBScript with ASP.  However, I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m still trying to figure out
how to use the two together.

What is occurring is that I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m working on a website, most of
which is already written with VBScript.  I have a rather
large function that I would like to integrate.  Is there
some way to call this function from within the VBScript?
Or, it would work if I can get numbers, generated by the
VBScript, into the Python function and the results of this
function back to the VBScript.

Thanks for you time.

jason milliron



* Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautiful


From lk at veriomail.com  Fri Jun  9 02:18:26 2000
From: lk at veriomail.com (Lawrence Kesteloot)
Date: Thu, 8 Jun 2000 23:18:26 -0700
Subject: = vs. == in Py3k
Message-ID: <F2005.859$Qf6.33080@nuq-read.news.verio.net>

If one of the goals of Python 3000 is to make it easy for new
programmers to learn, how about removing the distinction
between = and ==?  The only kinds of examples I can think of
where this is useful is something like:

    a = b = 5

vs.

    a = b == 5

But this isn't worth the trouble it gives new programmers.  The
syntax could be changed to make = and == equivalent tokens
and have the parser figure out which is meant from the context,
just as it's done in Basic.  The documentation would only ever
mention =, but the language would support == for backward
compatibility.

(One problem I can see with this proposal is that "a = b = 5" would
probably mean "set A to whether B is equal to 5 or not", and this
is the less useful of the two interpretations.  I guess the parser
could be written to use the other.  Or the whole construct could be
disallowed since it's not terrifically useful anyway.)

(Um, the other problem is named arguments to functions.  I'm not
sure how to get around that.  Any suggestions?  That might be enough
to justify not doing this...)

Lawrence





From neilh at scintilla.org  Wed Jun 28 20:20:07 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 29 Jun 2000 00:20:07 GMT
Subject: native Python text editor with Hebrew support?
References: <395A0B71.918A6673@nih.gov>
Message-ID: <XCw65.917$Tb7.5912@news-server.bigpond.net.au>

> I realize that this is getting a bit specialized, but I am looking for a
> text editor written in Python which can handle some rudimentary facets
> of Hebrew; namely, writing right-to-left and the Hebrew fonts (e.g., ISO
> 8859).
>
> Ah, yes, I need the source code so that I can make some changes, and it
> must run on Windoze.
>
> Is anything available?  I have looked through Parnassus, and realize
> that Scintilla and PythonWin might be candidates.

   While Scintilla/SciTE/PythonWin currently support Unicode and allow the
use of Hebrew characters they do not have bidi capability and so do not
layout Hebrew correctly. This is 'planned' for future implementation but so
far there has been little demand so it is quite a way off. Because there is
also a GTK+ version of Scintilla, more Unicode functionality will be worked
on when GTK+ 1.4 which includes the Pango international text module is
released. Then I can see how to approach these issues in a portable way.

   For performance, Python oriented editors are mostly written on two levels
with the basic editing widget written in C/C++ and Python used for higher
level control.

   On Windows, the rich text edit control can be used for international text
and I think this includes Hebrew. You should be able to use Richedit from
within PythonWin fairly easily. Before Scintilla, it used to be used for all
source editing within PythonWin.

   Neil





From sdm7g at virginia.edu  Fri Jun 16 14:46:18 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 16 Jun 2000 14:46:18 -0400 (EDT)
Subject: Lot's of one-time-pads available for $15 each! [was: Nth digit of PI]
In-Reply-To: <394a5109.80394991@10.1.1.28>
Message-ID: <Pine.A32.3.90.1000616142212.16829D-100000@elvis.med.Virginia.EDU>

On Fri, 16 Jun 2000, Tim Dixon wrote:

> The strongest possible encryption is a secret key one time pad.  A one
> time pad is an encryption method by which each byte (or couple bytes)
> is encrypted using a different key which is discarded and not used
> again. 

> 
> There are two problems:  one is that it's a secret key system and not
> a public key system, so it's not useful for many applications.  It
> doesn't work for something like SSL or PGP, for instance.  This is a
> characteristic of secret key systems.
> 
> The second is the enormous effort involved in producing one time pads.
> You need enough bits in your pad to encrypt the entire message;
> essentially for a 16K message, your key would have to be 16K long as
> well.  Naturally, you need to get the key to the recipient via some
> secure means and communicate what portion of the pad is used by a
> particular message.


Nowadays, Music CDs and DVDs are an easily available source of a stream
of bits. All you have to do is secretly agree upon a disk, and maybe
a starting track or location in the stream. ( There's also an infinite
set of algorithms to agree to run that stream thru. For example: 
break the stream into 3-bit chunks, take a group of 7 chunks and drop
all but the first and last, concatenate them all and rechunk them into
8-bit bytes, and XOR them with the data stream. That's not too complicated
for someone to memorize. ) 


Only problems is that encryption based on a physical artifact is 
vulnerable to physical snooping: someone dressed in black breaks into
your house in the middle of the night and catalogs your CD collection 
( or if you're not too smart, finds the one left sitting next to the
computer! ) Books used to be used as one time pads, and they had the
same vulnerability -- sometimes it was possible, by close inspection, 
to tell where the book had been frequently left open. ( The Bible 
was a popular choice, as there were reasons to make carrying one
around constantly not be considered suspicious behaviour. ) 


--- Steve Majewski  
    Chop wood, carry water.




From andre at beta.telenordia.se  Tue Jun 13 10:21:55 2000
From: andre at beta.telenordia.se (=?iso-8859-1?Q?Andr=E9_Dahlqvist?=)
Date: Tue, 13 Jun 2000 16:21:55 +0200
Subject: importing module through variable
In-Reply-To: <20000613160606.W20700@xs4all.nl>; from thomas@xs4all.net on Tue, Jun 13, 2000 at 04:06:06PM +0200
References: <8i5e9p$nm0$1@zingo.tninet.se> <20000613160606.W20700@xs4all.nl>
Message-ID: <20000613162155.A12328@beta.telenordia.se>

> You want the __import__ function:
> 
> module = __import__(var)

I misphrased the question. Inside my module I have a dictionary that I
need to access. And the name of this dictionary is stored in a
variable. If I do:

import the_module

How can I then access the dictionary inside this module, if the name of
the dictionary is stored in a variable? Or is there some equivalence to
"from the_module import the_dictionary" when you're using __import__?
-- 

// Andr?



From wjdandreta at worldnet.att.net  Thu Jun 29 11:39:46 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 29 Jun 2000 15:39:46 GMT
Subject: PyType_Type?
References: <Ebo45.13372$Uw3.865124@bgtnsc04-news.ops.worldnet.att.net> <slrn8l9p1v.elg.thomas@centurion.xs4all.nl> <qcp65.13064$kK.349754@bgtnsc05-news.ops.worldnet.att.net> <slrn8lmi2m.qb9.thomas@centurion.xs4all.nl>
Message-ID: <65K65.22276$pu6.502892@bgtnsc06-news.ops.worldnet.att.net>

Hi Thomas,

>Hm. I don't have the vaguest idea what 'overlays' are, nor how to compile
>for DOS, with or without DOS extenders ;-) But I can tell you where to find
>the symbols.


Overlays are used in DOS because of the 640K memory limit. You create a base
exe that has all the common stuff and break up the rest of the .c files into
smaller pieces called overlays that get swapped in and out of memory.

For example,  let's say the base exe takes 256K of ram, you allocate 128K
for the overlays, that leaves 256K for the heap.

This will be the first time I have tried to use overlays.

>This is the 'initstruct' function from the struct module. You don't seem to
>compile the struct module (it isn't in the list of compiled C files you
>included) so you might want to comment this out. Wether this'll result in a
>working Python is something else, I don't know what uses the struct module.
>

There are 2 similarly named modules structmodule.c and structmember.c. The
DOS name for these modules is struct~1. Since they are in different
directories, there shouldn't be a conflict but since all the .obj files are
in the same directory, one would over write the other. So compiling with
one, the other or both all caused errors. By naming the files differently
and including both, the related errors disappeared.


>The signal module. Same story as above, I guess, though I assume signals
>are entirely different under native DOS. Signals might live in an entirely
>different file, or they might be nonexistant. I do not know wether you need
>this module or not.
>


I had eliminated the signalmodule because it did not look like I needed it,
I needed to use compiler directives in config.c to eliminate the _initsignal
function.

It now all compiles and links without errors and I am ready to start on the
overlay part.

Thanks for the help.

Bill




From petersc at stud.ntnu.no  Fri Jun  2 21:22:10 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 03 Jun 2000 03:22:10 +0200
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <3937D345.CC0935D5@uab.edu>
Message-ID: <39385DC2.BC0B4853@stud.ntnu.no>

Shae Erisson wrote:
> 
> Martijn Faassen wrote:
> >
> ja, maar is het niet vanzelfsprekend dat alleen mensen die Nederlands
> spreken zich voor Python interesseren?

Maar alle mensen die Nederlands spreken horen Python liefhebben!
 
> or-at-least-lives-with-someone-who-speaks-dutch-ly y'rs

actually-being-a-german-living-near-the-dutch-border-and-so-
understanding-some-dutch-but-right-now-studying-in-norway-
and-having-a-belgium-dutch-speaking-friend-who-did-not-
help-me-with-this-one-so-excuse-any-errors-ly y'rs

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From michael at stroeder.com  Wed Jun 14 14:41:36 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 14 Jun 2000 20:41:36 +0200
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com> <8i86sf$e7c$0@216.39.141.36> <1VO15.607$i06.193222@ptah.visi.com>
Message-ID: <3947D1E0.8AFB0975@stroeder.com>

Grant Edwards wrote:
> 
> I doubt that anybody
> is going to install Python just to run this program.

That's the whole point. Download hits went up on my site when I
mentioned on the web site that the required software is present in
Linux distributions as pre-packaged RPMs etc.

Ciao, Michael.


From hjwidmai at foxboro.com  Thu Jun 15 06:24:52 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Thu, 15 Jun 2000 12:24:52 +0200
Subject: Entry widget with history/completion, again
Message-ID: <20000615122452.A4014@foxboro.com>

After digging in Tk docs for some time I've finally found how
to get the current index of the insertion character in an Entry
widget:

  self.index("insert")

will do the trick. Fair enough, "insert" isn't listed in the
INDICES section, just in a note (in the Text widget man page).
Or is "insert" a predifined mark?

Anyway, now that the biggest obstacle is out of the way, I can
actually begin implementing it.

-- 
Hans-Joachim Widmaier



From dana at mmi.oz.net  Wed Jun 14 16:00:11 2000
From: dana at mmi.oz.net (Dana Booth)
Date: 14 Jun 2000 20:00:11 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <85em60nozb.fsf@one.net.au>
Message-ID: <8i8o8b$83i$0@216.39.141.36>

Neurocrat <neurocrat at one.net.au> wrote:

>> i find i'm more enamored of elegance these days.

N: Me too.
N: Now if only I could find it in an OS ...

TRS-DOS or NEW-DOS? :)


-- 
-----
Dana Booth <dana at mmi.oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From tim at paradise.net.nz  Sat Jun 24 21:12:53 2000
From: tim at paradise.net.nz (Tim Evans)
Date: Sun, 25 Jun 2000 01:12:53 GMT
Subject: tkinter, grid and resize
References: <39550101.AD8C7956@uniserve.com>
Message-ID: <878zvuwokq.fsf@cassandra.evansnet>

Bob van der Poel <bvdpoel at uniserve.com> writes:

> I'm having a problem getting widgets managed by grid to resize
> themselves when the screen size changes. This is easy to do using the
> pack manager and 'expand', however 'expand' doesn't exist in grid and
> the docs seem to indicate that the 'news' sticky option should do the
> same. The following snippet creates a text widget with scrollbars. If
> you resize the main window, the widgets do not resize....
> 

You need to use the grid_rowconfigure and grid_columnconfigure
methods of the containing frame, specifically the 'weight=' option.

> from Tkinter import *
> 
> root=Tk()
> 
> ys = Scrollbar(root)
> xs = Scrollbar(root)
> 
> dsp = Text(root, yscrollcommand=ys.set, xscrollcommand=xs.set)
> xs.config(orient='hor', command=dsp.xview)
> ys.config(orient='vert', command=dsp.yview)
> xs.grid(row=1, column=1, sticky=E+W)
> ys.grid(row=0, column=0, sticky=N+S)
> dsp.grid(row=0, column=1, sticky=N+S+E+W)

root.grid_rowconfigure(0, weight=1)
root.grid_columnconfigure(1, weight=1)

> 
> root.mainloop()
> 
> What am I don't wrong? I've tried to insert columnspan stuff and played
> with the weight options--but I don't seem to be getting it.
> 
> Thanks.

See the Tcl/Tk documentation under 'grid' for more information,
including other useful options for the rowconfigure and
columnconfigure methods.

-- 
Tim Evans


From claird at starbase.neosoft.com  Wed Jun 28 17:32:43 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Jun 2000 16:32:43 -0500
Subject: Following a file, or cloning tail
References: <8jdj3t$or4$1@news1.xs4all.nl>
Message-ID: <56086CBD66F65B5B.564A00DC35DF4777.AC9C6557FC26BA58@lp.airnews.net>

In article <8jdj3t$or4$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
>I'm trying to write an application that follows the output of another
>app, analyzes it, and serves it up as html. Kind of filtering tail,
>really. Is there any accepted 'best' way of 'tailing' a file?
>
>I thought about having two threads, one doing the tailing, one
>doing the servering (instantiating new threads as needed), but I
>can't even get the tailing right!
			.
			.
			.
This is the platform-neutral way to do it:  open the file
for reading.  Read to the end.  Idle (whether by an event-
or thread-oriented mechanism).  Poll for new content, as
explained below.  Read to end (possibly with some seek()ing
and find()ing, depending on details of your algorithm).
Idle.  Repeat.

Some observers gag at the thought of polling.  That's part
of the robustness of our contemporary file-systems, though:
the reliable way to find out what's in them is to ask.

How do you check for new content?  Among the possibilities
are
1.  Read to end.  You're polling anyway (presumably
    at a frequency of a few hertz), so there's no
    particular harm in this.
2.  Check the size of the file for a change.

Do you want code, also?  I assume that, if I'm only available
for one of them today, you'd prefer the *idea*.
-- 

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


From jblazi at netsurf.de  Fri Jun 16 15:42:10 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Fri, 16 Jun 2000 21:42:10 +0200
Subject: initializing a class
References: <3948a566_5@goliath.newsfeeds.com> <8iajn1$aib$1@slb6.atl.mindspring.net>
Message-ID: <394a8298_1@goliath.newsfeeds.com>

> Well, I'm not going to address the issue of "return z(y.x)", because
> that gets into metaclass issues I'm not comfortable with.  However, I'm
> pretty sure you meant to type
>
>   self.t = a.t1 + a.t2
>

Yes, of course you are right.

J.B.




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


From andres at mail.mamey.com  Mon Jun 26 21:59:51 2000
From: andres at mail.mamey.com (Andres Corrada-Emmanuel)
Date: Mon, 26 Jun 2000 21:59:51 -0400
Subject: Controlling Acrobat Reader with COM
Message-ID: <200006262159.AA3991339306@mail.mamey.com>

Hi,

I'm trying to develop an application that will print (as in a paper copy) a PDF file automatically on Win32. I've used Mark Hammond's genpy utility to create a wrapper around "Acrobat Control for ActiveX". I'm able to get a handle on Acrobat but cannot execute any of its registered methods:

>>> from win32com.client import Dispatch
>>> acro = Dispatch('PDF.PdfCtrl.1')
>>> acro.LoadFile(fileName='C:\Temp\ReadMe.pdf')
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "C:\Program Files\Python\win32com\gen_py\CA8A9783-280D-11CF-A24D-444553540000x0x1x3.py", line 30, in LoadFile
    return self._oleobj_.InvokeTypes(0x2, LCID, 1, (11, 0), ((8, 0),),fileName)
com_error: (-2147418113, 'Unexpected failure', None, None)

I'm very much a COM newbie, so this traceback means very little to me. Could someone enlighten me on what is going wrong in this case? I have no problem controlling Word with my setup so I'm stumped on this one.

Andres Corrada



From redaelli at inc.it  Sun Jun  4 11:20:45 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 17:20:45 +0200
Subject: GTK broke things! Re: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home>
				<39397FD4.67C7FD73@inc.it> <867lc64a47.fsf@g.local> <393A2B6D.4B33A5C8@inc.it> <393A3902.F67CD1F8@stud.ntnu.no>
Message-ID: <393A73CD.4EAAB457@inc.it>

I was WRONG!
GTK broke things in a way I can't even wonder HOW....
I attack here two version. One functions and the other not. Little
problem. The ONLY difference detween them is a "import gtk" at the
beginning. Delete this line and it functions!!!!
I wonder WHY? It should create a "namespace" gtk.something

=================
[paolo at aldebaran Telaio]$ python pyplot.working.py 
Processing ['LCASE.001']
Case 1
Trying to parse '1' '0.000000E+00' '0.000000E+00'
Look for strange characthers [48, 46, 48, 48, 48, 48, 48, 48, 69, 43,
48, 48]
Parsed to '1' '0.0' '0.0'
Trying to parse '2' '0.000000E+00' '0.300000E+03'
Look for strange characthers [48, 46, 48, 48, 48, 48, 48, 48, 69, 43,
48, 48]
Parsed to '2' '0.0' '300.0'
Trying to parse '3' '0.500000E+03' '0.300000E+03'
Look for strange characthers [48, 46, 53, 48, 48, 48, 48, 48, 69, 43,
48, 51]
Parsed to '3' '500.0' '300.0'
Trying to parse '4' '0.500000E+03' '0.000000E+00'
Look for strange characthers [48, 46, 53, 48, 48, 48, 48, 48, 69, 43,
48, 51]
Parsed to '4' '500.0' '0.0'
Traceback (innermost last):
  File "pyplot.working.py", line 85, in ?
    c = CanvasExample()
NameError: CanvasExample
[paolo at aldebaran Telaio]$ python pyplot.working.py 
Processing ['LCASE.001']
Case 1
Trying to parse '1' '0.000000E+00' '0.000000E+00'
Look for strange characthers [48, 46, 48, 48, 48, 48, 48, 48, 69, 43,
48, 48]
Parsed to '1' '0.0' '0.0'
Trying to parse '2' '0.000000E+00' '0.300000E+03'
Look for strange characthers [48, 46, 48, 48, 48, 48, 48, 48, 69, 43,
48, 48]
Parsed to '2' '0.0' '300.0'
Trying to parse '3' '0.500000E+03' '0.300000E+03'
Look for strange characthers [48, 46, 53, 48, 48, 48, 48, 48, 69, 43,
48, 51]
Parsed to '3' '500.0' '300.0'
Trying to parse '4' '0.500000E+03' '0.000000E+00'
Look for strange characthers [48, 46, 53, 48, 48, 48, 48, 48, 69, 43,
48, 51]
Parsed to '4' '500.0' '0.0'
[paolo at aldebaran Telaio]$ python pyplot.py         
Processing ['LCASE.001']
Case 1
Trying to parse '1' '0.000000E+00' '0.000000E+00'
Look for strange characthers [48, 46, 48, 48, 48, 48, 48, 48, 69, 43,
48, 48]
Traceback (innermost last):
  File "pyplot.py", line 86, in ?
    t = Telaio()
  File "pyplot.py", line 78, in __init__
    x = string.atof(x)
ValueError: invalid literal for atof(): 0.000000E+00
[paolo at aldebaran Telaio]$ diff pyplot.working.py pyplot.py
10a11
> import gtk
80a82
>             ## End reading rods
[paolo at aldebaran Telaio]$ 
=============================
-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/
-------------- next part --------------
 1
 4 3
      1  0.000000E+00  0.000000E+00
      2  0.000000E+00  0.300000E+03
      3  0.500000E+03  0.300000E+03
      4  0.500000E+03  0.000000E+00
      1     1     2
      2     2     3
      3     3     4
      1  0.300000E+03  0.000000E+00  0.100000E+01
  0.000000E+00 0.000000E+00 0.000000E+00 0.250000E+04 -.146922E+03 0.652654E+04
  0.150000E+02 0.478073E-01 -.395235E-04 0.250000E+04 -.146922E+03 0.668798E+04
  0.300000E+02 0.956146E-01 -.174419E-03 0.250000E+04 -.146922E+03 0.720229E+04
  0.450000E+02 0.143422E+00 -.429172E-03 0.250000E+04 -.146922E+03 0.806948E+04
  0.600000E+02 0.191229E+00 -.828271E-03 0.250000E+04 -.146922E+03 0.928954E+04
  0.750000E+02 0.239036E+00 -.139620E-02 0.250000E+04 -.146922E+03 0.108625E+05
  0.900000E+02 0.286844E+00 -.215745E-02 0.250000E+04 -.146922E+03 0.127883E+05
  0.105000E+03 0.334651E+00 -.313651E-02 0.250000E+04 -.146922E+03 0.150670E+05
  0.120000E+03 0.382458E+00 -.435786E-02 0.250000E+04 -.146922E+03 0.176985E+05
  0.135000E+03 0.430266E+00 -.584599E-02 0.250000E+04 -.146922E+03 0.206830E+05
  0.150000E+03 0.478073E+00 -.762539E-02 0.250000E+04 -.146922E+03 0.240203E+05
  0.165000E+03 0.525880E+00 -.972055E-02 0.250000E+04 -.146922E+03 0.277105E+05
  0.180000E+03 0.573687E+00 -.121559E-01 0.250000E+04 -.146922E+03 0.317535E+05
  0.195000E+03 0.621495E+00 -.149561E-01 0.250000E+04 -.146922E+03 0.361495E+05
  0.210000E+03 0.669302E+00 -.181454E-01 0.250000E+04 -.146922E+03 0.408983E+05
  0.225000E+03 0.717109E+00 -.217484E-01 0.250000E+04 -.146922E+03 0.460000E+05
  0.240000E+03 0.764916E+00 -.257897E-01 0.250000E+04 -.146922E+03 0.514546E+05
  0.255000E+03 0.812724E+00 -.302936E-01 0.250000E+04 -.146922E+03 0.572620E+05
  0.270000E+03 0.860531E+00 -.352846E-01 0.250000E+04 -.146922E+03 0.634223E+05
  0.285000E+03 0.908338E+00 -.407874E-01 0.250000E+04 -.146922E+03 0.699355E+05
  0.300000E+03 0.956146E+00 -.468262E-01 0.250000E+04 -.146922E+03 0.768016E+05
      2  0.500000E+03  0.100000E+01  0.000000E+00
  0.000000E+00 -.956146E+00 0.105359E+06 0.000000E+00 -.810241E-12 0.118531E+15
  0.250000E+02 -.956146E+00 0.354593E+09 0.000000E+00 -.250000E+03 0.118531E+15
  0.500000E+02 -.956146E+00 0.553234E+09 0.000000E+00 -.500000E+03 0.118531E+15
  0.750000E+02 -.956146E+00 0.615784E+09 0.000000E+00 -.750000E+03 0.118531E+15
  0.100000E+03 -.956146E+00 0.561998E+09 0.000000E+00 -.100000E+04 0.118531E+15
  0.125000E+03 -.956146E+00 0.411631E+09 0.000000E+00 -.125000E+04 0.118531E+15
  0.150000E+03 -.956146E+00 0.184438E+09 0.000000E+00 -.150000E+04 0.118531E+15
  0.175000E+03 -.956146E+00 -.998256E+08 0.000000E+00 -.175000E+04 0.118531E+15
  0.200000E+03 -.956146E+00 -.421405E+09 0.000000E+00 -.200000E+04 0.118531E+15
  0.225000E+03 -.956146E+00 -.760544E+09 0.000000E+00 -.225000E+04 0.118531E+15
  0.250000E+03 -.956146E+00 -.109749E+10 0.000000E+00 -.250000E+04 0.118531E+15
  0.275000E+03 -.956146E+00 -.141248E+10 0.000000E+00 -.275000E+04 0.118531E+15
  0.300000E+03 -.956146E+00 -.168578E+10 0.000000E+00 -.300000E+04 0.118531E+15
  0.325000E+03 -.956146E+00 -.189761E+10 0.000000E+00 -.325000E+04 0.118531E+15
  0.350000E+03 -.956146E+00 -.202822E+10 0.000000E+00 -.350000E+04 0.118531E+15
  0.375000E+03 -.956146E+00 -.205787E+10 0.000000E+00 -.375000E+04 0.118531E+15
  0.400000E+03 -.956146E+00 -.196678E+10 0.000000E+00 -.400000E+04 0.118531E+15
  0.425000E+03 -.956146E+00 -.173522E+10 0.000000E+00 -.425000E+04 0.118531E+15
  0.450000E+03 -.956146E+00 -.134343E+10 0.000000E+00 -.450000E+04 0.118531E+15
  0.475000E+03 -.956146E+00 -.771639E+09 0.000000E+00 -.475000E+04 0.118531E+15
  0.500000E+03 -.956146E+00 -.105359E+06 0.000000E+00 -.500000E+04 0.118531E+15
      3  0.300000E+03  0.000000E+00  -.100000E+01
  0.000000E+00 0.000000E+00 -.129038E+07 0.000000E+00 0.000000E+00 0.000000E+00
  0.150000E+02 0.000000E+00 -.156394E+10 0.000000E+00 0.000000E+00 0.161438E+03
  0.300000E+02 0.000000E+00 -.243946E+10 0.000000E+00 0.000000E+00 0.675750E+03
  0.450000E+02 0.000000E+00 -.271495E+10 0.000000E+00 0.000000E+00 0.154294E+04
  0.600000E+02 0.000000E+00 -.247753E+10 0.000000E+00 0.000000E+00 0.276300E+04
  0.750000E+02 0.000000E+00 -.181429E+10 0.000000E+00 0.000000E+00 0.433594E+04
  0.900000E+02 0.000000E+00 -.812342E+09 0.000000E+00 0.000000E+00 0.626175E+04
  0.105000E+03 0.000000E+00 0.441215E+09 0.000000E+00 0.000000E+00 0.854044E+04
  0.120000E+03 0.000000E+00 0.185928E+10 0.000000E+00 0.000000E+00 0.111720E+05
  0.135000E+03 0.000000E+00 0.335473E+10 0.000000E+00 0.000000E+00 0.141564E+05
  0.150000E+03 0.000000E+00 0.484049E+10 0.000000E+00 0.000000E+00 0.174938E+05
  0.165000E+03 0.000000E+00 0.622943E+10 0.000000E+00 0.000000E+00 0.211839E+05
  0.180000E+03 0.000000E+00 0.743446E+10 0.000000E+00 0.000000E+00 0.252270E+05
  0.195000E+03 0.000000E+00 0.836847E+10 0.000000E+00 0.000000E+00 0.296229E+05
  0.210000E+03 0.000000E+00 0.894436E+10 0.000000E+00 0.000000E+00 0.343718E+05
  0.225000E+03 0.000000E+00 0.907502E+10 0.000000E+00 0.000000E+00 0.394735E+05
  0.240000E+03 0.000000E+00 0.867335E+10 0.000000E+00 0.000000E+00 0.449280E+05
  0.255000E+03 0.000000E+00 0.765225E+10 0.000000E+00 0.000000E+00 0.507355E+05
  0.270000E+03 0.000000E+00 0.592461E+10 0.000000E+00 0.000000E+00 0.568958E+05
  0.285000E+03 0.000000E+00 0.340332E+10 0.000000E+00 0.000000E+00 0.634090E+05
  0.300000E+03 0.000000E+00 0.129038E+07 0.000000E+00 0.000000E+00 0.702750E+05
-------------- next part --------------
#!/usr/bin/env python

## pyplot
## program to plot'n'show the results of "telaio", a fortran program
## 
# Author: Jesper Skov <jskov at cygnus.co.uk>
# A rewite of the C canvas example in the GNOME Developer's Information

## Importing GUI specific stuffs
import GDK
import gtk

import glob # to look for all LCASE.* files
import string ## to process informations
from UserList import UserList ## to create lists that starts counting from 1

class Node:
    def __init__(self,number,x,y):
        self.number = number
        self.x = x
        self.y = y
        
class Rod:
    def __init__(self, number, node1, node2):
        self.number = number
        self.node1 = node1
        self.node2 = node2
        self.sections = range(0,21) ## list of sections. Each section will be a list containg [position, u displacement, v displacement, n, t, m]. From index 0 to index 20
        
## Lista is a list with preallocated length that starts counting from 1. This is useful to avoid unnecessary mess within file processing expecially regarding the rods and nodes lists
class Lista(UserList):
    def __init__(self):
        UserList.__init__(self)
        self.data = []
    def __setattr__(self, attr, value):
        if attr == "size" and value>0:
            self.data = [None]*value
        else:
            self.__dict__[attr] = value
    def __getitem__(self,index):
        if index > 0:
            return self.data[index-1]
        else:
            raise IndexError
    def __setitem__(self,index,value):
        self.data[index-1] = value
    
class LoadCase:
    def __init__(self):
        self.nodes   = Lista() ## the list of nodes
        self.rods    = Lista() ## the list of rods

class Telaio:
    def __init__(self):
        ## Loading all the load cases 
        ## Looking for all LCASE.xxx files
        self.filelist = glob.glob("LCASE*")
        self.load_cases = [None] * len(self.filelist) 
        
        for filename in self.filelist:
            print "Processing %s" % [filename]
            file = open(filename)
            case_n = int( string.split(filename,".")[1] )
            print "Case",case_n
            ## Reading the trailing 1
            file.readline()
            ## read nodes and rods numbers
            number_list = string.split( file.readline() )
            numbers = map(int, number_list)
            nodes_n = numbers[0]
            rods_n  = numbers[1]
            for node_i in range(0,nodes_n):
                [node, x, y] = string.split( file.readline() )
                node = int(node);
                print "Trying to parse '%s' '%s' '%s'" % (node,x,y)
                #print "Looking for strange characters:",map(ord,x)
                print "Look for strange characthers",map(ord,x)
                x = string.atof(x)
                y = float(y) 
                print "Parsed to '%s' '%s' '%s'" % (node,x,y)
                ## Note: node_i is NOT used!
            ## End reading rods
        ## End reading files.
        
if __name__ == '__main__':
    t = Telaio()


-------------- next part --------------
#!/usr/bin/env python

## pyplot
## program to plot'n'show the results of "telaio", a fortran program
## 
# Author: Jesper Skov <jskov at cygnus.co.uk>
# A rewite of the C canvas example in the GNOME Developer's Information

## Importing GUI specific stuffs
import GDK

import glob # to look for all LCASE.* files
import string ## to process informations
from UserList import UserList ## to create lists that starts counting from 1

class Node:
    def __init__(self,number,x,y):
        self.number = number
        self.x = x
        self.y = y
        
class Rod:
    def __init__(self, number, node1, node2):
        self.number = number
        self.node1 = node1
        self.node2 = node2
        self.sections = range(0,21) ## list of sections. Each section will be a list containg [position, u displacement, v displacement, n, t, m]. From index 0 to index 20
        
## Lista is a list with preallocated length that starts counting from 1. This is useful to avoid unnecessary mess within file processing expecially regarding the rods and nodes lists
class Lista(UserList):
    def __init__(self):
        UserList.__init__(self)
        self.data = []
    def __setattr__(self, attr, value):
        if attr == "size" and value>0:
            self.data = [None]*value
        else:
            self.__dict__[attr] = value
    def __getitem__(self,index):
        if index > 0:
            return self.data[index-1]
        else:
            raise IndexError
    def __setitem__(self,index,value):
        self.data[index-1] = value
    
class LoadCase:
    def __init__(self):
        self.nodes   = Lista() ## the list of nodes
        self.rods    = Lista() ## the list of rods

class Telaio:
    def __init__(self):
        ## Loading all the load cases 
        ## Looking for all LCASE.xxx files
        self.filelist = glob.glob("LCASE*")
        self.load_cases = [None] * len(self.filelist) 
        
        for filename in self.filelist:
            print "Processing %s" % [filename]
            file = open(filename)
            case_n = int( string.split(filename,".")[1] )
            print "Case",case_n
            ## Reading the trailing 1
            file.readline()
            ## read nodes and rods numbers
            number_list = string.split( file.readline() )
            numbers = map(int, number_list)
            nodes_n = numbers[0]
            rods_n  = numbers[1]
            for node_i in range(0,nodes_n):
                [node, x, y] = string.split( file.readline() )
                node = int(node);
                print "Trying to parse '%s' '%s' '%s'" % (node,x,y)
                #print "Looking for strange characters:",map(ord,x)
                print "Look for strange characthers",map(ord,x)
                x = string.atof(x)
                y = float(y) 
                print "Parsed to '%s' '%s' '%s'" % (node,x,y)
                ## Note: node_i is NOT used!
        ## End reading files.
        
if __name__ == '__main__':
    t = Telaio()



From bjorn at roguewave.com  Thu Jun  8 17:42:37 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 08 Jun 2000 15:42:37 -0600
Subject: Obscure Exception message
References: <75F7304BB41CD411B06600A0C98414FC0A418B@ORSMSX54>
Message-ID: <3940134D.22284249@roguewave.com>

"Daley, MarkX" wrote:
> 
> Hi, all!
> 
> I'm still trying to catch up on my Python reading (>50 messages!), but I
> have a problem with a while loop I'm using.  Hope someone can help.
> 
> alphadict = {}
> for i in range(len(result)):
>         c = 1
>         serialnum = result[i][6]
>         try:
>                 while serialnum in alphadict[c]:
>                         c = c + 1
>                 alphadict[c] = alphadict[c] + [serialnum]
>         except KeyError:
>                 alphadict[c] = serialnum
> print len(alphadict)
> 
> The print len(alphadict) is there to let me know if it has done the job or
> not.  Basically, I anticipate the length being >1.  Here is the exception
> message.
> 
> Traceback (innermost last):
>   File "C:\Program Files\Python\Pythonwin\pywin\framework\scriptutils.py",
> line 237, in RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\Program Files\Python\cit.py", line 63, in ?
>     while `serialnum` in alphadict[c]:
> TypeError: string member test needs char left operand
> 
> I have no idea what this is telling me, although the 'char' comment makes me
> wish I had a flamethrower!

It seems to be telling you that serialnum (or is it really `serialnum`
as in the errormessage) is not a single character. I.e. you can't do
something like "if 'bar' in 'foo bar baz'" (you'll need string.index for
that).

hth,
-- bjorn



From M.Vitecek at sh.cvut.cz  Thu Jun 15 13:50:14 2000
From: M.Vitecek at sh.cvut.cz (Michal Vitecek)
Date: Thu, 15 Jun 2000 19:50:14 +0200
Subject: Python Productivity over C++
In-Reply-To: <39491494.671DC3D9@sightreader.com>; from kens@sightreader.com on Thu, Jun 15, 2000 at 10:38:28AM -0700
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com>
Message-ID: <20000615195014.A12612@fuf.sh.cvut.cz>

Ken Seehof <kens at sightreader.com> wrote:
>Encapsulation, access control and type strictness are often erroneously
>referred to as if they were features.  They aren't.  They are aspects of a
>design philosophy.  To make a system less error prone, constrain the programmer
>in order to prevent the programmer from writing bugs.  Of course, that's
>impossible, since it is always possible to write bugs anyway.  Because C/C++
>are extremely error-prone languages, they need stuff like encapsulation,
>access, control and type strictness.  Because python is less error prone, it
>doesn't need these "features".  Python is based on different philosophy: To
>make a system less error prone, make it simple.
>
>You have to ask, "what is the actual value of these features?"  The answer is
>that they prevent certain kinds of bugs from appearing.  Now ask, "what bugs
>would a C++ programmer encounter if encapsulation, access, control and type
>strictness were removed from C++."  Undoubtedly it would be a mess.  Get a
>picture in your mind of these bugs.  Now ask a python programmer how often
>these kinds of bugs come up.  Answer: "can't remember that ever happenning!"
>
>Look at the big picture.  Do encapsulation, access control and type strictness
>really make C++ really less error prone than python?
>
>I think not.

 but what are the method names with '_' at the beginning in modules for?
 aren't they there as a kind of hack to implement some way of access
 control?


-- 
			fuf


------------------------------ na IRC -------------------------------------
 BillGates [bgates at www.microsoft.com] has joined #LINUX
 ...
 mode/#linux [+b BillGates!*@*] by DoDad
 BillGates was kicked off #linux by DoDad (banned: We see enough of Bill
          Gates already.)
 




From jin.choung at gte.net  Thu Jun 15 01:30:04 2000
From: jin.choung at gte.net (jin choung)
Date: Thu, 15 Jun 2000 05:30:04 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <8i9cdf$t1c$1@ionews.ionet.net>
Message-ID: <wRZ15.3853$Gp6.258826@dfiatx1-snr1.gtei.net>

thanks for the links.  will follow up on them.  happy hunting to you as
well!

jin




From ge at nowhere.none  Mon Jun 26 11:08:50 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 26 Jun 2000 15:08:50 GMT
Subject: smtplib
References: <001501bfdf74$6dff1fe0$14b745c6@timaratz>
Message-ID: <6mK55.1713$iN5.345307@ptah.visi.com>

In article <001501bfdf74$6dff1fe0$14b745c6 at timaratz>, Peter Timaratz wrote:

>Yes, I figured out that smtplib requires the subject and
>headers to be imbedded in the message. I'm new to Python and I
>love the language. But I think that the design of this module
>reflects poorly upon Python.

The module impliments SMTP, hence the name.  SMTP does not
concern itself with message contents.  Writing smtplib to make
certain assumptions or force certain conventions regarding
message content would be just plain incorrect.  

>Almost every CGI programmer who looks at Python will want to
>use this module. It would only take a few lines of code to make
>it much more usable.

Maybe you should write a higher level e-mail module that does
what you are requesting.  It does not belong in smtplib.

>I'm saying this not to flame Python, but to encourage those
>responsible for it's distribution to look more critically at
>what they are putting out.

Perhaps you should look more critically at what you expect from
a library that claims only to impliment SMTP.  If you want a
layer of functionality on _top_ of SMTP, then that's something
another module can impliment.

>I hope that this module is atypical of the quality of this
>product. I'd like to see Python become more popular. There are
>many factors involved in that, but the quality of the modules
>is certainly one of them.

There is nothing wrong with smtplib.  You seem to be expecting
it to do things other than SMTP.

-- 
Grant Edwards                   grante             Yow!  I'm an East Side
                                  at               TYPE...
                               visi.com            


From jbower at lims.com  Wed Jun 14 07:41:31 2000
From: jbower at lims.com (Jim Bower)
Date: Wed, 14 Jun 2000 11:41:31 GMT
Subject: Python and Windows Scripting Host
References: <9Ov15.1411$vM.56343@typhoon1.ba-dsg.net> <39y15.839$Zh1.63369@news.easynews.com>
Message-ID: <LbK15.3205$vM.96895@typhoon1.ba-dsg.net>

Thanks, Roger. There is a test script
(D:\Python\win32comext\axscript\test\test.pys) which came from the Win32All
distribution. It also fails, and also contains wscript directly as a Python
variable. Did I miss something in the documentation?

Thanks,
"Jim Bower" <jbower at lims.com>

"Roger Upole" <rupole at compaq.net> wrote in message
news:39y15.839$Zh1.63369 at news.easynews.com...
> The Wscript object is not referenced directly as a Python variable.
> You create it via win32com.client.Dispatch:
>     import win32com.client
>     wsh=win32com.client.Dispatch('Wscript.Shell')
>     wnt =win32com.client.Dispatch('Wscript.Network')
>     ...
> Then you can use the their functions as
>     wsh.somefunction()
>     wnt.somefunction()
>     ...
> From Pythonwin, if you first run the makepy utility from the tools menu
> and select Windows Scripting Host, autocomplete will magically show
> you what functions are available.
>        HTH
>                 Roger Upole
>
> "Jim Bower" <jbower at lims.com> wrote in message
> news:9Ov15.1411$vM.56343 at typhoon1.ba-dsg.net...
> > Just getting started with Python. Downloaded the basic distribution and
> the
> > Win32All distribution. Python appears to register OK (ran pyscript.py
> > separately just to verify).
> >
> > However, I am getting "NameError: wscript" anywhere the wscript object
is
> > mentioned. Have I missed something in the install?
> >
> > Jim Bower jbower at lims.com
> >
> >
> >
>
>




From djc at itga.com.au  Tue Jun  6 20:17:42 2000
From: djc at itga.com.au (Dave Cole)
Date: 07 Jun 2000 10:17:42 +1000
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net>
Message-ID: <87em6as5sp.fsf@heresy.itga.com.au>

>>>>> "Drew" == Drew Fisher <drew at level3.nospam.net> writes:

Drew> Hello.  I am still learning Python, but I came across a need for
Drew> generating dynamic graphs (network load average, uptime, boring
Drew> stuff like this :).

Drew> Is the a module in Python that allows this?  If so, could
Drew> someone point me to it?

I started to write a module that would be used for the basis of a nice
graphing module.  It uses libart, freetype, and libpng.  You can see
where I even started documenting it here:

        http://www.object-craft.com.au/paint_doc.html

It is not complete, but it does work.  The source code is here:

        http://www.object-craft.com.au/paint.tar.gz

Object Craft is a very small consulting business that my friend and I
have just started.  I start fulltime next week.  We are looking for
work (hint).

- Dave



From ehagemann at home.com  Sun Jun  4 11:56:17 2000
From: ehagemann at home.com (Eric Hagemann)
Date: Sun, 04 Jun 2000 15:56:17 GMT
Subject: win32com and Excel
References: <3939D364.2EF22B13@zebrasoft.co.nz>
Message-ID: <B_u_4.111323$E85.2192336@news1.rdc1.md.home.com>

Matthew,
    Read Hammonds "Programming Python in Win32".  There is a very good
headstart there with an example of a class that eases the interaction
w/Excel via COM.  Next get "Writing Excel Macros" by O'Reilly.  It is all
about how to program Excel via Visual Basic but translates very well into
Python COM.

Cheers
Eric Hagemann

"Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
news:3939D364.2EF22B13 at zebrasoft.co.nz...
> hi everyone,
>
> A couple o0f questions:-
> 1)
> can you access an excel (.xls) file via com without having excel loaded
> on that machine?
> 2)
> once you've done:-
>  blah = win32com.client.Dispatch("Excel.Application")
> can anyone give details of
> a) howto open a particular s/s in the current directory
> and
> b) howto access particular cells in that particular s/s
> and
> c) howto close that s/s
>
> I have to admit that I'm hoping some kind soul will help before I'm
> consigned to grappling with what looks like grooling documentation!
> thanks alot matthew.




From aahz at netcom.com  Thu Jun 29 11:00:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jun 2000 15:00:00 GMT
Subject: exiting threads
References: <8jfcf9$2rp$1@nnrp1.deja.com> <8jfiuj$lf1$1@slb1.atl.mindspring.net> <8jfmdf$ah6$1@nnrp1.deja.com>
Message-ID: <8jfo9g$2fu$1@slb1.atl.mindspring.net>

In article <8jfmdf$ah6$1 at nnrp1.deja.com>,
Dale Burnett  <denalione at my-deja.com> wrote:
>In article <8jfiuj$lf1$1 at slb1.atl.mindspring.net>,
>  aahz at netcom.com (Aahz Maruch) wrote:
>> In article <8jfcf9$2rp$1 at nnrp1.deja.com>,
>> Dale Burnett  <denalione at my-deja.com> wrote:
>>>
>>>I am using thread.start_new_thread(self.foo,(x,y))
>>>
>>>Do I have to explicitly exit the thread or does it automatically exit
>>>when foo returns?
>>
>> I believe it does, but I strongly recommend that you subclass
>> threading.Thread instead.
>
>why?  because of the greater functionality of threading or is there
>something wrong with thread?

Because if you want to use *any* of the functionality of threading (e.g.
the queue class or semaphore), your threads need to be a subclass of
threading.Thread.  Sure, you say now that you don't need those features,
but why take chances?  threading just makes your life easier.

As a side note, the threading module emulates the Java model fairly
closely, so it helps if there's ever any chance you might use Java.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From al at servana.com  Tue Jun  6 17:52:48 2000
From: al at servana.com (al at servana.com)
Date: Tue, 06 Jun 2000 21:52:48 -0000
Subject: DEBUGGER question
Message-ID: <8hjrrg+tc6s@eGroups.com>

Hello!
I have read the pdb docs and can use PDB with python script functions 
just fine.

But, how do you use PDB with an 'executable' python script which
uses a MAIN()?

When I have a script with N functions, I can use pdb.run to debug any 
of the functions separately.

When I have a script with a MAIN that is envoked when you 'execute 
the script' (like: python scriptname.py -a -b -c arg1 arg2)
I cannot find a way to get it 'in' to PDB.

If I try pdb.run('scriptname.main()') it does not actually execute 
the script.
I tried pdb.runcall also, and also used the __main__ notataion - 
still no go.

If you look at the WEBCHECKER scripts which come with python, you can 
see scripts which are setup to be 'executed'.
I really want to peek inside some functions while running the script 
to see what is going on. (like 'addroot' in the Checker class of 
WEBCHECKER)

Any help would be appreciated,
Al G




From philippe.possemiers at the-ecorp.com  Mon Jun 26 04:43:54 2000
From: philippe.possemiers at the-ecorp.com (Philippe Possemiers)
Date: Mon, 26 Jun 2000 10:43:54 +0200
Subject: ppp-script
References: <39570BCE.5E5DEF97@the-ecorp.com> <slrn8le4gj.2p5.thomas@centurion.xs4all.nl>
Message-ID: <395717CA.BED460CD@the-ecorp.com>

Thanks, it works perfectly. Actually, I wonder why I haven't tried os.system()
yet. I guess I was too busy trying to alter the ppp script (and even trying to
rewrite it in python) to read the doc more carefully...

Philippe

Thomas Wouters wrote:

> On Mon, 26 Jun 2000 09:52:46 +0200, Philippe Possemiers
> <philippe.possemiers at the-ecorp.com> wrote:
>
> >My platform is Linux and I use a modem to dial my ISP. Therefore, I use
> >os.exec* in my script to activate /etc/ppp/ppp-on. The question is : how
> >do I return to my python script? From the moment I call 'ppp-on', I seem
> >to lose all control and cannot return to the python interpreter.
>
> os.exec*() replace the currently executing process, instead of creating a
> new process and waiting for it to finish (which is what you want, I
> assume.)
>
> You are looking for os.system(). It takes a single string as argument, which
> should be the command and any argument, just like you type them on the
> shell. (In fact, os.system() just spawns a shell and passes it the string ;)
>
> Alternatively, you can use os.fork() and os.exec*(), and then os.waitpid()
> for the child to finish, but that's only necessary if you want to do things
> while waiting, or if you want to avoid using the shell (because you want to
> fiddle with ARGV, or some such.)
>
> Argh-argv-argc-ly y'rs,
>         Thomas.



From emile at fenx.com  Thu Jun 22 17:32:58 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Jun 2000 14:32:58 -0700
Subject: boolean true and false values.
References: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk> <0b0801bfdc5f$b3cca3c0$1906a8c0@fc.fenx.com> <395280DF.7C76C1A4@stud.ntnu.no>
Message-ID: <0cbe01bfdc91$7c123580$1906a8c0@fc.fenx.com>

Of course you're right on 125 vs 122 octal.  My
personal problem with hex is that the 'high order'
values (08x-0Fx) have never been obvious without
some mental gymnastics, whereas with octal it's
always been easy.

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: "Peter Schneider-Kamp" <petersc at stud.ntnu.no>
To: "Emile van Sebille" <emile at fenx.com>
Cc: <python-list at python.org>
Sent: Thursday, June 22, 2000 2:10 PM
Subject: Re: boolean true and false values.


> Emile van Sebille wrote:
> >
> > Only when I want to know about the bits.  It's so
> > much easier to recognize 122 octal as having
> > alternate bits set than the equivalent in decimal
> > or hex.
>
> much easier than decimal - of course! But what is
> wrong about 52? By the way 122 is not alternating...
> 125 is. But then 55 is probably much better to read.
>
> or-did-you-mean-something-else-ly y'rs
> Peter
> --
> Peter Schneider-Kamp          ++47-7388-7331
> Herman Krags veg 51-11
mailto:peter at schneider-kamp.de
> N-7050 Trondheim              http://schneider-kamp.de
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From cut_me_out at hotmail.com  Tue Jun 13 21:18:18 2000
From: cut_me_out at hotmail.com (Alex)
Date: 13 Jun 2000 21:18:18 -0400
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de>
Message-ID: <etdk8ftm55x.fsf@w20-575-109.mit.edu>

Wow, thanks for looking that code over, Bernhard.  What a horrible bunch
of holes.

Alex.


From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 21 17:41:20 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 21 Jun 2000 14:41:20 -0700
Subject: Tkinter - image not found.
References: <Pine.BSO.4.10.10006211042360.18399-100000@gsb080.cs.ualberta.ca>
Message-ID: <02c98a88.6ea91b1a@usw-ex0106-045.remarq.com>

Hi Grzegorz,

There's a kind of quirky thing that causes images to disappear
depending if the reference you have isn't a global one. Search
comp.lang.python for 'unsettling tkinter bug' and you should
find some responses.

Richard

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From gmcm at hypernet.com  Thu Jun 29 19:28:56 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 23:28:56 GMT
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org> <8F62AF73Egmcmhypernetcom@199.171.54.194> <36D5F4DFCA9BCA46.B5AE15F2F19F393C.28F9B766F10B5E23@lp.airnews.net>
Message-ID: <8F62C20E2gmcmhypernetcom@199.171.54.154>

[posted and mailed]

Cameron Laird gets his tail in a knot: 

>Gordon McMillan <gmcm at hypernet.com> wrote:

>>Um, you said you had to be portable. Not only is tail not standard on 
>>Windows, but os.popen is flaky there too (depending on circumstances
>>too painful to enumerate).

>I think I'm going to start growling every time
>I hear that.  To be precise:  popen* flakiness
>is one of the chief discouragements in my use
>of Python (which is, I recognize, slightly
>skewed from other's).

Well, os.popen works fine between console processes on the NT family of 
Windows. However, Windows has problems with 16 bit DOS apps and Win95 has 
an MS acknowledged problem with any of this stuff. The win32 extensions fix 
most of these problems, (I think Bill Tutt even added a workaround for the 
Win95 bug), and I believe that 1.6 will use the win32 versions if they are 
installed (at least, this was discussed seriously at some point).

>>However, tail (usually - some implementations differ) does almost what
>>the following code does. Note that even this isn't portable, because
>>Windows doesn't have ino's to check. I have the code this is extracted
>>from 7x24 

>Guys, guys:  there are enough seriously hard
>problems in the world.  Don't make this one.
>
>Win* FSs don't have i-nodes.  So a portable
>formulation of the requirements shouldn't
>involve i-node.  That's OK.  Just poll for
>either readability or file size.  While I
>confess I haven't verified that Py-built-ins
>for these are properly portable, they REALLY
>need to be if they aren't already, and I'm
>willing to write test drivers and demonstra-
>tions if these remain in doubt for one more
>round of c.l.p dialogue.

The whole inode thing in that code is because it is a common Linux idiom to 
rely on the fact that once a process has opened a file, said file can be 
deleted by another process without the reading process noticing.

This idiom is completely impossible on Windows, so there's no sense in 
checking whether a new file with the same name has now been created.

So what'll it be - a portable version that's useless on Linux, or one that 
spends half it's time checking for something that can't occur on Windows 
<wink>?

on-Windows-once-you've-grabbed-the-tail-
 you've-got-the-whole-damn-dog-ly y'rs

- Gordon (who never uses popen on Windows anyway)


From moshez at math.huji.ac.il  Fri Jun  9 14:05:35 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 21:05:35 +0300 (IDT)
Subject: readlines() question
In-Reply-To: <8hrapd$e2g$1@nntp9.atl.mindspring.net>
Message-ID: <Pine.GSO.4.10.10006092103340.28247-100000@sundial>

[About...]
> >contents = open('file').readlines()
> >
> >is 'file' closed after readlines() finishes? If not, is it possible to
> >somehow get either the file descriptor or the file object of the opened
> >file? Thanks,

[Aahz] 
> Yes, it's closed.  The anonymous object created by open() gets
> deallocated after the assignment; part of the cleanup is to silently
> call close().

While what Aahz said can't be said to be untrue, it's hard to call it
true, either. That is what happens in the *current* *CPython*
implementation. In no way does the language require it, and specifically,
for JPython, it is not true.

just-picking-nits-here-people-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From etoffi at softhome.net  Mon Jun 19 04:00:23 2000
From: etoffi at softhome.net (Aluo Nowu)
Date: Mon, 19 Jun 2000 04:00:23 -0400
Subject: <OT> New filemanager needs people with files
Message-ID: <4.3.2.7.0.20000619035545.00aa1d10@pop.softhome.net>

I am going to be hosting a project on sourceforge called Newfie. Tell you the
truth I'm not good at naming things. (If I got a dog, I would name it "dog".
I'd rather not waste time thinking up names for filemanagers). On to more
important things, it is currently in the planning stage right now.  If you ever
had a wish for a file manager, feel free to tell me.  Python scripting
will more than likely be included. All ideas will be evaluated. You wont be
sorry (unless you're stuck on Explorer).

The main site is: http://sourceforge.net/projects/newfie/
The web site is: http://newfie.sourceforge.net
The mailing list is: mailto://newfie-devel at lists.sourceforge.com
The other mailing list is: mailto://newfie-wishlist at lists.sourceforge.com


--- Original Text  ---
Subject: New filemanager needs people with files
From: etoffi <etoffi at bigfoot.com>
Date: 2000-Jun-16 (1600) -0400

Ha ha ha.  Newfie (got a better name?) is a project which will be used to
surpass "* Commander" filemanagers.  It will have an extensible, pluggable
architecture, be crossplatform, among other great buzzword-type things.  Any
resources (suggestions, screenshots, sourcecode) are welcome.




From sdhyok at email.unc.edu  Wed Jun 14 21:43:14 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 21:43:14 -0400
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <20000614155102.E16232@freei.com>
Message-ID: <8i9cbh$mev$1@news2.isis.unc.edu>

Even though I corrected minor errors, it does not work.
Refer to the following comments.
>> class A:
> >     def __init__(self):
> >             self.x = 1  # this code calls __setattr__
                                  # when I change it into self.__x = 1, it
raised AttributeError in __setattr__
> >
> >     def __setattr__(self, name, value):
> >             if name == "x":
> >                     self.x = value # this code calls __setattr__
recursively
> >             else:
> >                 raise AttributeError
> >
> >     def __getattr__(self, name):
> >             if name == "x":
> >                 return self.x
> >             else:
> >                 raise AttributeError

I found one solution, but expect better one.

-------------- Solution, but not good ------------------
class A:
 def __init__(self):
   self.__dict__["x"]=1

 def __setattr__(self, name, value):
  if not self.__dict__.has_key(name):
   raise AttriubuteError
  else:
   self.__dict__[name] = value

 def __getattr__(self, name):
  if not self.__dict__.has_key(name):
   raise AttriubuteError
  else:
   return self.__dict__[name]

"Sean Blakey" <sblakey at freei.net> wrote in message
news:20000614155102.E16232 at freei.com...
> On Wed, Jun 14, 2000 at 05:48:45PM -0400, Daehyok Shin wrote:
> > How can I prevent illegal definition of a variable to an object?
> > For instance,
> >
> > class A:
> >     __init__(self):
> >             self.x = 1
> >
> >     __setattr__(self, name, value):
> >             if name == "x":
> >                     self.x = value
> >             else:
> >                 raise AttributeException
> >
> >     __getattr__(self, name):
> >             if name == "x":
> >                 return self.x
> >             else:
> >                 raise AttributeException
> >
> > >>> a = A()
> > >>> a.x = 10
> > >>> a.y = 20 # I want to raise an AttributeException.
> >
> > Peter
> >
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
>
> The code you posted seems to do what you want, provided you
> a)add the 'def' keyword to your function definitions
>     and
> b) define AttributeException (perhaps you meant AttributeError?)
>
> --
> Sean Blakey, sblakey at freei.com
> Software Developer, FreeInternet.com
> (253)796-6500x1025
> "Uncle Cosmo ... why do they call this a word processor?"
> "It's simple, Skyler ... you've seen what food processors do to food,
> right?"
> -- MacNelley, "Shoe"
>




From bjrubble at shell16.ba.best.com  Wed Jun 28 19:04:42 2000
From: bjrubble at shell16.ba.best.com (Adam Clark)
Date: 28 Jun 2000 23:04:42 GMT
Subject: <option> in xml.dom.HtmlBuilder
Message-ID: <8je0aa$pf3$1@nntp1.ba.best.com>

Hi all,

Does anyone know what the deal is with <option></option> in 
xml.dom.HtmlBuilder?  It looks like the problem is that <OPTION> is listed
in self.empties in the HtmlBuilder definition.  I think the proper fix 
would be something in self.infer_ends, but one of the inferred ends should
be </select> -- and I don't see <select> there at all?  Can someone
enlighten me?

Thanks
Adam


From helmit_world at hotmail.com  Tue Jun 27 08:30:28 2000
From: helmit_world at hotmail.com (Helmit)
Date: Tue, 27 Jun 2000 15:30:28 +0300
Subject: Books
Message-ID: <8jarl9$9mi1@news.qualitynet.net>

Can anyone sujest some good python books for beginners?




From t_e_sanders at my-deja.com  Fri Jun  2 18:39:54 2000
From: t_e_sanders at my-deja.com (t_e_sanders at my-deja.com)
Date: Fri, 02 Jun 2000 22:39:54 GMT
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h6tcv$o5g$1@slb0.atl.mindspring.net>
Message-ID: <8h9d3o$dnu$1@nnrp1.deja.com>

<clip>
> >Do the terms of Python's distribution license require me to allow
> >other users to copy and distribute my software?
<clip>
>
> What this license means is that your customers need to be free to
> distribute Python, but not your Python scripts.  As an example, use of
> gcc does not force you to allow your users access to your source code.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)

Makes sense to me. Any idea as to what happens if I make
modifications to the Python code and then distribute it? (E.g., if I
were to toss non-redistributable C code in with the Python sources, and
compile it all together.)

 - Thomas

Thomas Sanders
t.e.sanders at usa.net


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


From richard_chamberlain at ntlworld.com  Tue Jun 20 18:21:04 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 20 Jun 2000 23:21:04 +0100
Subject: Tkinter and setting focus
References: <394F860E.EF6042D6@bucodi.com>
Message-ID: <1_R35.5209$_55.108735@news2-win.server.ntlworld.com>

Hi Rony,

You could just generate your own tab event (I presume that's the behaviour
you want)

So:

def validate(self,event):
        # do some validation
        self.field.event_generate('<Tab>')

Richard

BuCoDI-Rony Steelandt <rsteelandt at bucodi.com> wrote in message
news:394F860E.EF6042D6 at bucodi.com...
>
> I have created a class wich expands the standard entry widget.
> In that class i did a override of the return key to do some validation
> of my entry
>
> self.entry.bind(<return>,self.validate)
>
> The validation routine works fine, but now comes my problem
>
> I want at the end of the validation, if validation is ok set the focus
> to the next entry widget in the form.
> I searched documentation but didn't find anything on this.
> The problem is since my class is a general class, i don't know in the
> class wich widget is the next to put the
> focus on.
> Here is some code to show what i'm doing:
> class numentry:
>  def
>
__init__(self,parent,rw,clmn,wdt,relf=SUNKEN,mask='$',min=0,max=999999999):
>
>   self.w = wdt
>   self.m = mask
>   self.min = min
>   self.max = max
>   self.field = Entry(parent,relief=SUNKEN,width=wdt)
>   self.field.grid(row=rw,column=clmn)
>   self.field.bind("<Return>",self.validate)
>   self.returnval=''
>   self.ind = 0
>   self.pointindex = 0
>
>
>  def validate(self,event):
>     # do some validation
> ===> here i want to set the focus to the next widget
>
>
> #main
> frame = Frame(root,width=100,height=100)
> # frame.bind("<Up>",callback)
> frame.pack()
> f=numentry(frame,1,1,10,SUNKEN,'0000.00',5)
> f2=numentry(frame,2,1,10,SUNKEN,'0000.00')
> root.mainloop()
>
>
> I'm realy sticked on that ?????
> Anyone has a idea how to do that
>
> Rony Steelandt
> --
> ======================================================
> Bureau de Conseil et de D?veloppement Informatique
> 2, Rue Henri Avril
> 22480 Saint - Nicolas du Pelem
>
> T?l : +33.2.96.29.79.08
> Fax : +33.2.96.29.79.09
>
> e-mail : bucodi at worldnet.fr
> ======================================================
>
>




From shahn at kits.cs.vu.nl  Fri Jun 16 04:44:29 2000
From: shahn at kits.cs.vu.nl (Hahn S)
Date: Fri, 16 Jun 2000 08:44:29 GMT
Subject: apply for objects
Message-ID: <8icpdd$8c7@cs.vu.nl>

Ok, i have figured out how to use apply on methods. You just have to get
the method from the class and use the im_func in the apply. The first
argument must be the one which should be the self object.

Example:
>>> class D:
	def start(self, hoi, daar=100):
		print self.aap
		print hoi
		print daar


>>> d = D()
>>> d.aap = 'Artis is leuk'
>>> apply(D.start.im_func, (d, 'Hello World'))
Artis is leuk
Hello World
100

My own problem is solved. I hope someone cares :-)

Good luck with Python!
Sander Hahn


From phil at geog.ubc.ca  Thu Jun  8 12:44:30 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 08 Jun 2000 09:44:30 -0700
Subject: Rich comparisons (was:  My wishlist for Python3K)
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3wvk06s29.fsf@brant.geog.ubc.ca>

hzhu at rocket.knowledgetrack.com (Huaiyu Zhu) writes:

> My wishlist for Python3K (This is not to start a flame war):

[snip]

An item not on this list, which is also of real importance for
numeric programers, is rich comparisons.  David Ascher has
written a fairly complete proposal

(http://starship.python.net/crew/da/proposals/richcmp.html)

and about two years ago it was part of the proposed V1.6 feature list:

http://www.foretec.com/python/workshops/1998-11/proceedings/guido/tsld007.htm

but I haven't heard it mentioned as part of 1.6/1.7 -- anyone know
the status of the rich comparison proposal?

Regards, Phil


From aa175 at torfree.net  Sat Jun  3 20:11:42 2000
From: aa175 at torfree.net (Ken McCracken)
Date: Sun, 4 Jun 2000 00:11:42 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org>
Message-ID: <FvLsJJ.CAC.0.bloor@torfree.net>

Not that this exactly belongs in a python ng but I'm told the only usable
onboard programming environment is quartus, an ansi forth system by Neal
Bridges. 

http://www.quartus.net 



From hamish_lawson at yahoo.co.uk  Fri Jun 16 19:29:32 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Fri, 16 Jun 2000 16:29:32 -0700
Subject: Augmented assignment, update
References: <20000616184514.Q26436@xs4all.nl>
Message-ID: <201d2030.4c839656@usw-ex0102-014.remarq.com>

The explanation given at http://www.xs4all.nl/~thomas/python/
states:

"Languages such as C++ and Java provide a mechanism for user-
defined classes to overload the augmented-assignment operators
specifically..."

But Java doesn't support operator overloading for user-defined
classes.

Hamish Lawson


Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From donn at u.washington.edu  Thu Jun 29 14:11:55 2000
From: donn at u.washington.edu (Donn Cave)
Date: 29 Jun 2000 18:11:55 GMT
Subject: Controlling an external program with Python
References: <1542456BE0832BEE.8A959B400FAE5129.26A3D7DF9AA9B305@lp.airnews.net> <nz855.1583$iN5.265679@ptah.visi.com> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com>
Message-ID: <8jg3hb$iic$1@nntp6.u.washington.edu>

Quoth claird at starbase.neosoft.com (Cameron Laird):
| In article <395B32D1.B926FDB3 at hursley.ibm.com>,
| Paul Duffin  <pduffin at hursley.ibm.com> wrote:
...
|> This sort of arrangement is only possible when people who are interested in
|> Expect functionality can transcend any language *issues*, by accepting that 
|> the language is basically irrelevant apart from the effect in may have on
|> the required functionality and work together.
...
| I believe I've heard Don express surprise at Expect's
| career.  He thought people would steal the ideas for
| other (languages') libraries.  What he least expected
| (sic), if I understand it correctly, was that the whole
| world would be content to invoke Expect as the standalone
| executable which we know utterly dominates its use.

Well, I'm surprised he's surprised!  Expect-Tcl works out
so well that it more or less cancels out the ``wrong language''
problem.  Mainly because Tcl (as I suppose Lisp/Scheme) can
support an "expect" statement in a way that doesn't fit in
Python (that I can think of - this is not a comment on the
existing Python expects, with which I am not familiar.)

And then I have looked at the source.  I'm sure it has no peer
when it comes to comprehensive cross-platform pseudotty support,
but a lot of barnacles come with that.  I would steal a younger
horse than this.

	Donn Cave, donn at u.washington.edu


From gchiaramonte at mail.com  Wed Jun 28 16:50:52 2000
From: gchiaramonte at mail.com (Gene)
Date: Wed, 28 Jun 2000 20:50:52 GMT
Subject: Why won't exec accept this?
References: <DD768E0BEA95D211B27A0008C74C86250A6B91@BEKEXCH1>
Message-ID: <Myt65.11940$_b3.289794@newsread1.prod.itd.earthlink.net>

Thanks. I stripped out the carrige returns and appended a newline and works
fine for me too.

Gene




From phd at phd.russ.ru  Fri Jun 30 03:37:15 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 30 Jun 2000 07:37:15 +0000 (GMT)
Subject: Detecting Window Resizing
In-Reply-To: <Pine.GHP.4.21.0006291102470.9839-100000@don.et.byu.edu>
Message-ID: <Pine.LNX.4.21.0006300736310.32043-100000@fep132.fep.ru>

On Thu, 29 Jun 2000, C. Porter Bassett wrote:
> Is it possible in python to detect when the virtual console that a curses
> program is running in is re-sized.  My program checks to see the size of
> the window to know how much it can draw, but it still gives me a
> segmentation fault if I shrink the window down.  I am running this on
> linux.

   Terminal emulator will send you SIGWINCH (signal for window changing)
upon resizing. Catch the signal...

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From Ray.Smith at fujitsu.com.au  Fri Jun  9 00:47:02 2000
From: Ray.Smith at fujitsu.com.au (Ray Smith)
Date: Fri, 9 Jun 2000 14:47:02 +1000
Subject: Newbie: Simple question about string slices
Message-ID: <8hpsl6$mve@newshost.fujitsu.com.au>

Hi there,

I'm completely new to Python and was reading the standard tutorial for
Python and came across string slices., The tutorial says:

"Strings can be subscripted (indexed); like in C, the first character of a
string has subscript (index) 0. There is no separate character type; a
character is simply a string of size one. Like in Icon, substrings can be
specified with the slice notation: two indices separated by a colon."

Then later:

"Here's a useful invariant of slice operations: s[:i] + s[i:]equals s."

Now,
if I have a string defined as

s = "abcdefg"

I assumed s[2:2] to be equal to 'c',
but when I tried it it equals '' (empty).
The way I read this is give me the slice from index 2 to index 2.

I also "feel like" s[:2] + s[2:] should be equal to 'abccdef'

Simple testing shows me
s[2:3] = 'c', and
s[:2] + s[2:] = "abcdef"

My question is:
If given a slice s[a:b]
can I assume index "b" means upto but not including the index "b"?

eg.  s="abcdef"

s[:3] = "abc", meaning from index 0, upto but not including index 3
s[3:] = "def", meaning from index 3 to the end of the string

I guess another way to ask my question is:
Why is s[a:a] always '' (empty) ?

Hope someone out there in Python land can make sense of my question!

Thanks

Ray Smith








From dmenns at home.com  Fri Jun  9 19:32:52 2000
From: dmenns at home.com (Darren Enns)
Date: Fri, 09 Jun 2000 23:32:52 GMT
Subject: Setting Port Speed in Linux Python: HOW?
Message-ID: <39417E33.C2780C16@home.com>

I checked Deja.com for answers to this question, and didn't find a
clear/short answer, so I will ask it myself:

What simple (Linux) Python code can be used to set the baud rate
of a serial port?

Here is the start of a short Python program that I have that works
just fine, and uses the 'cheater' method to set the baud rate of
the /dev/ttyS0 port:

#!/usr/bin/python
import sys
import os
import string
os.system("stty 4800 raw -F /dev/ttyS0")
udl = open('/dev/ttyS0','r')
...

I want to replace this with a method that does not rely on 'stty'
(since the version of 'stty' on my other Linux machine doesn't
seem to support the '-F' option).

Thanks

Dare

--
Darren Enns
EMAIL: dmenns at home.com
HTTP:  members.home.net/dmenns





From phd at phd.russ.ru  Fri Jun 16 04:06:00 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 16 Jun 2000 08:06:00 +0000 (GMT)
Subject: Suggestion: "Completing" HTTP support in Python.
In-Reply-To: <20000615213621.F30050@exeter.exeter.org>
Message-ID: <Pine.LNX.4.21.0006160804260.15416-100000@fep132.fep.ru>

On Thu, 15 Jun 2000, Karl G. Ulbrich wrote:
> Note that POST support in urllib lacks support for posting form contents
> that include files to be uploaded.  That means you're back to down and 
> dirty with the HTTPlib and some ugly http and mime boundary syntax.
> 
> What I need (and am writing as a 1st Python project) is something that'll
> let me post an arbitrary mix of normal fields and file-upload-fields+files
> to a URL.  (usage: POST a file, title, and description field).  
> 
> I'd love to see this support (and some of the other stuff in this thread)
> make it into urllib as easy-to-use methods.

   Why do you think someone (who do not neccessaryly needs this) will write
code that you need? In the world of free software it is simple - roll your
own.
   Patches are gladly accepted. Come on!

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From olczyk at interaccess.com  Thu Jun 29 13:21:12 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Thu, 29 Jun 2000 17:21:12 GMT
Subject: Problem with redirection, buffering and chaining scripts.
References: <395c3b52.183760656@nntp.interaccess.com> <gnF65.8680$Q7.225830@news-east.usenetserver.com>
Message-ID: <395b8570.268270515@nntp.interaccess.com>

On Thu, 29 Jun 2000 10:17:48 GMT, Andre van Straaten
<andre at UltraShell.Net> wrote:

>I never called a Python script from another Python script, but probably
>it might be helpful to flush the buffer with file.close().
>This is even necessary if you read and write to the same file within the
>same script.
>The buffers are flushed always when the script terminates, but .close()
>is sometimes necessary explicitely.
>
>-avs
And since the buffers already get flushed when the script closes, how
does this help?


From emile at fenx.com  Fri Jun  2 09:03:40 2000
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 2 Jun 2000 06:03:40 -0700
Subject: Status of List
Message-ID: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com>

I appear to be having problems with the list.  Yesterday
I saw only 10-20 posts, and overnight there were none.

I checked against usenet news server, but as I rarely read
there, it's somewhat hard to tell but I did spot only a
few 02-jun posts.

Can anyone confirm the problem? Or should I look on my
end.  Or is it just that quiet, and what I saw in usenet
only the result of propogation delay.

BTW, both the Zope and wxPython list are active, and I've
even seen more fnOrb posts than Python posts! ;-)

TIA,

Emile van Sebille
emile at fenx.com
-------------------






From boud at rempt.xs4all.nl  Thu Jun  8 18:21:12 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 8 Jun 2000 22:21:12 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com> <8hluop$mlo$1@news1.xs4all.nl> <2TT%4.178184$55.3794691@news2.rdc1.on.home.com>
Message-ID: <8hp68o$8hk$1@news1.xs4all.nl>

Tom <tom-main at removeme.home.com> wrote:

> "Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
> news:8hluop$mlo$1 at news1.xs4all.nl...
> ...
>> There's another option, of course. You could use Qt (and PyQt), which
>> works on Windows and Unix/X11, offers a great GUI and a good design.
>> And there are tons of programs using it you can look at & learn from.
>> (Not to mention the great documentation.)

> Actually, Qt sounds pretty good, but about six months ago I looked into
> trying it out, only to find that I would have to pay a big load of dough to
> do so (for C++ on Windows).

Yes, that's right - about $1550,- for a single developer license. The
interesting thing however is that you could easily distribute PyQt
for Windows for free - it's an application made with Qt, and the
PyQt license is X11-look-alike. 

-- 

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


From johann at physics.berkeley.edu  Thu Jun  1 16:51:26 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 01 Jun 2000 13:51:26 -0700
Subject: Is it just Syntactic Sugar ?
References: <k7lW4.96338$E85.1941352@news1.rdc1.md.home.com> <3929E47E.993727CF@roguewave.com> <3933F800.9CCE3B6D@cs.byu.edu> <20000530213957.E469@xs4all.nl> <mt4s7f7q62.fsf@astron.berkeley.edu> <m3u2ffydwy.fsf@atrus.jesus.cam.ac.uk> <mtog5n68ak.fsf@astron.berkeley.edu> <20000531090045.I469@xs4all.nl>
Message-ID: <mtr9ah9lbl.fsf@astron.berkeley.edu>

Sorry for the delay, but I've been busy.

Thomas Wouters writes:

> I'll reply to this post instead of the longer one, because you *almost*
> understand my meaning now ;-) I'm not asking for a 't1[t2].incr(e)'.
> Instead, I'm asking for a method exactly like __add__ that realizes that the
> caller actually wants to replace the current object with the new object. It
> shouldn't matter [to the definition] wether the object is changed in place,
> and returns self, or creates a new object just like __add__, and returns
> that.

I understand that much; that was what I was trying to spell with my
"incr" method, an in-place increment.

Guido's __add_ab__ fits what I was talking about; his a.extend is the
same idea as my a.incr.

My problem with this proposal is more how it fits with container
objects, such as lists and dictionaries.  Actually, I think I might
have just been thick about it.  I'll try to say it again, just so you
understand where I'm coming from.

If we assume mutable types, and a magic __add_ab__ method, the most
efficient way to do "lst[i] += 1" is to get the (mutable) object
referenced by lst[i], then increment that object in place.  This will
only call the __getitem__ method of the list, and the __add_ab__
method of the mutable object.

If we have nonmutable types, we have to get lst[i], using __getitem__,
compute the incremented value, using the default __add_ab__, then put
that value back into the containter, using __setitem__.

This is the asymmetry that I was objecting to.  The only way I can see
it being resolved is to always demands he full __getitem__,
__add_ab__, __setitem__ sequence, even if redundant.  The other
option, where we only call __getitem__ and __add_ab__, doesn't seem
(to me) like it would work with the way objects work in python.  How
would __add_ab__ know where the original object came from, so it can
put the new object back in, if the object is immutable.

If you demand the full sequence of get/add/set, then it works.  The
in-place addition simply becomes an object-creation optimization, not
a way to optimize away the sequence-set operation.  Does that make
sense?

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From agw at sobolev.dsm.fordham.edu  Thu Jun  8 11:55:33 2000
From: agw at sobolev.dsm.fordham.edu (Art Werschulz)
Date: Thu, 08 Jun 2000 15:55:33 GMT
Subject: MacOS: problems with tkinter
Message-ID: <ueya4pqux6.fsf@sobolev.dsm.fordham.edu>

Hi.

I'm trying to use python on a mac (233MHz Beige Power Mac G3 running
MacOS 9).  I have installed Python 1.5.2c1 and Tcl/Tk 8.3.0.

I am having trouble running various demos involving tkinter.  A
typical error traceback looks like the following:

Traceback (innermost last):
  File "Dogbert:Applications:Programming:Python
  1.5.2c1:Demo:tkinter:guido:hello.py", line 4, in ?
    from Tkinter import *
  File "Dogbert:Applications:Programming:Python
  1.5.2c1:Lib:lib-tk:Tkinter.py", line 8, in ?
    import _tkinter # If this fails your Python may not be configured for Tk
ImportError: No module named _tkinter

Please note that $(PYTHON):Lib:lib-tk is in my sys.path.

So how *does* one configure Mac python to work with Tk?
According to http://www.python.org/topics/tkinter/trouble.html#checking:

  If it fails with "No module named _tkinter", your Python
  configuration needs to be modified to include this module (which is
  an extension module implemented in C). Edit Modules/Setup (read the
  instructions at the top, search for _tkinter, read more instructions
  there, and follow the instructions) and rebuild and reinstall. Make
  sure to enable the _tkinter module as well as the TKPATH variable
  definition. 

Since I don't have a C compiler for my Mac, this isn't an option.
Suggestions?

Thanks.

-- 
Art Werschulz (8-{)}   "Metaphors be with you."  -- bumper sticker
GCS/M (GAT): d? -p+ c++ l u+(-) e--- m* s n+ h f g+ w+ t++ r- y? 
Net: agw at dsm.fordham.edu <a href="http://www.dsm.fordham.edu/~agw/">WWW</a>
Phone:   Fordham U. (212) 636-6325, Columbia U. (212) 939-7061


From MarkH at ActiveState.com  Fri Jun 30 21:35:53 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 01 Jul 2000 01:35:53 GMT
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org> <8F62AF73Egmcmhypernetcom@199.171.54.194> <36D5F4DFCA9BCA46.B5AE15F2F19F393C.28F9B766F10B5E23@lp.airnews.net> <8F62C20E2gmcmhypernetcom@199.171.54.154> <uk8f63fwy.fsf@ctwd0143.fitlinxx.com>
Message-ID: <ZVb75.2491$Tb7.18785@news-server.bigpond.net.au>

> I guess I'm just glad that my requirements to use Windows stops with
> NT, since I also think Python might be a non-starter for a lot of my
> administration apps if popen() didn't work reliably.  Of course, MS is
> far more to blame for this than Python :-)

And Bill Tutt is to be eternally thanked for Python solving it :-)

[And Gordon is correct - later builds have reliable popen on Win9x too,
also thanks to Bill!]

Mark.





From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Fri Jun 16 09:04:39 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Fri, 16 Jun 2000 06:04:39 -0700
Subject: Tkinter Widget command function call
References: <3949E71A.AF53A1CF@muc.das-werk.de>
Message-ID: <2f9fdd30.7dcfb414@usw-ex0105-035.remarq.com>

Hi Thomas,

This is not oop so you'll have to convert a little bit:

from Tkinter import *
root=Tk()

def buttonPressed(event):
    print event.widget
    if event.widget.cget('text')=='Hello':
        event.widget.configure(text='Press Me')
    else:
        event.widget.configure(text='Hello')

for i in range(0,10):
    myButton=Button(root,text="Press Me")
    myButton.bind('<Button>',buttonPressed)
    myButton.pack(side=TOP)

root.mainloop()

<EOF>

So basically instead of using command I'm binding an event (in
this case <Button>) to it and passing the method I want calling.

That way you get an event passed which you can use to get the
calling widget. Obviously you're not keeping any references for
the buttons around so this may not be all that useful but you
could get and set the configuration of them.

Richard

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From donn at u.washington.edu  Thu Jun 22 12:09:53 2000
From: donn at u.washington.edu (Donn Cave)
Date: 22 Jun 2000 16:09:53 GMT
Subject: apply for objects
References: <Pine.GSO.4.20.0006220932160.9460-100000@keg.cs.vu.nl>
Message-ID: <8itdoh$a7g6$1@nntp6.u.washington.edu>

Quoth Sander Hahn <shahn at cs.vu.nl>:
| I wanted to call a method of an object in Python.
...
| That works, however when you use apply on a method you get the following
| error:
|
| >>> class C:
| 	def m(x, y):
| 		print str(x) + str(y)
|
|
| >>> c = C()
|
| >>> apply(c.m, ('hello', ' bye now'), {})
| Traceback (innermost last):
|   File "<pyshell#12>", line 1, in ?
|     apply(c.m, ('hello', ' bye now'), {})
| TypeError: too many arguments; expected 2, got 3
|
| I previously got errors about no __call__ or something weird...!?!
| However i found out that you must use the im_func from the method to get
| the corresponding function:
|
| >>> apply(c.m.im_func, ('hello', ' bye now'), {})
| hello bye now
|
| So, my problem is solved for now :-)

No, it isn't!  Your problem is just that you defined your class
method without its 1st class instance parameter.

    class C:
        def m(self, x, y):
            print str(x) + str(y)

That's the third argument you didn't expect.  Define it this way
and it will work fine without any im_func hackery.

	Donn Cave, donn at u.washington.edu


From tiddlerdeja at my-deja.com  Mon Jun  5 11:19:45 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Mon, 05 Jun 2000 15:19:45 GMT
Subject: win32Service.OpenService not working
Message-ID: <8hggdu$4fa$1@nnrp1.deja.com>

Can you help?

I'm trying to get a handle on a service but can't. If I use the
following code, it works:
h = win32service.OpenService(hscm, "UPS",
win32service.SERVICE_ALL_ACCESS)

However, if I try to get a service with a space in the name, I get an
error:
>>> h = win32service.OpenService(hscm, "Windows installer",
win32service.SERVICE_ALL_ACCESS)

Traceback (innermost last):
  File "<stdin>", line 1, in ? pywintypes.api_error:
(1060, 'OpenService', 'The specified service does not exist as an
installed service.')


I even tried to convert the name to unicode with now luck:
>>> www = win32api.Unicode("Web Site Update")

Can anyone tell me what I'm going wrong.

I'm doing this with administrator privileges.

Any help greatly appreciated.


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


From morten at src.no  Mon Jun 26 10:01:29 2000
From: morten at src.no (Morten W. Petersen)
Date: Mon, 26 Jun 2000 16:01:29 +0200
Subject: Using MimeWriter
Message-ID: <Pine.LNX.4.21.0006261559480.23273-100000@lade.src.no>

Does anyone have a (somewhat simple) script (or know of one) where
MimeWriter is being used? Both for sending simple (body only) messages,
and more complex multipart messages with files attached?

-------------------------------------------------------------
How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From DOUGS at oceanic.com  Fri Jun  9 16:11:25 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Fri, 9 Jun 2000 10:11:25 -1000 
Subject: PATCH: Augmented assignment
Message-ID: <8457258D741DD411BD3D0050DA62365907A1FB@huina.oceanic.com>

[Thomas Wouters wrote:] 
> >As promised, here is my patch to add augmented assignment (+=, -= and
> >family) to Python. It's still Work In Progress, and probably 
> requires a
> >rewrite of some magnitude before Guido will accept it.
> 
> 
> Why would the Python community want this?  It obfuscates 
> code, in as much
> that newbies aren't likely to intuite what the symbols mean.

There are a lot of things that the newbie doesn't understand.  I still can't
get my brain around 're'.  That definately doesn't mean that it shouldn't be
available to those who can.

> We just had a thread suggesting that = and == are frequently 
> confused by
> newbies.   How much more confused would they be with a plethora of
> alternative symbols?

I disagree that confusion by newbies, in and of itself, is necessarily a bad
thing.  Programming has some inherent sticky problems.  The question should
always be whether a particular confusion is artificially introduced by the
language or inherent to what needs to be understood.  Having two symbols '='
and '==' and thus having to explicitly understand the difference is good
because that is a fundamental to programming.  Not having them introduces
confusion because of the implicit nature.  "Explicit is better then
implicit". - Tim Peters

> Python seems to stick to the "one right way" model of 
> programming: most
> people's code tends to look the same.  It increases 
> maintainability and
> readability.

If the clearest way is what is taught most of the code produced will end up
clear.  I am in the middle of producing an example proving the exception to
the rule of clarity of Python coding.  It definately doesn't look the same
as "most people's code". It is much clearer than what I might produce if I
was trying to do it in, say, QBASIC.  I promise to clean it up when I get
more time ;-)
  
> I think we'll end up with a confusion of styles if augmented 
> assignments
> are allowed.

As in the case with '=' and '==', if you know you need them you've had to
think about why and where they go.  Obfuscate the documentation enough and
no newbie will be tempted.  If a newbie is reading code the documentation
should be clear that these are for optimization.  Good way to introduce such
concepts.

> And it's not like it adds new functionality nor improves code 
> readibility.
> It just saves the lazy programmer a whopping one or two keystrokes.

I don't see it this way.  I think mostly it won't be used if like you say it
reduces readability.  Those who need it will really appreciate it being
there.  I'm too lazy to look up the reference, but its been stated that
there can be a performance improvement given certain problem spaces.  It
would be great if that optimization (whatever it is) could be done under the
covers but that won't happen (probably) in the Python 1.x lineage.  Maybe
these extra symbols can be optimized out in Py3K.

-Doug-



From gmcm at hypernet.com  Sun Jun  4 22:50:39 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 05 Jun 2000 02:50:39 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com>
Message-ID: <8F49E9873gmcmhypernetcom@199.171.54.154>

Courageous wrote:

>Thomas Wouters wrote:

>> Actually, it's very intuitive, if you see the start and end index of
>> the slice not as items, but as the gaps between the items.

>Well, yeah, but [0:3] as "element zero through three, inclusive"
>is about as intuitive as you can get. While I understand completely
>how the current implementation works (reading the source code and
>all, dontcha know), I still fail to see *why* this was chosen.
>I'm presuming there must be some conditions where this particular
>idiom was required or caused for nasty syntactic transformations
>if it wasn't used?

Actually, this being Usenet, you're both wrong! There's nothing "intuitive" 
about either one. It's all what you're used to.

I suspect you haven't used many repetitive slicing algorithms. It boils 
down to being nice because

 lst == lst[:x] + lst[x:]

That is, no horsing around with adding one. Once you get used to it, you'll 
find the "inclusive" way to be clunky and error prone.

- Gordon


From petersc at stud.ntnu.no  Sat Jun 10 02:15:08 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 10 Jun 2000 08:15:08 +0200
Subject: Programming Poll
References: <8hrim7$fjv$1@nnrp1.deja.com> <20000609210521.A6218@bigskytel.com>
Message-ID: <3941DCEC.5E5A0C06@stud.ntnu.no>

David Porter wrote:
> 
> > http://cupid.suninternet.com/~joeo/poll.html, with the intent of trying
> 
> it returns 404.

No it doesn't.

at-least-it-worked-for-me-yesterday-and-today-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From rhicks at nospam.rma.edu  Wed Jun  7 08:26:30 2000
From: rhicks at nospam.rma.edu (OPUS)
Date: Wed, 07 Jun 2000 12:26:30 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net>
Message-ID: <Wbr%4.69$vh.19893@news>

"Conrad Schneiker" <schneiker at jump.net> wrote in message
news:8hl866$bt5$1 at news.jump.net...

:     Ruby is a relatively new, very high level, fully OO language that
:     integrated many good ideas from Perl, Python, Smalltalk, Eiffel,
:     ADA, Clu, and Lisp. (Ruby is more fully OO than Python.)

I thought everything in Python was an object and that Python was designed
from the ground up to be object oriented...how can you be more "fully" OO?

And stop trolling this newgroup for Ruby converts! If someone wants to learn
it they will.







From jvickroy at sec.noaa.gov  Thu Jun 15 12:03:43 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Thu, 15 Jun 2000 10:03:43 -0600
Subject: Python 1.5.2 on HP-UX v11
Message-ID: <3948FE5E.5A433424@sec.noaa.gov>

Is anyone using the *thread* module in Python 1.5.2 on HP-UX v11?




From aahz at netcom.com  Fri Jun 23 15:58:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jun 2000 19:58:00 GMT
Subject: Posting spew
Message-ID: <8j0fg8$q9d$1@slb6.atl.mindspring.net>

I think I've found the problem.  If you look at the two pairs of headers
below, you'll see that they have the same Message-ID except for case.
The machines between news.gtei.net and news.maxwell.syr.edu are added to
the second Path, so I'm guessing that one of them is re-emitting
comp.lang.python posts with a mangled Message-ID.  If I had to guess
further, I bet it's a cranky old machine sitting in some AT&T lab.

Message-ID: <etd8zw15mvd.fsf at w20-575-109.mit.edu>
Path: mindspring!news.mindspring.net!cpk-news-hub1.bbnplanet.com!
  news.gtei.net!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!
  bloom-beacon.mit.edu!senator-bedfellow.mit.edu!dreaderd!not-for-mail

Message-ID: <ETD8ZW15MVD.FSF at W20-575-109.MIT.EDU>
Path: mindspring!news.mindspring.net!cpk-news-hub1.bbnplanet.com!
  news.gtei.net!feeder.qis.net!ptdnetP!newsgate.ptd.net!attbtf!attcg2!
  S1044A3M,attcg1!news.bcs.moore.com!attbt1!ip.att.net!news.ti.com!
  news.smu.edu!news.tamu.edu!news.sgi.com!news.maxwell.syr.edu!
  newsswitch.lcs.mit.edu!bloom-beacon.mit.edu!senator-bedfellow.mit.edu!
  dreaderd!not-for-mail
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Why don't they make poppasmarts or didjeridon'ts?  --Aahz


From smst at bigfoot.com  Fri Jun  2 09:33:54 2000
From: smst at bigfoot.com (Steve Tregidgo)
Date: Fri, 02 Jun 2000 14:33:54 +0100
Subject: Status of List
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com> <3937B59C.3AFE081E@bam.com>
Message-ID: <3937B7C2.A012D3B2@bigfoot.com>

Bill Scherer wrote:
> 
> Things have been very quite since yesterday from my point of view.
> I'm just glad I'm not the only one.
> 
> Emile van Sebille wrote:
> 
> > I appear to be having problems with the list.  Yesterday
> > I saw only 10-20 posts, and overnight there were none.
> >

Same here.  I checked the eGroups mirror of the list, and they list just
15 messages for June so far:

  http://www.egroups.com/list/python-list/

Perhaps the news->mail gateway is broken?  Mail->news seems okay --
Remarq is listing this thread...

Never mind, it leaves me more time to do some work ;^)

-- 
Steve Tregidgo
Software Developer
http://www.businesscollaborator.com



From aahz at netcom.com  Fri Jun  9 18:41:42 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jun 2000 22:41:42 GMT
Subject: readlines() question
References: <Pine.GSO.4.10.10006092339580.29560-100000@sundial>
Message-ID: <8hrrr6$fj4$1@slb2.atl.mindspring.net>

In article <Pine.GSO.4.10.10006092339580.29560-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>
>Ummmmm...I'm known as a professional paranoid in many places, so I usually
>code like:
>
>my_file = open("blah")
>try:
>	...my_file.read, my_file.readlines, etc.
>finally:
>	my_file.close()

And if you're *really* paranoid ;-)

try:
	my_file = open("blah")
	...my_file.read, my_file.readlines, etc.
finally:
	my_file.close()
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From mtbros at tin.it  Sat Jun 24 08:09:39 2000
From: mtbros at tin.it (Mario Tola)
Date: Sat, 24 Jun 2000 14:09:39 +0200
Subject: Can I detect if an app it's already running (Win)...
Message-ID: <8j28fg$1c5$1@nslave1.tin.it>

There's a way to detect (on win, both 98 and NT-2000) if a program (say
'notepad.exe')  is still running ? I have to start from python an
application only if it's not already running...




From stephen at cerebralmaelstrom.com  Thu Jun 29 02:54:19 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 29 Jun 2000 06:54:19 GMT
Subject: Module Name Problem
References: <eviltofu-C678D0.18293328062000@news.earthlink.net> <8F61CAE59gmcmhypernetcom@199.171.54.194> <eviltofu-E613EC.21014728062000@news.earthlink.net>
Message-ID: <voC65.98134$WS3.847668@typhoon.we.rr.com>

This doesn't sound very broken,.. you're creating the conflict-of-names.. :)

I read somewhere the sugguestion of an 'import name as name' syntax..don't
know where...don't know if its a 1.6 or a py3k thing...

--S


Jerome Chan <eviltofu at rocketmail.com> wrote in message
news:eviltofu-E613EC.21014728062000 at news.earthlink.net...
> In article <8F61CAE59gmcmhypernetcom at 199.171.54.194>, gmcm at hypernet.com
> (Gordon McMillan) wrote:
>
> > Jerome Chan wrote:
> >
> > >If I have a module called
> > >
> > >maxvu.os
> > >
> > >and I use
> > >
> > >import os
> > >
> > >Somehow, when I try to use the os calls or maxvu.os calls it gets
> > >confused. Is there a workaround besides renaming my module?
> >
> >
> > Not if you want access to os within maxvu.
> >
> > - Gordon
>
> Will this be fixed in 1.6?




From 12klat at sightreader.com  Mon Jun  5 18:02:47 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Mon, 05 Jun 2000 15:02:47 -0700
Subject: operator overloading
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com>
Message-ID: <393C2386.307A160D@sightreader.com>

Emile van Sebille wrote:

> Here's a quick example of using __cmp__
>
> >>> class Test:
>  def __init__(self,a):
>   self.a = a
>  def __cmp__(self,other):
>   return self.a < other.a
>
> >>> a = Test(3)
> >>> b = Test(4)
> >>> print a<b
> 0
> >>> print a>b
> 1
>
> HTH,
>
> Emile van Sebille
> emile at fenx.com
> -------------------

Um.  No.  Not exactly.  Usually 1 designates "true" and 0 designates
"false" unless I'm mistaken.  Also IMHO, > means "greater than" and <
means "less than".  BTW there is yet another comparison operator == than
means "equal to".  :-)

The return value of __cmp__ should be -1, 0, 1 for <, ==, >,
respectively.

The following more closely conforms to these conventions:

>>> class Test:
...  def __init__(self,a):
...   self.a = a
...  def __cmp__(self,other):
...   if self.a == other.a:
...    return 0
...   elif self.a < other.a:
...    return -1
...   else:
...    return 1
...
>>> a = Test(1)
>>> b = Test(2)
>>> c = Test(2)
>>> a<b, a>b, a==b, a!=b
(1, 0, 0, 1)
>>> b==c, b<c, b!=c
(1, 0, 0)

-- Ken Seehof
kens at sightreader.com

> ----- Original Message -----
> From: <kazan at kazan.fenx.com>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Monday, June 05, 2000 1:01 PM
> Subject: operator overloading
>
> > Hello
> >
> > How can I overload the less than operator?
> > learning python did mention something about __cmp__ but
> did
> > not provide any examples.
> >
> >
> > Thanks in advance
> >
> > Erling
> > --
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000605/98d62490/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000605/98d62490/attachment.vcf>

From fiona at sitegnome.com  Tue Jun 27 09:22:08 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 27 Jun 2000 13:22:08 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 27th, 2000
Message-ID: <20000627132208.10146.qmail@synop.com>

Hi Guys!

Below are the entries I've entered into http://python.faqts.com tonight.

Cheers, Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
Client side cookie
http://www.faqts.com/knowledge-base/view.phtml/aid/4012
-------------------------------------------------------------
Feng Tian



-------------------------------------------------------------
Where can I find informations about the Gnuplot module?
http://www.faqts.com/knowledge-base/view.phtml/aid/4025
-------------------------------------------------------------
Thomas Leclercq



-------------------------------------------------------------
can i make something like jtable with tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/4019
-------------------------------------------------------------
udo apolloner



## New Entries #################################################


-------------------------------------------------------------
Is there a way to differentiate between WinNT and Win9x using Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/4016
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

import win32api
win32api.GetVersionEx()

will return, e.g., the tuple:

(4, 0, 1381, 2, 'Service Pack 5')


The 4th element in the tuple ("2" here) has the specific info you 
require; it's 1 for Win/95 and Win/98, 2 for Win/NT.


-------------------------------------------------------------
Is there any way to control the function that is used during an 'in' search?
http://www.faqts.com/knowledge-base/view.phtml/aid/4020
-------------------------------------------------------------
Fiona Czuczman
Stephen Hansen, Thomas Wouters

I believe its the __contains__ method.

In 1.6, yes. In 1.5.2, overloading 'in' is not possible, except by
overloading __getitem__ -- the 'in' implementation for user classes and 
most builtin types simply __getsitems__ until it finds a match or 
reaches IndexError. In 1.6, it's possible to override it in both and 
provide a more efficient implementation. (See 'Objects/rangeobject.c' 
for an example ;-)


-------------------------------------------------------------
Is there any way to make a wrapper for a string object, like UserDict is a wrapper for the dict type?
http://www.faqts.com/knowledge-base/view.phtml/aid/4021
-------------------------------------------------------------
Fiona Czuczman
Thomas Wouters

There is the UserString module in the 1.6 alphas, and it seems to be 
fairly portable. (It defines all methods that strings have in 1.6, 
whereas they have none in 1.5.2, but that shouldn't bother portability. 
You can't use them unless you use 1.6 strings as well, though, because 
those methods are defined in terms of string methods ;-)

You can get the UserString module by downloading a 1.6 alpha, or by 
grabbing a recent CVS tree, or by browsing the CVS three through 
sourceforge:

http://www.sourceforge.net/cvs/?group_id=5470

The module should at least show that it's *possible* to write a string
wrapper ;-) though without the 1.6 additions to strings & the 
__contains__ overloader it might not be that efficient.


-------------------------------------------------------------
Is there a built in Python function that eliminates duplicates in a list?
http://www.faqts.com/knowledge-base/view.phtml/aid/4028
-------------------------------------------------------------
Fiona Czuczman
Bjorn Pettersen, Aahz Maruch, jerry_spicklemire

def unique(lst):
    d = {}
    for item in lst:
        d[item] = None
    return d.keys()

--------

The Timbot recommends

        d[item] = 1

because there's no lookup involved on 1, so it's fractionally faster.
(Any other constant should probably work the same.)

--------

Just a minor "heads up" for a possible "gotcha".

It looks to me as if this method, though speedy, will return the
cleaned list in a much different order than the original, since the

  dict.keys()

method outputs in the (seemingly random) order of the dicts internal
hash table.

The moral is, if the order of the list members matters, be sure to wait
until after the cleaning step to sort, or at least remember to re-sort!


## Edited Entries ##############################################


-------------------------------------------------------------
I update a class with 'reload (module)'. How can I update the instances too?
How to change class instances behavior via reload?
http://www.faqts.com/knowledge-base/view.phtml/aid/3347
-------------------------------------------------------------
Fiona Czuczman
Martin von Loewis,Laurent Szyster, J?rgen Hermann

You'd have to find the instance one-by-one. If you manage to get hold
of them, you can set their __class__ attribute to the new class:

>>> class A:
...   def doit(self):print "old"
... 
>>> a=A()
>>> class A:
...   def doit(self):print "new"
... 
>>> a.doit()
old
>>> a.__class__=A
>>> a.doit()
new

--- Alternate ---

You should implement static and dynamic features that you need in 
different modules and use proxy methods.

Like this

-- dtest.py --
def method(self):
    ...

-- test.py --
class test:
    def method(self):
        return dtest.method(self)

And then,

>>> import test, dtest
>>> t = test.test()
>>> reload(dtest)
>>> t.method()

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

Laurent showed you how to do this automatically. Another technique I 
used to DETECT whether the implementation of a class has changed is 
this:

class spam:
    def __init__(self):
        pass

    def checkReload(self):
        return self.__class__ != spam


-------------------------------------------------------------
What is the best and fastest way to scan a whole list of foldersto see if something has been added or not?
http://www.faqts.com/knowledge-base/view.phtml/aid/1326
-------------------------------------------------------------
Nathan Wallace, Aaron Digulla
Darrell

If your talking about NT then give this a try.

    http://www.dorb.com/darrell/win32WorkSvr/makeThumbsDG.py

def main():
    """
    Use events to avoid polling.
    A sleep is used to give most files a chance to finish
    being writen. Sort of a hack and might not be needed.
    The notification event LAST_WRITE seems to trigger
    at the start and end of writing. So a single file
    will trigger this twice.
    """
    hnd=win32file.FindFirstChangeNotification\
        (os.path.abspath('.')+os.sep+'incoming'\
        ,0, winnt.FILE_NOTIFY_CHANGE_LAST_WRITE )

    while 1:
        int = win32event.WaitForSingleObject( hnd, -1)
            # Try to give the file time to finish writing
        time.sleep(2)
        print 'run'
        try:
            passed, failed, skipped = makeThumbNails()
        except:
            if '-d' in sys.argv:
                traceback.print_exc()
            raise ThumbNailException
        win32file.FindNextChangeNotification(hnd)

On Unix, things are a bit more complicated (there is no way to get
notified when a directory is changed). Instead, you have to check
for the modification time (mtime). The mtime of a directory will
change if an item in that directory is added, deleted or renamed.
It will *not* change when an item is modified (for example, if a file
is overwritten or an item is added to a subdirectory).

You will have to use os.path.walk() to recusively scan the directory
tree if you don't have a list of folders to check. Since that can
take a long time, you should use a cache which contains the
names of all directories plus their mtimes (get them with
os.stat (path)[stat.ST_MTIME]). Next time, you should load
that chache and just check and re-read directories which have
changed.


-------------------------------------------------------------
How do I get a file list via an ftp connection?
http://www.faqts.com/knowledge-base/view.phtml/aid/3887
-------------------------------------------------------------
Fiona Czuczman, Kenroy Harrison
Thomas Weholt

Here's a sample from the FTPLIB Library Reference page:

http://www.python.org/doc/current/lib/module-ftplib.html

from ftplib import FTP
ftp = FTP('ftp.python.org')   # connect to host, default port
ftp.login()               # user anonymous, passwd user at hostname
ftp.retrlines('LIST')     # list directory contents
ftp.quit()


The file Tools/scripts/ftpmirror.py in the Python source distribution is
a script that can mirror FTP sites, or portions thereof, using the
ftplib module. It can be used as an extended example that applies this
module.






From dmost at magna.com.au  Mon Jun  5 16:02:19 2000
From: dmost at magna.com.au (dmost at magna.com.au)
Date: Mon, 05 Jun 2000 20:02:19 GMT
Subject: "dlist" -- beta is ready
References: <393B4144.4C1DC04C@san.rr.com>
Message-ID: <8hh0vt$i6c$1@nnrp1.deja.com>

This is a very interesting concept. It might be worthwhile exploring
the possibility of adding some heuristic which would detect whether the
list is being added to at the front or back, and adjusting the realloc
behaviour accordingly.

e.g. you'd start your list with a small constant overallocation at
either end. The overallocation at each end would be recorded
separately, with independant doubling behaviour as the list grows in
that direction. In this way, a list used only to add elements at the
start of the list wouldnt have its trailing overallocation increased.

In article <393B4144.4C1DC04C at san.rr.com>,
  Courageous <jkraska1 at san.rr.com> wrote:
>
> I have a functioning implementation of my "dlist":
> an ADT which is a vector similar to python's list,
> with the addition of a buffer zone on the front as
> well as the back. This gives the vector O(1)+k
> performance on add/remove from both the head and
> the tail, while maintaining constant time for random
> access (with a small additional constant cost of
> tracking a virtual index).
>
> The implementation seems stable, and is passing
> a small test suite I've written, so I think it works.
>
> There are three items which I will finish soon
> (perhaps sooner if someone asks for them):
>
> 1. sorting; probably shouldn't be too hard to add
>    in Guido's/Tim's binarysort/sample sort code.
>
> 2. reversing.
>
> 3. invariant structures, in conjunction with 1.
>
> This is a VC98 project, intended to compile/run
> on windows platforms, although it could with some
> small mods work on unix platforms with equal ease.
>
> Simply request the source, and I'll send you a
> copy of the whole project dir, including the small
> test suite I have.
>
> C/
>


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


From pinard at iro.umontreal.ca  Wed Jun 14 10:58:52 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 14 Jun 2000 10:58:52 -0400
Subject: python is hard to learn without c
In-Reply-To: thomas@cintra.no's message of "Wed, 14 Jun 2000 12:39:55 GMT"
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
	<394b3e7d.284293552@news.online.no>
	<zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net>
	<394c7c53.300124315@news.online.no>
Message-ID: <oqr9a0wbpv.fsf@titan.progiciels-bpi.ca>

[Thomas Weholt]
> On Wed, 14 Jun 2000 09:57:51 GMT, "jin choung" <jin.choung at gte.net>

> Well, I disagree.  Learning Python was fine.  I don`t know what you
> found so difficult.

I could easily imagine that for someone who really begins using computers,
and who is uncomfortable with the meaning of an assignment like `a = 3',
say, will find almost _any_ programming language difficult.  For someone
who had been exposed to programming in some way, Python should be easy.

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





From pharris at crosswinds.net  Sun Jun  4 08:50:53 2000
From: pharris at crosswinds.net (Phil)
Date: Sun, 4 Jun 2000 13:50:53 +0100
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <393a50cb@news.server.worldonline.co.uk>

print "Uploaded %s bytes\r" % byte_count ,

The imprtant bits are the \r and the comma at the end.  Miss one or the
other and it won't work.

hth

Phil
phil DOT harris AT zope DOT co DOT uk

"Bablos" <angband at blueyonder.co.uk> wrote in message
news:393A38E9.MD-1.4.4.angband at blueyonder.co.uk...
> Hi There...
>
> Now that I've got my FTP client doing some basic stuff, I now have a
> progress report which looks something like the following;
>
> uploaded X bytes
> uploaded Y bytes
> uploaded Z bytes
>
> Is there a way that I can get Python to print this all on the same
> line, so that it looks as if the line is being updated constantly, as
> opposed to having a ridiculously long list of 'uploaded...' lines?
>
> Hope someone can help.
>
>
> Bablos.




From digitig at cix.co.uk  Wed Jun 14 17:31:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 14 Jun 2000 22:31 +0100 (BST)
Subject: Python Productivity over C++
References: <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net>
Message-ID: <memo.20000614223157.31901B@digitig.compulink.co.uk>

In article 
<E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64 at lp.airnews.net>, 
claird at starbase.neosoft.com (Cameron Laird) wrote:

> On the other hand, I'm still searching for the best
> way to speak about the "well you Python people only
> find those errors at run-time" objection.  

Ask them why C and C++ are deprecated for safety critical applications. 
It's essentially because C and C++ programmers only find errors at 
run-time, too. The difference is that Python programmers know it and 
account for it.


From aahz at netcom.com  Sat Jun  3 10:49:32 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jun 2000 14:49:32 GMT
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de> <3938090C.9FB5D057@stud.ntnu.no> <8h9682$v4h$1@nntp9.atl.mindspring.net> <m3d7lzyvir.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8hb5ts$6mk$1@slb7.atl.mindspring.net>

In article <m3d7lzyvir.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>aahz at netcom.com (Aahz Maruch) writes:
>>
>> Let's be technically precise here: an exception is raised and caught by
>> the "for" handler; the for loop is then broken (precisely as if the
>> break statement was executed).  Like any other caught exception, the
>> exception is thrown away (not "error is cleared").
>
>Not quite "precisely as if the break statement was executed", wrt else
>blocks.  

Oh, right, forgot about that.  I never use else blocks for loops.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From aahz at netcom.com  Mon Jun 12 18:47:37 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jun 2000 22:47:37 GMT
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu>
Message-ID: <8i3pa9$lvl$1@slb6.atl.mindspring.net>

In article <jl7lbu4jdf.fsf at endor.bbb.caltech.edu>,
Michael Vanier  <mvanier at endor.bbb.caltech.edu> wrote:
>
>I wrote a little python script using the thread module the other day.  I
>wanted to be able to set a flag in the parent thread that would be read in
>the child thread.  So I set a flag variable (an integer in this case) in the
>parent, created the child thread, and then checked the value of the flag
>every time through a loop in the child thread to see if it had changed.  It
>turns out that the child thread apparently made a copy of the flag, because
>when it changed in the parent it didn't change in the child.
>
>To fix this, I changed the flag to be a list with one integer element.  Now,
>when I change the list element in the parent, the child sees it too.  I
>gather that what this means is that there is a fundamental distinction
>between atomic types like integers and reference types like lists or
>dictionaries.  However, since python treats everything as a reference this
>seems inconsistent to me.  So what I want to know is: what is the cause of
>this behavior, and what is the justification for it?  Or could it be a bug?

This actually has nothing to do with threads and everything to do with
the distinction between global and local variables, plus the distinction
between mutable and immutable objects.  Generally speaking, it's better
practice for what you're doing to create a class instance, pass it in
to the thread, and read/modify class attributes.  Note that you'll
likely want to use a semaphore to make sure that you don't get into
situations where you're reading the variable in one thread while you're
setting it in another.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From bvdpoel at uniserve.com  Sat Jun 24 17:47:30 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 24 Jun 2000 14:47:30 -0700
Subject: tkinter, grid and resize
References: <39550101.AD8C7956@uniserve.com> <gt955.953$F01.22132@news6-win.server.ntlworld.com>
Message-ID: <39552C72.6BB4FCE3@uniserve.com>

richard_chamberlain wrote:
> 
> Hi Bob,
> 
> Grid basically has an absence of fill, which is what pack uses to resize
> widgets when their master resizes.
> 


Ahh, that's the conclusion I was coming to. I had a look at some of the
tk example code, mainly the ccanvas, and it appears to use grid...but
it's written in tcl and I wasn't in the mood to trace what is really
happening there <g>.


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


From erik at src.no  Wed Jun 28 09:54:11 2000
From: erik at src.no (Erik Enge)
Date: 28 Jun 2000 15:54:11 +0200
Subject: Are you searching for a merchant's account?
References: <20000628102843.D8AC71D086@dinsdale.python.org>
Message-ID: <m3u2edvrlo.fsf@localhost.localdomain>

"ZWang Productions" <jimclark22 at msn.com> writes:

> this mail cannot be considered spam

Right.


From johann at physics.berkeley.edu  Tue Jun  6 13:10:23 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 06 Jun 2000 10:10:23 -0700
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it>
Message-ID: <mtwvk2ohvk.fsf@astron.berkeley.edu>

Paolo Redaelli writes:

> Look at the this short program:
> =======
> #!/usr/bin/env python

> s = "0.000000E+00"
> print "Converting",s," into: ", float(s)
> print "Importing gtk"

> import GDK
> from gtk import *
> from gnome.ui import *
> import GdkImlib
> print "Converting",s," into: ", float(s)


Try the following, just to find out if float is being redefined.

>>> import GTK
>>> import gtk
>>> import gnome.ui
>>> 
>>> for module in [GTK, gtk, gnome.ui]:
...   if 'float' in dir(module): print module

and see if it comes up with anything.

I don't have this problem, but perhaps it's a gnome.ui locale issue.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From bmcd at es.co.nz  Fri Jun 16 22:05:03 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 17 Jun 2000 02:05:03 GMT
Subject: numpy bug?
References: <slrn8klh53.56e.hzhu@rocket.knowledgetrack.com> <8ieis7$4nf@stochastic.eng.umd.edu>
Message-ID: <slrn8klnbj.120.bmcd@ripley.sulaco.net>

On 16 Jun 2000 21:05:11 -0400, Warren B. Focke <moron at Glue.umd.edu> wrote:
>
>Huaiyu Zhu sez:
>[Numeric and Python integer division don't agree.]
>
>	Numeric just calls the C operators for / and %.  C leaves the
>rounding direction for integer division up to the implementor if both
>operands are not positive, but does require that (a/b)*b + a%b == a as
>long as b != 0.  So NumPy gets it right from that point of view.

The problem is that NumPy shouldn't be consistent with your C implementation
but with Python!

>	Python does specify the rounding direction, in a manner that
>has been extensively argured in this group to be correct.  I don't
>really remember the reasons at the moment, check dejanews, it's on
>there somewhere, I'm sure.

I don't recall the discussion, but the Python choice has one clear weakness:
         (-5)/4 != -(5/4)
For the new C standard (C 1999) (as opposed to the implementation dependence of
C 1989), rounding is towards zero i.e. drop any fraction part. Then:
         (-5)/4 == -(5/4)
which is more consistent with algebra.

However, I prefer the Python choice because is it invariant under a
translation, that is, you always know your rounding direction if you know the
sign of your denominator (negative for positive denominator), even if you don't
know the sign of your numerator. This also gives the lovely property that your
remainder has the same sign as your denominator (or is zero). The C99 choice
doesn't have this.

If anyone still isn't convinced that Python's way is the right way, plot, or
even just look at:

print (arange(21)-10)/2

It has  kink in it! By comparison

for i in range(21): print (i-10)/2,

has two of each value in a smooth succession. Plot it (against i), and the line
is straight.

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From jamarijr at my-deja.com  Thu Jun 29 14:41:43 2000
From: jamarijr at my-deja.com (Arinté)
Date: Thu, 29 Jun 2000 18:41:43 GMT
Subject: stderr and stdout
References: <qzI65.16987$_b3.366473@newsread1.prod.itd.earthlink.net> <m3itusjswb.fsf@greebo.nodomain.de>
Message-ID: <8jg591$n9b$1@nnrp1.deja.com>

Ok, I took care of that stuff do you know of anyway to keep the stdout
and stderr in sync.


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


From jhe at webde-ag.de  Tue Jun  6 11:57:11 2000
From: jhe at webde-ag.de (Juergen Hermann)
Date: Tue, 06 Jun 2000 16:57:11 +0100
Subject: "syncmail" in CVS repository
Message-ID: <20000606145723586.AAA450.401@hermes.cinetic.de>

Hi!

I made the following change to the syncmail script (fixes an 
ugly error msg when committing a file for the first time).

--- syncmail    2000/05/20 17:50:34     1.2
+++ syncmail    2000/06/06 14:54:40
@@ -1,4 +1,4 @@
-#! /usr/bin/python
+#! /usr/bin/env python

 """Complicated notification for CVS checkins.

@@ -79,16 +79,26 @@
     except ValueError:
         # No diff to report
         return ''
-    # This /has/ to happen in the background, otherwise we'll run into CVS
-    # lock contention.  What a crock.
-    diffcmd = '/usr/bin/cvs -f diff -C 2 -r %s -r %s %s' % (
-        oldrev, newrev, file)
-    fp = os.popen(diffcmd)
-    lines = fp.readlines()
-    sts = fp.close()
-    # ignore the error code, it always seems to be 1 :(
-##    if sts:
-##        return 'Error code %d occurred during diff\n' % (sts >> 8)
+
+    if oldrev == "NONE":
+        try:
+            lines = open(file, "rt").readlines()
+            lines.insert(0, "--- NEW FILE ---\n")
+        except:
+            # Something's gone wrong, give up
+            return ''
+    else:
+        # This /has/ to happen in the background, otherwise we'll run into CVS
+        # lock contention.  What a crock.
+        diffcmd = '/usr/bin/cvs -f diff -u -r %s -r %s %s' % (
+                oldrev, newrev, file)
+        fp = os.popen(diffcmd)
+        lines = fp.readlines()
+        sts = fp.close()
+        # ignore the error code, it always seems to be 1 :(
+        ##    if sts:
+        ##        return 'Error code %d occurred during diff\n' % (sts >> 8)
+
     if len(lines) > DIFF_TRUNCATE_IF_LARGER:
         removedlines = len(lines) - DIFF_HEAD_LINES - DIFF_TAIL_LINES
         del lines[DIFF_HEAD_LINES:-DIFF_TAIL_LINES]



Ciao, J?rgen

--
J?rgen Hermann (jhe at webde-ag.de)
WEB.DE AG, Amalienbadstr.41, D-76227 Karlsruhe
Tel.: 0721/94329-0, Fax: 0721/94329-22





From cba at mediaone.net  Fri Jun 16 19:41:29 2000
From: cba at mediaone.net (Craig Allen)
Date: Fri, 16 Jun 2000 19:41:29 -0400
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au>
Message-ID: <394ABB29.E47E03EC@mediaone.net>

Jason wrote:
> where would  i put the original text file to run this script on it? etc...
> i am running on windoze 98 at work....

Forget about exporting from Notes and then importing - go right to the source. 
Don't be put off by the reference to Zope, this is about Python and Domino.  You
can read your Domino view from Python, use the Notes field names,...
http://www.zope.org/Members/cba/Domino_and_Zope

And some day I'll try to actually connect Zope and Domino...

Craig
-- 
Craig Allen  - Managing Partner - Mutual Alchemy
Web Architecture - http://alchemy.nu



From dsavitsk at e-coli.net  Fri Jun  2 12:23:06 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 02 Jun 2000 16:23:06 GMT
Subject: Python code editable/viewable by user
Message-ID: <KbRZ4.939$gr4.39365@news1.rdc1.il.home.com>

I am looking for advice from someone who may have tried this before...or who
thinks at least that it is not such a bad idea.
i am working on a program where a user is presented a Tkinter interface.
They push a button, some stuff happens.  The tricky part (i think) is that i
want some of the even handler code to be in a text box where the user can
edit it.

esentially the program gets a long string from another source.  it then must
parse it and place various parts into various database fields. where the
string is parsed, and into what fields changes every time.

either the user has to edit the source then run the program, or run the
program and edit the source while running.  since the end users will not
necessarilly know any more python than what i tell them, i would rather hide
as much complexity as possable.  (somehow in my mind only presenting the
ADO/OLEDB part counts as hiding th complexity...hmm)

esentially this is a recursive program editing itself or python being its
own scripting language.  i would also like to save their changes, but this
seems to add an extra layer of trickyness. (perhaps just the editable stuff
will be in a separate text file that is imported)

any help would be appreciated. thanks.

doug




From hohage at muenster.de  Wed Jun 14 10:02:47 2000
From: hohage at muenster.de (hohage at muenster.de)
Date: Wed, 14 Jun 2000 14:02:47 -0000
Subject: Newbie : Add to PYTHONPATH
Message-ID: <8i83a7+kraa@eGroups.com>

Hello,
this is an absolute newbiequestion.
I'd like to know how to add a path to sys.path, but not temporarily.
Python is running on Win 98.
 Thanks.




From olczyk at interaccess.com  Wed Jun  7 17:11:14 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 07 Jun 2000 21:11:14 GMT
Subject: Calling python from C++.
Message-ID: <393eb924.99556671@nntp.interaccess.com>

Is it possible for C++ to call Python routines and vica-versa?
I have to write apps in C++ that have a GUI. They have to run under
almost any platform, so I thought that a python front end might do the
trick.
TIA.


From chibaA at TinterlogD.Tcom  Fri Jun 23 22:56:06 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Sat, 24 Jun 2000 02:56:06 GMT
Subject: simple question...
References: <3953e13a.761314922@news1.on.sympatico.ca> <3954056E.E3C18DF@bioeng.ucsd.edu> <395405c4.770668532@news1.on.sympatico.ca> <u1z1nstyf.fsf@ctwd0143.fitlinxx.com>
Message-ID: <39542339.778149139@news1.on.sympatico.ca>

Sounds promising... I'll give it a shot.

Thanks.


On 23 Jun 2000 22:18:00 -0400, David Bolen <db3l at fitlinxx.com> wrote:

>chibaA at TinterlogD.Tcom writes:
>
>> At the end of a script A, I would like to initiate script B.  I don't
>> want to run script B as a function in A; rather, I want to quit A, and
>> start script B.  Is this possible?
>
>Try the os.exec* functions in the os module.  It sounds pretty much
>like what you want to do.  The exec* functions replace the current
>running executable with another executable - only one process slot is
>involved, and the original executing code does not exist following the
>exec call.
>
>Note however that since you'll be replacing your current process, you
>actually need to exec python giving it the new script as an argument,
>rather than just directly-exec()ing the script.
>
>You can use sys.executable to determine the python that is executing
>the current script and just exec() that with an argument of the new
>script.
>
>You don't note your platform, but I believe this is documented for
>both Windows and Unix.
>
>--
>-- David
>-- 
>/-----------------------------------------------------------------------\
> \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
> /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
>\-----------------------------------------------------------------------/



From mrfx at etang.com  Mon Jun 12 04:38:39 2000
From: mrfx at etang.com (mrfx)
Date: Mon, 12 Jun 2000 16:38:39 +0800
Subject: IDLE  can be rewrited using wxPython as GUI widget?
Message-ID: <8i27cc$29gs$2@news.cz.js.cn>

The current IDLE uses TCL/TK as GUI widget.
IDLE  can be rewrited using wxPython as GUI widget?










From frankc at colconsulting.com  Tue Jun 27 06:56:43 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Tue, 27 Jun 2000 06:56:43 -0400
Subject: OO Approach to file parsing with Python?
References: <3958294F.4B17E6F7@attglobal.net>
Message-ID: <3958886B.D5B366B3@colconsulting.com>

I saw somewhere, and excuse me for not remembering, someone wrote a
Python package that allows you to express the file layout using a
modified EBNF grammar, it took care of the rest except for of course
walking the file via the Python objects. I think it's name was Parsley.

Jon McLin wrote:
> 
> Our company has decided to try an initial deployment of Python in a file-parsing
> role.  We need to translate some customer-provided structured ASCII files into
> our standard schema for use by legacy applications.    A procedural
> implementation is obvious, but completely non-reusable, and I'd like to be able
> to leverage this in the future.  (different inputs, the same outputs.)  Can
> someone point me to examples of OO approaches to this type of problem,
> preferrably implemented in Python?
> 
> Some data, and thoughts:
> 
> The basic file structure is
> 
> Multi-Line Header
> One or More DataSets
>     where each Data Set consists of
>     a single line header, and
>     one or more lines of tab-delimited data (fixed structure across all sets).
> 
> The data sets are identified by the structure of their header lines.
> 
> A nested loop processor is the obvious procedural approach, assigning each input
> field to a corresponding field in an output object.
> 
> One seemingly-OO approach is to create classes for each construct on the input
> side.  The constructors would actually parse the file, and call subordinate
> constructors (member classes) as appropriate.   (This seems awkward).
> The output classes would match our schema, and of course these would be
> reusable.
> 
> Since there is not a one-for-one correspondence between input and output in
> either classes or records, the next step would be to derive the set of  outputs
> from the inputs (quantity of output instances).    These would be constructed,
> and the input structure would be walked, telling each instance to populate it's
> output analogue.  Finally, the output hierarchy would be walked, telling each
> instance to "Publish" itself.
> 
> With this basic model, the overall structure and the output classes would be
> reusable, while the input classes and the translators (if not part of the input
> classes) would be application specific.
> 
> Conceptual Code Snippet:
> #  Parse and Create an object hierarchy
> #    MyInput becomes an object containing member objects, dictionaries, etc..
> #    SourceData is a class in an ApplicationSpecific
> MyInputObject=SourceData("MyFile")
> 
> # Translate and create output hierarchy.
> #  The custom classes will need to import the standard OutputObject module,
> which contains constructors
> #    for all output classes.
> MyOutputObject=MyInputObject.Translate()
> 
> #  Create the desired (standard) output
> MyOutputObject.Publish()
> 
> But as I noted, parsing with this model seems awkward.
> 
> Any pointers/constructive criticisms/suggestions are much appreciated.
> Non-constructive criticism will be deflected without ego damage.   After all, I
> know that I know far less than I know not, and that furthermore far more is
> unknown than is known to be unknown.   Sigh.   ;>
> 
> Thanks,
> Jon

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From dalke at acm.org  Fri Jun  2 00:04:22 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 1 Jun 2000 22:04:22 -0600
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com>
Message-ID: <8h7blm$j1b$1@slb2.atl.mindspring.net>

t_e_sanders at my-deja.com wrote:
>Do the terms of Python's distribution license require me to allow other
>users to copy and distribute my software? I would guess not, but
>inclusion of this notice in my package explicitly states so. There's
>obviously something I'm missing,

The key word is "use".

The copyright only applies to the Python code.  You are free to add
your own code to Python, which is under your own copyright.  The
clause you quoted allows you to use the Python code with your own
code, regardless of the license applied to your own code.

The GPL, which you are probably thinking of, does not allow similarly
unfettered use of their code.  Packages compiled with GPL code
restrict their use to only GPL-compatible licenses.

The inclusion of the statement in code you distribute only applies to
Python code.  Someone could copy that portion of the source from your
program and use it elsewhere.  However, since it's already available
from python.org, there nothing new revealed.

But remember, I-am-not-a-lawyer.

                    Andrew
                    dalke at acm.org





From hzhu at mars.localdomain  Sat Jun 17 19:31:17 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Sat, 17 Jun 2000 23:31:17 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com> <028001bfd817$e9dc9ca0$6401a8c0@home>
Message-ID: <slrn8ko2ok.1vs.hzhu@mars.localdomain>

On Fri, 16 Jun 2000 21:52:57 -0700, Darrell Gallion <darrell at dorb.com> wrote:
>I'm a strong believer in expanding tabs into four spaces.
>Spaces reliably render even in email.
>
>Not to mention the pain to edit a file with the tabs screwed up.
>I use a script to strip the evil tabs and my life has been better for it :-)
>

Would you mind posting the script here?  Thanks.

Huaiyu


From lohgan at my-deja.com  Mon Jun 19 04:51:04 2000
From: lohgan at my-deja.com (lohgan at my-deja.com)
Date: Mon, 19 Jun 2000 08:51:04 GMT
Subject: win32gui DialogBox
Message-ID: <8ikmtp$rvu$1@nnrp2.deja.com>

Hi all.  I'm attempting to open a dialog box on a
click event from a task bar icon.  I do not have
much experience with Win32 programming and am
experiencing unexpected results.

I load the dialog template from a resource only
DLL.  Then when I try to show the dialog box, the
function "DialogBox" blocks after handling the
message "WM_INITDIALOG".  The main and child
(dialog) windows freeze up as if the message loop
is broken.  When I double click on the task bar
icon and the WM_QUIT message gets posted, all the
blocked DialogBox functions release control.

I hope this is something stupid like simply
needing another message loop somewhere.

Any help is appreciated.  Please CC me on
responses.  Mail me personally if you wish a copy
of 'Resources.dll' and other files.

Thanks,
-Sean

---> application.py
from win32api import *
from win32gui import *
import win32con

class MainWindow:
    def __init__(self):
        self.message_map = {
            win32con.WM_DESTROY: self.OnDestroy,
            win32con.WM_USER+20 :
self.OnTaskbarNotify,
        }

        # Register the Window class.
        wc = WNDCLASS()
        hinst = wc.hInstance = GetModuleHandle
(None)
        wc.lpszClassName = "PythonTaskbarDemo"
        wc.style = win32con.CS_VREDRAW |
win32con.CS_HREDRAW;
        wc.hCursor = LoadCursor( 0,
win32con.IDC_ARROW )
        wc.hbrBackground = win32con.COLOR_WINDOW
        wc.lpfnWndProc = self.message_map # could
also specify a wndproc.
        classAtom = RegisterClass(wc)

        # Create the Window.
        style = win32con.WS_OVERLAPPED |
win32con.WS_SYSMENU
        self.hwnd = CreateWindow(
classAtom, "Taskbar Demo", style, \
                 0, 0, win32con.CW_USEDEFAULT,
win32con.CW_USEDEFAULT, \
                 0, 0, hinst, None)
        UpdateWindow(self.hwnd)
        ShowWindow(self.hwnd, win32con.SW_SHOW)

        # Add the taskbar icon
        hicon = LoadIcon(0,
win32con.IDI_APPLICATION)
        flags = NIF_ICON | NIF_MESSAGE | NIF_TIP
        nid = (self.hwnd, 0, flags,
win32con.WM_USER+20, hicon, "Python Demo")
        Shell_NotifyIcon(NIM_ADD, nid)

    def OnDestroy(self, hwnd, msg, wparam,
lparam):
        nid = (self.hwnd, 0)
        Shell_NotifyIcon(NIM_DELETE, nid)
        PostQuitMessage(0) # Terminate the app.

    def OnTaskbarNotify(self, hwnd, msg, wparam,
lparam):
        if lparam==win32con.WM_LBUTTONUP:
            print "You clicked me."
            try:
                import ui_preferences
                prefs =
ui_preferences.UIPreferences(self.hwnd)
            except:
                import sys, traceback
                traceback.print_exc
(file=sys.stdout)
            print "Should be loaded?"
        elif lparam==win32con.WM_LBUTTONDBLCLK:
            print "You double-clicked me -
goodbye"
            PostQuitMessage(0)
        return 1

---> ui_preferences.py
from win32api import *
from win32gui import *
import win32con

RESOURCE_FILE = 'Resources.dll'
ID_DIALOG = 106
ID_ICON = 101

ID_USERNAME = 1002
ID_PASSWORD = 1003
ID_CLOSE = 1000

class UIPreferences:
    def __init__(self, parentHWND):
        message_map = {
            win32con.WM_INITDIALOG:
self.InitDialog,
            win32con.WM_DESTROY: self.OnDestroy,
        }

        self.parentHWND = parentHWND
        self.res = LoadLibrary(RESOURCE_FILE)
        self.id_dlg = DialogBox(self.res,
ID_DIALOG, self.parentHWND, message_map)
        print "ID:", self.id_dlg
        print "LastError", GetLastError()

    def InitDialog(self, hwnd, msg, wparam,
lparam):
        print "InitDialog"
        return 1

    def OnDestroy(self, hwnd, msg, wparam,
lparam):
        return 1




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


From petersc at stud.ntnu.no  Fri Jun  2 21:13:16 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 03 Jun 2000 03:13:16 +0200
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de> <3938090C.9FB5D057@stud.ntnu.no> <8h9682$v4h$1@nntp9.atl.mindspring.net>
Message-ID: <39385BAC.771293BE@stud.ntnu.no>

Aahz Maruch wrote:
> 
> >As far as I understand the for loop goes on until an IndexError is
> >raised. Then the error is cleared and the for loop ends.
> 
> Let's be technically precise here: an exception is raised and caught by
> the "for" handler; the for loop is then broken (precisely as if the
> break statement was executed).  Like any other caught exception, the
> exception is thrown away (not "error is cleared").

Okay, then we will talk source. From dist/src/Python/ceval.c of the
current CVS tree (haven't checked, but should be similar with 1.5.2):

static PyObject *
loop_subscript(v, w)
	PyObject *v, *w;
{
	PySequenceMethods *sq = v->ob_type->tp_as_sequence;
	int i;
	if (sq == NULL || sq->sq_item == NULL) {
		PyErr_SetString(PyExc_TypeError, "loop over non-sequence");
		return NULL;
	}
	i = PyInt_AsLong(w);
	v = (*sq->sq_item)(v, i);
	if (v)
            return v;
	if (PyErr_ExceptionMatches(PyExc_IndexError))
		PyErr_Clear();
	return NULL;
}

Note the line PyErr_Clear() which is a call into
dist/src/Python/errors.c
where it is defined as PyErr_Restore(NULL,NULL,NULL) which just puts
NULL
values into the type, value and traceback fields. In my opinion this
(besides the name) qualifies as "error is cleared".

am-I-missing-something-obvious-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From johnvert at my-deja.com  Fri Jun  9 21:02:56 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Sat, 10 Jun 2000 01:02:56 GMT
Subject: getting a URL
References: <8hop1b$c1b$1@nnrp1.deja.com>
Message-ID: <8hs43p$scq$1@nnrp1.deja.com>

In article <8hop1b$c1b$1 at nnrp1.deja.com>,
  johnvert at my-deja.com wrote:
> Hello,
>
> As an exercise I'm trying to write a simple cgi script that will get
the
> source code of an html file and echo it to the screen.  Here's what I
> have:
>
> import cgi, urllib
>
> (snip)
>
> 		url = urllib.urlopen("http://www.server.org/foo.html")
> 		print "<pre>"
> 		print cgi.escape(url.read()),
> 		print "</pre>"
>
> But nothing is printed.  The problematic line seems to be:
>
> url = urllib.urlopen("http://www.server.org/foo.html")
>
> But the URL I gave it in the actual program is correct.  I tried a
few.
> If I run this on the shell (it's meant to be a cgi) it raises IOError
> sayin 'network unreachable'.  What could be wrong?  Also, why does
> execution stop at:
>
>       print "<pre>"
>
> Why isn't the rest (print "</pre>" among others) printed?
>
> Thanks,
>  -- John

I changed the code a little bit and ran it on the shell:

print "Content-type: text/html"
print

#form = cgi.FieldStorage()
#if form.has_key("text1"):
#	if form["text1"].value != "":
#		print cgi.escape(form["text1"].value)
url = urllib.urlopen("http://www.tuxedo.org/~esr/sitemap.html")
print "<pre>"
#		print cgi.escape(url.read()),
print "</pre>"

And I get:
$ ./test.py
Content-type: text/html

<pre>
</pre>

Thanks,
 -- John


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


From mhammond at skippinet.com.au  Fri Jun  2 00:08:28 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 02 Jun 2000 04:08:28 GMT
Subject: Getting the directory size
References: <2B1262E83448D211AE4B00A0C9D61B030130118F@msgeuro1.creo.be>
Message-ID: <0rGZ4.3331$Hz.25757@news-server.bigpond.net.au>

"Pieter Claerhout" <PClaerhout at CREO.BE> wrote in message
news:2B1262E83448D211AE4B00A0C9D61B030130118F at msgeuro1.creo.be...
> Hi all,
>
> I want to be able to find the total size of a directory on my NT
machine,
> and I started with writing it using the os.path.walk function, and it
looks
> something like this:

I needed this too.  Attached is a program I wrote some time ago that walks
a directory tree and prints the size is its files, and the total size of
all children.  Its not that different to what the Unix du command can
dump.

It was cloned from my memory of an old utility I once had called
"dirsize" - dont know the origin or history of that tho.

Hope this is of some use.

Mark.


begin 666 dirsize.py
M:6UP;W)T(&]S#0II;7!O<G0@<W1A= T*:6UP;W)T('-Y<PT*:6UP;W)T(&=L
M;V(-"@T*;6%X4')I;G1,979E;#TW(",@<WES+FUA>&EN= T*#0ID968 at 0V%L
M8U-I>F4H9&ER+"!T:&ES3&5V96P],"DZ#0H);F%M97,@/2!G;&]B+F=L;V(H
M;W,N<&%T:"YJ;VEN*&1I<BP@(BHB*2D-"@ET:&ES4VEZ92 ](#!,#0H)8VAI
M;&1R96Y3:7IE(#T@,$P-"@EF;W(@;F%M92!I;B!N86UE<SH-"@D):68@;W,N
M<&%T:"YI<V1I<BAN86UE*3H-"@D)"6-H:6QD<F5N4VEZ92 ](&-H:6QD<F5N
M4VEZ92 K($-A;&-3:7IE*&YA;64L('1H:7-,979E;"LQ*0T*"0EE;'-E. at T*
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M9W8I/3TQ. at T*"0EP(#T@;W,N9V5T8W=D*"D-"@EE;'-E. at T*"0EI9B!S>7,N
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8+2TM+2TM(BD-"@E#86QC4VEZ92AP*0T*
`
end



From aahz at netcom.com  Thu Jun  8 16:45:36 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jun 2000 20:45:36 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <393D94F8.712427E6@san.rr.com> <memo.20000608213632.55497B@digitig.compulink.co.uk>
Message-ID: <8hp0lg$oij$1@slb7.atl.mindspring.net>

In article <memo.20000608213632.55497B at digitig.compulink.co.uk>,
Tim Rowe <tim at digitig.cix.co.uk> wrote:
>
>Ah, yes, I had been told that some cultures tend to miss irony...

Irony is the UK specialty; sarcasm is the USA specialty.  See rasseff*
for more details.

* rec.arts.sf.fandom
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From 12klat at sightreader.com  Tue Jun  6 07:39:29 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 04:39:29 -0700
Subject: C extension using Numeric arrays -- do I really need all this error
References: <etdpupvdccz.fsf@w20-575-109.mit.edu>
Message-ID: <393CE2F0.74B32153@sightreader.com>



Alex wrote:

> Hi.  I wrote a simple C extension using Numeric arrays today.  It takes
> an integer array and the coordinates of an element in that array, and
> increments it by one.  The code is below, if anyone's interested.
>
> I was wondering, do I really have to be so prissy about error checking?
> Is there anything I can do to make Numeric do more of the work?
>
> Alex.

In general it is a good idea to be very prissy about error checking.  In particular, it should be
impossible to crash an extension module, or corrupt memory, no matter how hard you try from
python.

> #include "Python.h"
> #include "arrayobject.h"
> #include <assert.h>
>
> void *die (PyObject *exception, char *error_message) {
>     PyErr_SetString (exception, error_message);
>     return NULL;
> }
>
> void *die_value (char *error_message) {
>     return die (PyExc_ValueError, error_message);
> }
>
> void *die_system (char * error_message) {
>     return die (PyExc_SystemError, error_message);
> }
>
> PyObject *None () {
>     Py_INCREF (Py_None);
>     return Py_None;
> }
>
> long offset (int dimension, int *strides, PyObject *coordinates) {
>   long int answer = 0, dimension_index;
>   PyObject *coordinate;
>   assert ((PySequence_Check (coordinates)) &&
>           (PySequence_Length (coordinates) == dimension));
>   for (dimension_index = 0; dimension_index < dimension; dimension_index++) {
>     coordinate = PySequence_GetItem (coordinates, dimension_index);
>     answer += strides[dimension_index] * PyInt_AsLong (coordinate);
>   }
>   return answer;
> }
>
> void PyArrayObject_Print (PyArrayObject *array) {
>   PyObject_Print ((PyObject *)array, stdout, 0);
>   printf ("\n");
> }
>
> static PyObject *my_arrays_inc (PyObject *self, PyObject *args) {
>   PyObject *indices, *increment;
>   PyArrayObject *array;
>   int dimension_index, cell_offset, coordinate;
>
>   if (!PyArg_ParseTuple(args, "O!O", &PyArray_Type, &array, &indices)) {
>     return NULL;
>   }
>
>   if (array->descr->type_num != PyArray_INT) {
>     return die_value ("Array must be of type int");
>   }
>
>   if (!PySequence_Check (indices)) {
>     return die_value ("Second argument should be a sequence.");
>   }
>
>   for (dimension_index = 0; dimension_index < PySequence_Length (indices);
>        dimension_index++) {
>     if (!PyInt_Check (PySequence_GetItem (indices, dimension_index))) {
>       return die_value ("Second argument should be a sequence of integers.");
>     }
>   }
>
>   if (array->nd != PySequence_Length (indices)) {
>     return die_value ("Length of indices must correspond to dimension of array.");
>   }
>
>   for (dimension_index = 0; dimension_index < array->nd; dimension_index++) {
>     coordinate = PyInt_AsLong (PySequence_GetItem (indices, dimension_index));
>     if ((coordinate < 0) || (array->dimensions[dimension_index] <= coordinate))  {
>       return die (PyExc_IndexError,
>                   "Values in indices must be nonegative and less than corresp. dimensions.");
>     }
>   }
>
>
>   cell_offset = offset (array->nd, array->strides, indices);
>   (int)(array->data[cell_offset]) += 1;
>
>   return None ();
> }
>
> static PyMethodDef my_arrays_methods[] = {
>   {"inc", my_arrays_inc, METH_VARARGS},
>   {NULL, NULL} /* Marked as "Sentinel" in manual. */
> };
>
> void initmy_arrays () {
>   (void) Py_InitModule ("my_arrays", my_arrays_methods);
> }
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/7b09173a/attachment.vcf>

From gregory.p.green at boeing.com  Tue Jun 27 20:03:53 2000
From: gregory.p.green at boeing.com (Greg Green)
Date: Wed, 28 Jun 2000 00:03:53 GMT
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com> <4f265.9121$227.218658@nnrp1.uunet.ca>
Message-ID: <395940E9.4CB2FC2C@boeing.com>

Warren Postma wrote:
> 
> I find callbacks from C to python to be no problem. I use a simple
> publish/subscribe event model implemented as a C extension that has a
> dictionary with the event names as the keys (strings), and the callbacks as
> a list of Python functions.  The C code is both portable and elegant, IMHO.
> 
> Adding COM just to get callbacks would be ugly for two reasons:
> 
> 1. You already have TWO sets of  primitive types (C builtin variable types,
> and Python's native types), why mix in a third?
> 
> 2. COM on Linux is a fudge, and not a very stable one at that. There is a
> COM implementation for Linux but it's not open source, so if you have a
> problem, you can't fix it, you're stuck.
> 
> Warren

That sounds interesting. Could you expand on it a little more? Is the
publish/subscribe done in the python module or in the C extension? Or
both?

-- 

Greg Green
Advanced Design Systems
Math & Computing Technology
Boeing
(425) 865-2790


From ame at swipnet.se  Sat Jun  3 14:02:40 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Sat, 03 Jun 2000 20:02:40 +0200
Subject: Combining two graphic images
References: <959996628.860207@centauri.ezy.net.au>
Message-ID: <n3aijsoefu73lgv2h2h37750di9ckk7uh6@4ax.com>

On Sat, 3 Jun 2000 11:45:12 +1000, "A J Wilson" <ajwilson at ezy.net.au>
wrote:

>Hi
>
>I would like to combine two BMP images so that one overlays the other.  Does
>anyone know how this could be done using Python.
>

When it comes to imaging and working with images you can't beat PIL!

PIL, Python Imaging Library, created by mr Fredrik Lundh is the best
thing since sliced bread!

http://www.pythonware.com/

You may want to look up the function 'composite' 

// Anders



From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Mon Jun 26 06:24:27 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Mon, 26 Jun 2000 03:24:27 -0700
Subject: tkinter font selector?
References: <39564255.65F7C155@uniserve.com>
Message-ID: <029c54f6.0e097830@usw-ex0101-005.remarq.com>

Hi Bob,

In tkFont there is a method called families which returns a list
of installed fonts. I've created a little script below which
pops them in a combobox and changes the font of a label
according to the item selected. It would be fairly easy to to
add size and style (but I'll let you do that ;) ).

from Tkinter import *
from tkFont import families
import Pmw
root = Tk()
Pmw.initialise()

choice = None
comboentries = families(root)
lblTest = Label(root, text='Test Font',padx=20, pady=10)
lblTest.pack(expand=1, fill=BOTH, padx=8, pady=8)

def selectFont(entry):
    lblTest.configure(font=(entry,10))
combobox = Pmw.ComboBox(root, label_text='Font:', labelpos='wn',
                        listbox_width=24, dropdown=1,
                        scrolledlist_items=comboentries,
                        selectioncommand=selectFont)
combobox.pack(fill=BOTH, expand=1, padx=8, pady=8)

combobox.selectitem(comboentries[0])

root.mainloop()

Richard

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From summer2000 at cybercamps.com  Mon Jun 12 19:12:16 2000
From: summer2000 at cybercamps.com (Cybercamps2000)
Date: Mon, 12 Jun 2000 16:12:16 -0700
Subject: Summer Computer Jobs
Message-ID: <f6e15.66$g03.23694@news03.micron.net>

Summer Jobs- Teach Computers to Children

CyberCamps, the nations leading computer camp for children is still hiring
for Summer 2000. If you have skills that you can teach to children visit our
web site and fill out our on-line application at www.cybercamps.com.



We are still hiring for our camps in:

Bellevue, WA

San Diego, CA

Berkeley, CA

Santa Clara, CA

Orange, CA



This is the perfect job for computermajors that want experience on their
resume or graduates looking for temp work while they are finding a permanent
position.



Applicants must have college experience, speak English fluently, and be able
to commute to camp daily.

 All questions should be directed to Recuiting Specialist at
summer2000 at cybercamps.com.





From claird at starbase.neosoft.com  Mon Jun 19 08:35:27 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 19 Jun 2000 07:35:27 -0500
Subject: Options database (was: Define & Include?)
References: <39438F94.62A8BA37@SPAMcwcom.net> <39468730.57BFCBDE@pinebush.com> <394A88BA.41D75348@SPAMcwcom.net> <8ikot4$ki$0@216.39.170.247>
Message-ID: <299B98E693AB9A18.92CD50B08422CED4.A0A16EC40ADCCD45@lp.airnews.net>

In article <8ikot4$ki$0 at 216.39.170.247>, Dave LeBlanc <whisper at oz.net> wrote:
			.
			.
			.
>The option database mechanism might also be the 'tcl way', but aside
>from reading about it in Welsh's book, i've not used them.
			.
			.
			.
It's worse than that--all we've seen in *any* of the books
is a paragraph, or a page, or, in the case of Chris Nelson's
book, three pages of reference material and a few examples
(which is, of course, exactly on-target for Chris's format).
This bothered Allen Flick; he knows the options database can
be more useful than is obvious from these dry descriptions.

So credit Allen with doing something about it.  We're in the
middle <URL:http://www.regularexpressions.com/#options>
now of a series on the options database that we intend as
useful to *all* Tk programmers, whether they happen to be
using the Perl, Python, Ruby, Tcl, ... binding.  This is a
good time, by the way, for comments and questions folks would
like to see addressed before we wrap up the series.

I've narrowed follow-ups a bit.  Adjust as appropriate.
-- 

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


From pinard at iro.umontreal.ca  Tue Jun 27 08:28:51 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 27 Jun 2000 08:28:51 -0400
Subject: Speed with Python
In-Reply-To: Ken Seehof's message of "Tue, 20 Jun 2000 15:29:41 -0700"
References: <QH645.94$QA1.13929@news.uswest.net>
	<394FF055.47163332@sightreader.com>
Message-ID: <oq7lbbmhoc.fsf@titan.progiciels-bpi.ca>

[Ken Seehof]

> I do all GUI work in python (wxPython) and I have not noticed a difference
> in speed as compared to C++, probably because most time is spent in
> library calls anyway.

I experimented some GUI writing in Python, and initially expected that things
would be slow.  Not at all: I was very pleasantly surprised by the speed.

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





From jkraska1 at san.rr.com  Sat Jun  3 22:35:26 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 02:35:26 GMT
Subject: PERFORMANCE: list versus a "dlist"
References: <39398DC0.C04E9784@san.rr.com> <X9B0IUAJCZO5Ew98@jessikat.freeserve.co.uk>
Message-ID: <3939C115.9D4F6D3@san.rr.com>

> so how about also doing some random tests (or assuming I'm perverse) the
> worst case for your dlist ie insert and remove in the middle.

I didn't do this, because the cost will be identical to the
python list implementation, with the exception of a little
bit of extra constant time which won't be measurable. The
only anticipated data points of interest were at the head,
hence I confined my measurements to that.

If you want insert/remove in the middle, you should use my
deque implementation, which supports safe (I think) write-
on-iterate. In order to acquire safety with write-on-iterate,
I only allow one outstanding iterator at a time and lock
the deque as long as an iterator is outstanding. This
prevents (I think, heh, heh) a few situations which could
cause python to crash if mutiple threads were accessing
the deque.

In any case, assuming any normal iteration over the deque,
at any location I, insertion at I offers a complexity guarantee
of constant time. Iterators support arbitrary forward or
backward iteration, and don't have any trouble with this
because of the dual-sentinel approach ala Sedgewick's
algorithm.

The deque also implements merge in constant time, with the
caveat that it breaks a sort of implicit covenant of python
and is destructive to the argument deque (it doesn't destroy
the deque per se, but rather removes all of its elements
wholesale).

I'll probably be putting a website up for all these sources
sometime soon, but in the meantime mail me and I can send
you the whole VC98 project zipped up if interested.

I'm implementing all of the sequence and invariant sequence
methods on the dlist object at the moment, and will probably
get back to completing the deque as soon as this is done.

Do keep in mind that my deque implements random access as
O(n), which is a major drawback if you have any desire to
access your data structure that way. I hesitate to even
implement the d[i] sequence operator for this reason; it's
almost better to have an error pop up than to allow a naive
program to index the array like that.


C/


From ullrich at math.okstate.edu  Thu Jun 15 11:01:36 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 15 Jun 2000 15:01:36 GMT
Subject: TEXT MANIPULATION (in CSV format)
References: <20000615133826.11516.qmail@web1502.mail.yahoo.com> <3948E40C.2425B170@bam.com>
Message-ID: <01bfd6da$d53ca800$cccf8ad1@daves-dell>


Bill Scherer <scherbi at bam.com> wrote in article
<3948E40C.2425B170 at bam.com>...
> 
> 
> Robin Porter wrote:
> 
> > I am wondering how difficult it would be to use PYTHON to go through a
> > CSV (Comma separated value) formatted file, take two of the fields and
> > merge them into one field.
> >
> > An example would be:
> >
> > Field 1             Field 2
> >
> > "K"           ,       "1234"
> >
> > I want to change to:
> >
> > Field 1
> >
> > K1234
> >
> 
> Easy:
> 
> -----------------------
> #untested...
> import string
> 
> split = string.split
[...] 
> This will work fine if your original file is not huge.

	Except that I'm pretty sure that CSV allows commas _in_
a field if the field is enclosed in quotes. (There's probably a
good reason for CSV, I guess. Seems much simpler to me
to use a separator that does not occur in text but which does
occur on keyboards, like "|", and then to _disallow_ the
separator in fields. Then string.split works.)


From jjlucsy at concentric.net  Thu Jun 22 08:56:24 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Thu, 22 Jun 2000 08:56:24 -0400
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <JEk45.262$Sn2.1571@news-server.bigpond.net.au> <slrn8l3rpi.8ku.r2d2@mao.acc.umu.se>
Message-ID: <m0o45.6859$5f7.130092@news-east.usenetserver.com>

Try my version of python15.dll at http://www.concentric.net/~jjlucsy/

"Niklas Frykholm" <r2d2 at mao.acc.umu.se> wrote in message
news:slrn8l3rpi.8ku.r2d2 at mao.acc.umu.se...
> >> However, on the NT system, the very same command takes 4.0 s.
> >
> >   One thing to try:
> >      set path=
> >      python -c "print 'hello'"
>
> It doesn't do the trick. The running time is still 4 seconds.
>
> // Niklas




From digitig at cix.co.uk  Mon Jun  5 18:04:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 5 Jun 2000 23:04 +0100 (BST)
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <5Ys_4.5311$Hz.39657@news-server.bigpond.net.au>
Message-ID: <memo.20000605230416.25437B@digitig.compulink.co.uk>

In article <5Ys_4.5311$Hz.39657 at news-server.bigpond.net.au>, 
neilh at scintilla.org (Neil Hodgson) wrote:

>    So it is a required parameter? In the original problem statement it 
> was
> an optional parameter, which is why the name is more prominent. I agree 
> that
> the primary subject of a call may sensibly follow a different naming 
> scheme
> than a bunch of optional arguments which generally indicate 
> 'flavourings' of
> a call.

In my particular case it is required, but if the method had a default 
Canvas, making canvas an optional parameter, I'd use the same convention.

>    Mmm. Fixing up Python's lack of strictly typed parameters ;)

AIUI, Python /does/ have strictly typed parameters. It strictly checks 
that all parameters are objects :-)

More seriously, if you start making changes like that you move Python in 
the direction of being C++ (with STL). If I want that, I /use/ C++!

>    I'm thinking about moving the opposite way (I sometimes currently use
> lower case class names for class instances) and starting to write 
> aCanvas
> rather than canvas as it doesn't seem much additional verbiage to 
> remove an
> opportunity for confusion.

It's not the additional verbiage; it's the fact that /every/ parameter has 
to have the prefix, or you have to remember which parameters have, which 
haven't. You then have a naming scheme that's inconsistent with the 
standard libraries, so you have to remember where a function comes from...

I found that the "a_" convention caused /more/ confusion than the 
upper/lower case convention, and that the "opportunity for confusion" in 
the latter was entirely theoretical. Of course, you may work in a 
different way and find the opposite, but there seemed to be an opinion in 
this thread that I should be prevented from working in a way /I/ find 
reliable and effective, to no benefit that I can see. By all means have 
editors or static checkers with the /option/ of looking for this, but no 
enforcement in the language, please!


From robin at jessikat.demon.co.uk  Thu Jun  1 03:04:44 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 1 Jun 2000 08:04:44 +0100
Subject: Calling Function in DLL on Windows
References: <dThZ4.101952$fF5.3223396@news1.rdc1.il.home.com>
 <3935F38E.FF0061DC@sightreader.com>
Message-ID: <xSu2lLAMsgN5Ewy0@jessikat.demon.co.uk>

In article <3935F38E.FF0061DC at sightreader.com>, Ken Seehof
<12klat at sightreader.com> writes
>dsavitsk wrote:
>
>> i need to call a DLL from within a python program. in VB i would declare the
>> function like...
>>
>> Declare Function LoadSFW Lib "Sfwlnk32" (ByVal strPath As String, ByVal
>> strConfig As String) As Long
>>
>> then use it later like...
>>
>> intErrorCode = LoadSFW("C:\SFW", 6)
>>
>> i don't know how to do this in Python, so any quick explanation would be
>> helpful.
>>
>> doug

try Sam Rushing's windll/calldll stuff; should be findable at medusa

>I don't know of any easy way to do it.  The hard way to do it would be to
>write an extension module that wraps your DLL function(s).  If you don't
>know how to do this, read about "Extending Python".  If you don't know
>C or C++, I'm afraid you are probably out of luck until someone implements
>a generic solution....
>
>Here's an interesting project for Windows programmers: Write a general
>purpose DLL caller.  Functions should be declared in python (like in VB)
>and the DLL should be accessible from python without requiring the user
>to write any C code.  This is doable (VB does it).  The hard part will be
>converting the DLL function arguments.
>
> - Ken Seehof
>
>

-- 
Robin Becker


From neilh at scintilla.org  Fri Jun 16 10:21:50 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 16 Jun 2000 14:21:50 GMT
Subject: Ret status technique popen, popen2 (NT)
References: <3949a528.1161643@news.pipeline.com>
Message-ID: <2Kq25.372$qz4.3272@news-server.bigpond.net.au>

> Can anyone give me a pointer on how to monitor the status of win32pipe
> popen & popen2 calls? I want to get the return stat of the popen job
> and to be able to detect if the popen2 job worked/started.

  I've reimplemented popen using calldll and the bare Win32 API. Its
available from
http://www.scintilla.org/piper.zip
   The zip file contains Sam Rushing's calldll.pyd which provides access to
Win32, piper.py which implements a PipedCommand class which is roughly
equivalent to popen, and some test/example scaffolding. There is no popen2
or popen3 equivalent included although it shouldn't be too hard to implement
them. The exit status is available and the source can be modified to access
other information if needed.

   Its been tested on both Windows 95 and Windows 2000 and is based on C++
code that has been published and working for 18 months although the
translation to Python was done today. So there may be some bugs in the
translation.

   Neil




From mfein at aplcomm.jhuapl.edu  Wed Jun 28 11:11:15 2000
From: mfein at aplcomm.jhuapl.edu (Matt Feinstein)
Date: Wed, 28 Jun 2000 15:11:15 GMT
Subject: Python vs. Matlab
References: <724s6sa0k5.fsf@rogn.ii.uib.no> <slrn8kno2k.vh.hzhu@mars.localdomain>
Message-ID: <395a1247.14185848@aplnews>

On Sat, 17 Jun 2000 20:28:53 GMT, hzhu at mars.localdomain (Huaiyu Zhu)
wrote:

>Hi,
>
>There are several options you might consider:
>

<Octave, NumPy info deleted...>

>So, there also came the MatPy package which I am developing, which provides
>wrappers around the packages Numpy, Gnuplot, and SpecialFunc.  It is an open
>source project hosted on SourceForge.  It is currently classified as in
>alpha stage, but it is usable in the features that I used or tested.  I no
>longer need to use NumPy directly now and I'd claim the syntax for linear
>algebra is much easier than NumPy.  However, for additional features you
>need to either add them yourself (not difficult I'd say), or write to me and
>wait for my free time. :-) Additional developers are very welcome.

This is really interesting-- though given the high cost of MATLAB and
the many deficiencies of the MATLAB language, maybe not too
surprising. I can't prove it, but I suspect that the MATLAB language
development group is currently suffering from a bad case of
Python-envy and would take a successful development of MATLAB-like
abilities in Python -very- seriously.


--
Matt Feinstein
mfein at aplcomm.jhuapl.edu
Organizational Department of Repeated
and Unnecessary Redundancy


From darrell at dorb.com  Tue Jun 20 05:47:32 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Tue, 20 Jun 2000 02:47:32 -0700
Subject: TABs (was: Waffling between Python and Ruby)
Message-ID: <015301bfda9c$8e822370$6401a8c0@home>

 From: "Paul Jackson" <pj at sgi.com>
 >
 > |> 'Tabs will be replaced with %d spaces\n Are you sure
 > |> Y/N:'%tabSize)
 >
 > But a tab isn't 8 spaces, but rather the number of
 > spaces needed to get to the next multiple of 8 column,
 > which might be less than 8 spaces (s/8/tabSize/g, if
 > you must).
 >
Good point.
Although I haven't had a problem and I use this quite a bit.
Maybe because most Python source isn't indented on anything other than \t, 4
or 8 spaces.
After running it I do have some clean up but it's easy to see in any editor
what's needed.

 >
 > The expand/unexpand solution presented in another reply
 > look more promising, in my view.
 >
 I'm mostly on a Windows machine so utils like this are less common.
 I find my self constantly recreating common Unix tools for windows in
 Python.
 Sure I have Cygwin installed but Python runs everywhere.

 --Darrell Gallion








From dag at orion.no  Fri Jun 23 04:23:33 2000
From: dag at orion.no (Dag Sunde)
Date: Fri, 23 Jun 2000 08:23:33 GMT
Subject: Current date and time
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com><20000620010356.A892@better.net> <oq4s6n89po.fsf@titan.progiciels-bpi.ca>
Message-ID: <98F45.14715$R7.2481487@juliett.dax.net>

Hear, hear!

I fully support that one...

ISO8601 is the _only_ sensible format to use when developing
new appligations:

    1. It is accepted by more and more national standardization orgs.
    2. It is very easy for human readers to "parse"
    3. It is 100% unambigious.
    4. It is a _lot_ easier to parse and handle in code
    5. _All_ major DB's accept an ISO date and/or time without
       any need for reformatting.

IMNSHO...

Dag Sunde


"Fran?ois Pinard" <pinard at iro.umontreal.ca> wrote in message
news:oq4s6n89po.fsf at titan.progiciels-bpi.ca...
> [William Park]
>
> > >>> import time
> > >>> time.ctime(time.time())
> > 'Tue Jun 20 01:02:03 2000'
>
> Let me advocate (or rather, let Markus Kuhn) for ISO 8601.  I always
found a
> bit miserable that, by default, dates are not international.  Let me
see...
> Yes, this was sent on 2000-02-24:
>
>
> From: Markus Kuhn <Markus.Kuhn at cl.cam.ac.uk>
> Subject: [tar 1.13.17] Problems with USE_OLD_CTIME
> To: ro at TechFak.Uni-Bielefeld.DE, bug-tar at gnu.org
> Cc: pinard at IRO.UMontreal.CA, karl at cs.umb.edu
>
>
> Rainer Orth <ro at TechFak.Uni-Bielefeld.DE> write 2000-02-03:
> >
> >    I've never understood why GNU tar changed to use the ISO time
> >    format instead of the traditional ctime-based format.
>
> The two arguments brought forwards at the time that I remember were:
>
>   - ISO 8601 is considered to be acceptable in practically all
Gregorian
>     locales, while the old format isn't, as it contains English words.
A
>     locale-dependent output format would make its automatic post
processing
>     horribly difficult, which is why locale-dependent date formatting
>     should be considered with caution.
>
>   - The all-numeric bigendian nature of ISO 8601 actually makes
>     comparing dates in columns by humans significantly faster and
>     easier, so we considered ISO 8601 not only more practical for
>     i18n purposes, but also more user-friendly for English users.
>     (Experiment: try to find the oldest and newest date/time in a
>     printed 100-entry column of dates/times both when written
>     in ISO 8601 and in the old ctime format.)
>
> http://www.cl.cam.ac.uk/~mgk25/iso-time.html
>
> Markus
>
> --
> Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard
>
>
>




From kens at sightreader.com  Wed Jun 14 07:06:11 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 14 Jun 2000 04:06:11 -0700
Subject: A simple question
References: <8i5gg5$ip$1@nnrp1.deja.com>
Message-ID: <39476722.1A171AC9@sightreader.com>

I'm not too familiar with Tkinter, but I recommend you look at PythonWin
and wxPython.  They both will do a better job of getting the Windows look.
PythonWin is better at integrating with legacy Visual C++ MFC GUI code, and
wxPython is more portable.  Also PythonWin and wxWindows paradigms are
likely to be more familiar to VB programmers.  Both are free of course.

There is nothing that VB can do that can't be done with python + PythonWin
or WxPython

"Michael A. Howard" wrote:

> I am developing an application using Python/Tkinter that needs to look
> similar to a VB application (a language I am far more familiar with) as
> it will extend some functionality & we want the user interface to be as
> close as possible to the existing system.
>
> What I need to do is have a tabbed switchboard type menu screen with
> buttons for each program that the user will interact with. This has
> already been developed and went far faster than doing the same thing
> with VB.
> From the switchboard the user will select the program they want to run
> and it should open a new window for data entry/edit.
>
> None of the code examples I look at show me how to do this.
>
> Am I missing something really simple or is this task beyond the
> capabilities of Tkinter or Python?
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000614/de4a2a40/attachment.vcf>

From paul at svensson.org  Mon Jun 19 09:51:44 2000
From: paul at svensson.org (Paul Svensson)
Date: 19 Jun 2000 13:51:44 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <3948D25D.7C83CD6@micron.net> <39499FC1.8C31F7B5@sightreader.com>
Message-ID: <8il8hg$bfn$1@newsy.ifm.liu.se>

Ken Seehof <kens at sightreader.com> writes:

>LOC (lines of code) is also a measure of programmer output.  The idea is
>that you evaluate your programmers by the sheer quantity of code they
>crank out.  The one who cranks out the most lines of code per week is
>clearly the best programmer (yeah right!).  Anyway, it is sometimes used
>as a rough extimate of productivity.

Very rough;
like estimating the productivity of a welder by the amount of acetylene used.

	/Paul


From thamza.nospam at geocities.com  Thu Jun 29 11:23:38 2000
From: thamza.nospam at geocities.com (Tim Hamza)
Date: Thu, 29 Jun 2000 15:23:38 GMT
Subject: Newbie making noises (play xxxHz tone on Win32?)
Message-ID: <_RJ65.1649$7%3.135352@news.flash.net>

First Post
Newbie

I have a need to play a tone of arbitrary frequency on a Win9x system
through the soundcard.

Something like play_tone(frequency, duration)

Also I would like to be able to play multiple tones at once.

ie play_chord(freq1, freq2, ..., duration)

I look around on the Vaults of Parnassus and through the standard libraries,
etc and found nothing like this.

Anybody: Been there done that?  Ideas?

Thanks,
                Tim




From jkraska1 at san.rr.com  Fri Jun 16 21:38:15 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 17 Jun 2000 01:38:15 GMT
Subject: Monitor on unix
References: <394AC7D0.EE82B737@student.liu.se>
Message-ID: <394AD720.B3726C24@san.rr.com>

"Charles W.W." wrote:
> 
> I want to write a monitor program to monitor the users running on unix
> system.

What do you want to monitor, exactly?



C/


From db3l at fitlinxx.com  Thu Jun 15 18:26:36 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 15 Jun 2000 18:26:36 -0400
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com> <ug0qflns9.fsf@ctwd0143.fitlinxx.com> <394883C0.AF4822BB@bigfoot.com>
Message-ID: <u66rawpgj.fsf@ctwd0143.fitlinxx.com>

Steve Tregidgo <smst at bigfoot.com> writes:

> David Bolen wrote:
> > 
> [much of `verify_class' snipped]
> >
> >                 except AttributeError, TypeError:
> >                     print "Method %s not supported" % curmethod
> 
> Note that this code won't pick up the TypeError -- to do that, make a
> slight change:
> 
>     except (AttributeError, TypeError):

Argh - thanks - I shouldn't have let that slip by - I decided to use
the separate error for the type mismatch at the very end, and didn't
explicitly test that case.

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


From scarblac-spamtrap at pino.selwerd.nl  Wed Jun 28 07:14:19 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 28 Jun 2000 11:14:19 GMT
Subject: Requiring a certain version of interpreter?
References: <Lqj65.4362$h8.209230@news-west.usenetserver.com> <s2k65.1080$MZ3.2911@newsfeed.slurp.net> <slrn8ljlin.l8b.thomas@centurion.xs4all.nl>
Message-ID: <slrn8ljna9.4ni.scarblac-spamtrap@flits104-37.flits.rug.nl>

Thomas Wouters wrote in comp.lang.python:
> Python 1.6 has sys.version_info, which is a tuple with version info:
> 
> >>> sys.version_info
> (1, 6, 0, 'alpha', 2)

Hmm, mine doesn't, it seems.

> But no, there is no builtin way to 'require' a version. Most new features
> break cleanly with old versions of Python (like keyword arguments) so the
> usual remedy is to say 'Python 1.5 required!' near the top of the docs, and
> point people there when they complain it doesn't work ;-)
> 
> And for the bug-prevention scheme, you can use 'if sys.version_info[1] < 6:'
> in Python 1.6 (which will, of course, be bugfree.)

The effbot pointed out the neat:

if sys.version_info < (1, 6, 0):

once. Tuples compare dictionary-style.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
'Oook?'


From thomas at xs4all.net  Sun Jun 11 12:54:53 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 11 Jun 2000 18:54:53 +0200
Subject: 8-bit cleanliness
In-Reply-To: <20000611175536.A3365@bcnartdirecte.com>; from rcm@bcnartdirecte.com on Sun, Jun 11, 2000 at 05:55:36PM +0200
References: <39430E84.2679DD45@micron.net> <20000611175536.A3365@bcnartdirecte.com>
Message-ID: <20000611185452.N20700@xs4all.nl>

On Sun, Jun 11, 2000 at 05:55:36PM +0200, Rafael Cordones Marcos wrote:

> I just started to use Python a few days ago because I got fed up of so much convoluted Perl code. ;)
> Anyway, I have to read some text files and process the words appearing in them. I have discovered, to
> my surprise, that accents like (?, ?, ...) get replaced by a 4 character code. Is there any class/module/option
> available to read *text* files with non english characters in them?

Those non-ASCII characters do not get replaced by that 4-character code,
those non-ASCII characters *are* that 4-character code ;-) If you use repr()
on strings, non-printable characters are expressed as an octal number, to be
able to reliably reproduce them:

>>> s = "????rgh"
>>> s			# which is the same as repr(s), in interactive mode
'\341\341\341\341rgh'

>>> print s
????rgh

\341 is the 'accurate' representation of the '?' character, it'll always be
converted in the same actual character, regardless of your font settings.
How it is displayed depends on your font or your locale settings, depending
on what you use to view it ;-)

Just treat your strings as data, as you should anyway, and all will end up
fine. Just be sure not to use 'repr()' (or ``) when you really mean 'print'
or 'str()'.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From andre at beta.telenordia.se  Thu Jun  1 15:39:10 2000
From: andre at beta.telenordia.se (André Dahlqvist)
Date: Thu, 01 Jun 2000 20:39:10 +0100
Subject: OT: too-tiny fonts in Netscape
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com> <39356936.7115883@news.telus.net> <39369663.495889@news.telus.net> <FvHKxv.B5M@world.std.com>
Message-ID: <8h6akp$qg3$1@cubacola.tninet.se>

> I actually run into this problem with Netscape pretty frequenly, on my
> Solaris box at work. Is there any way to prevent this, perhaps to set an
> option for "never use a font smaller than X, no matter what the HTML
> says"? I think the version of Netscape is 4-something. Thanks.

Netscape 4.x has always been worthless when it comes to font. The
problem is of course X to begin with. Mozilla is however quiet stable these
days, and fonts are handled much better  there in my opinion.
http://www.mozilla.org

// Andr?



From quinn at mark.ugcs.caltech.edu  Thu Jun 22 23:10:23 2000
From: quinn at mark.ugcs.caltech.edu (Quinn Dunkan)
Date: 23 Jun 2000 03:10:23 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <39529090.507147FD@zip.com.au>
Message-ID: <slrn8l5leh.4pd.quinn@mark.ugcs.caltech.edu>

On Thu, 22 Jun 2000 22:17:52 +0000, Erik de Castro Lopo <erikd at zip.com.au>
wrote:
>Niklas Frykholm wrote:
>> 
>> I've been noticing a significant difference in the start-up speed of
>> Python on my NT machine and on a Linux system where I have an account.
>> 
>> On the Linux system
>> 
>>         python -c "print 'hello'"
>> 
>> times to about 0.1s.
>> 
>> However, on the NT system, the very same command takes 4.0 s.
>> 
>> Can someone explain this difference. (The processor on the Linux system is
>> certainly not 40 times faster.)
>
>It could be Linux's disk caching that gives Linux a huge win here. If 

4 seconds is too long, even for NT.  There was a thread a while back in which
some people more knowledgable about windows discussed the reasons... I think
it had to do with DLLs and things, and someone had a stripped down
python15.dll that helped out...  anyway, I con't figure out how to use deja
news any more, but if you can, search for 'startup time' and 'windows' and you
oughta find it.

Sorry I can't be less vague :)


From frankn=nuws at cs.vu.nl  Sun Jun  4 11:44:37 2000
From: frankn=nuws at cs.vu.nl (Frank Niessink)
Date: Sun, 4 Jun 2000 15:44:37 GMT
Subject: How to find out height of a Canvas
References: <memo.20000603235615.14417A@digitig.compulink.co.uk>
Message-ID: <8hdth5$8i1@cs.vu.nl>

Tim Rowe <digitig at cix.co.uk> wrote:
> How do I find out the height of a Tkinter Canvas? canvas.itemcget looks 
> promising, but it needs two arguments (I get a Tcl error if I omit the 
> second argument, or set it to None), and I can't find any documentation on 
> what the second argument should be (or the first argument, for that 
> matter; I'm assuming /that/ should be a string containing the name of the 
> attribute).

Canvas.itemcget gets the options of an item on the canvas, not of the
canvas itself. Use Canvas.config for that, e.g.:

>>> from Tkinter import *
>>> root = Tk()
>>> c = Canvas(root)
>>> c.pack()
>>> c.config('height')
('height', 'height', 'Height', '7c', '250')

The second argument of itemcget is the option you want to know about, e.g.:
>>> tag = c.create_line(0, 0, 100, 100, arrow='both')
>>> c.itemcget(tag, 'arrow')
'both'

Cheers, Frank
-- 
There is a theory which states that if ever anyone discovers exactly what the
Universe is for and why it is here, it will instantly disappear and be
replaced by something even more bizarre and inexplicable. There is another 
theory which states that this has already happened.
	-- Douglas Adams, 'The Restaurant at the End of the Universe'


From Michaell.Taylor at reis.com  Thu Jun  8 10:27:27 2000
From: Michaell.Taylor at reis.com (Michaell Taylor)
Date: Thu, 8 Jun 2000 10:27:27 -0400
Subject: Position Announcement
Message-ID: <NDBBJMANPNIHDHDJJBHMMEOGCLAA.Michaell.Taylor@reis.com>


I hope that I don't step on too many toes by using this list to announce a
job opening.  I doing so, I appeal to your interest in python advocacy.

I have  been toying with Python and reading this list over the past few
weeks.  The experience has convinced me that a good deal of our corporate
production code should be rewritten in Python (currently in a specialized
statistical scripting language that none of you has heard of). If you are
interested, or know somebody who may be interested, please take a look at
the job description and drop me a note.

I am much less interested in formal creditials as I am in finding somebody
with a hacker mentality.

**********************

Job Announcement

Junior Programmer

Reis.com is a real estate investment consultancy located in Manhattan.  We
collect and statistically/numerically analysis proprietary data on real
estate markets around the U.S.  We are in the process of reorganizing and
updating many of our statistical and numerical analysis programs and
processes.  As part of our reorganization, we are converting key pieces of
analysis code to more efficient, and platform independent, environments.

We are flexible in term of whether this position if full or part time.
Could be a good place for a student.

Responsibilities

?	Converting existing scripted analysis code to C++ and/or Python.
?	Maintaining/extending simple SQL code.
?	Assisting in the administration and extension of Linux systems.
?	Suggesting technical solutions to increase productivity in the Economics
Research Group.
?	General project development to enhance our ability to deliver econometric
analysis results quickly, accurately, and simply to clients.
?	This position will not be involved in web development.

Qualifications

?	Experience in C++.  Experience in scripting languages such as Perl and
Python preferred.  Some familiarity and/or willingness to learn VBS and SQL
would also be beneficial.
?	Flexibility to learn new scripting, programming, or shell languages.
?	Innovative character with ability to suggest technical solutions to new
problems, or to automate existing processes.
?	Self-motivated, hacker mentality.
?	Interest in mathematical and statistical analysis and programming.
?	Experience in using (preferably administrating) Linux systems.

Questions and letters of application can be forwarded to Michaell Taylor at
Michaell.Taylor at reis.com .


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

Michaell Taylor
Senior Economist, Economics Research Group, Reis.com, New York City, USA
Associate Professor, Dept. of Political Science, NTNU, Trondheim, NORWAY
Adjunct Professor, University of Durban, South Africa




From gmcm at hypernet.com  Wed Jun 28 08:48:21 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 28 Jun 2000 12:48:21 GMT
Subject: Extending python
References: <8jccdb$qoh$1@nnrp2.deja.com>
Message-ID: <8F615B5D9gmcmhypernetcom@199.171.54.154>

daniellundqvist at my-deja.com wrote in <8jccdb$qoh$1 at nnrp2.deja.com>:

>Is there a way to create python classes in C and make them behave like
>ordinary classes,can be inherited and so on?


http://www.digicool.com/releases/ExtensionClass/

- Gordon


From bhoel at server.python.net  Sun Jun  4 07:02:40 2000
From: bhoel at server.python.net (Berthold=?iso-8859-1?q?_H=F6llmann?=)
Date: 04 Jun 2000 13:02:40 +0200
Subject: Upcoming Starship changes
In-Reply-To: "Andrew M. Kuchling"'s message of "Fri, 2 Jun 2000 11:01:43 -0400 (EDT)"
References: <200006021501.LAA11161@amarok.cnri.reston.va.us>
Message-ID: <m2aeh1n1yn.fsf@pchoel.privat.uni-hamburg.de>

Hello all,

Could it be, that changing the machine broke mail forwarding again?
Mail send to bhoel at starship.python.net seems to be lost again in
nowhere. (It neither arrives at the address where it should be
forwarded to, nor do I get an error message to the sending address.)

Thanks

Berthold H?llmann

alternative email address: se6y095 at public.uni-hamburg.de
-- 
bhoel at starship.python.net / http://starship.python.net/crew/bhoel/
        It is unlawful to use this email address for unsolicited ads
        (USC Title 47 Sec.227). I will assess a US$500 charge for
        reviewing and deleting each unsolicited ad.



From toby at team-rademacher.de  Sat Jun 10 20:13:02 2000
From: toby at team-rademacher.de (Tobias Rademacher)
Date: Sun, 11 Jun 2000 02:13:02 +0200
Subject: Starting with Python
Message-ID: <8hulcc$guh$14$1@news.t-online.com>

Hi,

I threres a lot of posivte resonance about Python and therefore I decide to
learn it as far as possible.
I'm developing tools in Java and written some simple Programms in C++.
Where can I get good Tuturials about JPython?

Regards & Thx
                Tobias




From aahz at netcom.com  Sat Jun 17 17:07:23 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jun 2000 21:07:23 GMT
Subject: Python 3000 -> No, but there's 'Ruby'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org>
Message-ID: <8igpab$7tf$1@slb2.atl.mindspring.net>

In article <20000617153652.A2727 at dophnic.yi.org>,
Derek Moeller  <moeller at dophnic.yi.org> wrote:
>On Sat, Jun 17, 2000 at 07:50:27PM +0000, Aahz Maruch wrote:
>>
>> Y'know, those of you in the Ruby world who make a point of going around
>> trashing other languages are probably doing more to hurt the acceptance
>> of Ruby over the long haul than anything else.  It's *rude*.  I was
>> almost interested enough in Ruby at one point to check it out, but I
>> have to say that the Attitude I've seen so far is even worse than the
>> Perlers.
>
>We seem to harp on the shortcomings of C, C++, Perl, Ruby, and a few
>other languages in c.l.py. Seeing as how this fellow mentioned some
>things that he felt he didn't like about Python, and what he did like
>about Ruby, I can't see the large difference between that and Pythoners
>spieling about their increased productivity compared to C++/C, the
>hairiness of Perl, whatever Ruby's got against it, etc.

It's possible that I'm oversensitive to Ruby posts right now due to the
spammed ads that have popped up over the last few weeks.  OTOH, it seems
to me that there's a fundamental difference between extolling the
virtues of Python in a Python newsgroup and extolling the virtues of
Ruby in a Python newsgroup.

>Now, I can't say I agree with posting that bit about Ruby as a follow-up to
>mentioning Python-3k, as this was not what the original poster asked for.

Yes, and that's exactly the kind of thing some Ruby people have been
doing for a while.  I'd feel exactly the same if someone posted a
message requesting help with text processing and another person followed
up with a message saying that Python would never be good at text
processing and Perl was the only way to go.  That goes triple if there
were a pattern of it.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From cut_me_out at hotmail.com  Sun Jun 11 13:29:12 2000
From: cut_me_out at hotmail.com (Alex)
Date: 11 Jun 2000 13:29:12 -0400
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil>
Message-ID: <etd4s7015zr.fsf@w20-575-109.mit.edu>

> Do keep digging. Ref-counting is important.

Ref-counting errors can be such a pain in the neck to track down that I
have taken to placing assertions at the end of routines to make sure
the ref-counts are what they ought to be.  I'll probably grow out of it
as I get the hang of it, though.

This is the sort of thing I've been doing:

int dict_add_pattern (PyObject *dict, char *string, int length, long pos[DEPTH]) {
  PyObject *pattern, *coords, *value, *tuple;
  int return_code, pattern_existed;

  assert (length <= DEPTH);

  coords = dict_set_coords (length - 1, pos);
  if (!coords) {goto fail;}
  
  pattern = Py_BuildValue ("s#N", string, length, coords);
  if (!pattern) {Py_DECREF (coords); goto fail;}

  value = PyDict_GetItem(dict, pattern);
  if (!PyDict_GetItem(dict, pattern)) {
    pattern_existed = 0;
    value = PyInt_FromLong( 1 );
    return_code = PyDict_SetItem(dict, pattern, value);
  } else {
    pattern_existed = 1;
    assert (PyInt_Check( value ));
    value = PyNumber_Add( value, PyInt_FromLong( 1 ));
    if (value == NULL) {goto fail;}
    return_code = PyDict_SetItem(dict, pattern, value);
  }
  /* If a tuple equal to pattern was already in dict, then pattern will not be
     referred to by dict, and should have a refcnt of 0.  If there was no such
     tuple, then PyDict_SetItem increased the refcnt of dict by one.  In
     either case, the refcnt has to be decremented, now.
     */
  Py_DECREF (pattern);
      
  if (return_code == -1) {
    goto fail;
  } else {
    if (DEBUG_REFCOUNTS) {
      printf ("coords: %i\npattern: %i\nvalue: %i\n",
	      coords->ob_refcnt, pattern->ob_refcnt, value->ob_refcnt);
    }
    if (DEBUG_LIST_PATTERNS) {
      PyObject_Print (dict, stdout, 0); printf ("\n");
    }
    goto pass;
  }

fail:
  return 0;
pass:
  assert (pattern->ob_refcnt == (!pattern_existed));
  assert ((coords->ob_refcnt == (!pattern_existed)) | (length == 1));
  return 1;
}


From root at rainerdeyke.com  Mon Jun  5 19:51:53 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 05 Jun 2000 23:51:53 GMT
Subject: indirect references
References: <8hha1n$phl$1@nnrp1.deja.com>
Message-ID: <t2X_4.6212$r7.1362105@news1.rdc2.tx.home.com>

<shaleh at debian.org> wrote in message news:8hha1n$phl$1 at nnrp1.deja.com...
> How do I do this in python:
>
> foo = 'string'
> bar = foo
>
> somehow assign to bar and change the value of foo.

This is probably the closest you'll get:

class Reference:
  def __init__(self, it):
    self.target = it

foo = Reference('string')
bar = foo
bar.target = 'another string'
print foo.target


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From thomas at xs4all.net  Sun Jun  4 17:06:08 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 4 Jun 2000 23:06:08 +0200
Subject: Questions regarding design decisions in listobject.c
In-Reply-To: <393ABA63.18E2F46@san.rr.com>; from jkraska1@san.rr.com on Sun, Jun 04, 2000 at 08:19:10PM +0000
References: <393ABA63.18E2F46@san.rr.com>
Message-ID: <20000604230608.U469@xs4all.nl>

On Sun, Jun 04, 2000 at 08:19:10PM +0000, Courageous wrote:

> 1. For slice operations, ilow/ihigh indices which undeflow
> or overflow the list in question are reset to the edges
> of the list. Why was it that PyExc_IndexError wasn't
> thrown instead?

I believe this was done to avoid having to use a lot of len() and 'or''s
inside indexes. If you want the first ten characters, or less, of a string,
currently you do:

s[:10]

If slicing raised IndexError, you'd have to do:

s[:(len(s) > 10 and 10 or len(s))] 

or wrap the entire thing in an if/else:

slen = len(s)
if slen < 10:
	substring = s[:slen]
else:
	substring = s[:10]

I believe it was decided, for the sake of simplicity, to make slices
'special' in that regard. It also fits fairly well with the 'gaps' thing
(see below), as a sequence of length 10 might not have items beyond the 10th
one, but it does have gaps, all the way into infinity ;) 

However, I Aint Guido, and I dont presume to channel Guido, so I may be
completely wrong.

> 2. In slice expressions of the form s[i:j], what was the
> design decision that lead to s[i:i] always returning an
> empty list and s[i:i+1] selecting one item. At first
> glance, it would appear that s[i:i] is the obvious and
> intuitive correct expression to return just one item on
> a slice, where s[i:i+1] ought to select two items and so
> forth. Was there some particular situation which arose
> which required this slightly counterintuitive idiom?

Actually, it's very intuitive, if you see the start and end index of the
slice not as items, but as the gaps between the items. In s[0:1] 0 is the
gap before the first item, 1 the gap between the first and the second item,
and so s[0:1] captures exactly one item, the first. This is a fairly
frequently asked question, by the way, but I can't find it in the FAQ...
Anyone know if it's in there or not, and if not, maybe someone can add it ?
I'm still wrestling with the overseas payment :P

Creditcard-less-European-ly yr's,
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jmwarfe at my-deja.com  Tue Jun 20 13:16:21 2000
From: jmwarfe at my-deja.com (jmwarfe at my-deja.com)
Date: Tue, 20 Jun 2000 17:16:21 GMT
Subject: Using Python to change Directory ACL's in NT
Message-ID: <8io8se$hjq$1@nnrp1.deja.com>


Hello, I have been using Python to do administrative tasks on my NT PDC.

Is there a Win32 module that I can use to set directory privileges, that

has the functionality of cacls.exe using python? I am currently using
cacls to set the privileges by doing a os.system command in my script.
Thanks for the help in advance!


Mike Warfe


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


From harry at infoscientific.com  Tue Jun 27 11:14:38 2000
From: harry at infoscientific.com (Harry Reed)
Date: Tue, 27 Jun 2000 08:14:38 -0700
Subject: ActiveX scripting and Python
Message-ID: <3958c4f6$0$4576@wodc7nh6.news.uu.net>

Hi, All!
    I am attempting to use the M$ ActiveX scripting with Python (under VB6)
to add scripting to an application. I've succeded in executing small Python
scripts via the ActiveX Scripting stuff included in the Python-1.5.2 +
win32all-132 distributions and things seem to, generally, work OK (w/ the
ActiveX scripting demo working under ie5)
.
    The problem I am having is that 1) how does one cause a Python script to
return a value under the ActiveX Scripting engine? When I try something like

    a = 1
    b = 2
    return a + b

I get a nasty message/error box from VB saying:

    Run-time error '-2147467259 (80004005)'
    Unexpected Python Error: pywintypes.com_error: (-2147467263, 'Not
implemented', None, None)'

and 2) when I try to define a simple function abc as shown:

def abc():
    return 1

I get the same nasty message.

Any ideas?

Thanks,
Harry Reed
doon at infoscientific.com





From pauldubois at home.com  Fri Jun 30 16:52:25 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Fri, 30 Jun 2000 20:52:25 GMT
Subject: New: Easier URL for Numerical and other Python-related sites
Message-ID: <dM775.17183$i5.179904@news1.frmt1.sfba.home.com>

Upgrades at SourceForge allow a new way to address the "Project Page" for a
given project. You can now go directly to the project pages for the
following projects rather than going through their home page.  Since there
is a link to the home page on the project page I believe most people will
want to do this. Some sites are:

http://sourceforge.net/projects/numpy -- Numerical Python
http://sourceforge.net/projects/pyfortran -- Pyfort
http://sourceforge.net/projects/cxx -- CXX

The home pages for projects are still projectname.sourceforge.net, e.g.
numpy.sourceforge.net.







From jkraska1 at san.rr.com  Sun Jun  4 00:03:08 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 04:03:08 GMT
Subject: Um? *boggle* Silly question about listobject.c
Message-ID: <3939D5A3.2B56E309@san.rr.com>

So why is it that most/all the list operations
(for example, list_item) are implemented with
int and not long? Was it forseen that nobody
would ever desire to index a list with more than
maxint entries?

Am I missing something or is this a critical
oversight in the definition of sequences in python?

????

C/


From rupole at compaq.net  Sun Jun 11 00:03:47 2000
From: rupole at compaq.net (Roger Upole)
Date: Sun, 11 Jun 2000 04:03:47 GMT
Subject: PythonWin and collections
References: <8hv12v$ol1$1@nnrp1.deja.com>
Message-ID: <DcE05.23269$wp.1594324@pouncer.easynews.com>

win32net.NetGroupGetUsers does exactly what you want.  It returns a
list of dictionaries containing info for users in the group.
        Roger Upole

<barryhart at my-deja.com> wrote in message news:8hv12v$ol1$1 at nnrp1.deja.com...
> I'm trying to port the following fairly simple code to Python. It uses
> ADSI to get a list of members in an NT group.
>
>     For Each member In GetObject("WinNT://usa/Developers").Members
>         Debug.Print member.ADsPath
>     Next
>
> I'm a fairly new Python user, and so far I've never figured out how to
> enumerate a COM collection object properly. In previous cases, I've
> been able to work around this by using the "Count" and "Item" methods
> to index over the collection, but this particular collection doesn't
> expose Count and Item methods.
>
> Here's my Python attempt at the above code:
>
>     from win32com.client import GetObject
>
>     for member in GetObject("WinNT://usa/Developers").Members:
>         print member.ADsPath
>
> The "for" loop fails with "TypeError: loop over non-sequence". I also
> tried adding "._NewEnum()" after the ".Members", but this fails
> with "AttributeError: _NewEnum".
>
> Help!
>
> -----------------------------
> Barry Hart
> Velant, Inc.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From wware at world.std.com  Sun Jun  4 17:15:42 2000
From: wware at world.std.com (Will Ware)
Date: Sun, 4 Jun 2000 21:15:42 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <slrn8jk9pa.mfl.andyl@azaal.dircon.co.uk>
Message-ID: <FvnF26.I1J@world.std.com>

Andy Leighton (andyl at azaal.dircon.co.uk) wrote:
> Well LispMe! is pretty good too.  I would provide a URL but I don't have
> it to hand.

Try http://www.lispme.de/lispme/

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From wjdandreta at worldnet.att.net  Thu Jun 22 11:27:45 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 22 Jun 2000 15:27:45 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net> <DXk45.79957$WS3.716413@typhoon.we.rr.com>
Message-ID: <Rfq45.15293$C44.879496@bgtnsc05-news.ops.worldnet.att.net>

Hi Stephen,

>I have absolutely *no* idea if it will run with Python 1.2, or under DOS,
>but when I was looking for database support awhile back, someone pointed me
>to MetaKit, and I absolutely *love* it...


I checked the website for MetaKit. There are no DOS binaries and I searched
the site for DOS but nothing useful showed up so I lost interest.

Bill




From porter at et.byu.edu  Thu Jun 29 13:06:00 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Thu, 29 Jun 2000 11:06:00 -0600 (MDT)
Subject: Detecting Window Resizing
Message-ID: <Pine.GHP.4.21.0006291102470.9839-100000@don.et.byu.edu>

Is it possible in python to detect when the virtual console that a curses
program is running in is re-sized.  My program checks to see the size of
the window to know how much it can draw, but it still gives me a
segmentation fault if I shrink the window down.  I am running this on
linux.

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------




From matthewm at zebrasoft.co.nz  Sat Jun 24 20:09:58 2000
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Sun, 25 Jun 2000 12:09:58 +1200
Subject: Serial port IO Python module?
References: <39554BAE.6C00C44E@es.co.nz>
Message-ID: <39554DD6.C9B45E88@zebrasoft.co.nz>

OK, OK, so I should have checked the ng before posting. the sio-151 is a
pretty good fit for what's needed. Out of interest though, anything else
out there?

Matthew Miller wrote:
> 
> Hi everyone,
> 
> Is there a python module for RS-232 comms? Thanks. Matthew.


From shapr at uab.edu  Thu Jun  8 08:10:30 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 08 Jun 2000 12:10:30 GMT
Subject: GUI editor
References: <393f4568$0$25329@wodc7nh0.news.uu.net> <vskujs867ghjso3aqi2r38n5mei0bg4lhq@4ax.com>
Message-ID: <393F8D02.9A87A6EF@uab.edu>

Anders M Eriksson wrote:

> On Thu, 8 Jun 2000 01:59:37 -0000, "Blu Dragoon"
> <eddthompson at ssi.parlorcity.com> wrote:
> 
> >Is there a gui editor written in python for tkinter?
> >The reason I ask is that I use python on 6 different platforms.
> >And tkinter seems to be the most prevailant of the gui's.
> >Or should I go ahead a take the PIL? and is there any gui editor
> >for PIL too.

> There is a 'pre-alpha' version for wxWindows, Boa Constrictor
> https://sourceforge.net/project/?group_id=1909
> 
> There are alot of 'pre-alpha' versions of GUI's and editors but non
> that works.... (please correct me on this!)

I'm confused. Are you talking about an editor for building GUIs, or are
you talking about IDLE/PythonWin/etc.?

> If you find any IDE/GUI editor that you like Please let me know! I
> also searching...

If you just want a nice way to edit Python source, check out IDLE. If
you look in your python install directory, it should be under
Tools/idle/
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From fstajano at uk.research.att.com  Wed Jun 28 12:05:36 2000
From: fstajano at uk.research.att.com (Frank Stajano)
Date: Wed, 28 Jun 2000 16:05:36 GMT
Subject: Q: memory ownership in python extensions
References: <3958e1ee.13405033@localhost>
Message-ID: <395a2179.23332871@localhost>

On Tue, 27 Jun 2000 17:34:25 GMT, fstajano at uk.research.att.com (Frank
Stajano) wrote:

>In the C code, I malloc some memory, perform some processing that
>writes a string in it, then turn this C string into a Python string
>via Py_BuildValue. The question is: Who is responsible for freeing
>that memory?

The answer comes from the NEXT version of the docs, quoted below
(thanks Fred) ;-)

When memory buffers are passed as parameters to supply data to build
objects, as for the \samp{s} and \samp{s#} formats, the required data
is copied.  Buffers provided by the caller are never referenced by the
objects created by \cfunction{Py_BuildValue()}.


From thomasm at legend.co.uk  Mon Jun  5 07:09:16 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Mon, 5 Jun 2000 12:09:16 +0100
Subject: Zope compatible with Windows 2000?
References: <393B485F.BEE013FF@micron.net>
Message-ID: <393b8a00.0@katana.legend.co.uk>

> Forgot to ask, does Zope have a version that runs well with Windows 2000
> server?

I had no problem at all ..




From mwh21 at cam.ac.uk  Sun Jun  4 15:20:32 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 04 Jun 2000 20:20:32 +0100
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk> <393A3A9A.FB9F2E37@stud.ntnu.no> <393A4C60.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <m3d7lxxngf.fsf@atrus.jesus.cam.ac.uk>

"Bablos" <angband at blueyonder.co.uk> writes:

> On Sun, 04 Jun 2000 13:16:42 +0200, Peter Schneider-Kamp wrote;
> > Bablos wrote:
> > > 
> > > Is there a way that I can get Python to print this all on the same
> > > line, so that it looks as if the line is being updated constantly, as
> > > opposed to having a ridiculously long list of 'uploaded...' lines?
> > 
> > For unix you could just write:
> > 
> > print "\ruploaded",x,"bytes",
> 
> This is buffered on the Amiga, I assume.  I see very infrequent
> updates, so I assume it saves up a large number of print statements
> and then flushes them when it reaches a certain size.  Irmen?
> 
> But hey, at least '\v' works for me.  If it's all so very platform
> dependant, surely it would be wise to standardise it somehow to
> maintain some form of standard method?

Because I suspect the behaviour common to all the platforms that
Python runs on is so restricted as to be useless.

Can curses do Win32?  There's slang too, isn't there?

character-mode-neophyte-ly y'rs
M.

-- 
  In many ways, it's a dull language, borrowing solid old concepts
  from many other languages & styles:  boring syntax, unsurprising
  semantics, few  automatic coercions, etc etc.  But that's one of
  the things I like about it.                 -- Tim Peters, 16 Sep 93


From hungjunglu at hotmail.com  Wed Jun 14 04:59:24 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Wed, 14 Jun 2000 01:59:24 PDT
Subject: SV: Python Productivity over C++
Message-ID: <20000614085924.80042.qmail@hotmail.com>

>From: Michal Vitecek <M.Vitecek at sh.cvut.cz>
> >    if ( a = getNumber() ) {...}
>  i have no idea why you call this a stupid feature - IMO this all makes
>  the program shorter and as such more readable. this is really something i
>  hate about python, missing ++ and -- operators plus having to do the
>  assignment stupidly before if comment - more & longer source.

Hey, I can recommend you a great language for shorter and notations: Perl. 
:) You've got all your lovely operators there. You are welcome to use it 
instead of Python. :) Why use Python if you've got Perl? :)

One thing we have learned from Python and Perl, it's that shorter notations 
are not better. Proper longer notations (not those exaggerated notations 
like in Cobol) are almost self documenting.

Your experience with C++ certainly is highly limited. Do you know that 
Metrowerk compiler by default issues a warning when it encounters 
assignments inside an "if" statement? Now, are you going to tell people in 
Metrowerk that they are dumb, too? :)

> >C++ programmer get bitten all the time with the == && || operators, and
> >refuse to acknowledge that there is a problem.
>
>    :) maybe C++ programmers like you?

Wow wow, someone trying to lecture me about C++, you've got some guts. I 
pushed out quite a few products distributed worldwide, managed teams of 
programmers with several Ph.D. degree holders. Dear sir, what have you done 
in your life?

>i personally never had any problem with ==, && and ||,

You are a newbie, and it shows. :)

regards,

Hung Jung

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From thomas at xs4all.net  Tue Jun 20 07:28:57 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 20 Jun 2000 13:28:57 +0200
Subject: converting hex number contained in a string to an integer
In-Reply-To: <8inhd0$7s$1@news.kolumbus.fi>; from torppa@polykoira.megabaud.fi on Tue, Jun 20, 2000 at 10:35:12AM +0000
References: <8ing7d$o6r$1@nnrp2.deja.com> <8inhd0$7s$1@news.kolumbus.fi>
Message-ID: <20000620132857.E26436@xs4all.nl>

On Tue, Jun 20, 2000 at 10:35:12AM +0000, Jarkko Torppa wrote:

> >I have a string containing a number. This is an
> >hex number. How can I convert this to an integer?

> >How can I do this? I am sure it is really easy,
> >but I am just beginnig with
> >Python.

> # Forces base to 16(hex)
> >>> string.atoi('32',16)
> 50
> # Guesses base from leading characters
> >>> string.atoi('0x32',0)
> 50

Note that in Python 1.6, you can do the above two things like this:

>>> int('0x32',16)
50
>>> int('0x32',0)
50

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From joseph at src.no  Fri Jun  2 12:29:52 2000
From: joseph at src.no (Joseph Santaniello)
Date: Fri, 2 Jun 2000 18:29:52 +0200
Subject: Currency format for floats?
In-Reply-To: <8h7ar6$nb8$1@slb6.atl.mindspring.net>
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no> <3936EC02.60768A66@sightreader.com> <Pine.LNX.4.21.0006020157430.25547-100000@lade.src.no> <8h7ar6$nb8$1@slb6.atl.mindspring.net>
Message-ID: <Pine.LNX.4.21.0006021828550.13804-100000@lade.src.no>

Nice!

I gave up on a locale way, but this is just what I needed. Thanks
everybody!

Joe

On Thu, 1 Jun 2000, Andrew Dalke wrote:

> 
> Joseph Santaniello wrote:
> >Ah, yes. But I'm still missing my commas [for 23,342.40]... A nice
> > clean way to do that?
> 
> I think the following is only somewhat tricky, but perhaps the
> cleanest.  (Okay, there's also supposed to be a way to do this with
> locale, but I don't know how.)
> 
> def dollar(f):
>   if f < 0:     # pull the sign out now so I know where to stop
>     sign = "-"  # when going bacwards
>     f = -f
>   else:
>     sign = ""
> 
>   s = "%.2f" % f
>   # -6 is the thousand's place  --  987654.21
>   # -3 goes back 1000 at a time
>   # stop at 0 instead of -1 so "999.99" doesn't lead with a ",".
>   for i in range(len(s)-6, 0, -3):  # -6 is the first thousands place
>     s = s[:i] + "," + s[i:]         # Go back 1000 at a time
>   return sign + s
> 
> >>> for i in (0, .9, 1, 10, 99.99, 100.00, 100.0, 999.99, 1000.,
>   1000.00, 1000000000L, -0.01, -0.9, -1, -99.9, -100, -999.99, -1000.00,
>   -123456789):
> ...  print i, dollar(i)
> ...
> 0 0.00
> 0.9 0.90
> 1 1.00
> 10 10.00
> 99.99 99.99
> 100.0 100.00
> 100.0 100.00
> 999.99 999.99
> 1000.0 1,000.00
> 1000.0 1,000.00
> 1000000000L 1,000,000,000.00
> -0.01 -0.01
> -0.9 -0.90
> -1 -1.00
> -99.9 -99.90
> -100 -100.00
> -999.99 -999.99
> -1000.0 -1,000.00
> -123456789 -123,456,789.00
> 
>                     Andrew
>                     dalke at acm.org
> 
> 
> 
> 



From matt at virtualspectator.com  Wed Jun  7 01:21:24 2000
From: matt at virtualspectator.com (matt)
Date: Wed, 7 Jun 2000 17:21:24 +1200
Subject: win python and graphics
In-Reply-To: <Bah%4.22817$Rq.2324142@pouncer.easynews.com>
References: <393692FB.8610EBCA@dallas.net> <393D7BFD.F50F9932@sightreader.com> <Bah%4.22817$Rq.2324142@pouncer.easynews.com>
Message-ID: <0006071724260A.00959@localhost.localdomain>

Hi, I pretty much only use Linux, but have to do some things on windows.  In
particular I need to take screen shots of the desktop or particular windows.  I
would like do this using python rather than in borland C++.  Can anyone help out
here ?

regards
Matt



From msoulier at nortelnetworks.com  Tue Jun  6 13:51:32 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 6 Jun 2000 17:51:32 GMT
Subject: popen in real time?
Message-ID: <8hjdn4$6kf$1@bmerhc5e.ca.nortel.com>

	Hey guys. I realize that you can loop on the output of popen if you want
to read from a pipe, but it would seem from experience that the output of the
child process is not available until the process exits. I've tried this with
the output of long-running processes like find. Is there a way to get the
output of a child in real time, as it's printed?

	Mike


-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From moshez at math.huji.ac.il  Mon Jun 12 13:10:48 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 12 Jun 2000 20:10:48 +0300 (IDT)
Subject: Testing for a file in python?
In-Reply-To: <39450724.231994953@nntp.interaccess.com>
Message-ID: <Pine.GSO.4.10.10006122010090.15238-100000@sundial>

On Mon, 12 Jun 2000, Thaddeus L. Olczyk wrote:

> How would you go about testing if a certain file exists in python?
> In most scripting languages you have the -f test:

I don't know what you mean by "most".

> stat doesn't work, it aborts if the file does not exist.

Why not use os.path.isfile()?

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From svenk at tzi.de  Tue Jun  6 15:30:15 2000
From: svenk at tzi.de (Sven Kuenzler)
Date: Tue, 6 Jun 2000 21:30:15 +0200
Subject: PyXML build error
Message-ID: <960319815.1685255680@News.CIS.DFN.de>

Hi,

building PyXML 0.5.5, I get the following error message

extensions/pyexpat.c: In function `newxmlparseobject':
extensions/pyexpat.c:474: parse error before `xmlparseobject'                   

I am using SuSE linux with python 1.52 build from source tarball and Distutils
0.8.2.

What's wrong?

    Sven.....

 -- 
We must admit that looking at progress bars does not qualify as working
	-- Apocalypse Geek



From paul.magwene at yale.edu  Thu Jun 15 14:39:27 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Thu, 15 Jun 2000 14:39:27 -0400
Subject: tkinter vs. Perl/Tk
References: <vg925.1551$Qf6.60136@nuq-read.news.verio.net>
Message-ID: <394922DF.E2B16473@yale.edu>

Paul Wilkinson wrote:
> 
> How does tkinter compare to Perl's implementation of Tk?  I'm a little
> frustrated with Perl/Tk - I have strange focus problems that I've never
> managed to tackle.  I never did manage to find any decent Perl/Tk
> documentation.  Will I be happier with Python?

I like Tkinter (Pythons interface to Tk) a lot.  And if you're looking
for good documentation than you're in luck - check out John Grayson's
book "Python and Tkinter Programming" -- highly recommended!



Paul Magwene			Postdoctoral Fellow	
paul.magwene at yale.edu		Dept. of Ecology & Evolutionary Bio.
				Yale University


From root at rainerdeyke.com  Tue Jun  6 15:44:03 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 06 Jun 2000 19:44:03 GMT
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <7wc%4.7928$r7.1560935@news1.rdc2.tx.home.com>

Richard P. Muller <rpm at wag.caltech.edu> wrote in message
news:393D2797.4E4EE7CC at wag.caltech.edu...
> What else should we be doing to make Python *the* programming language
> rather than *a* programming language?

Write an optimizing compiler that increase performance by a factor of at
least 100 and decreases memory usage by a factor of at least 100.  Python in
its current incarnation is too inefficient to be useful as primary language
for shrink-wrapped software.


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From m.faassen at vet.uu.nl  Wed Jun  7 19:51:42 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 23:51:42 GMT
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development  Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain> <393E84B6.D80CD9D2@ecs.soton.ac.uk> <8hm0pq$lc1$1@slb1.atl.mindspring.net>
Message-ID: <8hmn6e$ets$3@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <393E84B6.D80CD9D2 at ecs.soton.ac.uk>,
> Jacek Generowicz  <jmg at ecs.soton.ac.uk> wrote:
>>
>>De acuerdo, let's all revert to English !

> I'll say this: despite being an Ugly American, I do appreciate the
> extent to which comp.lang.python is much more internationally oriented
> than most newsgroups.

They only let in the Really Brilliant Furrners, who speak 50 languages.
Excluding me; I'm the exception here.. :)

Regards,

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


From mkx at excite.com  Mon Jun 12 01:07:02 2000
From: mkx at excite.com (mkx at excite.com)
Date: Mon, 12 Jun 2000 01:07:02 -0400
Subject: Does Python help with the no-Unix handicap?
Message-ID: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>

In many Perl vs. "other scripting language" discussions, I have often
seen Perlers state that those who dislike Perl's syntax and structure
are often non-Unix types, and thus are not used to common Unix
concepts. And, since Perl was born as a Unix admin tool, us "Windows
types" will likely be lost.

Well, I freely admit that I have little-to-no Unix/Linux experience,
and that I primarily use Windows in my profession, and even at home.
Since there is no 12-step program for this condition yet, I must push
on. 

After learning and experimenting with Perl on WinNT for a while, I
began to understand the concept stated above. While I could understand
most of the examples I came across, sometimes that understanding came
with great effort. Any increase in productivity using Perl in place of
some my other programming tools would only come after a great deal of
effort and trial-and-error educaton.

So, I come now to Python. After reading much about it's ease of use
and readability advantages, I see it may be a better alternative. In
most cases, I can actually get a concept of what the code is doing,
without picking up a reference book - after only a week of
experimentation.

So my question is this: Is Python a more "Windows-friendly" language?
Is it's base structure less related to any one specific operating
system? Is learning Python an exercise in learning the language, and
not a Unix tutorial? Can I be successful even if I do not use Unix ;)
?

[No offense to the Perl-types. Just looking for opinions.]


From s2mdalle at titan.vcu.edu  Sat Jun 17 23:37:34 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Sat, 17 Jun 2000 22:37:34 -0500
Subject: Overriding 'select'
Message-ID: <8ihgad$2qj$1@bob.news.rcn.net>

Hello,

I'm relatively new to python, and I've come up with a solution to
one of my problems that is not exactly optimal.  I was hoping to 
solicit advice on what might be better or 'prettier'.

The problem is this.  I've got a Connection object.  Each instance
of a Connection object has a field that is a socket.  The server
that uses these Connection objects wants to call select() on 
Connection objects, not their corresponding sockets.  This is
because the Connection object providesa bunch of methods 
specific to each connection, and holds a lot of data for the
server.  

But when implementing select() inside the Connection object,
I come across the problem - how to use the select() call from
the select module on the socket of the Connection, and then
match it up with the right object later on?

I first wanted to use circular references which didn't work
since sockets seem to be read only.  Say I had an instance
foobar with foobar.sock being a socket object.  I thought
I could do:

foobar.socket._circular_reference = foobar

so that when the sockets came out the other end of select()
I could just check that field for their corresponding object, but
that didn't work.  So I ended up writing this:

(the obj.socket() method returns the socket field from that object)

    def select(self, readers, writers, errors, timeout=0):
        # Do this with hashes instead of circular pointers.
        # make a hash of all the socket objects which hash
        # to their corresponding parents
        # when select returns, grab the hash values
        refhash = {}

        reader_sockets = []
        writer_sockets = []
        errors_sockets = []
    
        for r in readers:
            sock = r.socket()
            # The socket's value in the dictionary is the
            # object the socket belongs to for backreferencing.
            refhash[sock] = r
            if sock:
                reader_sockets.append(sock)
        for w in writers:
            sock = w.socket()
            refhash[sock] = w
            if sock:
                writer_sockets.append(sock)
        for e in errors:            sock = e.socket()
            refhash[sock] = e
            if sock:
                errors_sockets.append(sock)

        r, w, e = select.select(reader_sockets, writer_sockets,
                                errors_sockets, 10)

        rreturn = []
        wreturn = []
        ereturn = []

        # Find out which object each socket belongs to

        for x in r:
            rreturn.append(refhash[x])
        for x in w:
            wreturn.append(refhash[x])
        for x in e:
            ereturn.append(refhash[x])

        return rreturn, wreturn, ereturn


This isn't TOO bad...but it's ugly, and not as fast as
circular references might have been.  Creating a 
dictionary isn't too bad, but when you have 100 clients,
and this needs to be called a whole lot, it's an unneeded
penalty on CPU and memory to have to create a huge
dictionary every time select gets called (and then destroy
it)

Any ideas on speeding this up, or rewriting in a better
way?  Is there any way to get around the read-only problem
on sockets?

-- 
David Allen
http://opop.nols.com/
----------------------------------------
Have you noticed the way people's intelligence capabilities decline 
sharply the minute they start waving guns around? 
- Dr. Who 



From seefelds at magellan.umontreal.ca  Tue Jun  6 15:44:13 2000
From: seefelds at magellan.umontreal.ca (Stefan Seefeld)
Date: Tue, 06 Jun 2000 19:44:13 GMT
Subject: dictionary with complex keys
Message-ID: <393D54B6.EAF70234@magellan.umontreal.ca>

I'm trying to create a dictionary with the keys being lists.
How can I do that ? The docs say that only simple types are
possible as keys. Is there a way to achieve what I want ? May
be some special operator I need to define...

Thanks,	Stefan
_______________________________________________________              
              
Stefan Seefeld
Departement de Physique
Universite de Montreal
email: seefelds at magellan.umontreal.ca

_______________________________________________________

      ...ich hab' noch einen Koffer in Berlin...


From rjroy at takingcontrol.com  Sun Jun  4 14:19:43 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Sun, 04 Jun 2000 18:19:43 GMT
Subject: Combining two graphic images
References: <959996628.860207@centauri.ezy.net.au> <n3aijsoefu73lgv2h2h37750di9ckk7uh6@4ax.com> <960072929.170200@centauri.ezy.net.au>
Message-ID: <393a9bfc.921350265@news1.on.sympatico.ca>

On Sun, 4 Jun 2000 08:56:02 +1000, "A J Wilson" <ajwilson at ezy.net.au>
wrote:

>I downloaded PIL and tried to compile/install it.  However, when I try to
>import it in python to test it I get the following error.
>
>>>> import _imaging
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>ImportError: ./_imaging.so: undefined symbol: jpeg_resync_to_restart
>>>>
>
>Does anyone know what I am doing wrong.
>I am running Mandrake 7 on an intel box.
>
>Also what other rpm packages do I need to have installed to make it work?
>
>Thanks
>Austin
>
>

Not sure which rpm packages it takes but see the README. You will need
the jpeg and zlib packages. 

Bob



From richard_chamberlain at my-deja.com  Mon Jun 26 10:37:54 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Mon, 26 Jun 2000 14:37:54 GMT
Subject: Class instances
References: <20000626135336.A8180@bork>
Message-ID: <8j7ps0$cqa$1@nnrp1.deja.com>

Sorry Egbert I didn't answer all of your questions.

>   up till now I saw a class object as a template, that has to be
>   instantiated in order to be of any use. However here it seems to
>   lead its own life.

You are creating an instance of Knop by calling Knop(). Try doing that
interactively:

Knop
<class __main__.Knop at 8594c0>
>>> Knop()
<__main__.Knop instance at 85e590>

You are just not keeping a reference to it.

Sorry about the multiple postings, Deja lost it for a moment.

Richard




In article <20000626135336.A8180 at bork>,
  Egbert Bouwman <egbert at bork.demon.nl> wrote:
> Good afternoon,
> The following small script does what I intended it to do:
> it produces a button, and that button dissappears when i click on it.
>
>     from Tkinter import *
>     class Knop(Button):
>        def __init__(self,tekst="KNOPJE"):
>            Button.__init__(self,text=tekst,command=self.quit)
>            self.pack(side=TOP,fill=BOTH,expand=YES)
>     root=Tk()
>     k=Knop()
>     root.mainloop()
>
> However when I replace the last three lines with only:
>     Knop().mainloop()
> then the script still works.
>
> This leaves me with a number of questions (sorry, richard):
> - if 'root=Tk()' is dispensable, then what is its builtin replacement
>   and what does that replacement do
> - up till now I saw a class object as a template, that has to be
>   instantiated in order to be of any use. However here it seems to
>   lead its own life.
> - probably 'mainloop()' plays a crucial role here, but what are all
>   its functions ?  What is the difference between
>   'root.mainloop()' and 'Knop().mainloop()'
> I promised Richard to pose only one question per post, but these
> questions really are one big question about the Tkinter mechanism.
> egbert
>
> --
> Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020
6257991
>
========================================================================
>
>


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


From matt at null.net  Thu Jun 15 20:28:41 2000
From: matt at null.net (Matthew Schinckel)
Date: Thu, 15 Jun 2000 18:28:41 CST
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <qG025.4713$Gp6.293338@dfiatx1-snr1.gtei.net>
Message-ID: <20000615182843.905255@210.9.21.14>

"jin choung" <jin.choung at gte.net> wrote:
> you mean programming python while snowboarding or doing a "superman"
> on a bmx bike?

Somebody wrote:
> )

"jin choung" <jin.choung at gte.net> wrote:
> jin

Gee, and I thought staying up past midnight was '3XTR3M3!!'

On-a-school-night-and-i-am-the-teacher-ly your's,

Matt.

(That's my first Tim inspired sign-off. I feel like an extreme python
programmer just after doing that!)


From kens at sightreader.com  Wed Jun 21 09:19:30 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Jun 2000 06:19:30 -0700
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <3950C0E2.7532D38@sightreader.com>

Number of hours per programmer per month lost to indentation related
issues: 0.0263

Number of hours per programmer per month arguing about indentation related
issues that won't change anyway: 71.820

Me putting in my 71.820 hours for this month:

91.83% of all programming errors are related to the redundancy principle.
The redundancy principle applies to any situation where information is
represented in multiple forms, except when it is used in the context of
signal processing to combat noise.  The redundancy principle states that
redundancy leads to errors.  Where information is represented in multiple
forms, erroneous results can be expected.  Most erors occur as a result of
redudancy.

Often one representation of the information in question is
correct, and masks the error in the other representation,
making the error hard to find.  Often one representation of
of the information in question is correct, and masks the error
in the other representation, making the error hard to find.

This important principle applies to the utterly unimportant indentation
issue for the following reason:

Languages such as C/C++ that represent blocks redundantly, (braces for the
compiler and indentation, by convention, for humans to read), create the
potential for errors.  For example:

if (x)
    if (z)
        do_something();
else
{
     do_something_completely_different();
}

Brother Occam sharply states that one shouldn't create a multiplicity
needlessly.  This applies to programming as well as explaining existence.

As for the space/tab mixing issue, the answer is simple: Make tab an
illegal character in python.  Problem solved.

Haven't we gone over this before?

It's time to start a Whitespace-SIG.

"Thaddeus L. Olczyk" wrote:

> After using Python for a while I've come to the conclusion that
> using whitespaces to determine the begining and ending of blocks
> is bad for three reasons.
>
> Before I list those three reasons let me make two points. People say
> that the indentation requirements don't matter because you generally
> indent the same way. That may be true, but do you think it is a good
> idea to make conventions requirements ( remember that conventions
> start flame wars )? The second is that modern editors with good
> indentation rules ( like emac's python mode ) make indentation based
> coding easy. The thing is that by doing this you are abrogating
> control flow to the editor. When it was the editor indenting code
> which was well defined then there was no problem. By having
> the code depend on indentation, and the editor do the indentation
> for you, you give the editor responsiblity in how the code will flow.
>
> Now on to my points.
> 1) The first is rather stylistic and may not appeal to most, but I
> have often found it useful. When debugging code I often put in
> code to help me figure out whats happening. I also avoid indenting
> such code so that it easy to remember where it is later when I want to
> take it out. An example ( assuming I use begin and end to denote
> blocks ):
>
> def Average():
> begin
>       sum=0
> print "number of numbers=",len(numbers) #illegal
> i=0
>       for n in  numbers:
>       begin
> print "n=", n #illegal
>             sum=sum+n
> print "sum=",sum #illegal
>  i=i+1                     #illegal
> if sum>MaxNumber*I:#illegal
> begin                     #illegal
>   print "sum is too large"#illegal
> end                        #illegal
>      end
>      if len(numbers)!=0:
>      begin
>         return sum/len(numbers)
>      end
>      return 0
> end
>
> You can immediately see code that you used for debugging and remove
> it. However right now writing code that way makes it illegal.
>
> 2) You can easily insert a tab introducing a bug.  Consider the
> following code ( similar to a bug I recently fixed ):
>
> def FindInvoice(idnum):
>       for invoice in invoices:
>             if invoice.id=idnum:
>               return invoice
>             return None # should be indented one less tab.
>
> ....
>
> for lines in file.readline():
>      idnum=lines[:6]
>      invoice=FindInvoice(idnum)
>      if not invoice:
>               invoice=InvoiceClass(idnum)
>          invoices.append(invoice)
> ....
> Initially the line in error was indented correctly, but someplace
> along the line the extra tab was introduced. A lot of effort then had
> to go in to finding this bug. With begin...end or braces this is less
> likely to happen.
>
> 3) Finally. Python is a language where it is almost easy to write
> extensible programs. To illistrate what I mean consider this example
> of a python program I recently wrote:
> There is a C++ project which involves many project files and workspace
> files ( the project files are roughly equivalent to makefiles, the
> workspace files primarily are used to couple projects and define
> dependencies ). Version 1 simply read all the project files and
> generated a call to compile each one with all different configuartions
> (debug , release, alpha, beta etc..) . Since this built everything
> even if it was already built, and since this had to be run several
> times ( sometimes compiles would fail because a project that the
> present project was dependent on wasn't compiled ), I made version
> two. Version 2 read the workspace file and organised the projects
> into workspaces. Version 3 passed the string to a second script so
> that the second script could figure out if this was something that
> should be built ( for example whether a standard Win32 build or a
> WinCE build should be done ). Version 4 tracked dependencies more
> carefully ( it did a topological sort and orgainsied the build
> according to it ). Since the core stuff should be built first, I had
> to find out a way to figure out which was core. I used a definition of
> the core stuff is the stuff with the longest chain of dependencies.
> That was Version 5.
>
> As you can see I started with a simple program and made it more
> complex as I went on.  That's what I mean by extendible. Python is
> almost good at it. Almost. But that damned code indentation dependence
> backs its ugly head. In the process of extending code, I have to move
> blocks of code around. Sometimes put it in a loop, sometimes add an if
> clause to the top. Sometimes extract code into a subroutine. That
> means that I have to reindent the block. So as I reindent I have to
> try to remember how the indenting went. Something I stink at.
> Emacs also doesn't reindent correctly. So I have to go through the
> code, and try to figure out how it works. This distracts from my
> rewrite ( this also applies to refactorings ). I would rather think
> about how new code is supposed to work rather than how old code
> is supposed to work.
>
> Anyway I hope this causes TPTB to think about how blocks should be
> handled in Pyhton 3000.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000621/13df2b56/attachment.vcf>

From cg at gaia.cdg.acriter.nl  Thu Jun 29 15:29:05 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 29 Jun 2000 21:29:05 +0200
Subject: Aspect-oriented Python?
Message-ID: <8jg821$vll$1@gaia.cdg.acriter.nl>

I've been reading on aspect-oriented programming a bit, and it sounds really
interesting. A Java implementation, AspectJ (www.aspectj.org) looks promising,
but has the drawback that it only supports static weaving (wouldn't know
how to do dynamic weaving with Java, so it's not a big surprise). 

With Python, experimenting with aspect-oriented stuff would probably be
way easier, because it is easy enough to replace class methods etcetera. I've
tried Google, but didn't find anything. Is someone busy with aspect-oriented
programming for Python?

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From md9tim at mdstud.chalmers.se  Mon Jun  5 09:29:09 2000
From: md9tim at mdstud.chalmers.se (=?ISO-8859-1?Q?Tim_Gahnstr=F6m?=)
Date: Mon, 5 Jun 2000 15:29:09 +0200
Subject: How to run scripts from idle
Message-ID: <Pine.SOL.4.21.0006051523380.9521-100000@scooter.mdstud.chalmers.se>

When I start idle and have the window called
Pyton Shell
starting with 
>>> I canot load scripts I have created.

I can type >>>Print "Hi"
but If I create a file named hi.py containing print "hi"

And try to run it I just gett the error message 
---
Traceback (innermost last):
  File "<pyshell#5>", line 1, in ?
    test
NameError: test
---
It seems like idle wont find my scripts but I have tried put them al over
the computer :-) and the path seems correct.

Tim Gahnstr?m



From PClaerhout at CREO.BE  Thu Jun 22 08:27:30 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 14:27:30 +0200
Subject: List of currently loaded drivers under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301214@msgeuro1.creo.be>

> Hello,
> 
> does anyone know if it is possible to get a list of all currently loaded
> device drivers under Windows NT using Python? Is this somewhere in
> the registry or is it stored somewhere else?
> 
> I would like to get similar information like you get in the devices control
> panel under windows nt.
> 
> Kind regards,
> 
> 
> Pieter



From eosn at rahul.net  Sat Jun  3 13:20:21 2000
From: eosn at rahul.net (Ed Ohsone)
Date: 3 Jun 2000 17:20:21 GMT
Subject: Python GUI requires tcl?
Message-ID: <8hbeol$obo$1@samba.rahul.net>

I am a complete newbie in python, though I know some
about tcl.
Now I am interested in learning python.
But one of my concerns is its GUI.
I hear python requires the complete installation of
both tcl and tk in order to use its tk-based GUI.

Is it true?

Is there another graphic package which is more 
compact, I mean, which requires python + the graphic toolkit only
without the extra burden of loading another language like tcl?

How about the stability of the python's tk-based graphic package?
As far as I know, tk is intricately connected to tcl in its C source.
Therefore, the python's GUI may be buggier than the original tk based on tcl.
Of course, it can be less buggier :).
If you have experience in both, please let me know.

Thanks in advance.
-------


From ge at nowhere.none  Mon Jun 26 11:13:57 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 26 Jun 2000 15:13:57 GMT
Subject: SMTP with Python 1.6
References: <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com> <8j7n8d$4u64$1@www.univie.ac.at>
Message-ID: <VqK55.1715$iN5.345609@ptah.visi.com>

In article <8j7n8d$4u64$1 at www.univie.ac.at>, Matthias Huening wrote:
>> #!c:\python\python.exe
>>
>> import cgi, smtplib
>> s=smtplib.SMTP("localhost")
>
>'Localhost' will not work.
>You have to make this the address of a valid SMTP-server.

Gosh, on all my computers, "localhost" is the address of a valid
SMTP-server.  Are your computers all broken?

-- 
Grant Edwards                   grante             Yow!  WHY are we missing
                                  at               KOJAK?
                               visi.com            


From holger at fuessler-net.de  Thu Jun 29 17:26:58 2000
From: holger at fuessler-net.de (holger at fuessler-net.de)
Date: Thu, 29 Jun 2000 23:26:58 +0200
Subject: MD5 - Passwords with python
References: <6e7dj8.bvb.ln@iiab.soluma.de> <8jd74i$16vk@newton.cc.rl.ac.uk>
Message-ID: <2vegj8.ia5.ln@odin.hmf.de>

Hallo!

Now I've got it. It's f for followup and not r for reply mail (in TIN). I
was wondering if somebody kills my followups ;-)

Unfortunately, the standard MD5-module doesn't know about the structure of
the password-string so I can't encode a password with it. I found a perl
module which accomplishs my task and so I thought that there is also a python
module. I thought redhat uses a lot of python stuff, so there should be a
routine to generate suitable passwords. I'v already posted in
comp.redhat.security. When I find the details, I'll post them here (if I can
translate them to python). 

Thanks

Holger


From mwh21 at cam.ac.uk  Sat Jun  3 05:16:28 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Jun 2000 10:16:28 +0100
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de> <3938090C.9FB5D057@stud.ntnu.no> <8h9682$v4h$1@nntp9.atl.mindspring.net>
Message-ID: <m3d7lzyvir.fsf@atrus.jesus.cam.ac.uk>

aahz at netcom.com (Aahz Maruch) writes:

> In article <3938090C.9FB5D057 at stud.ntnu.no>,
> Peter Schneider-Kamp  <peter at schneider-kamp.de> wrote:
> >mlauer at trollinger-fe.rz.uni-frankfurt.de wrote:
> >> 
> >> But why ? How does the for loop knows
> >> how much elements the sequence contains ?
> >> I did not program the __len__ function,
> >> so... how ?
> >
> >As far as I understand the for loop goes on until an IndexError is
> >raised. Then the error is cleared and the for loop ends.
> 
> Let's be technically precise here: an exception is raised and caught by
> the "for" handler; the for loop is then broken (precisely as if the
> break statement was executed).  Like any other caught exception, the
> exception is thrown away (not "error is cleared").

Not quite "precisely as if the break statement was executed", wrt else
blocks.  Ponder the output of these two statements:

for i in range(10):
	print i
else:
	print "else"

for i in range(10):
	if i > 4: break
	print i
else:
	print "else"

but otherwise, yes.

Cheers,
M.

-- 
  The ability to quote is a serviceable substitute for wit.
                                                -- W. Somerset Maugham


From borlak at home.com  Thu Jun  1 08:12:16 2000
From: borlak at home.com (Michael Morrison)
Date: Thu, 01 Jun 2000 12:12:16 GMT
Subject: Dictionary size/speed limit?
References: <f3lZ4.557$r7.287786@news1.rdc2.tx.home.com> <uk8ga112t.fsf@ctwd0143.fitlinxx.com>
Message-ID: <AqsZ4.621$r7.342688@news1.rdc2.tx.home.com>

[snip]

Very helpful!  Thank you, David.

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




From cjc26 at nospam.cornell.edu  Sat Jun  3 11:01:29 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sat, 03 Jun 2000 15:01:29 GMT
Subject: Erm... MacOS X?
References: <eviltofu-1AEA18.02454303062000@news.earthlink.net> <usenet-6DC18E.10333803062000@news.visi.com>
Message-ID: <slrn8ji7hc.5b1.cjc26@localhost.mindriot.net>

* James Felix Black <usenet at inter-slice.com> menulis:
| 
| > Has anyone compiled Python for MacOS X?
| 
| Yes.  It builds right out of the box.  The only trap is if you build on 
| an HFS+ as opposed to UFS volume; then, you have to be careful about 
| case (HFS+ is case-insensitive but preserving.)

Specificially, if you use HFS+ and want it to "just work" when you run
make, then you need to rename the Python directory to something else (I
called it Python.d), then change Makefile.in, configure.in, and
Python/Makefile.in to use the new directory name.  Then run autoconf,
then configure, etc.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From andres at corrada.com  Wed Jun 28 11:11:27 2000
From: andres at corrada.com (andres at corrada.com)
Date: Wed, 28 Jun 2000 11:11:27 -0400
Subject: Controlling Acrobat Reader with COM
In-Reply-To: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EB75@phm-exch.phm.gov.au>; from Hirsch, John on Wed, Jun 28, 2000 at 02:17:41PM +1000
References: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EB75@phm-exch.phm.gov.au>
Message-ID: <20000628111127.D29827@corrada.com>

On Wed, Jun 28, 2000 at 02:17:41PM +1000, Hirsch, John wrote:
> "Robin Becker" <robin at jessikat.demon.co.uk> wrote in message
> news:Q38VarAdKTW5Ewcx at jessikat.demon.co.uk...
> >> ...
> 
> >> I tried already with correctly escaped filenames and always got the same
> >> error ie 'c:\\Python\\Installs\\test.pdf' got the same error as
> >> 'gnarg.pdf' (and I don't have any docs called gnarg.pdf :)).
> 
> >Its likely that Acrobat needs a full-blown control container.  You may
> >have to run it like the OCX demos in Pythonwin are run - check out
> >pywin\demos\ocx.
> 
> >Mark.
> 
> If you can use wxPython it has an ActiveX control container in which you can
> run Acrobat reader.
> There is a demo called ActiveXWrapper_Acrobat.py which does exactly that.
> It could be worth a look.
> 
> Jon

Thank you Mark and Jon for the hints. I'll look into it and report back

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From wlav at hpatl26.cern.ch  Fri Jun  9 10:06:52 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 9 Jun 2000 14:06:52 GMT
Subject: print without newline
References: <39410382.C83492D2@pfluegel.fsnet.co.uk>
Message-ID: <8hqtls$oig$1@sunnews.cern.ch>

Eckhard Pfluegel <Eckhard at pfluegel.fsnet.co.uk> writes:

>I want to print a string without a newline afterwards.

 [...]

>>> import sys
>>> sys.stdout.write( "foo" )
foo>>>

Best regards,
      Wim Lavrijsen




From redaelli at inc.it  Sun Jun  4 16:55:05 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 22:55:05 +0200
Subject: python and web applications
References: <393A7208.23AF3E23@tin.it>
Message-ID: <393AC229.C17C0328@inc.it>

Jilani Khaldi wrote:
> 
> Hi All,
> I am looking for some examples in python to init do some web programming
> with this nice nice new (for me) language.  Could you please tell me
> where to find some code?
www.zope.org a COMPLETE, Object oriented, enviroment with connection to
traditional SQL engines, with a Object oriented database and MANY other
things.
Forget about perl!
-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/


From jmassung at magpiesystems.com  Thu Jun 15 17:36:45 2000
From: jmassung at magpiesystems.com (Jeff Massung)
Date: Thu, 15 Jun 2000 15:36:45 -0600
Subject: Python Productivity over C++ (Encapsulation)
References: <20000615205922.99304.qmail@hotmail.com>
Message-ID: <O6c25.810$M85.40173@news.uswest.net>

Hung Jung Lu wrote in message <20000615205922.99304.qmail at hotmail.com>...
>>I can easily subvert encapsulation in many C++ classes simply with the
line
>>"#define private public" before I include any header files. Perhaps this
>>means
>>that C++ will in fact lead to the overthrow of evil and oppression in the
>>world?

Many times this won't work:

class Dummy {
  int x;
public:
  /* ... */
};

x is atuomatically private in this case - without the private keyword.

>private data members. In my world there are no variables that can't be made
>public, ha! The thing is, if you have to constantly recurr to subversion to
>add functionalities, there is a problem with the language.

No - there is a problem with the (original) programmer - not the language.
And while yes, all variables CAN be public, IMHO, this is bad programming,
just use struct { ... } instead of class { ... } if that is what you want.

Having private/protected members is VERY good for not letting data
"accidently" or "intentially" changing on the object without it knowing.
Example? A string class:

class String {
  char *data;
  int len;
public:
  /* ... */
};

What if the programmer has access to data? He can alter it, and then the
class will produce incorrect results when using len.

Encapsulation is GOOD - our friend.

Oh well, just my .02 worth ;)
Jeff





From l.szyster at ibm.net  Mon Jun 26 03:40:21 2000
From: l.szyster at ibm.net (Laurent Szyster)
Date: Mon, 26 Jun 2000 09:40:21 +0200
Subject: How to change class instances behavior via reload?
References: <961962342.297785615@news.libertysurf.fr>
Message-ID: <395708E5.80E0406F@ibm.net>

Jerome Quelin wrote:
> 
> I would like to change the behavior of all the existing instances
> of a class when using the reload statement. I wonder if this is
> possible?

No. You should implement static and dynamic features that you need
in different modules and use proxy methods.

Like this

-- dtest.py --
def method(self):
    ...

-- test.py --
class test:
    def method(self):
        return dtest.method(self)

And then,

>>> import test, dtest
>>> t = test.test()
>>> reload(dtest)
>>> t.method()



Laurent Szyster


From olczyk at interaccess.com  Wed Jun 21 02:38:57 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 21 Jun 2000 06:38:57 GMT
Subject: Output from a python script to a calling program.
References: <3951f692.948648187@nntp.interaccess.com> <M7T35.3891$qz4.30140@news-server.bigpond.net.au>
Message-ID: <395062db.976369718@nntp.interaccess.com>

On Tue, 20 Jun 2000 23:29:48 GMT, "Neil Hodgson" <neilh at scintilla.org>
wrote:

>> When I call a python script from another program, I notice several
>> things.
>>
>> 1) It sort of bunches the output. Specifically it sits and waits for a
>> large block of output, then prints, stops printing, sits and waits
>> then prints the next batch.
>> 2) Sometimes it seems that parts of the output are printed in
>> different order them I would expect.
>
>   1) The -u option turns off buffering. However, performance may be lower
>without buffering.
>
>   2) You may be seeing standard output being buffered and standard error
>not being buffered. So output to standard error appears before some earlier
>output to standard output.
>
>   Neil
>
>
Yep, that did the trick. The question now is can I turn off buffering
from inside the script?


From loredo at spacenet.tn.cornell.edu  Thu Jun 15 16:01:06 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 15 Jun 2000 16:01:06 -0400
Subject: Can't get NumPy to Install
References: <upqeksgfc39indo8mg3dsc4vfj8oghdr2d@4ax.com> <39477D0A.C2510C3C@stud.ntnu.no> <39489BF7.71CE3304@zip.com.au>
Message-ID: <39493602.49DFDFFF@spacenet.tn.cornell.edu>

Erik de Castro Lopo wrote:
> 
> and get Numerical-15.3 or later. I had this exact same problem a week
> ago and fixed it with this version.

Yes, and if you are doing any FFT work, make sure you get the latest
patches to fftpackmodule.  The "real_fft" function has been buggy for
many NumPy versions, and returns erroneous results.  Warren Focke posted
a fix.

Peace,
Tom Loredo


From bhella at ce.mediaone.net  Tue Jun  6 22:11:46 2000
From: bhella at ce.mediaone.net (Opinderjit Singh Bhella)
Date: Tue, 06 Jun 2000 21:11:46 -0500
Subject: Example of Python code (C code to Python code)
References: <Ych%4.758$0l2.6362@news2.tor.primus.ca>
Message-ID: <393DAF62.451C60E0@ce.mediaone.net>

Here you go, I hope this helps. Don't forget to TAB on the ifs.


TaxRate  = 0.23
MinPay   = 300.0
RegHours = 40.0

TaxPayable = 0
Overtime = 0
OvertimeHours = 0.0

HoursWorked = float(raw_input("Please enter the hours worked per week? "))
PayRate = float(raw_input("Please enter the rate of pay? "))

if HoursWorked > RegHours:
    OvertimeHours = HoursWorked - RegHours
    Overtime = (OvertimeHours) * (PayRate*1.5)

GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime

if GrossPay > MinPay:
  TaxPayable = (GrossPay-MinPay)*TaxRate

NetPay = (GrossPay-TaxPayable);

print "The gross pay is %f" % GrossPay
print "The net pay is %f" % NetPay


Robin Porter wrote:

> Could someone please give me an example of some Python code.  I am just new
> to this language and would appreciate if you could re-write this simple
> piece of C code in Python.
>
> Thank you,
>
> Robin
>
> #include "stdafx.h"
> #include <stdio.h>
> #include <math.h>
>
> float
> HoursWorked,PayRate,Overtime,IncomeTaxes,GrossPay,NetPay,TaxPayable,Overtime
> Hours;
>
> const float TaxRate=.23;
> const MinPay=300;
> const RegHours=40;
>
> main()
> {
>
> TaxPayable=0;
> Overtime=0;
> OvertimeHours=0;
>
>  printf("\nPlease enter the hours worked per week? ");
>  scanf("%f",&HoursWorked);
>
>  printf("\nPlease enter the rate of pay? ");
>  scanf("%f",&PayRate);
>
>  if (HoursWorked>RegHours)
>  {
>   OvertimeHours =HoursWorked-RegHours;
>   Overtime = (OvertimeHours)*(PayRate*1.5);
>  }
>
>  GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime;
>
>  if (GrossPay>MinPay)
>  {
>   TaxPayable = (GrossPay-MinPay)*TaxRate;
>  }
>
>  NetPay = (GrossPay-TaxPayable);
>
>  printf("\nThe gross pay is %f",GrossPay);
>  printf("\nThe net pay is %f",NetPay);
>
>  return 0;
> }



From karl-sve at dsv.su.se  Fri Jun 30 05:06:26 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Fri, 30 Jun 2000 11:06:26 +0200
Subject: RC4.py?
In-Reply-To: <395BC3EC.C8418EF@darwin.in-berlin.de>
References: <395BC3EC.C8418EF@darwin.in-berlin.de>
Message-ID: <Pine.GSO.4.21.0006301047230.10430-100000@luna.dsv.su.se>

Hi!

On Thu, 29 Jun 2000, Dinu C. Gherman wrote:

> does anybody have a working, free (or not free) implementa-
> tion of RC4, a symmetric encryption algorithm developped by
> RSA, Inc., written in pure Python?

Take a look at CipherSaber (http://ciphersaber.gurus.com/).
Short summary: CipherSaber is a system for symmetric encryption. It uses
the RC4 algorithm. You are encouraged to write your own implementation,
and there are lots of tips on how to do that.

Now, if you still don't want to write your own implementation (booh :)),
there is a list of implementations at
http://www.xs4all.nl/~cg/ciphersaber/, including one in python.

HTH,
  Kalle Svensson



From garry at sage.att.com  Thu Jun 15 15:04:02 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Thu, 15 Jun 2000 19:04:02 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com>
Message-ID: <394928A2.BEA2BC1@sage.att.com>

Mark Jackson wrote:
> 
> "Warren Postma" <embed at NOSPAM.geocities.com> writes:
> 
> > Having something
> > look correct on the screen, but not work because of misuse of tabs versus
> > spaces, or incorrect indentation, is something that still continues to bug
> > me.
> 
> Indulge my curiosity - how many times has this actually happened in
> your six months of using Python?

well, i'm not warren, but over the last several years of 
python use this has happened to me maybe twice.  each time i got
a syntax error, looked at it, said "oh.", and fixed it.
lost productive time: perhaps a minute and a half.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From matt at null.net  Fri Jun 16 21:21:34 2000
From: matt at null.net (Matthew Schinckel)
Date: Fri, 16 Jun 2000 19:21:34 CST
Subject: List of functions called in Python source files?
References: <3Xc25.3507$Uw3.210307@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <20000616192134.375851@210.9.20.69>

"William Dandreta" <wjdandreta at worldnet.att.net> wrote:
> I need to get a list of the external (ones not in the same source
> file) functions called in each Python source file. Does anyone know an
> easy way to do this?
> 
> Bill

You can try using the Python profiler - but it will give you all of the
functions, from this file and from any imports.

I have only used to it compare speeds, but it seems flexible enough to
be able to use it to list functions (and possibly where they came from).

Matt.


From sbrunning at trisystems.co.uk  Fri Jun  2 03:56:10 2000
From: sbrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 2 Jun 2000 08:56:10 +0100 
Subject: A Mountain of Perl Books + Python Advocacy
References: <20000405053920.20579.qmail@web2102.mail.yahoo.com> <slrn8hdj3i.47b.tsummerfelt1@co146453-a.kico1.on.wave.home.com> <ur9bc7i3p.fsf@ctwd0143.fitlinxx.com> <slrn8hggff.lv.tsummerfelt1@co146453-a.kico1.on.wave.home.com> <39184F02.E6AD745D@basho.fc.hp.com> <c3qdjs8c1msmr6i8hjp7p5fo862r15nu7k@4ax.com>
Message-ID: <34575A892FF6D1118F5800600846864D3D2A91@intrepid>

Chris Stromberger wrote in message ...
>>I would invite everybody to read Eric Raymond's article: "Why
Python?"
>>Honestly, I expected to see comments about it already in this news
group
>>. . . but maybe I was just late, and missed the thread.
>
>Thank you for the invitation.  Where might one find this article?

<http://www2.linuxjournal.com/lj-issues/issue73/3882.html>

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk
The opinions expressed are mine, and are not necessarily those of my
employer. All comments provided "as is" with no warranties of any
kind whatsoever.




From R.Brodie at rl.ac.uk  Wed Jun 14 08:10:24 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Wed, 14 Jun 2000 13:10:24 +0100
Subject: TCP packet size?
References: <CC59868A0A76D311B1F50090277B321C1F33BA@VRTMAIL> <3948AC51.98E5EAEC@rpgarchive.com>
Message-ID: <8i7snf$r2m@newton.cc.rl.ac.uk>

"chris" <chris at rpgarchive.com> wrote in message news:3948AC51.98E5EAEC at rpgarchive.com...

>The messages just aren't reassembled the way I thought TCP worked.
>I'm hoping to find a way to received one message at a time in its entirety,
>or determine the start and end of my messages

Sadly, it's supposed to work like that. TCP is a stream protocol; there is no concept
of a packet or message. The data stream comes out like a ticker-tape; if it stops
mid-message and carries on, or aggregates two messages, that's the programmer's
problem.

Now, one might think that it would be useful to have a protocol that allowed
arbitrary length messages to be sent and received atomically. TCP doesn't
do that. Easy to implement, harder to program for: it's the Unix way ;)

Everyone programming to the TCP/sockets API gets burnt sooner or later
by this. Curse three times, dance round the room widdershins, then put
a data length field in your message header. Check Stevens or any other
decent introduction to socket prograaming for details.




From clgonsal at keeshah.penguinpowered.com  Tue Jun 20 04:50:29 2000
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Tue, 20 Jun 2000 03:50:29 -0500
Subject: Pyexpat and iso-8859-1
References: <394629C7.4F7322D8@lib.ua.ac.be>
Message-ID: <slrn8kuc2o.ks.clgonsal@keeshah.penguinpowered.com>

On Tue, 13 Jun 2000 07:32:07 -0500, Marc Jeurissen <mjeuris at lib.ua.ac.be> wrote:
> When I parse a XML-file containing the declaration '<?xml version="1.0"
> encoding="iso-8859-1"?>' with Pyexpat, every iso-latin1 character is
> being replaced by 2 new characters, the first of wich is nearly always
> #195 (&Atilde;), the second one has a decimal value of 64 less than the
> original character value.
> 
> Some examples:
> 
> #233 (&eacute;) becomes #195 + #169 (&copy;)
> #231 (&ccedil;) becomes #195 + #167 (&sect;)
> #239 (&iuml;) becomes #195 + #175 (&macr;)
> 
> Anyone knows what to do about this?

I don't know what to do about it, but it looks like the characters are
getting UTF8 encoded. I have no experience with UTF8 in Python, but a
bit of searching on www.python.org turned up this:

    http://w1.132.telia.com/~u13208596/unicode.htm

HTH...

-- 
  C. Laurence Gonsalves                "Any sufficiently advanced
  clgonsal at kami.com                     technology is indistinguishable
  http://www.cryogen.com/~clgonsal/     from magic." -- Arthur C. Clarke



From gmcm at hypernet.com  Thu Jun 29 19:44:05 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 23:44:05 GMT
Subject: Questions about Installer/Builder/Standalone
References: <p9R55.17984$Yr4.303276@news1.rdc1.tn.home.com> <8F5FCDEB5gmcmhypernetcom@199.171.54.194> <grP65.21733$Yr4.417778@news1.rdc1.tn.home.com>
Message-ID: <8F62C0258gmcmhypernetcom@199.171.54.154>

Mike Callahan wrote: 

>I tried this but it did not work:
>
>1. I copied and pasted runTk.py to the top of my script.
>2. I ran simple.py to generate a proper Install_xxx.exe

In which case your xxx.exe already runs runTk.py automatically, but running 
it twice shouldn't hurt.

>3. I ran the install_xxx.exe and got the tk folder, the xxx.exe and the
>support files. The tk folder is at c:\xxx\tk
>4. I moved everything into a new test folder on a different machine with
>the tk folder under the test folder ie. c:\test\tk.
>5. I double click on xxx.exe and nothing happens.

If you pick up the entire installation directory tree, it should work. 
Nothing about the install gets magically remembered - it's all based on 
"where am I" at run time.

>I thought that if I moved the tk folder to the same relative position,
>directly under the main folder, and I ran runTk first that my app would
>use the tk at c:\test\tk. Is installutils.installCArchive() doing more
>than simply unpacking?

No.

>From your description, I can't tell what's going wrong. It's a pain, but I 
suggest you rebuild setting the debug flag on, and turning userunw off 
(both in the [MYSTANDALONE] section). Then go through your process. You'll 
get a DOS box and debugging messages when you try it.

- Gordon


From wjk at wjk.mv.com  Wed Jun 28 23:08:44 2000
From: wjk at wjk.mv.com (wjk)
Date: Wed, 28 Jun 2000 23:08:44 -0400
Subject: books????
References: <sla4ago0jev12@corp.supernews.com> <961961617.1744372225@news.rma.edu> <8j85gu$e86@autodesk.autodesk.com> <dp3cj8.21o.ln@kmself.nntp.ix.netcom.com>
Message-ID: <395ABDBB.1D5A6873@wjk.mv.com>

I liked the O'Reilly book and just picked up
Ivan Laningham's book: Sams Teach Yourself Python in 24 hours
and found it very informative and useful.. It is filling in some gaps
for me. (only the screen shots are a bit too small )

kmself at ix.netcom.com wrote:

> I'll second both suggestions.  The O'Reilly book is useful for guidance,
> the New Riders book is a good reference.  For a *very* abbreviated
> refernce, the ORA pocket reference is also handy.
>
> Akira Kiyomiya <akira.kiyomiya at autodesk.com> wrote:
> > I am using "Learning Python" from O'Reilly and "Python Essential Reference"
> > from New Riders.  They are both pretty good books for beginners.
>
> > Akira
>
> > "Robert" <rhicks at rma.edu> wrote in message
> > news:961961617.1744372225 at news.rma.edu...
> >> On Sat, 24 Jun 2000, SugarKing wrote:
> >> >Are there any good beginner books for Python? I'm itching to get my hands
> > on
> >> >any. Thanks
> >>
> >> There is:
> >>
> >> Learning Python
> >> Learning Python in 24 hours
> >> Python and Tkinter programming
> >>
> >> There should be links to these books in the library portion of the python
> >> website. I have not used the LP in 24 hours but have heard good things
> > about it.
> >>
> >> Robert
>
> --
> Karsten M. Self <kmself at ix.netcom.com>         http://www.netcom.com/~kmself
>   Evangelist, Opensales, Inc.                       http://www.opensales.org
>    What part of "Gestalt" don't you understand?      Debian GNU/Linux rocks!
>      http://gestalt-system.sourceforge.net/      K5: http://www.kuro5hin.org
> GPG fingerprint: F932 8B25 5FDD 2528 D595  DC61 3847 889F 55F2 B9B0



From matt at null.net  Mon Jun 19 20:10:48 2000
From: matt at null.net (Matthew Schinckel)
Date: Mon, 19 Jun 2000 18:10:48 CST
Subject: urllib question
References: <394B196F.5ED9E73@fizzylab.com>
Message-ID: <20000619181059.373117@210.9.22.17>

paul at fizzylab.com wrote:
> Why does the first example work, but not the second (actual URL not
> supplied, but you get the idea)?
> 
> I assume something to do with query string values and encoding: even
> if I was right, I have been unsuccessful at making it work.
> 
> #!/usr/bin/python import string, urllib
> 
> u = urllib.urlopen(string.join(["http://cnn.com/", "index.html"]))
> print u.read() u =
> urllib.urlopen(string.join(["http://some.host.com/",
> "montstrousQueryString?withParameters"])) print u.read()
> 
> My second day trying to use python . . . excuse my thickness of skull.

I would suggest using urlparse.urljoin(), it then makes sure it is a
valid URL, and you can also urllib.urlencode() a dictionary of values
rather than get a string.

Matt.


From 12klat at sightreader.com  Thu Jun  1 19:17:40 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Thu, 01 Jun 2000 16:17:40 -0700
Subject: search for a string?
References: <8h6211$ark$1@news-int.gatech.edu>
Message-ID: <3936EF14.524AE55F@sightreader.com>

Shengquan Liang wrote:

> how do i search for a string in a document
>
> and inseart content of another file before
> that string?

>>> import string
>>> f = open("todo.txt")
>>> s = f.read()
>>> print s
Today's Menu:
  spam and jelly sandwitch
  eggs
  spamburger

>>> n = string.find(s, 'eggs')
>>> s = s[:n] + 'spam and ' + s[n:]
>>> print s
Today's Menu:
  spam and jelly sandwitch
  spam and eggs
  spamburger





From jwbnews at scandaroon.com  Fri Jun 16 16:20:59 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 16 Jun 2000 13:20:59 -0700
Subject: Function that knows the names of its actual parameters
References: <394a6395.7217524@news.btx.dtag.de>
Message-ID: <jwbnews-9D73CE.13205916062000@news.olympus.net>

In article <394a6395.7217524 at news.btx.dtag.de>, spamfranke at bigfoot.de 
(Stefan Franke) wrote:

> I found myself in need for the following functionality a couple of times:
> 
> >>> a=1
> >>> b="foo"
> >>> c=b
> >>> magic(c, b, a)
> {'a': 1, 'b': 'foo', 'c': 'foo'}
> 
> that is, a function that goes one step further than **dict arguments and 
> returns
> a dictionary with names and values of its actual parameters.
> It's useful every time one tries to write generators for Python or other 
> code (right
> now I'm writing some SQL code emitting functions).

I would think that this method should point the way (you can have some 
positional arguments as well, along with named keyword ones, plus the 
catchall):

>>> def m(**args):
...   print args
... 
>>> m(a=5, b='ten')
{'b': 'ten', 'a': 5}

One example is on page 109 of "Learning Python",  with discussion 
preceeding and following.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From hinsen at cnrs-orleans.fr  Wed Jun  7 05:33:45 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 11:33:45 +0200
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
References: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD> <14647.58023.493846.225899@buffalo.fnal.gov>
Message-ID: <m3wvk1hm2u.fsf@chinon.cnrs-orleans.fr>

Charles G Waldman <cgw at fnal.gov> writes:

> On the one hand, Python's behavior makes sense because it simply
> reflects the behavior of the system math libraries.  On the other

I'd prefer a portable solution to one that reflects the system math
library. I had the same problem under AIX, and I solved it by patching
Python. I just wonder if someone has the patience to work out a
patch that would create the same behaviour independently of what the
math library does. Not me, unfortunately...
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From m.faassen at vet.uu.nl  Sat Jun 17 10:12:28 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 17 Jun 2000 14:12:28 GMT
Subject: Python Productivity over C++ (Encapsulation)
References: <20000615235703.24571.qmail@hotmail.com>
Message-ID: <8ig10c$nc0$1@newshost.accu.uu.nl>

Hung Jung Lu <hungjunglu at hotmail.com> wrote:
> --- In python-list at egroups.com, m.faassen at v... (Martijn Faassen) wrote:
>>Encapsulation is necessary if you want abstraction.

> Hmm... I guess we are talking about different encapsulations. I was talking 
> about the "private" keyword alone (hence data hiding.) Sorry if I have not 
> been more clear. Instead of "encapsulation" I should have said "data hiding" 
> or simply the "private" keyword.

I think 'data hiding' is still too broad. Abstraction needs interfaces
needs implementation hiding needs data hiding. The mistake people make is
though that by doing data hiding you get abstraction. That is of course
cargo cult abstraction. :) (and Python made me see that)

> I agree that data members are best accessed through accessor methods, and 
> that's the way I (and most people) program. But I really can't live with the 
> "private" keyword. If I have been criticizing "encapsulation", it's because 
> I was referring to the "private" keyword, not the accessor methods.

What's the terrible thing about the private keyword, then? That you are
forced to work with bad classes which can't do what you want? That there's
no escape hatch? That it's too easy to get painted into a corner?

> There is no need for the "private" keyword, if a program works through 
> accessor methods. If one doesn't want a particular accessor method to be 
> used by others, one can simply use a naming convention (like an initial 
> underscore _).

You can do naming conventions for namespaces too (prefixes), but that doesn't
mean that's ideal. You lose consistency between code bases, for instance.
This in this case I'm inclined to agree with you that Python's approach is
a good thing. At least not a *worse* thing.

>>That doesn't mean a Python programmer should forget about encapsulation.

> Not encapsulation... "abstraction" would be a more correct word, I guess. 
> Encapsulation conveys the image of data hiding and data ownership.

I don't think there's anything wrong with data hiding and data ownership;
if you work with interfaces you'll have to forget about data implementation
anyway, right?

It is a pain when it's used for other than engineering reasons, though. Or
when it's applied the wrong way so you're stuck with something unworkable
that's hard to work around

Regards,

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


From mwh21 at cam.ac.uk  Fri Jun  9 10:52:54 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 15:52:54 +0100
Subject: = vs. == in Py3k
References: <4.3.1.2.20000609101313.01add2f8@phear.dementian.com>
Message-ID: <m3n1kuvrcp.fsf@atrus.jesus.cam.ac.uk>

"Steven D. Arnold" <stevena at permanent.cc> writes:

> Or we could have `==' always be a comparison operator, but if `='
> was used in `==' context, it would be interpreted as `=='.  One
> possible negative is that this could lead to inconsistent coding and
> sloppiness, but it seems like a virtue of a good scripting language
> to be as understanding as possible about this sort of error. 

One of Python's unqualified Good Things is not trying to infer
programmer intent from half-wrong code.

> (I do think at a minimum a user should at least be able to set a
> flag to get compiler warnings, a la perl `-w', and when this flag is
> set, an `=' in an if clause should raise a warning.)

Rather better to give a SyntaxError *all* the time, no?  Then you
learn not to do it pretty quickly.  I've never really understood calls
for '-w'-type flags in Python; constructs of dubious character should
be outlawed, not discouraged.

(Actually, I can see some merit in flagging things that are *usually*
wrong, but have some virtue, eg. "do you *really* want to have that
mutable default argument", but that could be a more static thing).

Cheers,
M.

-- 
41. Some programming languages manage to absorb change, but 
    withstand progress.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From aahz at netcom.com  Tue Jun  6 15:52:40 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Jun 2000 19:52:40 GMT
Subject: dictionary with complex keys
References: <393D54B6.EAF70234@magellan.umontreal.ca>
Message-ID: <8hjkq8$svr$1@slb6.atl.mindspring.net>

In article <393D54B6.EAF70234 at magellan.umontreal.ca>,
Stefan Seefeld  <seefelds at magellan.umontreal.ca> wrote:
>
>I'm trying to create a dictionary with the keys being lists.  How can I
>do that ? The docs say that only simple types are possible as keys. Is
>there a way to achieve what I want ? May be some special operator I
>need to define...

Use tuples instead of lists.  The restriction isn't so much on "simple"
types as "immutable" types -- dict keys must be constants.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From kpmurphy at my-deja.com  Thu Jun 29 10:49:05 2000
From: kpmurphy at my-deja.com (kpmurphy at my-deja.com)
Date: Thu, 29 Jun 2000 14:49:05 GMT
Subject: PhotoImage example?
Message-ID: <8jfnkp$bh0$1@nnrp1.deja.com>

Does anyone have an example using PhotoImage on a canvas?  I'd like to
write one pixel at a time to the file (PhotoImage) embedded on the
canvas.  Thanks,

Keith


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


From wware at world.std.com  Mon Jun 26 15:28:00 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 26 Jun 2000 19:28:00 GMT
Subject: Final garbage collection?
References: <962028703.1248416950@news.libertysurf.fr> <FwrnM8.MMA@world.std.com> <962045446.1993274611@news.libertysurf.fr>
Message-ID: <Fws0qp.14n@world.std.com>

Jerome Quelin (jerome.quelin at insalien.org) wrote:
>        When a [Perl] interpreter thread finally shuts down (usually
>        when your program exits), then a rather costly but
>        complete mark-and-sweep style of garbage collection is
>        performed...
> I was wondering why Python does not implement such a mechanism, that
> seems to me a Good Thing (tm)

Guido (Python's designer) is the final authority on decisions like
this, so if you feel strongly about it, he's the person to talk to.
My humble speculation is that a case has not yet arisen where great
harm has been done by a failure to execute the __del__ methods for
all objects stuck in circular references. Off the top of my head it
sounds like it's probably a good idea, but I don't know Guido's
thinking on the matter.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From jflores at codeit.com  Mon Jun 26 19:41:23 2000
From: jflores at codeit.com (Grakka)
Date: Mon, 26 Jun 2000 16:41:23 -0700 (PDT)
Subject: Palm & Python
Message-ID: <Pine.LNX.4.21.0006261638270.1767-100000@startrek.codeit.com>

I saw this morning someone that was requesting any information regarding
python & palmtops, check this link out:

http://pda.tucows.com/palm/dev_unix.html

Item:

PalmPython:  An extended replacement for the Python interface provided 
with Pilot-Link...

Regards,
Julio




From ge at nowhere.none  Mon Jun 12 15:30:11 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 12 Jun 2000 19:30:11 GMT
Subject: Threads -- play well with others? If not, what instead?
References: <8i34mr$9hti$1@nntp6.u.washington.edu>
Message-ID: <7Ta15.315$i06.94722@ptah.visi.com>

In article <8i34mr$9hti$1 at nntp6.u.washington.edu>, Russell E. Owen wrote:

>David Beazley's excellent "Python Essential Reference" says in the 
>section on threads: "In addition, many of Python's most popular 
>extensions such as Tkinter may not work properly in a threaded 
>environment."
>
>I assume it's true, but it was quite a bombshell. I was hoping to write 
>a networked GUI client, hence:
>- read data from a socket and fill in a GUI display
>- accept input from the user and write data to the socket
>I assumed I'd use two threads, one for input, one for output. Now I have 
>no idea what to do. Any suggestions?

In Tk, you can assign read handlers to file objects.  Anytime
there is data available to be read, the handler will be called.

Just open the socket connection and assign a read-handler to
it.  Piece-o-cake.  Here's an excerpt from a program that uses
that technique to handle data from a popen2'd child process:

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

if cmd is None:
    exceptString = 'no executable specified'
    raise exceptString, cmd
                                        
self.__returnCode = -1
self.__child = popen2.Popen3(cmd)
self.__fd = self.__child.fromchild.fileno()

fcntl.fcntl(self.__fd, FCNTL.F_SETFD, FCNTL.O_NDELAY);


Tkinter.tkinter.createfilehandler(self.__child.fromchild,
                                  Tkinter.tkinter.READABLE,
                                  self.__stdoutHandler)
                                  
------------------------------------------------------------

self.__child.fromchild      is a file object connected to the "read" end of a pipe.
Tkinter.tkinter.READABLE    is a constant that tells Tk what you care about.
self.__stdoutHandler        is a function to call when the file object has data available.

-- 
Grant Edwards                   grante             Yow!  I want to kill
                                  at               everyone here with a cute
                               visi.com            colorful Hydrogen Bomb!!


From jwbnews at scandaroon.com  Wed Jun 28 15:50:57 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 28 Jun 2000 12:50:57 -0700
Subject: long lines, long string
References: <j9tcj8.i0b.ln@iiab.soluma.de> <8jcudd$gdk$1@slb7.atl.mindspring.net>
Message-ID: <jwbnews-C1AE30.12505728062000@news.olympus.net>

In article <8jcudd$gdk$1 at slb7.atl.mindspring.net>, aahz at netcom.com 
(Aahz Maruch) wrote:

> In article <j9tcj8.i0b.ln at iiab.soluma.de>,
> Holger M. F??ler <fuessler at solutec.de> wrote:
> >
> >Is there a possibility to put long lines of python-source-code in 
> >multiple
> >lines (for better clarity)?
> 
> myDict = {
>   'foo': 1,
>   'bar': 2,
>   'baz': 3
>   }
> 
> if myDict[xyzzy] == 'foo' \
>   or myDict[xyzzy] == 'bar' \
>   or myDict[xyzzy] == 'baz' :
> 
> >How can I accomplish the same with strings?
> 
> """
> This
> is 
> a 
> long
> string
> """

And, if you don't want the resulting returns, look into concatenation of 
"adjacent" strings:

>>> a = "this "\
... "is "\
... "a "\
... "long "\
... "string. " 
>>> a
'this is a long string. '
>>> 

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From scarblac-spamtrap at pino.selwerd.nl  Tue Jun 27 10:56:28 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 27 Jun 2000 14:56:28 GMT
Subject: problem with accessing methods?
References: <8F60916A3mikecat@192.168.10.38>
Message-ID: <slrn8lhfut.1m8.scarblac-spamtrap@flits104-37.flits.rug.nl>

Mike 'Cat' Perkonigg wrote in comp.lang.python:
> How can I access a method with variable amount of parameters?
> I have sort of an object management system which routes method invocations 
> from outside to the objects, but the management system don't know anything 
> about the methods.
> In C I would use a pointer to a struct, but this is Python :-)

You use apply(). apply(f, args) calls the function f with the arguments
given in the tuple 'args'.

To make a function that takes multiple arguments, use

def my_apply(func, *args):
   return apply(func, args)

(args is a tuple containing the values passed in).

To handle keyword arguments as well, use

def my_apply(func, *args, **kw_args):
   return apply(func, args, kw_args)
   
In this case, kw_args is a dictionary with keyword: value pairs. Just play
around in the interpreter a bit, it's quite simple but looks complicated the
first time you see it...

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From rkiendl at gmx.net  Mon Jun  5 11:40:26 2000
From: rkiendl at gmx.net (Robert Kiendl)
Date: Mon, 05 Jun 2000 17:40:26 +0200
Subject: embedding pythonwin: deadlock on PyWinInterpreterLock_Acquire
References: <ECEPKNMJLHAPFFJHDOJBOEEKCKAA.mhammond@skippinet.com.au>
Message-ID: <393BC9EA.95B2442F@gmx.net>

I've got a problem with CEnterLeavePython,
PyWinInterpreterLock_Acquire/Release (build 128):

windows messages come into python via CEnterLeavePython.
C-code (swig) is called from some python message handlers. 
and this C-code sometimes implicitly sends windows messages which again
flow into python via CEnterLeavePython. Thereby
PyWinInterpreterLock_Acquire crashes with a "PyEval_AcquireThread:
non-NULL old thread state"

shouldn't there be code in PyWinInterpreterLock_Acquire/Release to take
care if the same windows thread just does a recursion Win -> Python -> C
(->Win) -> Python. 
or am I completely wrong?

robert


From jin.choung at gte.net  Wed Jun 14 17:11:33 2000
From: jin.choung at gte.net (Jin)
Date: Wed, 14 Jun 2000 21:11:33 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net>

thanks much for all the input.

i have no doubt that C++ is far more complex and complicated.  you guys are
probably right that Python is a much easier language but i picked a bad book
(for me) to start with.

will look for the 24hour book.

but i think the reason the C++ book is working so well for me is that it
assumes nothing about prior knowledge in an OO language because C++ is the
base upon which everything else is built.  the book realizes this and so all
such concepts are explained in detail from scratch.

the learning python book however seems to assume a lot of things that i
would only know if i came from a C++ or perl background!

thanks again, and i will pursue both paths to enlightenment and hopefully
one language will bolster skills in the other.

jin




From programmedintegration at home.com  Mon Jun 26 11:05:20 2000
From: programmedintegration at home.com (ColBoy)
Date: Mon, 26 Jun 2000 15:05:20 GMT
Subject: SMTP with Python 1.6
References: <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com> <8j7n8d$4u64$1@www.univie.ac.at>
Message-ID: <QiK55.7891$pW4.215717@news1.rdc1.on.wave.home.com>

Thanks Matthias, problem now sorted. Easy when you know how :-)

Colin




From bror.johansson at esavionics.se  Wed Jun 14 07:24:43 2000
From: bror.johansson at esavionics.se (Bror Johansson)
Date: Wed, 14 Jun 2000 13:24:43 +0200
Subject: Problem assigning timevalue to Excel - COM
Message-ID: <8i7pp7$1vl$1@newstoo.ericsson.se>

I haven't found out what type and/or format to use when assigning a value to
a cell in MS Excel having a predefined time format like this:
1997-04-03  12:15:00

When reading - via COM - from a cell with a manually entered content, I get
a 'time object' back.
This 'time object' is recognized as being of 'TimeType' and can be converted
to an int.

How can I convert an int into something that can be assigned to the Value
property of the desired cell and be accepted according to the format as of
above?

TIA /BrorJ




From mjhand at concentric.net  Tue Jun 13 14:57:40 2000
From: mjhand at concentric.net (Manus Hand)
Date: Tue, 13 Jun 2000 12:57:40 -0600
Subject: Two newbie questions
References: <20000613165659.A7828@basajaun>
Message-ID: <39468424.BED211C2@concentric.net>

To sort the characters in a string:

import string
aList = list(yourString)
aList.sort()
yourString = string.join(aList, '')

To get a random number between -sys.maxint and sys.maxint:

import random
yourNumber = random.randrange(sys.maxint) * [1,-1][random.randrange(2)]

[Note that random.randrange(-sys.maxint, sys.maxint) should work but instead
functions as a request for an OverflowError.]

Hope this helps,
Manus




From hei at adtranzsig.de  Mon Jun 26 07:46:20 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Mon, 26 Jun 2000 13:46:20 +0200
Subject: How to change class instances behavior via reload?
References: <961962342.297785615@news.libertysurf.fr>
Message-ID: <8j7fp9$dr$1@desig-bs01-s04.adtranzsig.de>

I don't know if my suggestion solves your problem.
My problem was: I wanted to write an interactive
program in an embedded Python. When testing the
behaviour of its objects, i would discover a slight
malfunction in one of the existing objects.
So i would like to fix the erroneous method and
reload it, without destroying the living objects.

I solved this via an editor macro that would create
a temp .PY module for the method i fixed that looks
like this:

# explanation of the identifiers i use:
# originalmodule - name of .PY file where your class is defined in
# Method - name of the method that needs to be reloaded in a
#    fixed version
# tempmodule - name of temporary .PY file created
# Classname - name of the class in originalmodule.py
#   the Method is a member of

# import everything from the original module the
# method comes from:
import originalmodule
from originalmodule import *
# now a copy of my fixed version of the method:
# (notice that it's unindented now!)
def Method(self,bla):
  self.x = bla
     # example code
# eof

After saving this temp PY file (say, as tempmodule.py),
the following is done
in the Python interpreter:

del tempmodule
  # in case it already existed as an imported module
import tempmodule
reload(tempmodule)
  # don't ask me why i do this... maybe unnecessary
  # - my editor macro just does it this way
  # Attention, originalmodule must exist (already be imported)
  # in the main namespace for the following to work:
setattr(originalmodule.Classname,"Method",tempmodule.Method)


This results in modified behaviour for all living instances of
the class, and it proved to supply a real development boost
for me. I don't know if this answers your question, but
it's an interesting technique anyway. And Python is the only
OO-language i know of that provides this possibility.


(Your approach would create a second instance of the whole
module, as older objects still refer to the code in the old
module instance so it can't be deleted)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net


"Jerome Quelin" <jerome.quelin at insalien.org> schrieb im Newsbeitrag
news:961962342.297785615 at news.libertysurf.fr...
> Hi all,
>
> I would like to change the behavior of all the existing instances of a
> class when using the reload statement. I wonder if this is possible?
>
> Consider the following example:
>
> -- file test.py
> class Test:
>     def func(self):
>         print "first version"
> -- end file
> >>> import test
> >>> t1=test.Test()
> >>> t1.func()
> first version
> >>>
> -- Editing file test.py, without leaving interpreter
> class Test:
>     def func(self):
>         print "second version"
> --
> >>> reload(test)
> <module 'test' from 'test.py'>
> >>> t2=test.Test()
> >>> t2.func()
> second version
> >>> t1.func()
> first version
> >>>
>
> This seems correct, because t1 holds a reference to the code of the
> first version of test.py. But when reloading, I would like to change
> the behavior of all existing instances of the class Test, and not only
> the behavior of the new instances that I will create. That is, I want
> t1.func() to output "second version". I wonder if this is possible and
> how?
>
> Thanks,
> Jerome
> --
> jerome.quelin at insalien.org




From moshez at math.huji.ac.il  Sat Jun 10 01:47:19 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 08:47:19 +0300 (IDT)
Subject: Probably a dumb question
In-Reply-To: <39418D8E.34D59FC4@micron.net>
Message-ID: <Pine.GSO.4.10.10006100845350.1511-100000@sundial>

On Fri, 9 Jun 2000, Peter Immarco wrote:

> Moshe,
> 
> How close to Python is JPython, that is, will I spend significant amounts of
> time tracking the subtle differences between the two like Sun Java and MS
> Java, or will I be blissfully ignorant of the minor difference? :)

There's a difference table at JPython's home page. There are a few small
differences, for examples, destructors are not called in JPython. (This is
quite all right -- the language reference does not mandate that)

For the most part, code will work on both without trouble.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From mikael at isy.liu.se  Tue Jun 20 02:52:30 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 20 Jun 2000 08:52:30 +0200 (MET DST)
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <l03102800b574c9a15f7a@[193.78.237.171]>
Message-ID: <XFMail.000620085230.mikael@isy.liu.se>

On 20-Jun-00 Just van Rossum wrote:
 >  (Due to historic narrowmindedness, a point is the same as a pixel under
 >  MacOS, no matter what physical resolution the display has; I believe
 >  Windows has (had?) the same problem, except that a pixel is 72/96th
 >  points...)

Erm, isn't a pt as in 10 pt font the typographic unit 1/72 inch?

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jun-00
Time:    08:50:32

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

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



From neurocrat at one.net.au  Sun Jun 11 09:18:35 2000
From: neurocrat at one.net.au (Neurocrat)
Date: 11 Jun 2000 23:18:35 +1000
Subject: win32gui question - is my window the foreground window?
References: <39437A27.1020403@bigpond.com>
Message-ID: <85g0qkices.fsf@one.net.au>

Linus Chang <Linus_Chang1 at bigpond.com> writes:

> Hi all,
> 
> I'd like to detect if the window of my wxPython application is currently 
> the foreground window.
> 
> The way I'm thinking of doing this is to use the GetForegroundWindow 
> function to grab the handle of the current foreground window and compare 
> it with my own handle (got by using the GetActiveWindow function)

Could you elaborate on this?  Are you trying to determine whether your
application's main window is at the top of the Z-order? Are you trying
to determine this from *within* your application, or from another
process?


From pinard at iro.umontreal.ca  Tue Jun  6 15:25:27 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 15:25:27 -0400
Subject: operator overloading
In-Reply-To: david_ullrich@my-deja.com's message of "Tue, 06 Jun 2000 17:09:48 GMT"
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com>
	<393C2386.307A160D@sightreader.com> <20000605235010.C469@xs4all.nl>
	<8hj2f9$1rn$1@nnrp1.deja.com> <oqwvk2df1q.fsf@titan.progiciels-bpi.ca>
	<8hjb8c$96f$1@nnrp1.deja.com>
Message-ID: <oqog5eboig.fsf@titan.progiciels-bpi.ca>

david_ullrich at my-deja.com ?crit:

> > If one wants tools like `sort()' to work properly, `__cmp__' should
> > establish a proper order relation, yes.  Otherwise, rather cryptic
> > bugs might pop out.

> If so this should probably be mentioned in the docs.

_If_ the documentation does not make it clear, it should indeed.  Could
someone check and, if the documentation is improper, report it appropriately?

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





From PClaerhout at CREO.BE  Thu Jun 22 08:28:17 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 14:28:17 +0200
Subject: Labels of remote volumes under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301218@msgeuro1.creo.be>

> Hello,
> 
> is it possible with Python under NT, to find out the type and the labels
> of the drives of a remote computer?
> 
> Kind regards,
> 
> 
> Pieter
> 
> Pieter Claerhout
> Application Support - CreoScitex Europe
> 



From scherbi at bam.com  Thu Jun 15 10:41:29 2000
From: scherbi at bam.com (Bill Scherer)
Date: Thu, 15 Jun 2000 14:41:29 +0000
Subject: TEXT MANIPULATION (in CSV format)
References: <20000615133826.11516.qmail@web1502.mail.yahoo.com> <3948E40C.2425B170@bam.com> <130601bfd6d5$d29c4ce0$1906a8c0@fc.fenx.com>
Message-ID: <3948EB19.93F5666F@bam.com>

Emile van Sebille wrote:

> You'll need to be more careful than Bill pointed
> out in his reply.  If you know your source data
> is clean, you should have no problems.  Otherwise,
> proceed cautiously.  In particular, if you must parse
> CSV, watch out for embedded commas, eg: name fields
> like "Abe Lincoln, Pres" or address lines like
> "123 Main, Apt C".  You'd be better off if you
> could avoid CSV files by seeing if, for example,
> tab delimited files could be provided instead.

I agree completely.  I was leveraging his example, which was clean of
embedded commas.
If, when creating a csv file, everyone interpreted the 'c' as circumflex
(^) instead of comma (,), life would be better.

-Bill




From mvanier at endor.bbb.caltech.edu  Mon Jun 19 20:05:55 2000
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 19 Jun 2000 17:05:55 -0700
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au>
Message-ID: <jlzoohrzbw.fsf@endor.bbb.caltech.edu>

Toby J Sargeant <tjs at cs.monash.edu.au> writes:

> On Mon, Jun 19, 2000 at 01:05:13AM +0000, Neil Hodgson wrote:
> >    You can change the colour to silver to make it less heavy. Or hack the
> > code to any colour you want. Originally tabs where displayed as round
> > cornered rectangles but Mark thought this was too heavy. Perhaps vertical
> > lines at the start of tabs instead would be lighter and make it easy to
> > match up code.
> > 
> >    Neil
> 
> I saw a screenshot once of a vim session that used the ; (ascii 0273)
> character to display tabs. It was actually very visually appealing, and I was
> a little annoyed when I failed to get emacs to do the same thing.
> 
> Toby.
> 

Put this in your .emacs file:

;;; ---------------------------------------------------------------------------
;;; Functions to display tabs as ^I.
;;; ---------------------------------------------------------------------------

(setq display-table-with-visible-tabs (make-display-table))
(let ((i 0))
  (while (< i 32)
    (or (= i ?\n)
        (aset display-table-with-visible-tabs i (vector ?^ (+ i 64))))
    (setq i (1+ i)))
  (aset display-table-with-visible-tabs 127 (vector ?^ ??)))

(defun make-tabs-visible ()
  "Make tabs display as ^I in the current buffer."
  (interactive)
  (setq buffer-display-table display-table-with-visible-tabs))

(defun make-tabs-invisible ()
  "Make tabs display as whitespace in the current buffer."
  (interactive)
  (setq buffer-display-table standard-display-table))

;; End of .emacs additions

BTW I *loathe* tabs with a white-hot passion bordering on psychosis.  It
comes from having to work in a cross-platform environment where the same code
was edited using emacs and also the Visual C++ editor, which had been
configured to use two-space tabs (!).  I hate tabs so much I've written
scripts to replace them with four spaces, and I also hacked the emacs python
mode to insert spaces where tabs are normally used.  Since then I don't have
problems with tabs in python.

Also re the indentation wars:

Pro-indentation:

-- I like the look of indented code as opposed to code with {} or begin-end
   blocks.  I think most people agree with this.  We can all find a style of
   {} that we like too, but the key is that with python the SAME style suits
   almost everybody.

Anti-indentation:

-- For long functions/classes I like to see an end marker.  Right now I
   usually use a comment e.g.:

class foo:
    # lots of code

    # End of class foo.

... but this is a bit hokey.

-- For code that may go into an environment that can munge whitespace
   (e.g. an active server page, or something that will be processed by
   another program), having tab-based indentation is asking for trouble.

-- Indentation-based grouping makes it hard to write multi-line lambda
   statements.   Of course, begin-end isn't that much better in this
   regard, but {} is (as is parenthesis-based grouping like in lisp).

I suppose in a perfect world we'd be able to have a pragma that would allow
us to select different block options e.g.

#pragma blocks {}
#pragma blocks begin-end

but I wouldn't hold my breath :-)  On balance, I believe the current system
in python does more good than harm.


Mike



--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.


From gtalvola at NameConnector.com  Wed Jun 21 16:42:23 2000
From: gtalvola at NameConnector.com (Geoff Talvola)
Date: Wed, 21 Jun 2000 16:42:23 -0400
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <8iq78m$u5m$1@nnrp1.deja.com>
Message-ID: <395128AF.1826F2E8@NameConnector.com>

Fred Gansevles wrote:

> If you use enums only as a 'conveniant way to represent values', the
> next (quick-hack, untested) might be usefull:
>
> class Enum:
>     def __init__(self, **kw):
>         self.__dict__.update(kw)
>
> enum = Enum(Tag = 0, Date = 1, Ammount = 2)
>
> print "Date-field:", enum.Date

A class can be used to do this quite nicely:

class MyEnum:
    Tag = 0
    Date = 1
    Amount = 2

print MyEnum.Date

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com





From shapr at uab.edu  Sat Jun 17 08:34:18 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sat, 17 Jun 2000 08:34:18 -0400
Subject: performance benefits from byte code files?
References: <8idr3p$33h$1@nnrp1.deja.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E65@mail.sauder.com>

Dan Gindikin wrote:
> 
> I was wondering what the performance benefit is from having a .pyc file
> around.
> 
> Here an example that motivates this quiestion. Lets say I have a
> relatively large script.py CGI script. Since it is always executed
> directly, a script.pyc file is never generated. Is it then beneficial
> to write a tiny wrapper that does nothing but
>    import script
>    script.run()
> because then script.pyc does get generated?

Check out this url:
http://www.lyra.org/greg/python/#dev
Greg Stein gives a method of executing *.pyc files directly for CGI.

Personally, I'd like to know if *.pyo files also will work the same way.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre



From rsteelandt at bucodi.com  Mon Jun 19 12:15:58 2000
From: rsteelandt at bucodi.com (BuCoDI-Rony Steelandt)
Date: Mon, 19 Jun 2000 18:15:58 +0200
Subject: accessing access from Python on windows
References: <394E51D1.71382D39@nrc.ca>
Message-ID: <394E473D.F93DC725@bucodi.com>

www.python.org/windows/win32all

(i think :)

Marcin Kolbuszewski a ?crit :

> Hi there!
>
> I am running python 1.5.2 on Windows NT and am trying to
> access an Access database through ODBC. No lick though...:
>
> Traceback (innermost last):
>   File "C:\PROGRA~1\Python\Tools\idle\ScriptBinding.py", line 131, in
> run_module_event
>     execfile(filename, mod.__dict__)
>   File "D:\python\toc.py", line 3, in ?
>     import dbi, odbc
> ImportError: No module named dbi
>
> Where may I get the dbi and odbc modules?
>
> marcin Kolbuszewski

--
======================================================
Bureau de Conseil et de D?veloppement Informatique
2, Rue Henri Avril
22480 Saint - Nicolas du Pelem

T?l : +33.2.96.29.79.08
Fax : +33.2.96.29.79.09

e-mail : bucodi at worldnet.fr
======================================================




From moshez at math.huji.ac.il  Thu Jun  8 21:52:51 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 04:52:51 +0300 (IDT)
Subject: wishlist Python3K: subclass ints
In-Reply-To: <8hp0hi$oh4$1@slb3.atl.mindspring.net>
Message-ID: <Pine.GSO.4.10.10006090447300.19646-100000@sundial>

[Moshe]
> Actually, in the number case specifically, I'd go for the
> no-hierarchy-have-a-good-model-optimize-behind-the-covers strategy:

[Aahz]
> You have a point, but I do still think there should be some sort of
> virtual base class behind all numbers.  That way, it's at least clear
> what operations should be available in any number class.

No, you're missing my point. At the Python level, there should be a single
number type. Asking whether a number is integral is fundemantelly
different then asking whether it is even, and yet nobody advocates having
odd and even numbers seperate classes deriving from the same base class.
A number is a number is a number -- and should be a single type. This is
where "optimize behind the scenes" comes in: we should have a "number"
type, s.t., isinstance(1, number) as wel as isinstance(1.5, numer), is
true. However, current implementation details say that it
would be convinient implementation-wise if type(1)!=type(1.5) -- which is
fine, just say that type(number) is implementation dependant, and cannot
be relied upon.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From jam at quark.emich.edu  Tue Jun 20 22:05:32 2000
From: jam at quark.emich.edu (Jeff)
Date: Tue, 20 Jun 2000 22:05:32 -0400
Subject: Current date and time
In-Reply-To: <395017fb@newsprime.tidalwave.net>; from nospam@quark.emich.edu on Tue, Jun 20, 2000 at 09:18:51PM -0400
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com> <20000620010356.A892@better.net> <395017fb@newsprime.tidalwave.net>
Message-ID: <20000620220532.A18093@quark.emich.edu>

On Tue, Jun 20, 2000 at 09:18:51PM -0400, Aeneas <shan_mu wrote:
> How do you format the result as--
> "Tuesday  20 June 2000  21:18hr" ?
> Thanks.
> 

greetings,

there are many ways to do this (as with everything), but here's what I came
up with:

[~] [9:53pm] [jam at toast-pts/8] % python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import time
>>> t = time.localtime(time.time())
>>> print time.strftime("%A %d %B %Y %H:%Mhr", t)
Tuesday 20 June 2000 21:55hr

for examples of the parameters available in strftime, check the man page
('man strftime'). you can also find documentation from the python website
about the 'time' module here:
http://www.python.org/doc/current/lib/module-time.html

hope that helps a little.. have fun! ;)

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 



From adams_s at lab.eng.usyd.edu.au  Wed Jun  7 22:33:49 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Thu, 8 Jun 2000 12:33:49 +1000
Subject: Newbie Q: CGI
Message-ID: <8hn0c1$1a22$1@news1.wire.net.au>

Hi,

How do you get the IP number from a request for a web page? gethostbyname()
and gethostbyaddress() weren't mentioned anywhere in the cgi module docs
(that I could see, did I miss it?)

simple problem, but I just can't find the answer...

thanks Steven




From aahz at netcom.com  Mon Jun 19 09:23:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jun 2000 13:23:34 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <20000619184000.639516@210.9.22.17>
Message-ID: <8il6sm$g00$1@slb3.atl.mindspring.net>

In article <20000619184000.639516 at 210.9.22.17>,
Matthew Schinckel  <matt at null.net> wrote:
>"William Dandreta" <wjdandreta at worldnet.att.net> wrote:
>>
>> I recently read a message that suggested that using regular
>> expressions might be faster than the find function. I've heard the
>> term regular expression used from time to time but I don't know what
>> it means. Could someone give me a simple explanation?
>
>I have come across one example where re was much faster than using
>string.find: I was stripping out html tags, and doing a one line
>re.sub('<*?>','') was much faster than using string.find() and slice
>notation.

Sure!  But then you're not really doing just a string.find().  The point
is that for any operation where string.find() alone can be used, it will
be faster than an re.  Any time you add complexity, such as find/replace
with a pattern rather than a fixed string, chances are good that the re
will be faster.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From robin at jessikat.demon.co.uk  Sat Jun 17 09:48:36 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 17 Jun 2000 14:48:36 +0100
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
 <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com>
Message-ID: <+FEUyXA0G4S5EwL0@jessikat.demon.co.uk>

In article <3eomks8pah3at7h9ponni0r7s8ekt4ukh9 at 4ax.com>, Bob Hays,
Computer Geek <rhays at interaccess.com> writes
....
>
>My interesting tale is a consultant I have right now working on some
>production Python code.  He started making changes using vi and
>realized that we had carefully used spaces for indention (I had that
>holy war in a previous life [at a shrink wrap vendor that had to run
>on almost anything], and we finally agreed that tabs were evil in a
>cross-platform environment).  He started using spaces and now isn't
>having problems.
>
...

Funny, being old and used to long comms delays I've always preferred
tabs. It's true that poor windowing/widget systems sometimes copy tabs
to spaces, but there are always a few glitches. A two second response
delay often converts space users to tabbers.
-- 
Robin Becker


From jkraska1 at san.rr.com  Fri Jun 16 23:57:39 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 17 Jun 2000 03:57:39 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <3948D25D.7C83CD6@micron.net> <39499FC1.8C31F7B5@sightreader.com>
Message-ID: <394AF7C9.275D5FC6@san.rr.com>

> > Courageous wrote:
> > > Most of your C++ friends would be picking their tongues
> > > up off the floor if they saw the LOC of python programs
> > > in comparison to their C++ counterparts.

> Peter Immarco wrote:
> >What's the LOC acronym  stand for?

Ken Seehof wrote:
>  LOC (lines of code) is also a measure of programmer output.  The idea is
> that you evaluate your programmers by the sheer quantity of code they
> crank out.  The one who cranks out the most lines of code per week is
> clearly the best programmer (yeah right!).  Anyway, it is sometimes used
> as a rough extimate of productivity.

"As a young programmer, I measured productivity in lines of
code written. As an experienced programmer, I learned to
measure productivity in lines of code NOT written."

C/

(who thinks he deserves to have his words carved in marble somewhere)


From gmcm at hypernet.com  Thu Jun 15 14:20:27 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jun 2000 18:20:27 GMT
Subject: Can I build a single file containing all the contents of LIB?
References: <Vg525.3907$227.93950@nnrp1.uunet.ca>
Message-ID: <8F549C81Dgmcmhypernetcom@199.171.54.195>

Warren Postma wrote: 

>I have an embedded system which has no support for long filenames. It's
>also a pain to upload the whole LIB subdirectory to this system.  The
>embedded system has a FAT filesystem and only supports 8 character
>filenames with a 3 character extension. 

There make8x3.py in the PC directory for long file name problems. You also 
need the environment variable LFN set to N (does that still work?).

>The first problem then is with
>the "exceptions.py" and from there on.    

You'll have to freeze that in, or have the 8x3 version sitting on your 
PYTHONPATH. It's imported from C during Py_Initialize(), so you've got no 
chance of interfering otherwise.

>Since my system recompiles for
>both native NT/Win32 and embedded mode, what I would ideally like to do
>is have a python script "bundle up" the standard Python library into a
>file, and then just upload that bundle. Then I would like it if python
>would search the \LIB directory FIRST for xxx.py and if not found, then
>search in the compressed (or not) file for it.   I would like the file
>to be separate from the executable, not linked in, so I can change the
>python library without recompiling the executable.

Start my Jim Ahlstrom's zipfile, or the .pyz stuff I do in my Installer 
(both of which rely on Greg Stein's imputils). But you'll have to do some 
hacking.

- Gordon


From thomas at xs4all.nl  Mon Jun 26 03:35:55 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 07:35:55 GMT
Subject: String wrapper?
References: <B1D55.2657$Gh6.168003@news-west.usenetserver.com>
Message-ID: <slrn8le1uf.2oh.thomas@centurion.xs4all.nl>

On Sun, 25 Jun 2000 23:44:24 -0800, Matthew Cline <matt at nightrealms.com> wrote:
>Is there any way to make a wrapper for a string object,
>like UserDict is a wrapper for the dict type?

There is the UserString module in the 1.6 alphas, and it seems to be fairly
portable. (It defines all methods that strings have in 1.6, whereas they
have none in 1.5.2, but that shouldn't bother portability. You can't use
them unless you use 1.6 strings as well, though, because those methods are
defined in terms of string methods ;-)

You can get the UserString module by downloading a 1.6 alpha, or by grabbing
a recent CVS tree, or by browsing the CVS three through sourceforge:

http://www.sourceforge.net/cvs/?group_id=5470

The module should at least show that it's *possible* to write a string
wrapper ;-) though without the 1.6 additions to strings & the __contains__
overloader it might not be that efficient.

Regards,
Thomas.


From mvanier at endor.bbb.caltech.edu  Wed Jun 14 21:41:55 2000
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 14 Jun 2000 18:41:55 -0700
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu> <8F51D9DB3gmcmhypernetcom@199.171.54.154> <jl7lbtm60s.fsf@endor.bbb.caltech.edu> <8F52DC18Agmcmhypernetcom@199.171.54.194>
Message-ID: <jl1z1zaffg.fsf@endor.bbb.caltech.edu>


gmcm at hypernet.com (Gordon McMillan) writes:

> [posted and mailed]
> 
> Michael Vanier wrote: 
> 
> >gmcm at hypernet.com (Gordon McMillan) writes:
> >
> >> Michael Vanier: 
> 
> [snip]
> 
> >> >To fix this, I changed the flag to be a list with one integer
> >> >element. Now, when I change the list element in the parent, the child
> >> >sees it too.  
> >> 
> >> 
> >> Let me guess. You did something like make "flag" a module level
> >> global, and then did "from ... import *" in the other.
> >
> >Even dumber than this; it was all in the same file.  I set the global in
> >the parent thread, and expected that the new thread would see the new
> >value when it was changed.
> 
> And did the routine that set it say "global flag"? If not, it set a local 
> (within the routine).
> 
> >> 
> >> If you had referenced "module.flag", all would have worked. But that 
> >> stinks, too. Your thread should have a flag member, and the parent
> >> thread should set it (through a setter, if you really want to be
> >> virtuous <wink>). That way the flag becomes part of the thread's API,
> >> instead of a bass- ackwards dependency.
> >> 
> >
> >Well, the thread wasn't an object; I just called a function and expected
> >that it would see the same data (see below).
> 
> Ah. I assumed that you were using threading, not thread. The latter is the 
> low level primitives. The former is the one to use.
> 
> >So what you're saying is that a thread's namespace is not the same as
> >the global namespace of the parent.  This is puzzling, since that's what
> >I thought a thread was: effectively a new process which shares the
> >parent's namespace.  Is this in a FAQ somewhere?  Or am I just not
> >getting it? 
> 
> No you're not. This has NOTHING to do with threads.
> 
> >To further complicate things, here is a stripped-down version of the
> >code that actually works!
> 
> If you'd post the code that *doesn't* work, then we could quit this 
> guessing game!
> 
> 
> 5-to-1-on-the-missing-global-ly y'rs
> 
> - Gordon

Right you are; sorry.  Here is the code that doesn't work:

#! /usr/bin/env python

import os, sys, string, thread
from Tkinter import *


def run():
    i = 0

    while 1:
        print "in child thread"
        if stop_flag:
            print "leaving child thread"
            break


#
# Main body of program.
#

# Globals.

tkgeometry        = "200x200+0+0"      # Default; can be overridden.
start_time        = 0.075
root = Tk()
root.geometry(tkgeometry)
stop_flag = 0
thread_on = 0

def callback(event):
    global thread_on
    key = event.char

    if key == "s":
        if not thread_on:
            stop_flag = 0
            thread_on = 1
            thread.start_new_thread(run, ())
        else:
            print "halting thread"
            stop_flag = 1
            thread_on = 0
    elif key == "q":
        raise SystemExit


frame = Frame(root, width=200, height=200)
frame.bind_all("<Key>", callback)
frame.pack()
os.system("sleep 2")
root.mainloop()

# End of script.

To test it, hit "s" in the Tk window to start and again to (attempt to)
stop.  It won't work, but if stop_flag is a list with an integer element and
the code is suitably modified, it will.  Also, without the Tkinter code, it
does work.

Sorry-to-belabor-the-point-ly y'rs,

Mike


--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.


From chibaA at TinterlogD.Tcom  Fri Jun 23 18:16:59 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Fri, 23 Jun 2000 22:16:59 GMT
Subject: simple question... 
Message-ID: <3953e13a.761314922@news1.on.sympatico.ca>

Hello,

I'm sure this is a simple question...

how would I call another python script from within a script.  ie. I
want to basically stop the script that's running, and start running
another.  I don't have any particular functions in the target script,
and I don't want to continue running the script I'm currently in (ie.
I don't want to use import xxxxx).  I would also like to avoid a
system call if possible.

Any ideas? 

TIA!




From digitig at cix.co.uk  Wed Jun 14 17:31:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 14 Jun 2000 22:31 +0100 (BST)
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <memo.20000614223158.31901C@digitig.compulink.co.uk>

In article <F2005.859$Qf6.33080 at nuq-read.news.verio.net>, lk at veriomail.com 
(Lawrence Kesteloot) wrote:

> 
> If one of the goals of Python 3000 is to make it easy for new
> programmers to learn, how about removing the distinction
> between = and ==?  The only kinds of examples I can think of

<snip>

The construct
	a = a + 1
can confuse newbies too, provoking a response along the lines of "not for 
finite 'a' it doesn't!"

I reckon that's sufficient ground for a distinction between assignment and 
equality. '=' possibly isn't the best choice for assignment, though I much 
prefer it to "ADD 1 TO a GIVING a".


From jkraska1 at san.rr.com  Fri Jun 16 12:33:59 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 16 Jun 2000 16:33:59 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <394a5109.80394991@10.1.1.28>
Message-ID: <394A5795.F4DE98FB@san.rr.com>

> Suppose, however, that you could get digit 'n' of pi directly (or e,
> or any other inifinte, nonrepeating sequence).  All I have to do is
> communicate which digit of pi the message key starts at, and I can
> generate the rest of the key.

This is a collosally bad idea, IMO.




C/


From hzhu at knowledgetrack.com  Wed Jun 21 19:39:33 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 21 Jun 2000 23:39:33 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com> <uitv9qb72.fsf@ctwd0143.fitlinxx.com> <slrn8kobd5.1vs.hzhu@mars.localdomain> <vaf4s6rvtow.fsf@lucy.cs.uni-dortmund.de>
Message-ID: <slrn8l2kn5.q2.hzhu@rocket.knowledgetrack.com>

Thanks a lot.  It really helps.

On 18 Jun 2000 16:29:37 GMT, Kai Gro?johann
<Kai.Grossjohann at CS.Uni-Dortmund.DE> wrote: 
>hzhu at mars.localdomain (Huaiyu Zhu) writes:
>
>> This is simple to do for one file, but not for many files.  I'd prefer
>> running something in batch mode on a list of files.
>
>There are Unix programs `expand' and `unexpand'.  Presumably, running
>the following commands on a file foo would change tab stops from 4 to
>8:
>
>    expand -4 foo > foo.new
>    unexpand -8 foo.new > foo
>    rm -f foo.new
>
>You could easily do this in a loop:
>
>    for f in *.c
>    do
>        expand -4 $f > $f.new
>        unexpand -8 $f.new > $f
>        rm -f $f.new
>    done
>
>Does this help?
>
>kai


-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From klaus.bruns at teleatlas.com  Tue Jun 13 02:16:30 2000
From: klaus.bruns at teleatlas.com (Klaus Bruns)
Date: Tue, 13 Jun 2000 08:16:30 +0200
Subject: Thanks
Message-ID: <3945D1BE.B4706667@teleatlas.com>

Hi Greg,

Thanks for your answer! That is exactly what I want to do.

Bye
	Klaus


From t_e_sanders at my-deja.com  Thu Jun  1 19:24:52 2000
From: t_e_sanders at my-deja.com (t_e_sanders at my-deja.com)
Date: Thu, 01 Jun 2000 23:24:52 GMT
Subject: Copyright Notice in Commercial Software
Message-ID: <8h6rbt$ksv$1@nnrp1.deja.com>

Howdy. I'm interested in using Python within a commercial software
package. Python's license seems to be really good about this, but I
have a question about the copyright notice:

> Permission to use, copy, modify, and distribute this software and its
> documentation for any purpose and without fee is hereby granted,
> provided that the above copyright notice appear in all copies and
> that both that copyright notice and this permission notice appear in
> supporting documentation...

Do the terms of Python's distribution license require me to allow other
users to copy and distribute my software? I would guess not, but
inclusion of this notice in my package explicitly states so. There's
obviously something I'm missing, (one of the downsides of being a
developer, rather than a lawyer? :) Can anyone shed some light on this?

Thanks much,
Thomas Sanders
t.e.sanders at usa.net



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


From hat at se-46.wpa.wtb.tue.nl  Fri Jun 23 03:56:39 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 23 Jun 2000 07:56:39 GMT
Subject: boolean true and false values.
References: <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl> <memo.20000622231736.45677A@digitig.compulink.co.uk>
Message-ID: <slrn8l69kp.geg.hat@se-46.wpa.wtb.tue.nl>

On Thu, 22 Jun 2000 23:17 +0100 (BST), Tim Rowe <digitig at cix.co.uk> wrote:
>In article <slrn8l1oc2.3c1.hat at se-46.wpa.wtb.tue.nl>, 
>hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:
>
>> I read a number of replies, but didn't read an answer that seems 
>> satisfactory.
>> I ran in the same problem, and was also not able to find a elegant 
>> solution.
>> Using 0 and 1 or `true` and None is not correct imho, there should be a 
>> native
>> way to use true and false for indicating boolean values.
>
>There is not an elegant solution, short of making very fundamental changes 
>to the language that would IMHO break its nature. There is no problem with 

I also fear that it would break things if you'd make changes to something
fundamental as this.

To be honest, I tried to write a reply yesterday, but couldn't really argue why
I feel bad about not having boolean values, so I did a bit of thinking first.

Today, I think I may have a part of the answer.
My work is language and tools development of a specification language aimed at
users with almost no experience in programming.
Unlike Python, our specification language is statically typed.
For my work, I program in C++.
Obviously, that makes me biased towards a statically typed language :-)

My trouble with python is that there are types, and at the same time they are
not there, which is confusing to me.

As an example:

a="a'slrk;jwhg"
a=1

works, since Python dynamically adapts the type of a (this may be not entirely
semantically correct, but I hope you understand my point).

a=1
print "a"+a

does not work, as Python keeps telling me every time I try.

Maybe I make a reasoning mistake somewehere, but to me a sometimes adapts
dynamically, and sometimes not.
As a language aimed at 'the masses', I think this is hard to explain.

Then the lack of booleans.
In my work I noticed that having clean and orthogonal concepts helps starting
programmers. In our language design the language tries to force users into
thinking cleanly and orderly, otherwise they end up with a messy program which
the machine doesn't like.
Having a separate concept 'boolean' forces users to really write tests, rather
than arbitrary expressions which happen to deliver the right result.
An example (or rather, 2 examples :-) ):

>>> if []:
...   print "yes"
... else:
...   print "no"
... 
no
>>> if [1]:
...   print "yes"
... else:
...   print "no"
... 
yes
>>>

I'd really hate to explain that a list has a truth value to a beginner.
(in fact, I would even argue that [] should deliver true, since it _IS_ a list
instead of None).

Last but not least, I am not sure whether you'd really want the dynamic nature
of current Python. [ what do you mean, a migration path ? :-) ]
>From programming theory, I learned that there are invariants which have to be
preserved. Something like "the variable 'total' contains the total sum of
elements dealt with upto now". Obviously, total has a single type here, which
makes the entire dynamic typing concept with sequences like 'a="rg"; a=1' unwanted.

I realize that introducing new variables on the fly is extremely important for
ease of hacking^H^H^H^H^H^H^Hprogramming, but letting it mess up the type
concept seems an unwanted side effect to me.

BTW: the semicolon in the sequence above is there not by accident, I'd really
     like to see Python move towards multi-statement lines, as it makes programs
     more readable if used correctly.
     I personally use the idea of 'first statement on a line messes up an
     invariant, the remainder of the line restores the invariant' in C++, which
     works great.

>assignment, but there is a serious danger with testing. You would have the 
>situation that
>	if foo:
>could follow the true branch, but
>	if foo == true:
>would follow the false branch. This has been a perennial C/C++ issue, and 

I know, just defining false=0 and true=1 will not work.

>C++ now has boolean types, and that problem has been largely resolved, but 
>that's only been possible because C++ is (fairly) strongly typed and will 

For C/C++, mixing boolean with integers is not bad, as C/C++ has a different
attitude towards the user. (Still, seeing C++ introducing bool means to me that
there is apparently a need for this concept even in C/C++).

>> Also, I am missing the data type 'set'. It can be faked with a 
>> dictionary, but it
>> is just not the same. Any one else got this problem ?
>
>It can be implemented very easily as a class wrapper around a list, that 
>checks for presence before adding an element, and defines union, 

Yes, I know, but why is it not in the language ?
Quite a lot of algorithms rely on having sets. If you put them in a list, there
is an order which is not always wanted.
Also, implementing a set in anything else than a set is conceptually not clean.
(to be fair, the list operations are quite extensive, so living without sets is
doable).

>intersection, &c. It raises interesting questions of how deep the equality 
>test should be, and I suspect different applications would need different 
>answers, so I'm not sure it should be a built in. Anyway, all the paired 

The user should be able to customize it, but this is really about the equality
test rather than the set concept.

>delimiters on my keyboard are already used -- what would you delimit it 
>with? :-)

{x,y,z} ?

(it may be possible to extend the dictionary concept with 'x' meaning 'x:none')

[ Before I my mailbox gets filled with flames, I do love Python, and I am
  pushing it as the standard glue language in our group. The comments above are
  solely arguments for improving Python further
]

Albert
---
Look ma, windows without Windows !!


From steffen at biomedcomp.com  Sat Jun 17 16:30:56 2000
From: steffen at biomedcomp.com (David Steffen)
Date: Sat, 17 Jun 2000 15:30:56 -0500
Subject: Problem Installing/Using Windows Extensions
Message-ID: <2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1@lp.airnews.net>

Howdy:

  I'm trying to use Pythonwin with Python 1.52 under Windows 95 and am
consistently running into a problem that I don't understand at all,
probably because I know relatively little about Windows.

  When I selected "COM Makepy Utility" from the Tools menu of
Pythonwin, I got the following error message:

"""
>>> Failed to execute command:
from win32com.client import makepy;makepy.main()
Traceback (innermost last):
  File "C:\Program Files\Python\Pythonwin\pywin\framework\toolmenu.py"\
, line 103, in HandleToolCommand
    exec "%s\n" % pyCmd  File "<string>", line 1, in ?
  File "C:\Program Files\Python\win32com\client\makepy.py", line 46, \
in ?
    import genpy, string, sys, os, types, pythoncom
  File "C:\Program Files\Python\win32com\client\genpy.py", line 16, in ?
    import pythoncom
    Import Error: DLL load failed: A device attached to the system is \
not functioning.
"""

  I noted that what I had was version 1.25, and that version 1.32 was
up on Mark Hammond's site on starship python, so I installed the newer
version.  The installed completed, but with three error messages near
the end:

"""
Registering COM Server AXScript Engine

Registration of the AXScript Engine COM server failed.

Installation will continue, but this server will require manual
registration before it will function.

exceptions.ImportError: DLL load failed: A device attached to the
system is not functioning.
"""

...with the other two messages being similar ones about the Python
Interpreter COM server and the Dictionary COM server.

I'd be delighted to RTFM, if I could find the relevant M (I've looked
in the included Python docs and on the python.org and starship python
websites).

I'd be happy to register the servers manually, if that would solve the
problem and if I had a clue as to what than meant.

If there is a device not functioning on my laptop, I'd love to know
that, but a disk scan indicates no problems and I haven't noticed any
other problems with my laptop.  I'm thinking I may be missing a DLL,
driver, or other software component; if anyone could suggest which one,
I'd be pleased as punch.

Thanks in advance!

-David Steffen-


From m.faassen at vet.uu.nl  Thu Jun  1 18:46:47 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 1 Jun 2000 22:46:47 GMT
Subject: *blink* That's neat.
References: <GXJY4.28606$WS3.320758@typhoon.we.rr.com> <SZKY4.6213$wYl.227247104@newsb.telia.net> <8h07fv$bnc$3@newshost.accu.uu.nl> <UINY4.6227$wYl.223307264@newsb.telia.net> <8h0ne2$70i$2@newshost.accu.uu.nl> <3934242B.ECFDBAAE@endea.demon.nl> <8h1k0q$h04$2@newshost.accu.uu.nl> <wUfZ4.6380$wYl.229799424@newsb.telia.net> <8h4djp$pkc$1@newshost.accu.uu.nl> <jXnZ4.6415$wYl.227858944@newsb.telia.net>
Message-ID: <8h6p4n$ecq$1@newshost.accu.uu.nl>

Fredrik Lundh <effbot at telia.com> wrote:
> Martijn Faassen wrote:
>> > I'm probably just dense, but how would you design such a
>> > double-blind test?
>> 
>> Construct two Python interpreters (or environments). One is case-sensitive,
>> the other is case-insensitive. Then randomly distribute them to your
>> newbie programmers.

> how do you make sure that the *observers* won't be able to
> figure out what kind of interpreter any given user is using?

With observers, do you mean the people helping the newbies? I was
imagining some controlled circumstances where there aren't any 
helpers around when the newbies get down to the actual business of
programming. Of course that biases your tests, *too*. 

Or do you mean the newbies themselves? Obviously they can figure it
out for themselves, but that's part of the test, isn't it? 

> (if you don't, it's not really a double-blind test, right?)

Anyway, you may be right and it's not a genuine double-blind test; but
if you want to do actual measurement with some semblance of objectivity
you're going to have to devise something like this. Perhaps some
messing around with various circumstances and statistics can allow one
to get some decent results that way.

Anyway, I was making the point that good empirical tests on this are
hard to do; perhaps you're right and double-blind tests are impossible.

Regards,

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


From morganea at bellatlantic.net  Tue Jun 13 18:29:14 2000
From: morganea at bellatlantic.net (Michel Orengo)
Date: Tue, 13 Jun 2000 22:29:14 GMT
Subject: need feedback on ADO and Python
Message-ID: <_Ay15.2965$Wt2.50447@typhoon2.ba-dsg.net>

Hi all!
I've been playing around with ADO (Microsoft ActiveX), and I came up with a
little module to give a more pythonic feel to the interface.
Now, I have a very limited knowledge of COM and Active X, so I'd like to
know if what I've done can be done more elegantly and if this module is
useful or not.
So here is the module.


I know that it is a very limited version but that's a beginning...
Thanks for your feedback.


begin 666 ADOpy.py
M(R M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2T-"B, at 041/<'DZ
M($5A<WD at 26YT97)F86-E(&9O<B!!1$\-"B, at +2TM+2TM+2TM+2TM+2TM+2TM
M+2TM+2TM+2TM+2TM+2TM#0HC#0HC(%=R:71T96X at 8GD@36EC:&5L($]R96YG
M;RX at 4'5B;&EC($1O;6%I;BX-"B, at 3F\@0V]P>7)I9VAT("P@;F\@4FEG:'1S
M(%)E<V5R=F5D+"!A;F0@;F\@5V%R<F%N=&EE<R!W:&%T<V]E=F5R+ at T*(PT*
M(R!V97)S:6]N(# N,0T*#0HC(R!4:&ES(&UO9'5L92!I;G1E;F1S('1O('!R
M;W9I9&4 at 82!M;W)E(")P>71H;VYI8R(@:6YT97)F86-E('1O($%$3PT*(R,@
M041/(#T at 36EC<F]S;V9TPJX at 06-T:79E6,*N($1A=&$@3V)J96-T<PT*(R,@
M5&AI<R!I<R!A;B!E>&%M<&QE(&)A<V5D('5S:6YG(&YA=&EV92!P>71H;VYC
M;VT@:6YT97)F86-E. at T*(R,@:6UP;W)T('=I;C,R8V]M+F-L:65N= T*(R,@
M8VYN(#T@=VEN,S)C;VTN8VQI96YT+D1I<W!A=&-H*"=!1$]$0BY#;VYN96-T
M:6]N)RD-"B,C(&-N;BY0<F]P97)T:65S*")$871A(%-O=7)C92(I+E9A;'5E
M(#T@)TUY1&%T86)A<V4G#0HC(R!C;FXN4')O<&5R=&EE<R at B57-E<B!)1"(I
M+E9A;'5E(#T@)U5S97).86UE)PT*(R, at 8VXN4')O<&5R=&EE<R at B4&%S<W=O
M<F0B*2Y686QU92 ]("=0<W=R9"<-"B,C(&-N;BY/<&5N*"D-"B,C(')S=" ]
M(&-N;BY%>&5C=71E*"=S96QE8W0@*B!F<F]M($UY5&%B;&4G*0T*(R, at 9FQD
M(#T@<G-T+D9I96QD<R at G8V]L,2<I("-C86X at 86QS;R!B93H at 9FQD(#T@<G-T
M*"DH)V-O;#$G*0T*(R,@<')I;G0 at 9FQD+E9A;'5E("-C86X at 86QS;R!B93H@
M<')I;G0@<G-T*"DH)V-O;#$G*0T*(R,-"B,C($YO=RP@=&AE('-A;64@=VET
M:"!!1$]P>3H-"B,C(&EM<&]R="!!1$]P>0T*(R, at 8VYN(#T at 041/<'DN0V]N
M;F5C=&EO;B at I#0HC(R!C;FXN9&%T85-O=7)C92 ]("=->41A=&%B87-E)PT*
M(R, at 8VYN+G5S97))1" ]("=5<V5R3F%M92<-"B,C(&-N+G!A<W-W;W)D(#T@
M)U!S=W)D)PT*(R, at 8VYN+F]P96XH*0T*(R,@<G-T(#T at 041/<'DN4F5C;W)D
M<V5T*&-N;BYE>&5C=71E*"=S96QE8W0@*B!F<F]M($UY5&%B;&4G*2D-"B,C
M(&9L9" ](')S=%LG8V]L,2==#0HC(R!P<FEN="!F;&0-"B,C#0HC(R!!<R!Y
M;W4 at 8V%N('-E92!!1$]P>2!I;G1R;V1U8V5S('-O;64@<VAO<G1C=71S+"!A
M;F0 at 9&5A;',@=VET:"!T:&4 at 9'EN86UI8R!P<F]P97)T:65S(&%S#0HC(R!I
M9B!T:&5Y('=E<F4@<W1A=&EC('!R;W!E<G1I97,-"B,C($YO=&4Z(&1Y;F%M
M:6,@<')O<&5R=&EE<R!A<F4 at 9&5P96YD:6YG(&]N('1H92 B4')O=FED97(B
M+ at T*(R, at 5&AE>2!A<F4 at 86-C97-S:6)L92!B>2!T:&4 at 8V]L;&5C=&EO;B!O
M8FIE8W0N4')O<&5R=&EE<PT*#0H-"FEM<&]R="!S=')I;F<L(')E#0II;7!O
M<G0@=VEN,S)C;VTN8VQI96YT+"!P>71H;VYC;VT-"F9R;VT@=VEN,S)C;VTN
M8VQI96YT+D-O;G-T86YT<R!I;7!O<G0@*@T*#0I?=V]R9',@/2!R92YC;VUP
M:6QE*"<H/U \=V0^6T$M6EU;02U:72M\6T$M6EU;82UZ,"TY72LI)RD-"@T*
M8VQA<W, at 7T%$3V-L87-S. at T*(" @(&1E9B!?7V=E=&%T=')?7RAS96QF+"!N
M86UE*3H-"B @(" @(" @7VYA;64@/2!S=')I;F<N=7!P97(H;F%M95LP72DK
M;F%M95LQ.ET-"B @(" @(" @=')Y. at T*(" @(" @(" @(" @=F%L=64@/2!G
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M<FEN9RYU<'!E<BAN86UE6S!=*2MN86UE6S$Z70T*(" @(" @(" @=')Y. at T*
M(" @(" @(" @(" @<V5T871T<BAS96QF+D%$3V]B:F5C="Q?;F%M92P@=F%L
M=64I#0H@(" @(" @("!E>&-E<'0 at 071T<FEB=71E17)R;W(Z#0H@(" @(" @
M(" @("!?;F%M92 ]('-T<FEN9RYJ;VEN*%]W;W)D<RYF:6YD86QL*%]N86UE
M*2PG("<I#0H@(" @(" @(" @("!T<GDZ#0H@(" @(" @(" @(" @(" @<V5L
M9BY!1$]O8FIE8W0N4')O<&5R=&EE<RA?;F%M92DN5F%L=64@/2!V86QU90T*
M(" @(" @(" @(" @97AC97!T('!Y=&AO;F-O;2YC;VU?97)R;W(Z#0H@(" @
M(" @(" @(" @(" @<F%I<V4 at 071T<FEB=71E17)R;W(H;F%M92D-"B @("!D
M968 at 7V-A<W0H<V5L9BP at 041/;V)J96-T*3H-"B @(" @(" @<V5L9BY?7V1I
M8W1?7ULG041/;V)J96-T)UT@/2!!1$]O8FIE8W0-"B @(" @(" @#0IC;&%S
M<R!#;VYN96-T:6]N*%]!1$]C;&%S<RDZ#0H@(" @9&5F(%]?:6YI=%]?*'-E
M;&8L('!R;W9I9&5R/2<G*3H-"B @(" @(" @<V5L9BY?7V1I8W1?7ULG041/
M;V)J96-T)UT@/2!W:6XS,F-O;2YC;&EE;G0N1&ES<&%T8V at H(D%$3T1"+D-O
M;FYE8W1I;VXB*0T*(" @(" @("!I9B!P<F]V:61E<B A/2 G)SH-"B @(" @
M(" @(" @('-E;&8N041/;V)J96-T+E!R;W9I9&5R(#T@<')O=FED97(-"@T*
M8VQA<W, at 0V]M;6%N9"A?041/8VQA<W,I. at T*(" @(&1E9B!?7VEN:71?7RAS
M96QF+"!!1$]O8FIE8W0]3F]N92DZ#0H@(" @(" @(&EF($%$3V]B:F5C="!I
M<R!.;VYE. at T*(" @(" @(" @(" @<V5L9BY?7V1I8W1?7ULG041/;V)J96-T
M)UT@/2!W:6XS,F-O;2YC;&EE;G0N1&ES<&%T8V at H(D%$3T1"+D-O;6UA;F0B
M*0T*(" @(" @("!E;'-E. at T*(" @(" @(" @(" @<V5L9BY?8V%S="A!1$]O
M8FIE8W0I#0H-"F-L87-S(%)E8V]R9'-E="A?041/8VQA<W,I. at T*(" @(&1E
M9B!?7VEN:71?7RAS96QF+"!!1$]O8FIE8W0]3F]N92DZ#0H@(" @(" @(&EF
M($%$3V]B:F5C="!I<R!.;VYE. at T*(" @(" @(" @(" @<V5L9BY?7V1I8W1?
M7ULG041/;V)J96-T)UT@/2!W:6XS,F-O;2YC;&EE;G0N1&ES<&%T8V at H(D%$
M3T1"+E)E8V]R9'-E="(I#0H@(" @(" @(&5L<V4Z#0H@(" @(" @(" @("!S
M96QF+E]C87-T*$%$3V]B:F5C="D-"B @("!D968 at 7U]G971I=&5M7U\H<V5L
M9BP@:V5Y*3H-"B @(" @(" @<F5T=7)N($9I96QD*'-E;&8N041/;V)J96-T
M+D9I96QD<RAK97DI*0T*#0IC;&%S<R!&:65L9"A?041/8VQA<W,I. at T*(" @
M(&1E9B!?7VEN:71?7RAS96QF+"!!1$]O8FIE8W0]3F]N92DZ#0H@(" @(" @
M('-E;&8N7V-A<W0H041/;V)J96-T*0T*(" @(&1E9B!?7W-T<E]?*'-E;&8I
M. at T*(" @(" @("!R971U<FX@<W1R*'-E;&8N041/;V)J96-T+E9A;'5E*0H-
A"B @(" @(" @(" @( T*(" @( T*(" @(" @(" @(" @
`
end



From thomas at xs4all.net  Thu Jun 15 10:20:40 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 15 Jun 2000 16:20:40 +0200
Subject: PATCH: Augmented assignment
In-Reply-To: <8ianhh$dro$3@newshost.accu.uu.nl>; from m.faassen@vet.uu.nl on Thu, Jun 15, 2000 at 02:00:17PM +0000
References: <20000609163937.D20700@xs4all.nl> <8hs4d8$33b$1@animus.fel.iae.nl> <20000610120811.H20700@xs4all.nl> <8ianhh$dro$3@newshost.accu.uu.nl>
Message-ID: <20000615162040.I26436@xs4all.nl>

On Thu, Jun 15, 2000 at 02:00:17PM +0000, Martijn Faassen wrote:
> Thomas Wouters <thomas at xs4all.net> wrote:
> [discussion on naming]

> Since there's a discussion about naming anyway, I'd like to ask what
> the 'ad' stands for in __add_ad__? This may be another datapoint
> in the discussion. :)


To quote Guido (whitespaceeatingnanovirussed some):
"""
I know I shouldn't be posting in this thread, and I won't be there to read
the responses, but here's what I thought would be cool.

x+=y is syntactic sugar for x=x.__add_ab__(y); the "ab" means "and becomes"
(an old Algol-68 naming convention; we could pick something better later but
this will do for the explanation).
"""

See http://www.python.org/pipermail/python-list/2000-June/059556.html

Note that I took Guido's comment " ... what I thought would be cool." as a
confirmation that he actually *wants* augmented assignment ;) If he doesn't,
well, it was a great learning experience.

While I'm on the subject, I am still working on the patch. The version
currently on my laptop adds support for inplace-operators to extension (and
builtin) types, fixes a couple of refcounting bugs, sanitizes the new
2-argument opcodes, un-breaks the optimizer and adds support for the new
opcodes in the disassembler. I'm still not sure wether Guido will accept the
patch, but I'm getting ready to send the patch (in a bunch of small diffs,
actually) to the patches list and forget about it ;-)

I'll put up an updated version of the patch later today, I think, at
http://www.xs4all.nl/~thomas/python/

(The current page is a bit old, as I'm no longer stuck trying to find out
how to keep binary compatibility. I love Python !)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas at xs4all.nl  Tue Jun 27 07:38:14 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 27 Jun 2000 11:38:14 GMT
Subject: profiler -- numeric overflow?
References: <h579j8.grg.ln@kmself.nntp.ix.netcom.com> <31i9j8.dbh.ln@kmself.nntp.ix.netcom.com>
Message-ID: <slrn8lh4go.kg2.thomas@centurion.xs4all.nl>

On Mon, 26 Jun 2000 23:36:19 -0700, kmself at ix.netcom.com
<kmself at ix.netcom.com> wrote:
>kmself at ix.netcom.com wrote:
>> I'm new to Python.

>> ...yes, I've also considered running a smaller sample of the log, but I
>> suspect that runtime may grow nonlinearly, large input samples are good.

>Followup to Self:

>...well, at least things appear to grow linearly.  Runtimes varying
>input lines:
>
>  Input     Runtime (sec)   Lines/s
>  -----     -------------   -------
>      100        0.210        476
>    1,000        3.860        259
>   10,000        41.30        242
>  100,000      415.670        240

>...which is darned close to straight-line growth.  However, performance
>is about an order of magnitude less than I'd like to see.

It's very likely to be linear, yes, unless you store things in lists or
dictionaries larger than available memory. (swapping massacres performance,
even in Python.) However, Python performance in the area of strings, and
regexps in particular, is nowhere near Perl or C. It's likely that critical
paths of the code can be optimized some, but that depends on code layout
(and ofcourse how optimal it is, already ;)

What *might* prove a significant improvement is using Python 1.6 and 'sre'
instead of 're'. The 'sre' module isn't finished, and I'm not sure how buggy
it is, but it's rumoured to be faster, in certain areas. (/F gave some
impressive numbers, but those are old, and might've been in another problem
domain entirely.)

Thomas.


From hgg9140 at seanet.com  Sun Jun  4 01:11:40 2000
From: hgg9140 at seanet.com (Harry George)
Date: 03 Jun 2000 22:11:40 -0700
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee> <m3g0qv91f2.fsf@wilma.george.org> <393855E8.DF87D1DB@home.com> <8han19$4uc$1@news1.xs4all.nl>
Message-ID: <m3aeh2ni7n.fsf@wilma.george.org>

boud at rempt.xs4all.nl (Boudewijn Rempt) writes:

> Ivan Van Laningham <ivanlan at home.com> wrote:
> 
> > My brain is hurting from trying to understand XML in two days; please
> > post *anything* you can to help out!
> 
> 
> I was lucky when I was in your position - Lars Marius Garshol was just
> asking for reviewers for his draft of his Python/XML book, and I got
> to read it. It was very, very helpful ;-). Hope it gets published
> soon. 
> 
> However, you're free to download my code - I use XML a lot in:
> 
> dbobj - the repository is XML based
> kura  - webserver that takes XML templates and spews out HTML, and an
>         XML data import utility.

Oddly enough, I just posted a languages page (nothing to do with
either Python or Dutch :-)):
  http://www.seanet.com/~hgg9140/languages/index.html

> kxml  - a small PyKDE xml file browser (unfinished!)
> 
> You can get them all at: http://www.valdyas.org/python/index.html
> -- 
> 
> Boudewijn Rempt  | http://www.valdyas.org

-- 
Harry George
hgg9140 at seanet.com


From darrell at dorb.com  Sun Jun 18 16:04:47 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 18 Jun 2000 13:04:47 -0700
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com> <uitv9qb72.fsf@ctwd0143.fitlinxx.com> <slrn8kobd5.1vs.hzhu@mars.localdomain>
Message-ID: <04b401bfd960$7a9f95f0$6401a8c0@home>

From: "Huaiyu Zhu" <hzhu at mars.localdomain>
> >One fairly simple way is to untabify your file, change the tab width,
> >and then if you want re-tabify.  Emacs will use a mixture of TABs
> >(using the new tab width definition) and spaces to achieve the same
> >spacing you already have.
> >
>
> This is simple to do for one file, but not for many files.  I'd prefer
> running something in batch mode on a list of files.
>
I'm sure there are some better utils floating around, but this one seems to
work well for me.

import string, sys, glob

def main():
    tabSize=8 # I start with 8. Starting with 4 will break some files.
    g=glob.glob(sys.argv[1])
    print g
    a=raw_input('Tabs will be replaced with %d spaces\n Are you sure
Y/N:'%tabSize)
    if not a in ['y','Y']:
        return
    for a in g:
        buf=open(a).read()
        buf=string.replace(buf,'\t',' '*tabSize)
        open(a,'w+').write(buf)

main()


--Darrell




From hgg9140 at seanet.com  Sun Jun  4 01:09:11 2000
From: hgg9140 at seanet.com (Harry George)
Date: 03 Jun 2000 22:09:11 -0700
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee> <m3g0qv91f2.fsf@wilma.george.org> <393855E8.DF87D1DB@home.com> <8han19$4uc$1@news1.xs4all.nl>
Message-ID: <m3bt1inibs.fsf@wilma.george.org>

boud at rempt.xs4all.nl (Boudewijn Rempt) writes:

> Ivan Van Laningham <ivanlan at home.com> wrote:
> 
> > My brain is hurting from trying to understand XML in two days; please
> > post *anything* you can to help out!
> 
> 
> I was lucky when I was in your position - Lars Marius Garshol was just
> asking for reviewers for his draft of his Python/XML book, and I got
> to read it. It was very, very helpful ;-). Hope it gets published
> soon. 
> 
> However, you're free to download my code - I use XML a lot in:
> 
> dbobj - the repository is XML based
> kura  - webserver that takes XML templates and spews out HTML, and an
>         XML data import utility.
> kxml  - a small PyKDE xml file browser (unfinished!)
> 
> You can get them all at: http://www.valdyas.org/python/index.html
> -- 
> 
> Boudewijn Rempt  | http://www.valdyas.org

I also just uploaded pyAbiword.  A really simplistic reader/writer for
Abiword.  Available at:
  http://www.seanet.com/~hgg9140/comp/index.html

-- 
Harry George
hgg9140 at seanet.com


From invalid.address at 127.0.0.1  Mon Jun 19 13:13:21 2000
From: invalid.address at 127.0.0.1 (David)
Date: Mon, 19 Jun 2000 17:13:21 GMT
Subject: Python 3000 -> No, but there's 'R***'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <8iib7d$73f$1@news.jump.net> <394D26D5.E69282A7@uab.edu>
Message-ID: <394e547f.39131982@news.telus.net>

On Sun, 18 Jun 2000 19:42:57 GMT, Shae Erisson <shapr at uab.edu> wrote:

>The point as I see it, is that comp.lang.python is not
>comp.lang.recruiting. There are times when it's completely natural to
>talk about Ruby on comp.lang.python. But, small irritations add up, and
>repeated off-topic Ruby advocacy, even of a very small caliber, is not
>what I want to read on comp.lang.python.

Agreed.  Proselytizing the Way of Ruby in comp.lang.python isn't
appropriate.


From owen at astro.washington.edu  Wed Jun 14 13:57:28 2000
From: owen at astro.washington.edu (Russell E. Owen)
Date: Wed, 14 Jun 2000 10:57:28 -0700
Subject: Threads -- play well with others? If not, what instead? (Summary)
References: <8i34mr$9hti$1@nntp6.u.washington.edu> <7Ta15.315$i06.94722@ptah.visi.com>
Message-ID: <owen-82FC7D.10572814062000@news.u.washington.edu>

In article <7Ta15.315$i06.94722 at ptah.visi.com>, Grant Edwards wrote:

>>...
>In Tk, you can assign read handlers to file objects.  Anytime
>there is data available to be read, the handler will be called.
>
>Just open the socket connection and assign a read-handler to
>it.  Piece-o-cake.  Here's an excerpt from a program that uses
>that technique to handle data from a popen2'd child process:
>
>------------------------------------------------------------
>
>if cmd is None:
>    exceptString = 'no executable specified'
>    raise exceptString, cmd
>                                        
>self.__returnCode = -1
>self.__child = popen2.Popen3(cmd)
>self.__fd = self.__child.fromchild.fileno()
>
>fcntl.fcntl(self.__fd, FCNTL.F_SETFD, FCNTL.O_NDELAY);
>
>
>Tkinter.tkinter.createfilehandler(self.__child.fromchild,
>                                  Tkinter.tkinter.READABLE,
>                                  self.__stdoutHandler)
>                                  
>------------------------------------------------------------
>
>self.__child.fromchild      is a file object connected to the "read" end 
>of a pipe.
>Tkinter.tkinter.READABLE    is a constant that tells Tk what you care 
>about.
>self.__stdoutHandler        is a function to call when the file object has 
>data available.
>...

Thank you very, very much. That was exactly the information I needed! I 
got a basic telnet application running on my mac pretty quickly once I 
had that, and can now (finally) see how to design a user interface 
system.

Oddly enough, I have "Python and TKInter Programming", and it says 
nothing at all about file handlers. I'm wondering what else I'm missing 
in Tkinter. If anyone can recommend any documentation that discusses 
createfilehandler and such, I'd love to know about it. (I did find it in 
a book on Tk, once I knew what to look for, but without this example I 
doubt I'd ever have figured out how to call it from Python even if I'd 
known about it.)

Regards,

-- Russell


From paul at prescod.net  Tue Jun 20 07:56:47 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 20 Jun 2000 13:56:47 +0200
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no>
Message-ID: <394F5BFF.36342B7C@prescod.net>

Thomas Weholt wrote:
> 
>...
>
> I want to convince alot of my co-workers that OpenSource software,
> especially Python, is ready for big business. ( is it ?! )

It sure is! Even after the general market decline, Red Hat is almost a 4
Billion dollar company. That's with a "B". I don't know how they'll
justify that market cap but clearly Wall Street has bought into Open
Source.
  
> How do I sell an OpenSource solution like Python to hardheaded
> closedsource-fanatics? I know they care about money, so that would be
> my best angle. If anybody has any good examples of companies using
> Python, or similar technology, with profit, that would be great.

Luckily there is a group dedicated to selling open source to companies: 

http://www.opensource.org/

Lookit, any company that thinks that a product is not reliable or
supported because it is open source must be willfully closing their eyes
to the fact that the entire Internet depends on open source technologies
like Apache, bind, sendmail and the scripting languages. 

It sounds like your guys want to be "mainstream". Tell them that Yahoo
and Amazon both use free software to serve their websites. IBM is a big
backer of free software also.

http://www.netcraft.com/whats/?host=www.amazon.com (apache)
http://www.netcraft.com/whats/?host=www.yahoo.com (freebsd)
http://www.netcraft.com/whats/?host=www.hotmail.com (apache/freebsd)
http://www.netcraft.com/whats/?host=www.google.com (Python/Linux)

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"Music is the stuff between the notes." - Claude Debussy



From aahz at netcom.com  Tue Jun 20 09:55:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jun 2000 13:55:34 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <8inq3u$p0i$1@newshost.accu.uu.nl> <39586f3d.366407626@news.online.no>
Message-ID: <8int4m$2u4$1@nntp9.atl.mindspring.net>

[posted & e-mailed]

[Thomas, please keep attributions; otherwise, it's hard to keep track of
the conversation.]

In article <39586f3d.366407626 at news.online.no>,
Thomas Weholt <thomas at cintra.no> wrote:
>Attribution restored: Martijn Faassen
>>Attribution restored: Thomas Weholt
>>>
>>> Linux with Windows  2000/ NT 4.0.
>>
>>Why switch from Linux to Windows NT/2000, though? ;)
>
>Well, first off it`s my OS of choice, but mostly it *was* because it
>costs money. I`ve realized later that the cost of an os is scraps in a
>huge project so that`s not a very important issue anymore. But I would
>still go for Linux. It has remote-administration out-of-the-box with
>telnet/ssh, ready-and-running web-server ( Apache ) , dns, ftp - well,
>do I need to go on. AND PYTHON by default.

Um, Martijn was making fun of your English; maybe you should switch to
Dutch?  The way you wrote your sentence, it looks like your choice was
NT.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From root at rainerdeyke.com  Thu Jun  8 19:51:15 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 08 Jun 2000 23:51:15 GMT
Subject: Why can't I restore my redirected stdout?
References: <8hoo05$fc$1@news5.jaring.my> <20000608150314.A26601@corrada.com>
Message-ID: <TjW%4.9350$r7.1981176@news1.rdc2.tx.home.com>

<andres at corrada.com> wrote in message
news:20000608150314.A26601 at corrada.com...
> On Fri, Jun 09, 2000 at 01:52:52AM +0800, Raymond Ng Tong Leng wrote:
> > class Quiet:
> >     def write(self, data):
> >         pass
> > >>> sys.stdout = Quiet()
> > >>> print 'hello'
> > >>> sys.stdout = sys.__stdout__
> > >>> print 'hello'

> sys.stdout and sys.__stdout__ point to the same object. Save the default
> sdtout before you redefine it:

If sys.stdout and sys.__stdout__ point to the same object, and you assign
some other object to sys.stdout, it is be possible to restore it by
assigning sys.__stdout__ to sys.stdout.


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From weiwa592 at student.liu.se  Mon Jun 12 10:47:34 2000
From: weiwa592 at student.liu.se (Wang Wei)
Date: Mon, 12 Jun 2000 16:47:34 +0200
Subject: Jpython
Message-ID: <3944F806.36BE6A29@student.liu.se>

Hi,
I want to implement my project of Unix progress monitor in Jpython or
Java with python. Please kindly send me some examples code so that I can
refer to and read to study the ender's language.

Thank you
!

WW

Return-Path: <moeller at dophnic.yi.org>
Delivered-To: python-list at python.org
Received: from dophnic.yi.org (P11-cj17.cjnetworks.com [199.240.71.162])
	by dinsdale.python.org (Postfix) with ESMTP id B5BC11CE74
	for <python-list at python.org>; Sun, 11 Jun 2000 21:28:23 -0400 (EDT)
Received: from moeller by dophnic.yi.org with local (Exim 3.12 #1 (Debian))
	id 131J1b-0003gw-00
	for <python-list at python.org>; Sun, 11 Jun 2000 20:28:23 -0500
Date: Sun, 11 Jun 2000 20:27:28 -0500
To: Toy <gee308 at mediaone.net>
Subject: Re: quick newbie question on variables. please answer
Message-ID: <20000611202728.C13539 at dophnic.yi.org>
References: <39401B5F.122C6DB4 at mediaone.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2i
In-Reply-To: <39401B5F.122C6DB4 at mediaone.net>; from gee308 at mediaone.net on Thu, Jun 08, 2000 at 09:56:35PM +0000
Resent-From: moeller at dophnic.yi.org
Resent-Date: Sun, 11 Jun 2000 20:28:23 -0500
Resent-To: python-list at python.org
Resent-Message-Id: <E131J1b-0003gw-00 at dophnic.yi.org>
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta3
Precedence: bulk
List-Id: General discussion list for the Python programming language <python-list.python.org>

On Thu, Jun 08, 2000 at 09:56:35PM +0000, Toy wrote:
> Im a programming newbie, so most of these concepts are basic to me.  I'm
> trying to write really basic  CGI programs just to get started.  How
> would I pass a text variable that a user entered to find a man page for
> a tool?  This is the code I wrote (startx is just a command I wrote in,
> but how would I write it so any tool could be inputed) the 3rd line od
> code is written wrong for sure:
> 
> #!/usr/bin/python
> import commands
> b = commands.getstatusoutput('man %x!') % startx
> print = b
> 
>   I know that this is a real basic question, but I am 100% new to
> programming and I am self teaching.  Also, are ther large security risks
> if this program is on a web server(if so, how could I fix it?)?  Thanks
> for your time in advance.
> 

Yes, there is a security risk with this. For example, what if someone wrote
(for the variable) '""; cat /etc/passwd'? The shell would first execute
'man ""', and then after that it would execute 'cat /etc/passwd'. Then the
site user has instant knowledge of all users on the system. That's just
a minor example, of course, it can get much worse if the user running the
script has any priviledges. One way of dealing with this is sanitizing
the input, ie, strip pretty much anything except alphanumeric characters.
Also, make sure the script is run with very limited permissions. Those
are the basics, some other people may have more suggestions or better ways
to do the above.

And limit the variable length. Don't want some 'man' buffer overflow causing
you problems (and limit the length in python, as html-enforced limits can
be easily overridden).

-- 
Derek Moeller

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \



From kain at cableadmin.com  Tue Jun 20 18:17:20 2000
From: kain at cableadmin.com (kain at cableadmin.com)
Date: Tue, 20 Jun 2000 22:17:20 GMT
Subject: Python in home automation
References: <fsteele-E5789E.13533715062000@news.inet-systems.net> <cBd25.38645$cH1.355549@typhoon.austin.rr.com> <m3r99swt4x.fsf@hare.demon.co.uk> <fsteele-DDF9A7.16453020062000@news.inet-systems.net>
Message-ID: <slrn8kvrup.ld1.kain@cableadmin.com>

I recently took over maintenance of XAL which is the X10 Abstraction
Library, as well as began writing Python modules for it.  The XAL
site went up at sourceforge today (http://xal.sourceforge.net) and
the pyX10 page will go up in a couple days at http://pyx10.sourceforge.net
I would be very interested in working with anyone on python home automation.

-Scott
sbarron @ ansic.net

In article <fsteele-DDF9A7.16453020062000 at news.inet-systems.net>, 
Frank Steele wrote:
>In article <m3r99swt4x.fsf at hare.demon.co.uk>, Les Smithson 
><lsmithso at hare.demon.co.uk> wrote:
>
>> I have some Python scripts that control a TWSER & TW523. This includes
>> low-level drivers, and a higher-level server that lets you control X10
>> over a network. I also wrote a simple TkInter GUI for controlling my
>> X10 devices.
>> 
>> A while ago I spoke to someone who hacked my code to work with a
>> CM12U. We planned to rework this code to properly support both TW523
>> and CM12U, but nothing ever came of it.
>> 
>> Is it time to resurrect this?
>> 
>
>
>I would certainly be interested in having a look and modifying these for 
>my own (sinister) purposes, after which I would be happy to share 
>changes back into the community.
>
>I had one other response, so I suppose we could actually try to work 
>together in an organized way to move the project forward....
>
>Any other takers?
>
>please-take-one-step-forward'ly y'rs,
>
>Frank
>fsteele at mindspring.com
>
><snip>
>
>>     >> So, is anyone aware of Python modules or extensions to do X10
>>     >> home automation? I'm interested hearing about them on _any_
>>     >> platform, but I intend to implement on Linux and MacOS....
>>     >> 
>>     >> Thanks, Frank fsteele at mindspring.com


From mfletch at tpresence.com  Mon Jun 12 11:50:20 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 12 Jun 2000 11:50:20 -0400
Subject: IDLE  can be rewrited using wxPython?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33B1@VRTMAIL>

Notes:
	Newest versions of wxPython include the scintilla control that's
used in PythonWin's editor.
	There is a project (Boa Constructor) to create a wxPython IDE,
adapting that to your needs (in concert with Riaan and Robin) might be more
appropriate than attempting to rewrite IDLE.  Boa aims at also providing GUI
editing (though of the static dialog/form type, rather than the more common
dynamically-expanding-packed-widgets form last I looked (I haven't seen any
support for editing placers, would love to be wrong)).  Boa has been largely
a personal project up to now (i.e. Riaan is the only listed developer,
progress seems to go forward when he has the time to work on it).  Maybe
it's time to start putting some tasks on the list and getting people applied
to them?

	http://sourceforge.net/project/?group_id=1909

Enjoy,
Mike

-----Original Message-----
From: Pedro Vale Lima [mailto:eq3pvl at eq.uc.pt]
Sent: Monday, June 12, 2000 11:47 AM
To: python-list at python.org
Subject: Re: IDLE can be rewrited using wxPython?


mrfx wrote:

> The current IDLE uses TCL/TK as GUI widget.
> IDLE  can be rewrited using wxPython as GUI widget?

Sure. But not an easy task, replacing a GUI toolkit is always
a pain. Also it seems that the text widget is better in tk than
in wx.

pvl

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



From rjroy at takingcontrol.com  Sat Jun  3 10:12:02 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Sat, 03 Jun 2000 14:12:02 GMT
Subject: Combining two graphic images
References: <959996628.860207@centauri.ezy.net.au>
Message-ID: <393910bb.820165343@news1.on.sympatico.ca>

On Sat, 3 Jun 2000 11:45:12 +1000, "A J Wilson" <ajwilson at ezy.net.au>
wrote:

>Hi
>
>I would like to combine two BMP images so that one overlays the other.  Does
>anyone know how this could be done using Python.
>
>Thanks
>Austin
>
>

I think PIL might be what you need.

http://www.pythonware.com/products/pil/index.htm

This page of the tutorial talks about pasting images and transparency

http://www.pythonware.com/library/pil/handbook/intro01.htm


Good Luck!

Bob


From c.hintze at gmx.net  Fri Jun  9 02:46:54 2000
From: c.hintze at gmx.net (Clemens Hintze)
Date: Fri, 9 Jun 2000 06:46:54 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news> <8hlrkn$5t2$1@nnrp1.deja.com> <1ebvwau.af3wl21301xdcN%bparsia@email.unc.edu>
Message-ID: <m366rjtkpt.fsf@qiao.localnet>

>>>>> "Bijan" == Bijan Parsia <bparsia at email.unc.edu> writes:

    Bijan> <david_ullrich at my-deja.com> wrote: [snip]
    >>  I don't know anything about Ruby, but a language could
    >> certainly be more fully OO than Python. For example in Python
    >> although everything is an object not every type is a class (you
    >> can't subclass list, for example).
    Bijan> [snip]

    Bijan> Of course, this presumes that the more "class" you have the
    Bijan> more fully "OOPY" you are. *Really* ools have *class*. Yuk
    Bijan> yuk.

:-) :-) :-)

I suppose it should be reformulated a little bit ;-) Perhaps to: Ruby
*is* a OOL (very close to Smalltalk), whereas Python *has* OO
constructs in it (means OOP is possible like in e.g. Perl).

    Bijan> Of course, this ignores prototype based systems (e.g., Self
    Bijan> & NewtonScript), which, some would argue, are *more* oopy
    Bijan> than class based systems.

Exact! I do not think we could do comparations what lang is more OOP
than others. We should only mention, if a lang is a hybrid (means
contains a OO system) or whether it is based/build around OO ...

Further I would allow to call Self a OOL, because OO means only
Object Oriented! OO includes things like: polymorphism, inheritance
and encapsulation (did I forget something? :-/) ...

It say nothing about how to realize it. It happens only that most OOL
use class instance scheme until today ... ;-)

You could simulate/emulate class <-> instance relations with
prototypes without any problem, though!

    Bijan> By *that* measure, Python would be *more* oopy than Ruby ;)

No! Because Ruby has both: Prototypes *and* Class/Instance scheme ;-)

    Bijan> (Of course, it's not clear to me that being unsubclassable
    Bijan> makes you not a class. Think of Javaesque "final" classes,
    Bijan> or Dylan's "sealed" classes.)

That was not the anchor, IMHO. I would not say Python is less OOL than 
Ruby. Python's is a hybrid, IMO. That internally there are objects to
realize the "primitives" for strings or lists make Python not OO.

If it would, then TCL would also be a OOL ;-)

The main point seems to be, that there are objects that are not
instance from a class, nor was they derived from a prototype. Like
'int' in C++. A hybrid!

    Bijan> Are Ruby's *classes*, themselves, instances of classes (aka
    Bijan> metaclasses)? Like, oh, let's see, *Smalltalk*!

They are! And modules also, and numbers, strings, ... All datatypes
are. Only the control structures aren't.

...

Clemens.


From heath7952 at my-deja.com  Wed Jun  7 12:56:12 2000
From: heath7952 at my-deja.com (heath7952 at my-deja.com)
Date: Wed, 07 Jun 2000 16:56:12 GMT
Subject: WxPython & MOTIF (Not GTK)
Message-ID: <8hlur3$8fg$1@nnrp1.deja.com>

Dear All,

	I would like to use WxPython and MOTIF only (not GTK!). I currently use
WxWindows MOTIF Only, and would like to do the same with WxPython.

Is this possible. If so how.

Heath.


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


From darrell at dorb.com  Tue Jun 27 00:15:21 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 26 Jun 2000 21:15:21 -0700
Subject: Now for the flames...
References: <39549529.F5FFBD89@mindspring.com>
Message-ID: <037c01bfdfee$4fc8bf70$6401a8c0@home>

Ben's is more readable. Except for the slicing of the extension, I don't
think it's too bad.
Also some people don't like letting reference counting close the files for
you. Jpython can't handle this because Java can't handle it.

import os.path

def catFiles(args, dirname, names):
    sliceLen= -len(args[1]) # Get the length of the extension we're looking
for
    for name in names:
        if args[1] == name[sliceLen:]:
            args[0].write(open(os.path.join(dirname, name)).read())

def doit(dir, outputfile):
    os.path.walk(dir, catFiles, (open(outputfile, 'w'),'.py'))

if __name__ == '__main__':
    doit('.', 'outputfile')



"David Broadwell" <dbroadwell at mindspring.com> wrote in message
news:39549529.F5FFBD89 at mindspring.com...
> Ok, now, what else am i doing that really isn't necessary?
>    (ie: point out my c hangups, please.)
>
> Ignore the exre print statements, most of those i'll comment out or
> otherwise handle. i am looking for actuall programing advice..
> specifically on if i have used any non-platform independent stuff.. (so
> that when i later migrate to haveing this same file structure that i am
> designing my toold for, they will still function.)
>





From joseph at src.no  Thu Jun  1 19:58:57 2000
From: joseph at src.no (Joseph Santaniello)
Date: Fri, 2 Jun 2000 01:58:57 +0200
Subject: Currency format for floats?
In-Reply-To: <3936EC02.60768A66@sightreader.com>
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no> <3936EC02.60768A66@sightreader.com>
Message-ID: <Pine.LNX.4.21.0006020157430.25547-100000@lade.src.no>

Ah, yes. But I'm still missing my commas... A nice clean way to do that?

Thanks for the tip!

Joe

On Thu, 1 Jun 2000, Ken Seehof wrote:

> >>> x = 34.9
> >>> s = "$%.2f" % x
> >>> print s
> $34.90
> 
> Read more in "More String Operations" in the "Python Library Reference"
> 
> Joseph Santaniello wrote:
> 
> > Hi,
> >
> > I have an SQL query which I can format to give me a nice value like
> > 23,342.40 which Python squashes into 23342.4 when I stick it into a
> > list element. Is there a nice way to format floats so they look like
> > (US) currency?
> >
> > Thanks,
> >
> > Joe
> 
> 



From shindich at my-deja.com  Thu Jun 29 01:30:47 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Thu, 29 Jun 2000 05:30:47 GMT
Subject: Books
References: <8jarl9$9mi1@news.qualitynet.net>
Message-ID: <8jemu8$jop$1@nnrp1.deja.com>

In article <8jarl9$9mi1 at news.qualitynet.net>,
  "Helmit" <helmit_world at hotmail.com> wrote:
> Can anyone sujest some good python books for beginners?
>
>
I agree with Thomas, "Learning Python" is a very good book for a
beginner.

Alex Shindich
http://www.shindich.com/


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


From darrell at dorb.com  Sat Jun  3 20:37:54 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 3 Jun 2000 17:37:54 -0700
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it>
Message-ID: <04ce01bfcdbd$1fa80000$6401a8c0@home>

Paolo Redaelli wrote:
> but if I run a program that does x=float(x) where x = "0.000000E+00 "
> (being a string) here's the output. 
> 
> ===========================
> [paolo at aldebaran Telaio]$ ./pyplot.py
> x(<type 'string'>)='0.000000E+00' y(<type 'string'>)='0.000000E+00'
> Traceback (innermost last):
>   File "./pyplot.py", line 198, in ?
>     t = Telaio()
>   File "./pyplot.py", line 172, in __init__
>     x = float(x)
> ===========================

Pasted this into a .py file and it worked fine.

a = "0.000000E+00"
print float(a)


--Darrell Gallion





From jam at quark.emich.edu  Mon Jun 12 20:43:49 2000
From: jam at quark.emich.edu (Jeff)
Date: Mon, 12 Jun 2000 20:43:49 -0400
Subject: Testing for a file in python?
In-Reply-To: <m3wvjusenz.fsf@wilma.george.org>; from hgg9140@seanet.com on Mon, Jun 12, 2000 at 03:44:16PM -0700
References: <39450724.231994953@nntp.interaccess.com> <m3wvjusenz.fsf@wilma.george.org>
Message-ID: <20000612204349.B21364@quark.emich.edu>

olczyk at interaccess.com (Thaddeus L. Olczyk) writes:
> How would you go about testing if a certain file exists in python?
> In most scripting languages you have the -f test:
> 
> if -f filename do
>    whatever_you_were_going_to_do
>    fi
> else do
>    print "Error" file not found
>    fi
> 
> stat doesn't work, it aborts if the file does not exist.

how about 'os.path.isfile()'? it returns a 0 if the file does not exist, and
a 1 if it does. my guess is this will also work under win32, but I have not
personally tested that platform.

import os
if os.path.isfile("filename"):
  # whatever_you_were_going_to_do
  pass
else:
  print "Error" # file not found


more information regarding the 'os.path' module is here:
http://www.python.org/doc/current/lib/module-os.path.html

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 



From kirill at xyz.donetsk.ua  Thu Jun 29 02:45:45 2000
From: kirill at xyz.donetsk.ua (Kirill Simonov)
Date: Thu, 29 Jun 2000 09:45:45 +0300
Subject: Bindings for game library
Message-ID: <20000629094545.A1234@xyz.donetsk.ua>

On Wed, Jun 28, 2000 at 05:51:15AM -0500, Chris Armstrong wrote:
> Hey all, been a while since i've posted.. Anyway, I've recently been
> getting into game coding with C++, but I'd really love to get back to
> Python. If anyone can recommend (for instance) a library with Py bindings
> that can support some intense 2d sprite rendering,  I'd be real happy.

There's a fine game library at http://www.libsdl.org with Python
bindings at http://pysdl.sourceforge.net

--
Kirill



From jerome.quelin at insalien.org  Fri Jun 30 08:15:19 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 14:15:19 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com>
Message-ID: <962367319.379353174@news.libertysurf.fr>

"Alex Martelli" <alex at magenta.com> wrote:
>By calling the accessors, rather than accessing the data
>directly, you ensure that if somebody overrides the
>accessor functions in a derived class you'll be calling
>the derived-class versions of those functions.  If that is
>what you want (one direction along which you want to let
>inheritors of your class extend or tweak behaviour), then
>it's probably worth the overhead you mention.
The class isn't to be derived any further. It's a class that I use in an
application, and that may be imported in some other applications, but not to be
inherited.

Then, is it better/cleaner to access private members with accessors or not?
Is it a matter of style? Or are there hidden caveheats using (or not using)
accessors? It _seems_ cleaner to use accessors, but man, it's quite awkward.

Jerome
--
jerome.quelin at insalien.org


From thiele at muc.das-werk.de  Fri Jun 16 10:01:31 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Fri, 16 Jun 2000 16:01:31 +0200
Subject: Tkinter Widget command function call
References: <3949E71A.AF53A1CF@muc.das-werk.de> <2f9fdd30.7dcfb414@usw-ex0105-035.remarq.com>
Message-ID: <394A333A.8E5804E3@muc.das-werk.de>

Thanks



From dgrisby at uk.research.att.com  Mon Jun 12 13:46:47 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 12 Jun 2000 17:46:47 GMT
Subject: Fun with sys.modules
Message-ID: <8i37m7$t96$1@pegasus.csx.cam.ac.uk>

I'm attempting to make a module appear under two different names, and
I've come up with the following. It does what I want, but I'm a bit
worried by one of the side effects...

# a.py
print "a started"
answer = 42

-------

# b.py
print "b started"
answer = 1234

# Now the evil bit
import sys, a
sys.modules["b"] = sys.modules["a"]

# This is what worries me
print "sys is", sys
print "answer is", answer

print "b finished"

-------

$ python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16) 
 [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import b
b started
a started
sys is None
answer is None
b finished


What happened to the globals in the running module!?  Presumably the
entry in sys.modules was the only reference to module b, so it was
deleted while it was still running. Is this safe?  My guess is that
it's OK since it's basically the same situation as destructors which
run after a module has been deleted.

Any comments?  Am I totally crazy to be considering this?  If it is
safe at the moment, what are the chances that a future Python release
will behave differently?

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From aahz at netcom.com  Fri Jun 16 14:48:01 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jun 2000 18:48:01 GMT
Subject: performance benefits from byte code files?
References: <8idr3p$33h$1@nnrp1.deja.com>
Message-ID: <8idsp1$ofs$1@slb3.atl.mindspring.net>

In article <8idr3p$33h$1 at nnrp1.deja.com>,
Dan Gindikin  <dan at netrics.com> wrote:
>
>Here an example that motivates this quiestion. Lets say I have a
>relatively large script.py CGI script. Since it is always executed
>directly, a script.pyc file is never generated. Is it then beneficial
>to write a tiny wrapper that does nothing but
>   import script
>   script.run()
>because then script.pyc does get generated?

Yes, it is beneficial.  Usually, it's not a huge time savings unless the
vast majority of your code never gets executed, but it does save the
time of compiling the byte codes.  Note that finding some way other than
CGI of executing the script will likely save even more time (e.g.
FastCGI, mod_python, Zope, and so on).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From jmg at ecs.soton.ac.uk  Fri Jun  2 07:00:00 2000
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: Fri, 02 Jun 2000 12:00:00 +0100
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl>
Message-ID: <393793B0.83E25F53@ecs.soton.ac.uk>

Martijn Faassen wrote:

> This again proves my Python-Dutch connection conspiracy theory: all
> Python programmers really also speak Dutch.
>
> First an individual with a german-sounding name posting from Norway
> posts Dutch, and then someone with an eastern european sounding name
> posting from the UK understands it.

Might as well change the official language of comp.lang.python to Dutch.

We beginnen meteen.

Python en Nederlands: wereldtalen voor de eenentwintigste eeuw !

Jacek



From mike at pl.net  Sun Jun 11 23:25:47 2000
From: mike at pl.net (mike at pl.net)
Date: Mon, 12 Jun 2000 03:25:47 GMT
Subject: Adding Directory to PYTHONPATH
Message-ID: <8i1l7j$d01$1@nnrp1.deja.com>

Hi All !
Probably been asked before but how do you add a new library module
directory to PYTHONPATH?

I know you can do it dynamically in Python, but in this case I need to
have a particular module directory available everytime that Python is
loaded.

I tried including a .pth file but can't see the new directory when I
print sys.path

Grateful for any pointers
Best regards
Aren't we lucky for having Python!
Mike


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


From daniel.dittmar at sap.com  Thu Jun 15 04:20:51 2000
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Thu, 15 Jun 2000 10:20:51 +0200
Subject: boolean true and false values.
References: <3947f922.425016156@nntp.interaccess.com>
Message-ID: <394891E3.B4D0357D@sap.com>

How about

true = 'true'
false = None

this prints much nicer than 1 and 0.

-- 
Daniel Dittmar
SAP AG, Basisentwicklung Berlin
daniel.dittmar at sap.com


From charlie at intelligenesis.net  Fri Jun 16 10:16:49 2000
From: charlie at intelligenesis.net (Charlie Derr)
Date: Fri, 16 Jun 2000 10:16:49 -0400
Subject: [FAQTS] Python Knowledge Base Update -- June 16th, 2000
In-Reply-To: <20000616111056.18561.qmail@synop.com>
Message-ID: <LOBBJCAMDNLNCGCCHGEIEEHGFJAA.charlie@intelligenesis.net>

~ 
~ 
~ Hi!
~ 
~ My arrangement with FAQTS was to get 225 entries into 
~ http://python.faqts.com by June 16.  I've done it! With a lot of help 
~ from you guys.  Thanks everyone.
~ 
~ This doesn't mean my daily entries will cease, I'm going to continue 
~ working on the FAQTS Python Knowledge base for a while to come.  Its fun 
~ and educational :-)
~ 
~ Regards,
~ 
~ Fiona Czuczman
~ 


I think what you're doing is fantastic.  Thanx!!!

	~c



From danunder at columbus.rr.com  Sat Jun 17 09:42:36 2000
From: danunder at columbus.rr.com (Dan Underwood)
Date: Sat, 17 Jun 2000 09:42:36 -0400
Subject: Pythonwin: editor nits
References: <39439B04.76F15F39@columbus.rr.com> <bWB25.700$qz4.6064@news-server.bigpond.net.au>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E77@mail.sauder.com>

Mark,
Thanks, turning off Smart Tabs makes it do what I want... 
rather obvious now that I know it!
I find Pythonwin to be an excellent piece of work!
Dan.

Mark Hammond wrote:
> 
> Check out the documentation in the Pythonwin help file about the tabs (in
> the overviews section).  If you can't find it, then update your build!
> 
> Basically, "Smart Tabs" is designed to work the way you are seeing it
> work - it uses the first indent in the source file to set the options for
> the file.  If you want your options to be in effect _always_, turn
> smart-tabs off.
> 
> This is identical to the way IDLE works - we share the same auto-indent
> code (courtesy of the timbot)
> 
> Mark.
> 
> "Dan Underwood" <danunder at columbus.rr.com> wrote in message
> news:39439B04.76F15F39 at columbus.rr.com...
> > I'm using build 129 on Win98SE. I find that the editor does not
> > (usually) remember my tab and indent settings. Nor does it remember that
> > I set "Use Tabs" on and "Show Whitespace" on. Actually, the options
> > dialog shows these settings as remembered but the editor seems to ignore
> > them for existing files. I do not want to deal with questions about why
> > I change tab and indent ;-) but I would like to know if I am just not
> > setting something right or are these just some bugs to be eventually
> > fixed. If a later build has already fixed them, "never mind".
> > Thanks.
> > Dan.



From aahz at netcom.com  Tue Jun 20 22:19:11 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jun 2000 02:19:11 GMT
Subject: Dictionary vs if-elif?
References: <UlS35.15101$Xx5.781176@bgtnsc06-news.ops.worldnet.att.net> <rg10ls89rd67ufpp48s2pr8j78q657kbdm@4ax.com>
Message-ID: <8ip8mv$89b$1@slb6.atl.mindspring.net>

In article <rg10ls89rd67ufpp48s2pr8j78q657kbdm at 4ax.com>,
Ralf Canis  <ralf at silicon.org> wrote:
>
>Don't know if this is really better but how about:
>
>bogusCodes = ( '1.', '31', '90', 'AZ' ...... )
>...
>if item in bogusCodes:
>  do something useful

That's a Bad Idea.  It's a linear search instead of a hash.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From db3l at fitlinxx.com  Mon Jun  5 23:43:44 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Jun 2000 23:43:44 -0400
Subject: Network byte ordering question ...
References: <8hhjo6$8ev$1@watserv3.uwaterloo.ca>
Message-ID: <uog5fsccv.fsf@ctwd0143.fitlinxx.com>

istevens at calum.csclub.uwaterloo.ca (ian stevens) writes:

> I am writing a Python application which makes use of a protocol which
> requires information to be in network byte order.  Specifically, a
> message header takes on the following form:
> 
>     bytes   type
> 
>     0-15    Message ID ... all numbers
>     16      Numeric value
>     17      Numeric value
>     18      Numeric value
>     19-22   Numeric value
> 
> What is the best way to extricate the above information and convert it
> to the byte ordering of the host machine?  To convert a header in the
> byte ordering of the host machine to network byte ordering?  How do
> I determine the type of machine the program is running on and, hence,
> its byte ordering?

I know Python is supposed to have one obvious way to do things, but I
think there are several possibilities here, with varying degrees of
readability and no doubt performance.  So I thought I'd offer a few
suggestions.

As you indicated you had looked at, you could probably use the struct
module, and that may be the simplest approach.  One caution is that
the module doesn't really guarantee much about the alignment of
individual elements in a structure, so I would suggest slicing off the
fields you need to decode and then unpacking them individually.

One other wrinkle to stay aware of - if your 4-byte value is an
"unsigned" value, then you need to work with Python long integers and
not regular integers, which are signed, to ensure that you can
represent all possible values.  The structure unpacking automatically
returns integers as Python long integers, but any manual conversion
would need to take that into account.  If you know the values are
going to be less than 0x8000, or that having the numeric value as
negative won't hurt, you can ignore that part.  I've assumed long
integer results in code below to ensure we can treat the entire
spectrum of 4-byte unsigned values.

In your case, the structure seems very straight forward, and there's
only one field for which the byte ordering matters (your final
4-byte/32-bit field).  I'm also assuming that by numeric value you are
talking true binary value (e.g., 0x00 for 0, and not 0x30, or the
ASCII '0').

I'm not sure how the data is already represented in memory, but
presuming that it's a string in "header", you could easily do
something like to pick up the first set of fields:

    message_id = header[:16]                    # Bytes 0-15
    values     = map(ord,header[16:19])         # Bytes 16-18 as ints

That would give you your numeric fields as values[0] through
values[3].  You could then just append (values.append) the 4-byte
field, or store it separately, however you like.

Decoding that final field using the struct module (as mentioned above)
could be done by slicing off the field, and use that as a single field
structure - to remove any padding dependency.  The following:

    struct.unpack("!I",header[19:23])

would result in a tuple of 1 with the value as a Python long int.  The
unpack string is exclamation-mark (network order), capital eye (for
unsigned).  If you wanted to use that in the above code for appending
to the values list, you'd need to subscript [0] to get the actual
object to append to the list, as in:

    values.append(struct.unpack("!I",header[19:23])[0])

Alternatively you could just write some direct code to handle the
conversion, by accessing one byte at a time from the field:

    values.append((long(ord(header[19])) << 24) +     # Network order 4-byte
		  (long(ord(header[20])) << 16) +     #   value at 19-22
		  (long(ord(header[21])) << 8)  +
		  long(ord(header[22])))

The conversion from a 4-byte field to a single integer value is a
simple way to convert network ordering to native ordering without
worrying about determining the native ordering.  You can do the
reverse on output - shift right and mask by 0xFF for each byte.  There
needs to be a few parentheses to ensure proper precedence for the
shifts versus addition, but it's basically just four operations.

If you wanted to use a more functional approach to the latter
computation using some of Python's support for that, you can consider
the above computation as a constant shift left (8 bites) of a running
total of each byte.  This can be accomplished with a
reduce/map/lambda approach with something like:

    values.append(reduce(lambda x,y: (x<<8)+y,
                         map(long,map(ord,header[19:23]))))

which turns bytes 19-22 of the header into a list of long integers,
which are then reduced by adding each to a running total which is
shifted left prior to each new addition.  Neat, eh?  :-)

Oh, and if you did want to determine the native ordering (to, for
example, short circuit the conversion if it matched network order),
one approach is to compare the internal representation of a number to
a fixed constant (memory buffer) to decide which endian you are.  This
can be done with the struct module - you can find an example in the
wave.py file in the standard library, which I'll quote below:

    # Determine endian-ness
    import struct
    if struct.pack("h", 1) == "\000\001":
	    big_endian = 1
    else:
	    big_endian = 0

Hope this helps.

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


From dsavitsk at e-coli.net  Mon Jun 19 15:46:23 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 19 Jun 2000 19:46:23 GMT
Subject: string parameter replacement question
References: <Pine.LNX.4.10.10006191432420.3161-100000@worf.mcnc.org>
Message-ID: <jMu35.9242$dF.414724@news1.rdc1.il.home.com>

>>> def fun1(str):
>>>   fun2("input=" + str)


"oliver" <fwang2 at yahoo.com> wrote in message
news:Pine.LNX.4.10.10006191432420.3161-100000 at worf.mcnc.org...
> hi,
>
> I have a function defined to accept one string parameter, inside this
> funtion, I need to call another function to take this string as part of
> the parameter, I am confused how to make the replacement happen. See
> following code:
>
> def fun1(str):
>   fun2("input=str")
>
> Noted that fun2 is from a moudle and it is expecting such a string. But
> the replacement won't happen with above structure, for example, fun1("me")
> won't produce a fun2("input=me"). Can anyone help me out here?
>
> Thanks a lot
>
>




From robinporter72 at yahoo.ca  Wed Jun 14 09:33:24 2000
From: robinporter72 at yahoo.ca (Robin Porter)
Date: Wed, 14 Jun 2000 09:33:24 -0400 (EDT)
Subject: Pointers in PYTHON
Message-ID: <20000614133324.10894.qmail@web1502.mail.yahoo.com>


I have been reading everything I can possibly get my hands on regarding PYTHON.  I have not yet come accross anything regarding the use of pointers for linked and double linked lists.  I am just new to this language so please forgive my ignorance if it has been mentioned here before.

In addition to what ever information you can give me on pointers in PYTHON, could you please provide some info on combining C, Assembler, and Python for a project.

Thank you in advance.

Robin



Mr. Robin Porter
St. Lawrence College
Computer Engineering

Database Coordinator - Raiser's Edge

Alumni Development

 



---------------------------------
Do You Yahoo!?
Get your free @yahoo.ca address at Yahoo! Mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000614/3cd050a0/attachment.html>

From corg at copernic.com  Fri Jun  9 12:00:09 2000
From: corg at copernic.com (Gaetan Corneau)
Date: Fri, 9 Jun 2000 12:00:09 -0400 
Subject: Why is Python so slow?
Message-ID: <C5A7720BDEB7D3118A5200E018C200A051E7E2@EXCHANGE>

William,

> > I am just learning Python and I wrote a little program to 
> modify a text file
> > with 120,000 lines. It took 1hr and 30 mins. I wrote a C 
> program that
> > created the text file from a .dbf and it only took 2 mins.

Please post your code, something is wrong. I processed hundred of files,
containing over a million of C++ code in much less than 1h30.
______________________________________________________
   Gaetan Corneau
   Software Developer 
   Copernic.com 
   http://www.copernic.com
   E-mail: corg at copernic.com                         
______________________________________________________
"Golf is a game whose sole aim is to hit a very small
ball into an even smaller hole, with weapons
singularly ill designed for that purpose." -- Winston Churchill

 



From thomas at xs4all.nl  Mon Jun 26 04:19:44 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 08:19:44 GMT
Subject: ppp-script
References: <39570BCE.5E5DEF97@the-ecorp.com>
Message-ID: <slrn8le4gj.2p5.thomas@centurion.xs4all.nl>

On Mon, 26 Jun 2000 09:52:46 +0200, Philippe Possemiers
<philippe.possemiers at the-ecorp.com> wrote:

>My platform is Linux and I use a modem to dial my ISP. Therefore, I use
>os.exec* in my script to activate /etc/ppp/ppp-on. The question is : how
>do I return to my python script? From the moment I call 'ppp-on', I seem
>to lose all control and cannot return to the python interpreter.

os.exec*() replace the currently executing process, instead of creating a
new process and waiting for it to finish (which is what you want, I
assume.)

You are looking for os.system(). It takes a single string as argument, which
should be the command and any argument, just like you type them on the
shell. (In fact, os.system() just spawns a shell and passes it the string ;)

Alternatively, you can use os.fork() and os.exec*(), and then os.waitpid()
for the child to finish, but that's only necessary if you want to do things
while waiting, or if you want to avoid using the shell (because you want to
fiddle with ARGV, or some such.)

Argh-argv-argc-ly y'rs,
	Thomas.


From pduffin at hursley.ibm.com  Thu Jun 22 05:49:28 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 22 Jun 2000 10:49:28 +0100
Subject: Regex engines
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org> <8igpab$7tf$1@slb2.atl.mindspring.net> <3950B2BE.AF756B9F@hursley.ibm.com> <3dd7laes4x.fsf_-_@mems-exchange.org>
Message-ID: <3951E128.6BF49923@hursley.ibm.com>

Andrew Kuchling wrote:
> 
> [Posted only to c.l.tcl and comp.lang.python; comp.lang.perl doesn't exist.]
> 
> Paul Duffin <pduffin at hursley.ibm.com> writes:
> > front end to a C++ program. Note that is Tcl, not Tk. And someone posted
> > don't use Tcl, use Python because it has a Tk binding and "Tcl is crap".
> 
> Ummm... If you mean the thread "C++ and Tcl", that person was writing
> about Perl, not Python.
> 

You are correct and I apologise for the unintentional slur on Pythoneers....


From pinard at iro.umontreal.ca  Wed Jun 21 15:00:03 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 21 Jun 2000 15:00:03 -0400
Subject: Transcendentals and Encryption (was Re: Nth digit of PI)
In-Reply-To: ansible@typhoon.xnet.com's message of "21 Jun 2000 14:21:04 GMT"
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com>
	<394a5ed7.83928743@10.1.1.28> <8idoj0$lea$1@nntp.itservices.ubc.ca>
	<394a680a.86283939@10.1.1.28> <8iqj0f$kk5$1@flood.xnet.com>
Message-ID: <oqem5q6erw.fsf@titan.progiciels-bpi.ca>

[James Graves]

> I am under the impression that there are an infinite number of
> transcendental numbers.

There is an aleph1 number of real numbers and an aleph0 number of non
transcendental numbers (since we can enumerate all algebraic equations).
Aleph1 > aleph0, an aleph1 number of transcendental numbers remain.  Right?

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





From see at message.body  Sun Jun 18 18:58:24 2000
From: see at message.body (Sam Penrose)
Date: Sun, 18 Jun 2000 15:58:24 -0700
Subject: Optimizing Object Instantiation (was Re: How Fast Does Python Run?)
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net> <39499C91.3A1586BE@sightreader.com> <8if21t$4u5$1@slb1.atl.mindspring.net> <20000617120003.S26436@xs4all.nl> <see-0F7889.18054217062000@news.dnai.com> <E0Y25.1541$qz4.13200@news-server.bigpond.net.au>
Message-ID: <see-262267.15582418062000@news.dnai.com>

In article <E0Y25.1541$qz4.13200 at news-server.bigpond.net.au>, "Mark 
Hammond" <mhammond at skippinet.com.au> wrote:

> "Sam Penrose" <see at message.body> wrote in message
> news:see-0F7889.18054217062000 at news.dnai.com...
> 
> > 1) Does informed opinion agree that object instantiation is
> > noticeably performance-expensive in CPython? (Unfortunately,
> > I can't seem to find the article in question.)
> 
> Sorry - no informed opinions here :-)
> 
> But I would guess that "type" creation (ie, creation of objects
> implemented in extension modules) would be quite fast, and mainly
> dependent on the native malloc() speed.
> 
> I could well believe that "instance" creation (ie, creation of user
> defined class instances) would be quite slow, especially if the object in
> question has an __init__ function!
> 
> Mark.

The code in question is based around nested __init__ calls:
a main working class which is based on an abstract class
(2 __inits__) and which calls a wrapper class (3) around the
MySQLdb module (4).

I'm starting to think for other reasons that I've been
overcoding, but it would be nice to know whether the
performance issue has any prospects of being addressed.

OTOH, it's amazing how much faster $2,500 worth of current
hardware is than the system we spent $2,500 on two years ago;
perhaps the point will prove moot.


From greg at cosc.canterbury.ac.nz  Tue Jun 27 23:00:12 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Wed, 28 Jun 2000 15:00:12 +1200
Subject: boolean true and false values.
References: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl> <memo.20000626221037.43263C@digitig.compulink.co.uk> <slrn8lgsjq.us.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <39596A3C.D3A1A824@cosc.canterbury.ac.nz>

Albert Hofkamp wrote:
> 
> Until now, I like the basic idea of Python of variables not having a type, but
> I don't really see the advantage.

The flexibility it offers seems to help with rapid prototyping.
You can change part of your design easily without having to
wear yourself out convincing the compiler that you know what
you're doing.

Perhaps there is a middle ground. G++ has an extension called
a "signature", which can be used to describe a type in terms
of the methods it supports. This seems to be a lot like the
notion of type that Python programmers think about. Perhaps a 
static type system for Python 3000 could be based on
something like that?

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


From cut_me_out at hotmail.com  Thu Jun  8 09:01:23 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Jun 2000 09:01:23 -0400
Subject: Some questions about C extensions
References: <etdu2f69rhz.fsf@w20-575-109.mit.edu> <080620000748310603%pecora@anvil.nrl.navy.mil>
Message-ID: <etdd7ls2uos.fsf@w20-575-109.mit.edu>

Thanks for the advice about PyArg_ParseTuple, Louis.  That cleared
things up a lot.

I think I figured out what's going on with PyTuple_SetItem, by the way.
You don't have to do a Py_INCREF in every case.  The thing is, it
doesn't affect the ref. count at all, so if you are putting a newly
created object in there, the reference that object already has will
suffice, but if you are putting in an object that you got from somewhere
else by a process that didn't increase the object's ref count, you need
to increase it explicitly.

Alex.




From richard_chamberlain at ntlworld.com  Sat Jun 17 10:43:01 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 17 Jun 2000 15:43:01 +0100
Subject: tkinter: making multiple command bindings
References: <394AF971.CCA0C410@uniserve.com>
Message-ID: <o%L25.637$_55.12844@news2-win.server.ntlworld.com>

Hi Bob,

from Tkinter import *
root=Tk()
button=Button(root,text='Click Me')
button.pack()
def Hello(event):
    print 'Hello from ',event.widget
def Goodbye(event):
    print 'Goodbye from ',event.widget
button.bind('<Button>',lambda event:(Hello(event),Goodbye(event)))
root.mainloop()

<EOF>

I believe that is what you mean, I probably wouldn't do it that way though.
I'd create a method which calls the others which is a bit more verbose but
more useful I think.

Richard

Bob van der Poel <bvdpoel at uniserve.com> wrote in message
news:394AF971.CCA0C410 at uniserve.com...
>
> In tcl/tk I can bind a sequence of commands to an event-binding:
>
> bind $w.e <Return> {destroyWin $w; func1; func2 $parm}
>
> Is it possible to do the same in tkinter? I'm sure it has to do with the
> use of lambda (learned that last week in creating a function call with
> an arg...thanks!).
>
>
> --
>    __
>   /  )      /         Bob van der Poel
>  /--<  ____/__        bvdpoel at uniserve.com
> /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>




From scorder at incigna.com  Fri Jun 23 11:53:25 2000
From: scorder at incigna.com (Sam)
Date: Fri, 23 Jun 2000 11:53:25 -0400
Subject: python & excel and 1 remark
References: <39533572.AB90FD1F@tcomhp43.epfl.ch> <086ff63a.3e6f872e@usw-ex0105-035.remarq.com> <39533D22.B6C50A18@tcomhp43.epfl.ch>
Message-ID: <jp17lsoq0389cl24cje63626uhcavie2qj@4ax.com>

On Fri, 23 Jun 2000 12:34:10 +0200, Cedric Lefeuvre
<Cedric.Lefeuvre at tcomhp43.epfl.ch> wrote:

>Richard Chamberlain wrote:
>
>> Hi Cedric,
>>
>> It's possible to generate an excel file using win32com
>> extensions on a windows box. Get the extensions and a bit of
>> documentation at:
>>
>> http://starship.python.net/crew/mhammond/
>>
>> You might want to pick up a copy of Python Programming on Win 32
>> By Mark Hammond, Andy Robinson which has a chapter on this very
>> subject.
>>
>> Richard
>>
>
>Ok,Richard, but without using python on windows system????
If you don't care about formulas and formatting then you can generate
a csv (comma  seperated value) file pretty easily.  On windows csv's
are even associated with excel so all you have to do is double click
on it.  Otherwise you may look into KSpread from the KOffice suite.  I
believe they are using python as their scripting language.  I don't
know if the filters are in place yet though.

-Sam


From tim_one at email.msn.com  Sun Jun  4 16:58:54 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 4 Jun 2000 16:58:54 -0400
Subject: Um? *boggle* Silly question about listobject.c
In-Reply-To: <tolot.09oq@utolot.toppoint.de>
Message-ID: <000001bfce67$b1f8f720$b02d153f@tim>

[Tim]
> ... In the meantime, if you really need lists with more than 2
> billion elements, your hardware budget is so large you can afford
> to pay someone to patch you up a custom Python <wink>.

[Marc Christiansen]
> What makes you sure, that sizeof(int)==4?

How tedious do you want to make this?  In point of fact, sizeof(int) may be
1 in ANSI C, and still allow expressing 2 billion non-negative values.  The
std only guarantees that an int can hold a value at least as large as could
be held in a 2-byte 2's-comp integer on a machine where a byte is 8 bits (in
particular, it does not guarantee that sizeof(int) >= 2, or that the units
returned by sizeof represent 8 bits each).

If we're done being tedious now <wink>, my *observation* is that an int is
large enough to express at least 2**31 billion non-negative values on every
platform Python runs on.  It has certainly been ported to platforms where
ints are much larger than that, but AFAIK never to a platform where ints are
smaller than that.  If one such pops up, I'll refer porting questions to you
<wink>.

there's-a-reason-most-people-don't-read-stds-ly y'rs  - tim





From "dbroadwell" at spam.mindspring.com  Thu Jun  1 18:19:03 2000
From: "dbroadwell" at spam.mindspring.com (David Broadwell (remove spam.))
Date: Thu, 01 Jun 2000 18:19:03 -0400
Subject: search for a string?
References: <8h6211$ark$1@news-int.gatech.edu> <LbBZ4.31403$WS3.364156@typhoon.we.rr.com>
Message-ID: <8h6na4$ql7$1@nntp9.atl.mindspring.net>

> 
>     OldString = File.read() + OldString
> 
>     I perfer, though: OldString = "%s %s" % (File.read(),OldString)
> 
The second once certianly looks more .. programmer. Now which is more
efficent? (faced with to apparenly equal choices) and why?


From daniel at visionart.com  Tue Jun 13 18:10:56 2000
From: daniel at visionart.com (Daniel Kramer)
Date: Tue, 13 Jun 2000 15:10:56 -0700
Subject: tcl/tk vs. python/Tkinter - Example attached
Message-ID: <3946B170.17A2630@visionart.com>

I have just started to use python this last week.. I wanted to port a
tcl/tk app to it and see how I like working with it.  As far as
programming with it I like it more.. but just as I got started on my app
I noticed some speed differences in TK.  The tcl/tk app is much more
responsive then the python/Tkinter app.

I have put together a simple example done both ways.. it's attached to
this mail.

Within the tar you will find:

canvas.tcl - This is just a simple tcl/tk prog.. 2 rectangles are added
to a canvas and you can move them around with the mouse.

canvas.py - This is the same thing written in python/Tkinter.
nodeUI.py - This contains most of the classes/functions to drive the
canvas.py.. it should be in the same dir as canvas.py


Since I am new to this lang I was just wondering if there was a way to
get better python/TK performance... if anyone could try out the examples
and let me know.. for me the tcl/tk one feels much more responsive but I
don't have any hard numbers to prove that.

I have been reading that wxWindows is the much faster solution.. but I
am very familiar with TK and it's plenty fast for what I need.. but I
worry that the Tkinter ver won't be fast enough once I add alot more to
it.


-Daniel


begin 666 canvas.tar.gz
M'XL("'FP1CD``V-A;G9A<RYT87(`[5C=;Z-&$+]7[U\Q(CX)4G`!AZ0ZU95Z
MO9<^)`]M3]=3A"P,:QL=!@O6"2AW_WMG/_ at PMI.+JMRU$B,YQO.QS,[,;W8V
M89#>!<5D6[UZ.;)M^_+R$O#;=:87_!N?//G-R?.F8%\YKN.Y5^[4`]N9>H[[
M"NP7]*FA7<&"'%V)@C2FR6F]N[B(L_2T7&X%FN__"1&RS+,-_/4I3AG-(=YL
MLYS!.5$/:1;1][\3<G;V=A<G$?PFRH6`(L*R[8<XA1DNH!LDQ`>IH4N!N0C"
M3ZL\VZ41BK153JLK6S-AD>41S>_CB*V1[YJPIO%JS?#YPK9-R&D2TR6WR.-H
M1=&@5D4QX'LF6UQ7#])PG>5<+:3<>\VDY380KW),6,9)PF6+C*TU at Q"RJ?AN
MD"4W-;G!+STT-?[300TIGX0Y#1C5'<\V\6/4=NYQ0[<Q=&M+%RU=:1G$:9)E
M6YV_GGSO5!^E4.*?A8^4_K^E)_`_]3Q/X7]J8P_@^+^\<`;\?PL"60``DQ"L
M#EHE5)'50:H+EL(IAZ&E4"H@"I94X0("')U\0;`41"4^P5+X=,`2Z.30!*N(
M$978+["=D(+*GC,OPB"AX$QD.%M^NMN`0\@VST)X"**(,[_``X3`8H86]["&
M$BI`'B&C59(M at H1WC]7^PO5:9,17+N%V'*I(E#`N?<FN.NP*QI6/;5#1KU$$
M;X."`M_.3;"ATF+!6:(OC.LWW!8LC],5Y$&*8=+F7$4#&V@:^9JTRFG(GK#B
M*CVK%-_ZA!57Z5OQ%KAOI20+GO<";HMM$J,[MYBK'/3Q_8^N<3YN0W=N.3[4
MPO5CPK[E:3-?PXCS4(L&*N*!.\`TW29Q&E%,B7+.]L,#GG.$YQ[A39%GL6!5
M((>G298 at KW//AKI^72S")I/SM?(X6!1ZO9PHB0Q1H=;Q<6'KM-#QC;V],5HR
MS`C"1SR-9=4JOT1*6[]^0BQQ4RQSE(,(&-8R6\M?HNI.BOGK]\2BR@[%F^R.
M*MLQ!JM"$))1G(9Y at Y$1^4(4XF)&-W]BOV3O<ESD(>10VT-:$A3L;_'W(X2[
M/!=_YMQ,AE7PNM$"!*;8/)=@EP`>LE97F,)3%K9_"L=\2#F!9<&6_H[+]M='
M%(K`Y$&,53(6'N\YL,1'")+$%[@B,CQM?)X?&FCFJ=.;J*4G]G$0UUZ0CD7U
M474>4O0=2V$)#W5?V00L7,ON<2X:V;DF`R06].'S9SBJRDMU7Y5'9M26GW1?
M8&U<6F,1JKI=C"O)^.B/T)U>UGII^Z*2L>`IP@/H9^<7T/:+%KFO2WA=::W2
M6\>ZSAB>[DI9Z'44"5$'#>=I-V)<Q*/)@\F5!S at CBL^!CMOJX#3(/^0_F'\U
MTG[/^]_4O7#K^]_E].*2SW\7[G#_^R9T=B8K`($8[1)ZXC:(%&+E%L`+^XTH
M2*PF,4GA78N,5)7=R/N5HQ0BG`[G\SB-V7RN%S19FK)N3=X/C#<"O<ERPB%1
MHIG=%'K+KX[PU<0Z4RCH2<49.A/344\B3GTI at 1_J2:RG(P[EC@[_W='Y(`;)
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MUA#OWPL(^R0'M@/=6]LW^LK\_`NS=!FOCJFWW<&L_XF4TT at X1+I!>;\]",DS
M''ZV#PJ.QF%VKIN*:5P1*L]Q1Q34$4?,.K/[%58GME=HSRL*<4QT=Q/11!T5
LPL$M'CEDR/]+YO][SQ\##3300`,---!``PTTT$`##330R],_QS*[S0`H```5
`
end


--
-----------------------------------------------------------------------------
Daniel Kramer
http://www.visionart.com/~daniel/
daniel at visionart.com:           VisionArt Design and Animation





From bwinton at tor.dhs.org  Thu Jun  8 22:29:10 2000
From: bwinton at tor.dhs.org (Blake Winton)
Date: Fri, 09 Jun 2000 02:29:10 GMT
Subject: OT: nederlands (Ans something else).
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <PCp%4.109780$Qp2.291904@afrodite.telenet-ops.be> <m3pupth63m.fsf@chinon.cnrs-orleans.fr> <393E9A33.AD5282C1@dulcesol.es>
Message-ID: <slrn8ju171.qgs.bwinton@tor.dhs.org>

On Wed, 07 Jun 2000 19:53:39 +0100, Pablo Prieto wrote:
>In VB (Please, put the knife down!) I can use an object called Printer.

In VB you can do a lot of stuff.
Like set a attribute on a property to -4, and all of a sudden, your
class can be used anywhere a Collection is used...  Or assign an array
to a Variant, and use it instead.  And don't get me started on VB's
"inheritance"...  :P

Writing-a-VB-to-Java-translator-ly y'rs,
Blake.
-- 
10:13pm up 16 days, 13:23, 1 user, load average: 1.10, 1.04, 1.01


From scoota at my-deja.com  Wed Jun 28 09:25:32 2000
From: scoota at my-deja.com (scoota at my-deja.com)
Date: Wed, 28 Jun 2000 13:25:32 GMT
Subject: Python COM for Linux?
Message-ID: <8jcubq$7qv$1@nnrp1.deja.com>

Warren,

thanks for your response.

Please don't get too hung up on the COM thing.  I only suggested COM as
an example as I have already implemented a similar system in NT using
VBScript/VB/C++/COM/ActiveX.  In this development all the user (script
author) had to do to handle an "event" was to write a single function
using the prescribed (simple) syntax (e.g. button_click(args)). There
was no need to write any initialisation code, no object reference
passing etc - easy!.
That's the sort of script simplicity I'm looking for in my Python
project. I don't mind if it makes my job (i.e. as the designer of the
development environment) harder, and I don't mind hiding some sneaky
code in a C++ extension object if that makes the end users life easier.

I would appreciate further details on your approach - it sound like
what I'm looking for.

Scoota


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


From eviltofu at rocketmail.com  Thu Jun 29 07:16:16 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Thu, 29 Jun 2000 11:16:16 GMT
Subject: Module Name Problem
References: <eviltofu-C678D0.18293328062000@news.earthlink.net> <8F61CAE59gmcmhypernetcom@199.171.54.194> <eviltofu-E613EC.21014728062000@news.earthlink.net> <voC65.98134$WS3.847668@typhoon.we.rr.com>
Message-ID: <eviltofu-FC3391.07161429062000@news.earthlink.net>

In article <voC65.98134$WS3.847668 at typhoon.we.rr.com>, "Stephen Hansen" 
<stephen at cerebralmaelstrom.com> wrote:

> This doesn't sound very broken,.. you're creating the conflict-of-names.. 
> :)
> 
> I read somewhere the sugguestion of an 'import name as name' 
> syntax..don't
> know where...don't know if its a 1.6 or a py3k thing...
> 
> --S

In java I would be able to access methods and classes in both modules 
without any difficulty

maxvu.os.blahblah
os.blahblah


From paul at prescod.net  Wed Jun 28 20:13:35 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 28 Jun 2000 17:13:35 -0700
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <395A94AF.DB0428A0@prescod.net>

Steve Juranich wrote:
> 
> I am *brand* new to python (as of Sunday), and I just got to the section
> in the tutorial about classes.  I was wondering why there really isn't
> such an idea as a "private" member of classes?  I understand that members
> can be hidden, but not really protected from the user.  I was wondering
> what the benefit of this would be?  Wouldn't it open the possibility of
> security holes?  

Private methods cannot really be used for security purposes. In any
decent language there is a way around the private declaration. In Java
it is reflection. In C++ it is pointer manipulation. Python says: "Why
pretend to provide security that isn't really there?"

> Also, what benefit is gained by allowing a user to
> add/delete members from existing classes?  Wouldn't it be more desirable
> to create his/her own derived class?

It's sometimes useful and more effort to prevent than allow. Consider a
set of objects that you are trying to serialize to disk. You must attach
a flag to in order to say "they've been processed." You can't always
inherit because you aren't trying to make new objects -- you're trying
to change existing objects.
-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski




From phil at geog.ubc.ca  Thu Jun 22 13:01:50 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 22 Jun 2000 10:01:50 -0700
Subject: Any other good tutorials?
References: <sl2i51h0e7f149@corp.supernews.com> <20000621192418.00540.00001208@ng-fl1.aol.com>
Message-ID: <m3bt0todj5.fsf@brant.geog.ubc.ca>


dparsavand at aol.com (Dparsavand) writes:

> Unfortunately, if you type "Python Tutorial" into most search engines, you get
> many mirrors of Guido's tutorial (which I like, but it's nice to have
> alternatives).  However, if you follow http://www.pythonlabs.com/ to
> http://www.vex.net/parnassus/ and then choose the  Info/Books/Tutorials link
> then Tutorials, you will get quite a few alternatives. I will include two more
> that are not listed there:
> 
> http://www.networkcomputing.com/unixworld/tutorial/005/005.html
> 
> and a link that definitely should be listed, but isn't yet,
> http://yhslug.tux.org/obp/thinkCS/thinkCSpy
> 
> Good luck, I hope you find something you like. - Dara

Some others that would be good to have on that page:

1) The current numpy tutorial (supercedes the one that's cited on Parassus):

http://numpy.sourceforge.net/


2) Richard P. Muller's python short course:

http://www.wag.caltech.edu/home/rpm/python_course/

3) Konrad Hinsen's Python for Science tutorial

http://starship.python.net/crew/hinsen/



From mhammond at skippinet.com.au  Wed Jun 14 18:43:53 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 22:43:53 GMT
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com> <dCL15.4673$cQ.38267@news-server.bigpond.net.au> <8i8m11$bni$1@nnrp1.deja.com> <01cf01bfd63b$d6ad0120$4500a8c0@thomasnb>
Message-ID: <8i91r6$l6t$1@nnrp1.deja.com>

In article <01cf01bfd63b$d6ad0120$4500a8c0 at thomasnb>,
  "Thomas Heller" <thomas.heller at ion-tof.com> wrote:

> I have some mixed feelings with these hidden dependencies in the core
> distribution of optional extension modules.

Agreed.  Better solutions welcome :-)

> What would happen if win32api is not installed? Fail silently?
> Print a warning? Require the installation of win32all?

In the case of os.abspath(), it does the "best it can", using the same
code Unix uses.  It wont fail, but it is not guaranteed to provide a
fully qualified filename in that case.

In the case of os.popen, its not clear what to do, as the core Python
cant work at all without it.  Whatever we come up with, it is surely
better than the current situation, which often yields an exception,
with the error message "The command completed successfully" :-(

Of course, the best solution would be for MS to simply fix popen!

Mark.


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


From mjackson at wc.eso.mc.xerox.com  Fri Jun  9 12:50:24 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 9 Jun 2000 16:50:24 GMT
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <8hr78g$n9u$1@news.wrc.xerox.com>

"Lawrence Kesteloot" <lk at veriomail.com> writes:
> 
> If one of the goals of Python 3000 is to make it easy for new
> programmers to learn, how about removing the distinction
> between = and ==?  The only kinds of examples I can think of
> where this is useful is something like:
> 
>     a = b = 5
> 
> vs.
> 
>     a = b == 5
> 
> But this isn't worth the trouble it gives new programmers.  The
> syntax could be changed to make = and == equivalent tokens
> and have the parser figure out which is meant from the context,

But "=" and "==" have different *meanings*, and in Python the
difference is even bigger than in most languages.  Anything that
muddies the meaning of assignment is no favor to new Python
programmers.

> just as it's done in Basic.

Another reason against. . .

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	You can never solve all difficulties at once.
				- Paul Dirac




From digitig at cix.co.uk  Tue Jun 20 17:25:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 20 Jun 2000 22:25 +0100 (BST)
Subject: Conversion from proprietary satellite control script language
References: <394F99D4.C28E013D@hursley.ibm.com>
Message-ID: <memo.20000620222509.43609B@digitig.compulink.co.uk>

In article <394F99D4.C28E013D at hursley.ibm.com>, pduffin at hursley.ibm.com 
(Paul Duffin) wrote:

> TJ Judson wrote:
> > 
> > I'm afraid my only other answer is to use a language that supports
> > "goto", like Tcl.
> > 
> 
> Tcl does not support goto.
> 
> I think that it should be possible to analyse all the possible paths
> through your satellite scripting language so that you create a graph
> of flow, in those cases where you see a goto you *simply* replace
> it with something else which does the job.
> 

I once had to port some BASIC code that had a pseudocode structure of the 
form:

def foo:
   # some code
label1:   
   # some code
   if test1:
      goto label2
   # some code
   
def bar:
   # some code
label2:   
   # some code
   if test1:
      goto label1
   # some code
   
It took me ages to unwrap. Rather than try to port such code, I would 
always assume that it's probably buggy anyway and re-implement it!


From info at pythonware.com  Wed Jun  7 10:07:45 2000
From: info at pythonware.com (PythonWare)
Date: Wed, 7 Jun 2000 16:07:45 +0200
Subject: ANNOUNCEMENT: The Python Imaging Library, version 1.1 (june 7, 2000)
Message-ID: <014a01bfd089$d32b4af0$0500a8c0@secret.pythonware.com>

yet another release from the labs:

    The Python Imaging Library (PIL) adds image processing capabilities
    to your Python interpreter. This library supports many file formats,
    and provides powerful image processing and graphics capabilities.

    Version 1.1 moves the Tkinter support to a separate binary module,
    and fixes a couple of problems caused by incompatible changes in
    Python 1.6.

    Get your copy of the source kit here:

        http://www.pythonware.com/downloads.htm#pil11

    For more information, including support options, see the product
    site:

        http://www.pythonware.com/products/pil/

enjoy,
the pil team

    "Secret Labs -- makers of fine pythonware since 1997."




From richard_chamberlain at ntlworld.com  Tue Jun 13 16:55:17 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 13 Jun 2000 21:55:17 +0100
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com> <2af14c7e.01880953@usw-ex0104-028.remarq.com> <ge0dks0erbc0ki9g4pdelolgn17fcmvnon@4ax.com>
Message-ID: <s4x15.911$9K.23580@news6-win.server.ntlworld.com>

James,

Pythonwin is a marvellous little Python / Windows IDE.

You can get the latest version and more information from Mark Hammond's page
at the python starship:

http://starship.python.net/crew/mhammond/

or go straight for the downloads:

http://starship.python.net/crew/mhammond/win32/Downloads.html

Richard

jim roush <jar at mminternet.com> wrote in message
news:ge0dks0erbc0ki9g4pdelolgn17fcmvnon at 4ax.com...
> On Tue, 13 Jun 2000 05:44:24 -0700, Richard Chamberlain
> <richard_chamberlainNOriSPAM at ntlworld.com.invalid> wrote:
>
> >Jim,
> >
> >It works fine for me on Windows - except in IDLE. So either run
> >it from the command prompt or from PythonWin.
> >
> >In PythonWin you'll get the result back to the interactive
> >window, which you may not be expecting.
> >
>
> I upgraded to IDLE 0.5 and it works fine.  What is PythonWin and where
> do I find it?
>
> Thanks.
>
> jar at mminternet.com
> James A Roush




From andrew.henshaw at gtri.gatech.edu  Wed Jun 21 03:22:20 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Wed, 21 Jun 2000 03:22:20 -0400
Subject: Removing items from a list in a loop
References: <3952fdf6.950540640@nntp.interaccess.com>
Message-ID: <8ipcln$bmd$1@news-int.gatech.edu>

Thaddeus L. Olczyk wrote:
>I would like to remove a bunch of items from a list *efficiently*.
>Essentially  I would like to do this:
>
>   for x in list:
>        if Function(x):
>	list.remove(x)
>
...snip...
>Is there a safe way to iterate through a list and remove some items?
>
If you can negate the sense of Function(x) such that Function(x) returns
false when you want to remove x and true otherwise, then you can use the very
fast 'filter' built-in:

list= filter(Function, list)

-- 
Andy Henshaw


From aahz at netcom.com  Tue Jun 27 10:12:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 14:12:00 GMT
Subject: Comments: full-text indexing of RDBMS
References: <395b7627.97196711@news.online.no>
Message-ID: <8jacng$hb4$1@nntp9.atl.mindspring.net>

In article <395b7627.97196711 at news.online.no>,
Thomas Weholt <thomas at cintra.no> wrote:
>
>I`ve cooked up some plans for a simple, note! simple, full-text
>index-engine and would like some comments. I got little or no
>experience in the matter, some in database and a pretty happy
>python-camper, so the things here might be dumb and ineffective, but
>since it`s not outthere yet I have to do it myself.

You might want to take a look at the source for ht://Dig and/or agrep to
get some ideas.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From akuchlin at mems-exchange.org  Tue Jun 27 17:28:18 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 27 Jun 2000 17:28:18 -0400
Subject: Darwin (Not Python, yet...)
References: <Pine.A32.3.90.1000624142526.18059A-100000@elvis.med.Virginia.EDU> <8j7skg$1f$1@bob.news.rcn.net> <Pine.A32.3.90.1000626195403.20434A-100000@elvis.med.Virginia.EDU> <8jahjq$fd8$1@bob.news.rcn.net>
Message-ID: <3dbt0mx18t.fsf@kronos.cnri.reston.va.us>

"Jeffrey P Shell" <jeffrey at digicool.com> quoted:
> >  A binary package for Mac OS X Server from Scott Anguish.
> >  It doesn't seem to be a Debian package, which is what Darwin wants.
> >  I would guess it used whatever packaging NeXT used.
> >  Unpacking in manually and trying to run python I get:

So Darwin uses Debian packages?  Does this mean that MacOS X will also
support installing .deb files?  If so, the importance of supporting
.deb files in the Distutils may jump up a notch...

--amk


From matt at virtualspectator.com  Fri Jun  2 21:16:41 2000
From: matt at virtualspectator.com (matt)
Date: Sat, 3 Jun 2000 13:16:41 +1200
Subject: scoping problem with lambda
In-Reply-To: <39353A47.10EC15FB@do.not.use>
References: <3933484B.B471017A@san.rr.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use>
Message-ID: <00060313245601.00960@localhost.localdomain>

The following code works in global space(Please note, some \n and extra tabs
inserted for viewing sake),

tmp = [[['a'],['one','two','three','four']],
	[['b'],['one','two','four','five']],
	[['c'],['one','two','three','six']],
	[['d'],['one','two','two','four']],
	[['e'],['one','ten','three','four']]]  

kformat = ['one','two','three','four']

tmp2 = tmp[:]
for i in (0,2,3):
	tmp2 = filter(lambda x:x[1][i] in (kformat[i],),tmp2)

the result is :
>>> tmp2
[[['a'], ['one', 'two', 'three', 'four']], [['e'], ['one', 'ten', 'three', 'four']]]
Which is correct.

Now if I put it into a class I get an error:
#!/usr/bin/python

class trying_it:
    def __init__(self):
	self.tmp = [[['a'],['one','two','three','four']],
		[['b'],['one','two','four','five']],
		[['c'],['one','two','three','six']],
		[['d'],['one','two','two','four']],
		[['e'],['one','ten','three','four']]] 
	
	self.kformat = ['one','two','three','four']

    def try_it(self):
	tmp2 = self.tmp[:]
	for i in (0,2,3):
	    tmp2 = filter(lambda x:x[1][i] in (self.kformat[i],),tmp2)
	return tmp2


a = trying_it()
a.try_it()

I get the error :
Traceback (innermost last):
  File "test_it.py", line 16, in ?
    a.try_it()
  File "test_it.py", line 11, in try_it
    tmp2 = filter(lambda x:x[1][i] in (self.kformat[i],),tmp2)
  File "test_it.py", line 11, in <lambda>
    tmp2 = filter(lambda x:x[1][i] in (self.kformat[i],),tmp2)
NameError: i 

I know this is a scoping problem, but what is the best way around it ?

regards
Matt



From aahz at netcom.com  Thu Jun 15 08:57:53 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 12:57:53 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <3948D25D.7C83CD6@micron.net>
Message-ID: <8iajsh$8ui$1@nntp9.atl.mindspring.net>

In article <3948D25D.7C83CD6 at micron.net>,
Peter Immarco  <android at micron.net> wrote:
>
>LOC acronym stand for?

Lines of Code.  A rough measure of how "high-level" a language is; it
takes fewer lines of code to express something in C than assembley, for
example.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From pehr at pehr.net  Mon Jun 26 22:28:47 2000
From: pehr at pehr.net (pehr anderson)
Date: Tue, 27 Jun 2000 02:28:47 GMT
Subject: zope, python vs apache, perl
References: <8ish5s$kd4$1@nnrp1.deja.com>
Message-ID: <3958115E.C763890E@pehr.net>

Zope is great for remotely administering a simple site.
You can build a well structured zope site with remarkably 
little effort.  Using python to make cgi scripts running
under apache gives you more direct control but you pay
for that with complexity. The most complex of all is
trying to extend zope or implement somthing in perl.

If you plan on an easy-to-maintain site, 
You definitely want Zope for content and 
Python for the backend.  This works just
fine when served through apache.

	-pehr


amitr at w-o-i.com wrote:
> 
> hi
> 
> I want to evaluate zope, python vs apache, perl.
> 
> which one can be used where!
> strong points and pitfalls.
> 
> thanks in advance
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From Jonathan_Epstein at nih.gov  Wed Jun 14 15:02:13 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Wed, 14 Jun 2000 15:02:13 -0400
Subject: callbacks in python
Message-ID: <3947D6B5.664ABA2B@nih.gov>

I'm working on my first Python project, and am encountering the issue of
callbacks.  In particular, whenever my class undergoes a change which
modifies its print representation, I would like to notify a callback
routine which has been specified by the user of the class.  This
callback routine would usually be some sort of GUI.

Is this the right way to do things in python?  I read a reference in
_Python Programming on Win32_ which seemed to suggest that this was not
the right way to do things, but I am not sure.

Thanks in advance for any help, especially including any pointers to
examples...

- Jonathan


From mikael at isy.liu.se  Tue Jun  6 08:49:10 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 06 Jun 2000 14:49:10 +0200 (MET DST)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
In-Reply-To: <8hiq8d$ngm$1@newshost.accu.uu.nl>
Message-ID: <XFMail.000606144910.mikael@isy.liu.se>

On 06-Jun-00 Martijn Faassen wrote:
 >  It's more inflexible. I can't do:
 >  
 >    start()
 >      do_a()
 >      do_b()
 >    commit()
 >  
 >  And occasionally that would be nice. 

But you can do 

  if 1:
    start()
    do_a()
    do_b()
    commit()

which gives you your wanted but unnecessary indentation.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    06-Jun-00
Time:    14:47:36

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


From mnot at pobox.com  Sun Jun 25 23:05:33 2000
From: mnot at pobox.com (Mark Nottingham)
Date: Sun, 25 Jun 2000 20:05:33 -0700
Subject: ANN: cgi-buffer 0.1
Message-ID: <20000625200532.A684@mnot.net>

cgi_buffer is a Python libary that may be used to improve performance of CGI
scripts in some circumstances, by applying HTTP mechanisms that are
typically not supported for them.

It may be used both with Python and other CGI programs; Python may use it as
a simple library, while others may use it as a wrapper program.

Version 0.1 is a alpha release. Please send feedback, comments and
questions!

http://www.mnot.net/cgi-buffer/


-- 
Mark Nottingham
http://www.mnot.net/



From mikael at isy.liu.se  Tue Jun 13 07:33:26 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 13 Jun 2000 13:33:26 +0200 (MET DST)
Subject: No Boolean type in "types"
In-Reply-To: <394612a2.0@katana.legend.co.uk>
Message-ID: <XFMail.000613133326.mikael@isy.liu.se>

On 13-Jun-00 Thomas Mangin wrote:
 >  Can someone explain me why the module "types" have no boolean type defined
 >  ?

Probably because there is no boolean type. Anything empty or zero is
considered false, and anything else is considered true. 

Exampes:
   0, '', [], {}              are all false
   1, ' ', [0], {'false':0}   are all true

You can tamper with your own classes to make instances be considered 
true or false, to your liking. See the manual.

IIRC,

   true = (0==0)
   false = not true

would actually give you the integers 1 and 0 in the current 
implementation.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    13-Jun-00
Time:    13:23:35

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


From ge at nowhere.none  Wed Jun 14 17:51:17 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 14 Jun 2000 21:51:17 GMT
Subject: TCP packet size?
References: <39487D5B.FC96BE3F@rpgarchive.com>
Message-ID: <p7T15.658$i06.207009@ptah.visi.com>

In article <39487D5B.FC96BE3F at rpgarchive.com>, chris wrote:

> Is there a way to ensure or predict packet size when using TCP?

Nope.

TCP/IP guarantees that the bytes get there in the right order.
It neither knows nor cares about any sort of boundaries between
bytes.

> I'm developing a Multi-user application that functions much
> like a muli-player game.  The applications are sharing xml
> data.  The xml data can get very big 3000-6000 bytes.
> Unfortunately, when the size of the network message gets over
> about 2000, its to get broken up into smaller packets.  So,
> when I send a 3000 byte message it gets cut up into 2-3
> packets.  Anther problem I get is when I send two small
> messages back to back, they are sometimes received together.
> Is there something I'm missing here?

That's the way TCP/IP works.  You're using a stream protocol to
try to send datagrams.  That's like trying to juggle tennis
balls while wearing shoes on your hands.  Shoes are a damn fine
invention, but you've got to use them for the right thing.

You want to use a datagram service instead of a stream service.

There's UDP, but it's unreliable, so you've got to add your own
ack/retry scheme -- bit of pain.  On Linux "man socket" claims
to support reliable datagrams:

   SOCK_SEQPACKET

       Provides a sequenced, reliable, two-way connection-
       based data transmission path for datagrams of fixed
       maximum length; a consumer is required to read an entire
       packet with each read system call.

That sounds like what you want, except it's not implemented for
address type AF_INET (which is probably what you're using).
Bummer.

There's also plain old unreliable datagrams (SOCK_DGRAM), and
reliable, unordered datagrams (SOCK_RDM). Probably the only
ones you can depend on being available are SOCK_STREAM (TCP/IP:
which is what you're using) and SOCK_DGRAM (UDP/IP: unreliable
datagrams).

>I tried using headers and breaking up messages into smaller chunks, but
>the size of the receiving packets doesn't always match what I sent.  I'd
>appreciate any help or suggestions on how I can tackle this problem.
>I'd be happy to send someone my network code.

Maybe you can find somebody who's already written a reliable
datagrram layer that runs on top of UDP/IP?

-- 
Grant Edwards                   grante             Yow!  Yow! We're going to
                                  at               a new disco!
                               visi.com            


From oblom at barak-online.net  Mon Jun  5 05:59:53 2000
From: oblom at barak-online.net (Evgeny Zemlerub)
Date: Mon, 05 Jun 2000 12:59:53 +0300
Subject: Python's RE
Message-ID: <393B7A18.D0F1AA78@barak-online.net>

Hi.

How exectly do I something that reassembles perl's :

$string = "qwerwe kkljlksdfg (first string) asdkjflkjg
(secondstring)jlskjlfdkg(one more string)";
@words = ($string =~ /\((.+?)\)/smg);
foreach(@words){
        print "$_\n";
}
                                            ?

the closest thing that i got is :
 print re.search(r'\((.+?)\)',
'aaa(foo)bbb(bar)asdfsadf',re.S|re.M).groups()

but it doesn't returns what i expected

Thanks in advance,
    Evgeny



From jkraska1 at san.rr.com  Sun Jun  4 15:46:27 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 19:46:27 GMT
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com> <39398EEE.C18EC987@tismer.com> <3939903D.79BD8D2D@san.rr.com> <393AA136.36C34915@tismer.com> <393AAC0F.FFB4B65@san.rr.com> <7sbljso0tvth02de6ql5at6100pns97f3h@4ax.com>
Message-ID: <393AB2B8.1D1B3A87@san.rr.com>

Ben Wolfson wrote:
> 
> On Sun, 04 Jun 2000 19:18:03 GMT, Courageous <jkraska1 at san.rr.com>
> wrote:
> 
> >it would also be possible to allow my deque
> >implementation to be concatenated in as well, but I didn't do it,
> >because I didn't want to force the module dependency.
> 
> Surely that code could be wrapped in #ifdef/#endif?

Well, of course. I tend to think about these things as binary
releases, but I suppose I could put conditional compilation code
in yes. That's a good idea. :)-

Another option would be to create a single ADT module which
contained *all* the container classes: deque, dlist, pqueue,
and anything else I come up with. Right now, the python idiom
for using these things is usually something like:

import deque

d = deque.create()

... but it could just as easily be ...

import adt

dq = adt.deque()
pq = adt.pqueue(n)
dl = adt.dlist(n)

BTW, I was trying to figure out how to pass 'optional' args
to external C functions and couldn't grok it. Anyone know how?
(I wanted pqueue and dlist to default to 10 or something if no
n was offered).



C/


From root at rainerdeyke.com  Wed Jun  7 12:54:14 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 07 Jun 2000 16:54:14 GMT
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <7wc%4.7928$r7.1560935@news1.rdc2.tx.home.com> <393DDDF8.DE0401BB@sightreader.com>
Message-ID: <W6v%4.8415$r7.1741613@news1.rdc2.tx.home.com>

Ken Seehof <12klat at sightreader.com> wrote in message
news:393DDDF8.DE0401BB at sightreader.com...
> Rainer Deyke wrote:
> > Write an optimizing compiler that increase performance by a factor of at
> > least 100 and decreases memory usage by a factor of at least 100.
Python in
> > its current incarnation is too inefficient to be useful as primary
language
> > for shrink-wrapped software.
>
> Because of it's dynamic nature, Python simply can't be optimized all that
much
> (read about all that type inference stuff and you'll see what I mean).

A truly good optimizer might do wonders by analysing the entire system.  For
example, it sees that a certain name can only reference objects of certain
type, so it bypasses the generic access mechanism.  It sees that instances
of class X only get attributes from a certain set, so the dictionary is
replaced by a more efficient data structrure.  In general, if a human can
read a python program and produce an equivalent efficient assembly program,
a computer should be able to get close.

> Fortunately that just doesn't matter!
>
> 98% of code does not need to run fast.  You write the other 2% in C.
> 98% of all objects (on a per-object basis) do not need to be packed tight.
You
> implement the other 2% in C.
>
> Conclusion: The primary language does not need to be efficient.

But this involves complicated maneuvering between Python and C.  In many
instances it is easier to just write the program in straight C.


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From tjjudson at yahoo.com  Mon Jun 26 14:08:34 2000
From: tjjudson at yahoo.com (TJ Judson)
Date: Mon, 26 Jun 2000 11:08:34 -0700 (PDT)
Subject: Vault of Parnassus
Message-ID: <20000626180834.25892.qmail@web2002.mail.yahoo.com>

I have only recently (last month) been searching for Python code from
the "Vault of Parnassus" http://www.vex.net/parnassus/

I have found that this site is unavailable far more than it is
available.  I have tried different browsers, different times of day,
etc.

1) Can anyone else access the site OK?  (today, for instance)

2) Is there a mirror site somewhere?  Today, I'm looking for a new
console GUI -- need an independent command line where the user at any
time can start/abort and pause/resume a script.


Thanks,
-TJ Judson

__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail - Free email you can access from anywhere!
http://mail.yahoo.com/



From rzantowNOrzSPAM at usa.net.invalid  Fri Jun 23 10:25:44 2000
From: rzantowNOrzSPAM at usa.net.invalid (RickZ)
Date: Fri, 23 Jun 2000 07:25:44 -0700
Subject: Countdown
References: <3952952D.88E3588D@asme.org>
Message-ID: <1a34d4c8.35960038@usw-ex0105-040.remarq.com>

Depending on what you're doing, how about:

for i in range( maxamt ):
   print item[ maxamt - i ]

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From pinard at iro.umontreal.ca  Sat Jun 10 12:00:51 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 10 Jun 2000 12:00:51 -0400
Subject: Python Productivity over C++
In-Reply-To: wlav@hpatl26.cern.ch's message of "9 Jun 2000 16:36:30 GMT"
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
	<394113F2.8D6D0833@lemburg.com> <8hr6ee$st8$1@sunnews.cern.ch>
Message-ID: <oq3dmlbk5o.fsf@titan.progiciels-bpi.ca>

[wlav at hpatl26.cern.ch (Wim Lavrijsen)]
> ["M.-A. Lemburg" <mal at lemburg.com>]

> >reduces LOCs, scales well (from 5-100k LOCs if needed -- which is
> >roughly equivalent to 25-250k LOCs in C++ from my own experience).

> This is interesting.  Could other people confirm this?  Ie. that the
> trade-off for Python is larger at small projects (factor ~5) than at medium
> sized projects (factor ~2.5)?  And what is the cause of the difference?

I would be tempted to say than for big projects, you usually build specific
libraries and tools for your project, kind of another language layer.
One this is done, and if done well, it increases the expressivity of the
language you use, whatever the language is, and so, reduces the difference
between languages.  For medium sized projects, the layer of specific
libraries is thinner, and you then observe more directly the expressivity
of the bare programming languages.

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





From cfelling at iae.nl  Sat Jun 10 18:07:03 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 11 Jun 2000 00:07:03 +0200
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <8hs4d8$33b$1@animus.fel.iae.nl> <20000610120811.H20700@xs4all.nl>
Message-ID: <8hue67$10l$1@animus.fel.iae.nl>

Thomas Wouters <thomas at xs4all.net> wrote:
> On Sat, Jun 10, 2000 at 03:07:52AM +0200, Carel Fellinger wrote:

>> What about __add_is__? This sounds like addish, a kind of add; it resembles
>> its use, plus followed by is; would prevent confusion of where to place the
>> r like in __r_inplace_add__ versus __inplace_radd__; and still is concise:)

> Well, that last bit is a non-issue, at least as far as I'm concerned,
> because an augmented assignment doesn't need a 'right-hand' hook: The
> new hook is called to change the left-hand-side object in-place, and you
> shouldn't expect the right-hand-side object to know how to do that to the
> lhs object. *that* would be confusing ;-)

Au, au, do I feel stupid!
-- 
groetjes, carel


From tim_one at email.msn.com  Tue Jun  6 13:38:34 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 6 Jun 2000 13:38:34 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
In-Reply-To: <m3wvk2wy1u.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <001101bfcfde$0a6a6f20$48a2143f@tim>

[Michael Hudson]
> ...
> Tangentially, does anybody know of any implementations of JavaScript
> (not necessarily connected to a browser) that let you type interactive
> commands into a listener and genereally muck about with the language?

Unsure about NS, but under IE5 you can type JavaScript expressions directly
into the address bar (as if entering a URL), but prefixed with
"javascript:".  For example, type this in:

javascript: 1+2

then hit ENTER; IE5 echoes back

javascript:%201+2

in the address bar and displays

3

in the main window.  Unfortunately(?), vrbl assignments do not persist, so
you're really limited to a single line of expressions:

javascript: x=3; y=4; x*x+y*y

So the answer to your question is "no" <wink>.  However, as with the Python
shell, it's a convenient way to see what a line of JS does.

not-sure-this-new-fangled-web-stuff-will-ever-catch-on-ly y'rs  - tim





From soundhack at hotmail.com  Thu Jun  1 05:36:16 2000
From: soundhack at hotmail.com (Robert)
Date: Thu, 1 Jun 2000 05:36:16 -0400
Subject: ANNOUNCEMENT: PythonWorks 1.0 now available (may 31, 2000)
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com> <0CmZ4.30723$WS3.355848@typhoon.we.rr.com>
Message-ID: <39362f52$0$10880@senator-bedfellow.mit.edu>

Stephen Hansen wrote in message
<0CmZ4.30723$WS3.355848 at typhoon.we.rr.com>...
>shop.microsoft.com, but the Learning edition is also about $100) [ MS
VC++'s
>standard is about $100 I think, also, but i'm not including it in this
>because I have a hard time calilng it a 'RAD' environment... that
>splitting-of-a-window thing it makes you do is irritating, I want to be
able
>to draw a whole form :P ]


I have my own ideas about pricing, but I'm curious about this. could
you please explain to me what you mean by the 'splitting-of-a-window thing'?
I agree that VC++ is not a RAD environment btw.

Robert





From aahz at netcom.com  Fri Jun 16 21:38:03 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jun 2000 01:38:03 GMT
Subject: Are list mutations thread-safe?
References: <etd8zw5qfdv.fsf@w20-575-109.mit.edu>
Message-ID: <8iekpr$inp$1@nntp9.atl.mindspring.net>

In article <etd8zw5qfdv.fsf at w20-575-109.mit.edu>,
Alex  <cut_me_out at hotmail.com> wrote:
>
>Hi.  Can two threads get in each others' way if they simultaneously try
>to mutate a list?  How could I make a reliable test to answer a question
>like this?

It depends how they mutate it.  I believe that single builtin functions
(e.g. [].append()) are thread-safe.  OTOH, [].sort() may not be
thread-safe, particularly if you've got user-defined classes on the list
with __cmp__() defined.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From stephen at cerebralmaelstrom.com  Mon Jun 26 03:05:46 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Mon, 26 Jun 2000 07:05:46 GMT
Subject: Control cmp used with 'in'?
References: <52D55.2660$Gh6.168003@news-west.usenetserver.com>
Message-ID: <ehD55.95136$WS3.791305@typhoon.we.rr.com>

I believe its the __contains__ method.

--Stephen

Matthew Cline <matt at nightrealms.com> wrote in message
news:52D55.2660$Gh6.168003 at news-west.usenetserver.com...
> Is there any way to control the function that is used during
> an 'in' search?
>
> Thanks in advance.
>
>
>




From embed at NOSPAM.geocities.com  Wed Jun 28 12:46:29 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 12:46:29 -0400
Subject: what is lambda?
References: <8jd8uu$gnh$1@nnrp1.deja.com>
Message-ID: <2Zp65.11236$227.231939@nnrp1.uunet.ca>

> I am a beginner at python and I would like to know what lambda does.
> Could someone post an in-your-own-words definition and possibly an
> example?  Thanks,

1) Meta-Answer: Read the FAQ and Manual sections.

2) Instant Gratification Answer

here's a standard function definition:

def func(x):
    return x *2

here's a way to write the same thing, in effect, but using Lambda:

func = lambda (x): x * 2

While that example isn't useful, it gives you the idea.  Think of lambda as
a way to generate a function and return that function as an object, without
having to give it a name. I primariyl use it to pass an algorithm or
expression as a parameter to a function, or other places where I want to
pass code in a variable instead of passing a reference to a method
containing that code. Clear


Warren Postma




From bjorn at roguewave.com  Wed Jun  7 16:45:02 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 07 Jun 2000 14:45:02 -0600
Subject: SQL Variable Substitution Question
References: <8hmacu$ia9$1@news-int.gatech.edu>
Message-ID: <393EB44E.80BEDCEE@roguewave.com>

Shengquan Liang wrote:
> 
> 'm trying to make a query using varible in a
> Python program.
> 
> say i have yyyy = 1000
> and i want to
> 
>  select * from attt
>  where bbbb = yyyy
> 
> in the documentation of DC Oracle, it says that
> i shoul d use 'select * from attt'
>                 'where bbbb = :yyyy'
> 
> to make it work.
> 
> BUT, when i tried it this way, the
> iDC oracle reponded by:
> 
>   oci.err:(1008,'ORA-1008: not all variables bound\012')
> 
> what's the problem with that?

You need to pass the yyyy varible, so it can be bound by the database
module.  I don't have it handy so I forget the syntax, but it should
look something like:

  cn.execute( 'select * from attt where bbb = :yyyy', (yyyy,) )

hth,
--bjorn



From borlak at home.com  Mon Jun 19 18:48:46 2000
From: borlak at home.com (Michael Morrison)
Date: Mon, 19 Jun 2000 22:48:46 GMT
Subject: Binary Input: read how?
Message-ID: <IRX35.917$UL.59581@NEWS1.RDC2.TX.HOME.COM>

Here's the basic problem: I can read a binary file using open(...,'rb') but
when I assign it to a object (b = file.read()) it stores it as an escape
character or \300 - whatever range.

I want to test this byte of binary data by comparing it's bits.  b[0] & 1 ..
b[0] & 2
for x in range(30):
    if b[0] & 2**x:
        print 1,
    else
        print 0,

This doesn't work.  Tried all sorts of things.... Can't get it to work :(

What am I missing here?

thanks




From shapr at uab.edu  Fri Jun  2 05:40:03 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 02 Jun 2000 09:40:03 GMT
Subject: Styling Temporary Code
References: <3936A9BB.FBEF9201@visionart.com> <3936C52F.520B04FD@uab.edu> <joEZ4.27$vh.14427@news>
Message-ID: <39378741.BFA99FE3@uab.edu>

Robert wrote:
> 
> What IS PyUnit...I just read a "bit" about eXtreme programming and PyUnit
> was mentioned.
> 
> Bob (my-name-is-a-palindrome)

You can find it at: http://pyunit.sourceforge.net
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius On fIre


From eduard.hiti at t-online.de  Thu Jun  1 10:23:11 2000
From: eduard.hiti at t-online.de (Eduard Hiti)
Date: Thu, 1 Jun 2000 16:23:11 +0200
Subject: str(None) == '' ?
Message-ID: <8h5rpb$fl3$14$1@news.t-online.com>

I would like to suggest that

    str(None)

should NOT return the string 'None', since most of the time, it complicates
things.

Then code like

    value = str(get_stuff())
    if value:
        # Do something to value

would not stumble over the occasional None.

At least one of the string converting functions repr|str should return a
empty string on None.

Is there a reason for str(None) == 'None' that I'm missing?





From wjdandreta at worldnet.att.net  Mon Jun 19 21:30:19 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 20 Jun 2000 01:30:19 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <LOz35.14014$Xx5.685990@bgtnsc06-news.ops.worldnet.att.net>

Hi all,

I want to thank everyone who contributed to this thread, it has helped me
understand Python a little better.

All my thoughts on what was causing my Python script to run slow are false.

There is no speed penalty for comments.

There is no speed penalty for DOC strings.

There is no speed penalty for function calls. (At least in this case)

My error was that I did the following:

from strop import joinfields, splitfields
from string import replace

I erroneously assumed that replace would use the joinfields, splitfields
from strop, but it really used the joinfields, splitfields from string which
are very slow.

Bill
William Dandreta wrote in message ...
>Let me preface my remarks with:
>
>--I am using the Python 1.2 binary for DOS with the Python 1.5.2 library
>which causes some problems. I avioded some problems by importing functions
>directly from strop instead of string. Replace is not in strop so I used
the
>one in string but I had to comment out (put an octothorpe at the start of
>each line) the function's descriptive string to get it to work.--
>
>A while ago I posted a message called Why is Python so slow? At that time
>this Python script took 90 minutes to run. By making some relatively minor
>changes, I was able to reduced the time to 6 minutes.
>
>The biggest improvement came (about a factor of 6) when I changed the
>replace(x,y) function in string with
>joinfields(splitfields(x,y),''). Considering that this is exactly what the
>replace function in string does, I was quite surprised at the results.
>Essentially I reduced 3 nested function calls to 2.
>
>The only thing that makes sense to me is that Python is spending 85% of the
>time processing the comment lines and 15% of the time doing useful work.
>There is a string.pyc so I would have assumed that it would not contain any
>comments or descriptive strings but apparently that's not true.
>
>I am not very familiar with interpretive languages, Python is the first one
>I have used, but it seems that comments can cause a serious perfomance hit
>because they have to be processed along with the code.
>
>Bill
>
>




From fisher at r.hi-fi-net.com  Fri Jun 16 07:49:23 2000
From: fisher at r.hi-fi-net.com (fisher at r.hi-fi-net.com)
Date: Fri, 16 Jun 2000 20:49:23 +0900
Subject: i made  mailer (Python CGI)
Message-ID: <394A1443.9738C5F3@r.hi-fi-net.com>

I made mailer written in python.

download:
http://fisher.r.hi-fi-net.com/python/tori/index.html

doc:
http://fisher.r.hi-fi-net.com/python/tori/readme-us.html

Try it if you have interest.



From matz at netlab.co.jp  Fri Jun  9 00:23:01 2000
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 09 Jun 2000 13:23:01 +0900
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news>
 <8hlrkn$5t2$1@nnrp1.deja.com>
 <1ebvwau.af3wl21301xdcN%bparsia@email.unc.edu>
Message-ID: <87em67fpp6.fsf@ev.netlab.co.jp>

bparsia at email.unc.edu (Bijan Parsia) writes:

|Are Ruby's *classes*, themselves, instances of classes (aka
|metaclasses)? Like, oh, let's see, *Smalltalk*!

I know it's off-topic, so in sort answer:

  yes
							matz.


From thomas at xs4all.net  Mon Jun  5 10:51:09 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Jun 2000 16:51:09 +0200
Subject: Newbie Questions
In-Reply-To: <8hgdpf$2fn$1@nnrp1.deja.com>; from arun_rajappa@my-deja.com on Mon, Jun 05, 2000 at 02:34:24PM +0000
References: <8hgdpf$2fn$1@nnrp1.deja.com>
Message-ID: <20000605165108.X469@xs4all.nl>

On Mon, Jun 05, 2000 at 02:34:24PM +0000, arun_rajappa at my-deja.com wrote:

> i've been working with python for the last week and have gone through
> the tutorial by guido. i have a few questions, and hope i will find
> help and enlightenment here... ~~ :-)

> a) Does Python support multi-threading ?

Yes.

> What are the thread and threading modules for ?

Multi-threading :-) the 'thread' module is the low-level thread interface,
whereas the threading module is a higher-level, more object-oriented,
Java-inspired (I do believe) interface. Threading is fairly basic, however,
and not very widely available (partly because not all platforms support
(proper) threading, and partly because it has to be explicitly enabled by
whoever compiles Python, on the target machine.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From think2 at pd.jaring.my  Thu Jun  8 15:44:57 2000
From: think2 at pd.jaring.my (Raymond Ng Tong Leng)
Date: Fri, 9 Jun 2000 03:44:57 +0800
Subject: Why can't I restore my redirected stdout?
References: <8hm5od$lki$1@news5.jaring.my> <slrn8jtgg8.oe2.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8hotma$48l$1@news5.jaring.my>

Remco Gerlich wrote in message ...
>
>I just typed this in, and it worked fine! Something else must be happening.
>Are you in idle or so? Does it work inside a normal python interpreter?
>What's your platform?
>

Yes, I was using IDLE 0.4 with Python 1.5.2 on Win NT. When I tried it
inside a normal Python interpreter like you suggested, it worked fine. I
hope Guido fixed this bug... sorry, improved on this feature :) in the
latest version of IDLE.

>
>Seems to me your solution is the easiest way to get a cross platform null
>device, but maybe there's something like it in the library already, I don't
>know.
>

Tried looking in the documentation but couldn't find any reference to a null
device.








From erno at iki.fi  Wed Jun 28 16:33:30 2000
From: erno at iki.fi (Erno Kuusela)
Date: 28 Jun 2000 23:33:30 +0300
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com>
Message-ID: <ku66qtv945.fsf@lasipalatsi.fi>

>>>>> "Eric" == Eric Lee Green <eric at estinc.com> writes:

    Eric> Insert speed is horrible -- inserting data takes
    Eric> forever and a half, though PostGreSQL 7.0 now has a bulk
    Eric> import facility that may make it faster (I haven't had a
    Eric> chance to benchmark it yet).

hmm. there are some things that can make inserts slow:

1) there is a switch telling postgresql whether it should flush to
disk after each transaction. this calls the flush() system call. its
implementation is very inefficient on linux.  if you are running on
linux and can live with the reduced robustness, turn that feature off.

2) if you are using sql insert statements without surrounding
begin/commit (ie in autocommit mode), a transaction for each insert
makes things slow.

3) using sql insert statements is not itself the fastest thing int the
world. the pygresql connection objects have a inserttable() method for
doing something similar (i think) to what COPY does with the "psql"
interactive interactive terminal.

try these, i bet you will have much better luck.

    Eric> And, finally, the documentation is cryptic, incomplete, and
    Eric> scattered all over the place... I had to reference two
    Eric> different manuals in order to figure out how to create (and
    Eric> destroy) a PL/SQL stored procedure, for example.

the book on the postgresql web site covers all of thiss stuff
pretty nicely.  <URL:http://www.postgresql.org/docs/awbook.html>.

    Eric> The biggest minus for the back-end of a b2b web site,
    Eric> though, would be the fact that it is bog-slow at opening a
    Eric> connection. PostGreSQL forks a new process for each
    Eric> connection, and the new process must do all sorts of things
    Eric> to establish authentication etc...

well, usually if you do not want to wait for seasons to change
when using the web site you do not use cgi anyway. even without
connecting to a sql rdbms it is dog slow.

    Eric> PHP3's PostGres support gets around this by creating
    Eric> "persistent" connections that hang around with the Apache
    Eric> process, but I have no idea if Python's Apache module
    Eric> support is robust enough to handle this situation.

yeah. i've been using httpdapy for many projects and it works fine.
mod_python (http://www.modpython.org/) is a rewrite of it by
the same author...

zope also works very well.

    Eric> Not to mention that Python's Apache module support is almost
    Eric> entirely undocumented... it is quite unclear, for example,
    Eric> how to run your former Python CGI scripts via the PyApache
    Eric> module.

i don't know about the documentation of PyApache (it's about as slow
as CGI anyway), but the mod_python and zope docs are pretty nice
(httpdapy docs are not totally nonexistant either).

  -- erno


From mwh21 at cam.ac.uk  Tue Jun  6 11:32:40 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Jun 2000 16:32:40 +0100
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl> <393D1155.D067A821@decrc.abb.de>
Message-ID: <m33dmqygdj.fsf@atrus.jesus.cam.ac.uk>

Peter Mueller <pmueller at decrc.abb.de> writes:

> Hi,
> 
> it looks like that I have the same problem. I use the pickler. When
> I import gtk/pygnome modules, the pickler can't read my pickled
> object (containing floats) anymore.
> This behaviour is new since redhat 6.2. On redhat 6.1 the same code
> with the same pickled object works without any problem. The python
> versions on both redhat 6.1/6.2 are the same. There must be a
> difference one of the modules.

Hmm, odd.  I can't get it to break here, on what was once a Redhat 6.1
box, but I've upgraded almost everything except libc. What versions
of python, gtk+, pygtk, pygnome and glibc are you running?

I have:

[mwh21 at atrus cricapplet]$ rpm -q gtk+ pygtk pygnome glibc
gtk+-1.2.7-1_helix_1
pygtk-0.6.3-1
pygnome-1.0.50-1
glibc-2.1.2-11

and python's from CVS just before the unicode landed (so, a couple
months back).

My money's on so bizzarre interaction with libc, myself.

Cheers,
Michael

-- 
  Of the four project development variables - scope, cost, time and
  quality - quality isn't really a free variable.  The only possible
  values are "excellent" and "insanely excellent", depending on
  whether lives are at stake.              -- Kent Beck, XP Explained


From m.faassen at vet.uu.nl  Thu Jun 15 18:27:58 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 22:27:58 GMT
Subject: Python Productivity over C++ (Encapsulation)
References: <20000615183521.59984.qmail@hotmail.com>
Message-ID: <8ibl9e$mf8$1@newshost.accu.uu.nl>

Hung Jung Lu <hungjunglu at hotmail.com> wrote:
[snip]
> One of the biggest lies in modern computer language theory is that 
> encapsulation is there for a good technical reason.
[snip]
> Encapsulation is not there for any technical reason.

Encapsulation is necessary if you want abstraction. There is no strictly
*technical* reason for wanting abstraction; but source is not written
for the computer alone. Source is written at least as much for other
humans. And humans want to be able to forget about details so they
can focus on the problem at hand.

By encapsulating the details (into a function, module or class), you can
use the functionality without having to think or worry about the details.
You don't care about the implementation. That's abstracted, encapsulated.

So there's a human reason. Or a software engineering reason, at least.

[snip]
> If one day Java becomes a legacy language and loses the battle against 
> open-source languages like Python, it's exactly because people are waking up 
> and liberating themselves from the corporate handcuffs.

This is perhaps a slightly different issue. The difference is subtle.
Python does not have as many tools for 'enforcing' encapsulation, just like
it doesn't have so many tools for enforcing type-safety. Python trusts
the programmer more in both cases.

That doesn't mean a Python programmer should forget about encapsulation.
Python programmers care about interfaces, and if you want to have abstract
interfaces, you'll have to stay away from implementation details, and
respect the encapsulations. But if you know what you're doing and you
really want it, Python doesn't get in your way. So you can't get painted
into a corner as easily with Python.

Regards,

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


From godzilla at netmeg.net  Sun Jun  4 09:51:24 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sun, 04 Jun 2000 13:51:24 GMT
Subject: HTMLParser: getting one extra space around &gt;/&lt;/^M tags
References: <m21z2grsut.fsf@netmeg.net>
Message-ID: <m2og5h1rfm.fsf@netmeg.net>

okay then.

does anyone know who WROTE htmllib.py and/or formatter.py? maybe i can
ask them.

les schaffer


From jajs at NOSPAMretemail.es  Wed Jun 28 07:40:21 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Wed, 28 Jun 2000 13:40:21 +0200
Subject: "No module named _locale" (1.5.2 w/ RH 6.2)
References: <16600709@oberon.noris.de>
Message-ID: <slrn8ljp15.25e.jajs@Pascual.es>

hartmut Goebel dijo sobre "No module named _locale" (1.5.2 w/ RH 6.2),
en fecha: Tue, 27 Jun 2000 22:28:22 +0100

|Hallo,
|
|when trying to import 'local' a get an error. How can I solve it?
|
|>>> import locale
|Import Error: No module named _locale
|
|Well, this is true, since there is no _locale.* on my system.
|I'm running RedHat 6.2 with Python 1.5.2 (from ReadHat's 6.2 rpm)

In Mandrake 7.1 the python rpm didn't have that file (it was a bug).
Probably the same happens in Red Hat (in the Mandrake 7.2 python rpm the
bug is fixed).


From Kai.Grossjohann at CS.Uni-Dortmund.DE  Sun Jun 18 12:29:37 2000
From: Kai.Grossjohann at CS.Uni-Dortmund.DE (Kai =?iso-8859-1?q?Gro=DFjohann?=)
Date: 18 Jun 2000 16:29:37 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com> <uitv9qb72.fsf@ctwd0143.fitlinxx.com> <slrn8kobd5.1vs.hzhu@mars.localdomain>
Message-ID: <vaf4s6rvtow.fsf@lucy.cs.uni-dortmund.de>

hzhu at mars.localdomain (Huaiyu Zhu) writes:

> This is simple to do for one file, but not for many files.  I'd prefer
> running something in batch mode on a list of files.

There are Unix programs `expand' and `unexpand'.  Presumably, running
the following commands on a file foo would change tab stops from 4 to
8:

    expand -4 foo > foo.new
    unexpand -8 foo.new > foo
    rm -f foo.new

You could easily do this in a loop:

    for f in *.c
    do
        expand -4 $f > $f.new
        unexpand -8 $f.new > $f
        rm -f $f.new
    done

Does this help?

kai
-- 
I like BOTH kinds of music.


From neelk at brick.cswv.com  Sun Jun 11 18:11:23 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 11 Jun 2000 22:11:23 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net>
Message-ID: <slrn8k85j2.fe6.neelk@brick.cswv.com>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> 
> On the other hand, I'm still searching for the best way to speak
> about the "well you Python people only find those errors at
> run-time" objection.

IMO, this objection is only valid when coming from an ML or Haskell
programmer, because misfeatures in the design of C++ and Java's type
systems undermine their semantic soundness to the point that their
static typing doesn't offer any benefit to speak of.

C++ has unsafe casts to kill whatever pretensions to type-safety the
language has. Java fixed this by raising exceptions on invalid casts,
but followed C++'s braindamaged example by including a null as a
member of every object type. 

Argh! Argh! Argh! This means that just about every invariant you care
to name no longer holds, because a variable with a particular object
type can't be guaranteed to behave sanely when a method that's a part
of that class's protocol is invoked on it. Now, what was that about
strong typing in Java reducing runtime errors?

We've known how to do better for decades, but we don't. I think our
only hope would be to give CAR Hoare absolute veto power over the
design of any new strongly-typed language....


Neel


From blablu at gmx.net  Tue Jun 27 09:25:27 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 27 Jun 2000 13:25:27 GMT
Subject: problem with accessing methods?
Message-ID: <8F60916A3mikecat@192.168.10.38>

Hi!

How can I access a method with variable amount of parameters?
I have sort of an object management system which routes method invocations 
from outside to the objects, but the management system don't know anything 
about the methods.
In C I would use a pointer to a struct, but this is Python :-)

Thanks in advance,
Mike


From tim_one at email.msn.com  Wed Jun 21 12:32:39 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 21 Jun 2000 12:32:39 -0400
Subject: 123.4+2=125.40000000000001
In-Reply-To: <Pine.GSO.4.20.0006211310100.17041-100000@luna.dsv.su.se>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEGCGFAA.tim_one@email.msn.com>

[Peter Koppatz]
> On my calculator: 123.4+2 = 125.4 !?

[Kalle Svensson]
> In fact, this probably isn't true. It's just that your calculator rounds
> the answer before handing it to you, just like print. Maybe. I should
> probably change my name to Search Deja... :)

Actually, on Peter's calculator it probably is true:  almost all hand
calculators today lack floating-point hardware, and use integer hardware to
simulate *decimal* floating-point via software.  The REXX language also uses
decimal floating point.  On "real computers", that's also possible, but much
slower than using their hardware binary floating point.

> Also note that there has been a lot of discussion on this topic here, and
> that the behavior might change in a later version of python.

It would sure be nice to have an alternative.  In a deep sense, it's
basically futile:  the deadly surprises in computer real arithmetic derive
from using a fixed amount of storage for each real value.  That gets you
into approximations ("round off error"), overflow & underflow, whether the
underlying system is binary, decimal, fixed-width rational, log or log-log,
...  The advantage to decimal floating point is mostly that people are
*accustomed* to the simpler kinds of errors it makes, thanks to calculator
and pencil-&-paper experience.  For example, nobody thinks to complain about
10/3 = 3.3333333 on an 8-digit hand calculator, or the later 3.3333333 * 3 =
9.9999999.  But do a similar thing with binary floating point, and they
suddenly feel violated <0.9 wink>.

floating-point-is-a-harsh-lover-ly y'rs  - tim





From Aline.Dumay at mc2.fr  Wed Jun 28 09:36:23 2000
From: Aline.Dumay at mc2.fr (Aline Dumay)
Date: Wed, 28 Jun 2000 15:36:23 +0200
Subject: Problem with os module
Message-ID: <3959FF57.85E1A503@mc2.fr>

Hi!
I'm new on the list, and I'm french, so excuse my english, please.

I'm using python with IntraSpect, and I've got a problem.
When I use the following instructions:

from os import *
if os.path.isfile("c:\test"):
   # whatever_you_were_going_to_do
   pass
 else:
   print "Error" # file not found

I've always the following message :  class 'org.python.modules.os' has
no attribute 'path' .
I've got the class in my classpath. I don't understand where is the
problem.
Can anyone help me?

Regards,
Aline.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Aline.Dumay.vcf
Type: text/x-vcard
Size: 316 bytes
Desc: Carte pour Aline Dumay
URL: <http://mail.python.org/pipermail/python-list/attachments/20000628/e5bfac34/attachment.vcf>

From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Fri Jun 23 10:13:15 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Fri, 23 Jun 2000 07:13:15 -0700
Subject: Disabling Comboboxes
References: <8iqqb5$d4s$1@nnrp1.deja.com>
Message-ID: <16ef54c0.7dd47049@usw-ex0105-035.remarq.com>

Hi Kapil,

In the ToDo list for Pmw disabling is mentioned so hopefully
we'll have a solution from the source before long. In the
meantime...

This is a bit of hack (in the derogatory sense of the word) but
hey!

So you basically want to import the myComboBox class and create
an instance just as you would with Pmw. I've added a couple of
methods - disable and enable suprisingly enough.

Included below is some test code.

It works in the same way as you mentioned, by unbinding the
events, and rebinding them to enable it.

Richard


from Tkinter import *
import Pmw
root = Tk()
root.title('Disabling ComboBox Hack')
Pmw.initialise()

class myComboBox(Pmw.ComboBox):
    def disable(self):
        # Rebind things to my empty handler
        self.component('arrowbutton').bind('<1>',self.handler)
        self.component('arrowbutton').bind('<3>',self.handler)
        self.component('arrowbutton').bind('<Shift-
3>',self.handler)
        self.component('entryfield_entry').configure
(state='disabled',fg='grey')

    def enable(self):
        # bind the events back up to the original methods
        self.component('arrowbutton').bind('<1>',self._postList)
        self.component('arrowbutton').bind('<3>',self._next)
        self.component('arrowbutton').bind('<Shift-3>',
self._previous)
        self.component('arrowbutton').configure(takefocus=1)
        self.component('entryfield_entry').configure
(state='normal',fg='black')

    def handler(self,event):
        # so it doesn't propagate the event
        return('break')

# Now test it

comboentries = ("Pmw Should", "have a nicer way", "to
disable", "a combo box")
theEnabler = Button(root, text='Enabled',padx=20, pady=10)
theEnabler.pack(expand=1, fill=BOTH, padx=8, pady=8)

combobox = myComboBox(root, label_text='Disabling Combobox:',
labelpos='wn',
                        listbox_width=24, dropdown=1,
                        scrolledlist_items=comboentries)
combobox.pack(fill=BOTH, expand=1, padx=8, pady=8)

combobox.selectitem(comboentries[0])

def disable(event):
    if theEnabler['text']=='Enabled':
        combobox.disable()
        theEnabler.configure(text='Disabled')
    else:
        combobox.enable()
        theEnabler.configure(text='Enabled')
theEnabler.bind('<Button-1>',disable)
root.mainloop()

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From embed at NOSPAM.geocities.com  Wed Jun 28 09:49:49 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 09:49:49 -0400
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com> <4f265.9121$227.218658@nnrp1.uunet.ca> <395940E9.4CB2FC2C@boeing.com>
Message-ID: <xpn65.10994$227.229599@nnrp1.uunet.ca>

> That sounds interesting. Could you expand on it a little more? Is the
> publish/subscribe done in the python module or in the C extension? Or
> both?

It's all in the C extension.  I didn't need to wrap it in a python module
(at least not yet), because it was a fairly quick hack designed for my
specific embedded application.

The basic idea is that you have a dictionary (created and maintained in the
C extension), and the methods are:

events.dict()       - returns a list of event names, and the parameters they
take, and the documentation strings for each event
ie:
    {  'event1':  ( '(iis)', 'event1(Foo,Bar,Baz) - something called event1
happened and it has data Foo,Bar,Baz' ),
        'event2':  ( '(O)', 'event1(FooObj) - something called event1
happened and it has data some arbitrary python object FooObj' ),
    }
events.subscribe( 'eventname', myself.callme ) - pass it any callable object
(an instance method for example)
events.unsubscribe( 'eventname', myself.callme ) - undo the subscription

On the C side of it, there is a publish API, which looks like this:

publish( char *name, char *argFormat /*ie: "(iis)", "(O)", etc */, char
*docstring);
unpublish( char *name);
event( char *name, ... ); // var args, like printf, format specified in
publish,


You register and unregister interest from whatever python modules you add
on. In my case, I start whatever python objects are needed by modifying
site.py to have it start up threads, etcetera.  In this case, only C
functions generate and publish and only Python subscribes, but if I was
trying to be more generic, I would certainly implement that either one could
publish or subscribe.

The other trick I have implemented in C is that I generate a thread in C,
and an associated thread state for Python, and an anonymous pipe which I use
as a  processing queue.   The events are handled in that thread context,
sequentially, but asynchronously to whatever thread registered the event
callback.   All events are handled by calling them from this queue.  If you
are interested more, email me and I can send you more code-intensive
samples.

Warren




From mwh21 at cam.ac.uk  Sat Jun 17 20:02:52 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 18 Jun 2000 01:02:52 +0100
Subject: Python 3000
References: <394c2b15.634411296@nntp.interaccess.com> <20000617150333.T26436@xs4all.nl> <395103d2.689896234@nntp.interaccess.com>
Message-ID: <m3d7lfx3df.fsf@atrus.jesus.cam.ac.uk>

olczyk at interaccess.com (Thaddeus L. Olczyk) writes:

> On Sat, 17 Jun 2000 15:03:33 +0200, Thomas Wouters <thomas at xs4all.net>
> wrote:
> 
> >Nope, because Py3K is just a name and an idea. A vision, so to speak, and a
> >place to dump great ideas [for Python] that aren't possible without breaking
> >old code. But, now that Guido & a couple of hardcore insides have moved/are
> >moving to PythonLabs, we might just start seeing concrete ideas ;)
> 
> Yeah, but is there a place they kept a list of those ideas?

No, or at least, not one that us mortals can see (I mean, there might
be one in Guido's head or on his desk, but that doesn't really count
does it?).

Cheers,
M.

-- 
  Imagine if every Thursday your shoes exploded if you tied them 
  the usual way.  This happens to us all the time with computers, 
  and nobody thinks of complaining.                     -- Jeff Raskin


From nickgeo at merle.acns.nwu.edu  Wed Jun 14 18:47:15 2000
From: nickgeo at merle.acns.nwu.edu (Nicholas Geovanis)
Date: Wed, 14 Jun 2000 17:47:15 -0500
Subject: Rude question about BeOpen.com
Message-ID: <Pine.HPX.4.10.10006141744570.3588-100000@merle.acns.nwu.edu>

How will the move of the Python crew to a commercial venture affect the
Computer Programming For Everybody (CP4E) project?

+-------------------->
| Nick Geovanis            Optimists tend to be promoted, so the
| IT Computing Svcs          higher up in the organization you are,
| Northwestern Univ          the more optimistic you tend to be.
V n-geovanis at nwu.edu         - "Davis's Law" (Tom Davis, SGI)



From gmcm at hypernet.com  Wed Jun 28 18:51:18 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 28 Jun 2000 22:51:18 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>
Message-ID: <8F61BBD91gmcmhypernetcom@199.171.54.194>

Huaiyu Zhu wrote: 

>Given the following directory structure,
>
>python/
>   AAA.py
>   BBB/
>      __init__.py
>      AAA.py
>      
>if sys.path contains 'python', then 
>
>from AAA import *
>from BBB.AAA import * 
>
>would do the right thing at most locations, except when pwd is python/
>and when '.' is before 'python' in sys.path, in which case they both
>import from python/AAA.py.
>Is this a fair description of current situation?

There's a problem (actually a couple) but this isn't the right description. 
If you're in python/BBB then "import AAA" and "import BBB.AAA" both get 
BBB.AAA.

>...If so this makes
>package name spaces essentially flat. That is, in order to avoid name
>clashes one has to avoid identical module names even if they are buried
>deep inside packages.  

Yes, (except I wouldn't call that "flat" - I would interpret "flat" as no 
subpackages).

>This is even worse on Windows - its file names
>are caseless - as one has to avoid filenames with different cases as
>well. 

Too true...

>This situation occurs quite often in practice.  For example, AAA could
>be a module for doing something.  BBB could be a wrapper package for
>doing things somewhat differently.  So within BBB/AAA.py one calls
>'import AAA' to get the real job done, and oops, it imports itself!

Besides avoiding using the same name, you could make the top level a 
package, then python.BBB.AAA could import python.AAA and it would be 
completely unambiguous. Which is a good reason to avoid relative imports - 
they are by nature ambiguous.

>Although it is not advisable to put '.' in path, especially at front, a
>lot of people actually do. And the functionality of a package should not
>depend on this detail.  Otherwise what happens if one imports several
>packages each expecting a different ordering of sys.path?

Actually it is very useful, since it allows you to have a script have some 
private support modules that don't clutter up sys.path for other scripts.

>So now to my proposed solution.  Is it possible to have something like
>this? 
>
>sys.rootpath  = ['/usr/lib/python1.5',
>'/usr/lib/python1.5/site-packages'] from rootpath.AAA import *
>from rootpath.BBB.AAA.import *

You could use Greg Stein's imputils, but in this case just packagizing the 
whole thing is a more straightforward solution.

- Gordon


From aahz at netcom.com  Wed Jun 21 09:54:43 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jun 2000 13:54:43 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <8iqhf3$4pl$1@slb2.atl.mindspring.net>

In article <ShZ35.11603$Uw3.753134 at bgtnsc04-news.ops.worldnet.att.net>,
William Dandreta <wjdandreta at worldnet.att.net> wrote:
>
>What I am doing is reading a comma delimited text file like this
>
>line = AB12345,20000621,3.56,...
>
>fields = splitfields(line,',')
>
>recognizing that
>
>fields[Date][:5]  is the year
>
>is more obvious and less cryptic than
>
>fields[1][:5]
>
>Trying to put this into a dictionary seems a contrivance to me that doesn't
>make things clearer and only adds overhead.

Sure, but at this point you're not using an enum in the narrow sense;
you're using an enum as a substitute for getting what you want.  What
you should really be doing in this case, probably, is writing a function
that copies the fields into a dict.  Then you can use fields['Date'][:5]
and the rest of your code will work even if the data layout changes.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From moshez at math.huji.ac.il  Sat Jun  3 01:25:57 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 3 Jun 2000 08:25:57 +0300 (IDT)
Subject: Is it just Syntactic Sugar ?
In-Reply-To: <3935B383.469032AE@python.org>
Message-ID: <Pine.GSO.4.10.10006030823290.14749-100000@sundial>

On Wed, 31 May 2000, Guido van Rossum wrote:

> User classes can do whatever their author likes.
> 
> A base class can be provided that defines __add_ab__, __sub_ab__ etc.
> in terms of self.__add__, self.__sub__, etc.

This is the only problem I have with this suggestion: I think the
implementation of the ADD_AB opcode should use the __add__ interface if 
__add_ab__ isn't available.

IOW,

x += y is equivalent to x = x.__add_ab__(y) if x has __add_ab__, otherwise
to x = x+y

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From Cedric.Lefeuvre at tcomhp43.epfl.ch  Fri Jun 23 06:34:10 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Fri, 23 Jun 2000 12:34:10 +0200
Subject: python & excel and 1 remark
References: <39533572.AB90FD1F@tcomhp43.epfl.ch> <086ff63a.3e6f872e@usw-ex0105-035.remarq.com>
Message-ID: <39533D22.B6C50A18@tcomhp43.epfl.ch>

Richard Chamberlain wrote:

> Hi Cedric,
>
> It's possible to generate an excel file using win32com
> extensions on a windows box. Get the extensions and a bit of
> documentation at:
>
> http://starship.python.net/crew/mhammond/
>
> You might want to pick up a copy of Python Programming on Win 32
> By Mark Hammond, Andy Robinson which has a chapter on this very
> subject.
>
> Richard
>

Ok,Richard, but without using python on windows system????



--
C?dric Lefeuvre
SUPAERO 2A-Stagiaire EPFL
0041 (0)21 693 26 16
Fax: 0041 (0)21 693 26 83
 _ _ _/_  _
( (-(// /(


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000623/d8648059/attachment.html>

From MarkH at ActiveState.com  Tue Jun 27 20:06:40 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 28 Jun 2000 00:06:40 GMT
Subject: Controlling Acrobat Reader with COM
References: <200006262159.AA3991339306@mail.mamey.com><s7khls0in7rrfbqnqdo4cha6chh9qqhstr@4ax.com> <Q38VarAdKTW5Ewcx@jessikat.demon.co.uk>
Message-ID: <kkb65.30$Tb7.278@news-server.bigpond.net.au>

"Robin Becker" <robin at jessikat.demon.co.uk> wrote in message
news:Q38VarAdKTW5Ewcx at jessikat.demon.co.uk...
> ...

> I tried already with correctly escaped filenames and always got the same
> error ie 'c:\\Python\\Installs\\test.pdf' got the same error as
> 'gnarg.pdf' (and I don't have any docs called gnarg.pdf :)).

Its likely that Acrobat needs a full-blown control container.  You may
have to run it like the OCX demos in Pythonwin are run - check out
pywin\demos\ocx.

Mark.




From kazan at kazan.  Mon Jun 12 09:28:40 2000
From: kazan at kazan. (kazan at kazan.)
Date: Mon, 12 Jun 2000 13:28:40 GMT
Subject: problems with cgi script
Message-ID: <lfq1g8.p6.ln@kazan>

Hello

I have written a small script that record uptimes on a server.
the classes used works just fine when I use them in scripts I run from the
command line
but I get this error when I run them as a cgi script

in netscape:
500 Server Error
The server encountered an internal error and could not complete your request.

and this output in the error log file:
[12/Jun/2000:20:35:53 GMT] cgi_header: unable to find LFLF

i use the boa webserver
the error occurs when I call a member funcion in class rekorder

like

rek = rekorder()
rek.lesFraFil()

if i just instantiates the object the rest of the script run just fine.

complete cgi script:

#!/usr/bin/python

import os

from oppe import rekorder

rek = rekorder()

rek.lesFraFil()         # remove this and the script works

hostname = os.popen( "hostname" ).read()

print "Content-type: text/html\n\n"
print "<HTML>\n"
print "<TABLE BGCOLOR=ffcc00>"
print "<TR>"
print "<TD>"
print "<H1>"
print "Oppetid rekorder for"
print hostname
print "</H1>"
print "</TD>"
print "</TR>"
print "</TABLE>" 
print "</HTML>"


# the oppe module file

import os,string,re,pickle

# definisjon av klasser

class Tid:
   def __init__( self, dag, time, minutt ):
      self.dag = dag
      self.time = time
      self.minutt = minutt 

   def samanlikn( self, other ):
      if ( self.dag == other.dag and self.time == other.time and
	   self.minutt == other.minutt ):
	 return 0	# like

      elif self.dag < other.dag:
	 return -1	# den som kaller er mindre

      elif self.dag == other.dag and self.time < other.time: 
	 return -1	# ditto

      elif ( self.dag == other.dag and self.time == other.time and
	     self.minutt < other.minutt ):
	 return -1	# ditto

      else:
	 return 1	# den som kaller er stoerre


class rekorder:

   def __init__( self ):
      self.oppeTider = []	# for aa lagre dei 3 oppetidene

   def finnOppeTid( self ):	# returner objekt av type Tid med oppetida naa

      oppeTidStreng = os.popen( "uptime" ).readline()

      # regular expressions for aa finne ulike typer oppetid
      # f.eks med dager, timer, minutt
      # timer + minutt eller berre minutt

      timerMinutt = re.compile( r"""up\s*\d:\d""", re.IGNORECASE | re.DOTALL |
	 re.VERBOSE ) 

      dagarTimerMinutt = re.compile( r"""\d.days""", re.IGNORECASE | re.DOTALL 
	 | re.VERBOSE )

      minutt = re.compile( r"""\d.min""", re.IGNORECASE | re.DOTALL |    
	 re.VERBOSE )

      # manipuler resultatet fra uptime kommandoen ettersom
      # det er dagar, timer, minutt eller timer, minutt eller minutt
      # og returner nytt Tid objekt med oppetida naa

      if timerMinutt.search( oppeTidStreng ):
	 timerMinutt = string.replace( oppeTidStreng, ',',"" )
	 timerMinutt = string.split( timerMinutt )
	 timerMinutt = timerMinutt[ 2 ]
	 timerMinutt = string.split( timerMinutt, ':' )

	 dagar = 0
	 timer = string.atoi( timerMinutt[ 0 ] )
	 minutt = string.atoi( timerMinutt[ 1 ] ) 

	 oppeTidNaa = Tid( dagar, timer, minutt )
	 return oppeTidNaa 

      elif minutt.search( oppeTidStreng ):
	 oppeTidStreng = string.split( oppeTidStreng )
	 dagar = 0
	 timer = 0
	 minutt = string.atoi( oppeTidStreng[ 2 ] )

	 oppeTidNaa = Tid( dagar, timer, minutt )
	 return oppeTidNaa


      elif dagarTimerMinutt.search( oppeTidStreng ):
	 oppeTidStreng = string.split( oppeTidStreng )

	 timerMinutt = oppeTidStreng[ 4 ]
	 timerMinutt = string.replace( timerMinutt, ',',"" )
	 timerMinutt = string.split( timerMinutt, ':' )

	 dagar = string.atoi( oppeTidStreng[ 2 ] )
	 timer = string.atoi( timerMinutt[ 0 ] )
	 minutt = string.atoi( timerMinutt[ 1 ] )

	 oppeTidNaa = Tid( dagar,timer, minutt )
	 return oppeTidNaa

   def skrivTilFil(self):

      utfil = open( "./usr/local/etc/oppetider",'w' )

      p = pickle.Pickler( utfil )

      p.dump( self.oppeTider[ 0 ] )
      p.dump( self.oppeTider[ 1 ] )
      p.dump( self.oppeTider[ 2 ] )


   def lesFraFil(self): 

      #innfil = open( "./usr/local/etc/oppetider",'r' )
      innfil = open( "./oppetider",'r' )

      p = pickle.Unpickler( innfil )

      self.oppeTider.append( p.load() )
      self.oppeTider.append( p.load() )
      self.oppeTider.append( p.load() )

   def sjekkOmRekord(self, nyTid ):  # sjekk om ny oppetid er ny rekord
      if ( nyTid.samanlikn( self.oppeTider[ 0 ] ) == 1 ):
	 self.oppeTider[ 0 ] = nyTid

      elif ( nyTid.samanlikn( self.oppeTider[ 0 ]) == -1 and 
	    nyTid.samanlikn( self.oppeTider[ 1 ] ) == 1 ):
	 self.oppeTider[ 1 ] = nyTid

      elif ( nyTid.samanlikn( self.oppeTider[ 1 ]) == -1 and 
	 nyTid.samanlikn( self.oppeTider[ 2 ]) == 1 ):
	 self.oppeTider[ 2 ] = nyTid


thanks in advance


	Erling R Elvsrud


From moshez at math.huji.ac.il  Thu Jun  8 04:12:28 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 8 Jun 2000 11:12:28 +0300 (IDT)
Subject: I need a lesson.
In-Reply-To: <8hmn80$4vu$1@ionews.ionet.net>
Message-ID: <Pine.GSO.4.10.10006081111340.3146-100000@sundial>

On Wed, 7 Jun 2000, F. G. Brown wrote:

> a = input("What is the distance you traveled in miles?  ")
> b = input("How many seconds did it require?  ")
> c = a*3600.00/b
> print "You were going %d mph!" %c
> if c > 75:
>  print "Hey you better slow down!"
> d = raw_input ("Would you like to do another? (y)es or (n)o")
> if d == 'y':
>  print 'Here we go!'
> else:
>  print 'See ya!'

more = 1
while more:
	....input.....calculations...etc
	d = raw_input("Would you like another? (y) (n)")
	more = (d=='y')
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From moshez at math.huji.ac.il  Wed Jun  7 01:52:32 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 7 Jun 2000 08:52:32 +0300 (IDT)
Subject: __getattr__ and __setattr__ troubles
In-Reply-To: <oq1z2ad4wi.fsf@titan.progiciels-bpi.ca>
Message-ID: <Pine.GSO.4.10.10006070851230.7757-100000@sundial>

On 6 Jun 2000, [ISO-8859-1] Fran?ois Pinard wrote:

> class See:
> 
>     def __getattr__(self, key):
>         print '**', key
>         return None

Ummm... __getattr__ should do something *based on the key*....

> * So, as soon as I use `__getattr__', I suddenly get a lot of other things to
>   define as well?  This would make `__getattr__' less attractive.

This is a good thing. It makes using __getattr__ less common, which makes
programs more readable.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From tim_one at email.msn.com  Sun Jun 11 14:15:18 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 11 Jun 2000 14:15:18 -0400
Subject: spaces in filenames
In-Reply-To: <LrM05.166678$Qp2.427323@afrodite.telenet-ops.be>
Message-ID: <000001bfd3d1$001a1c40$27a2143f@tim>

[ben . de . rydt at pandora . be]
> Try opening a "some bitmap with a space.bmp" in Paint, by
> double-clicking (Win95/98).

OK ... and it worked fine (Win95a -- the first, albeit up-to-date with
service packs), whether double-clicking from within Paint -> Open, or
double-clicking from within Explorer, or even typing the name (spaces and
all) by hand from within Paint -> Open.  It's only from a DOS box that I
need to slap quotes around the name.

> And try explaining people why this doesn't work, when you're
> teaching an introductory course to Win95/98.

Well, c.l.py is full of posts insisting that things that work are broken --
just reword one of those <wink>.





From ychoi at pipeline.com  Sun Jun  4 23:00:44 2000
From: ychoi at pipeline.com (Yoon S. Choi)
Date: Mon, 05 Jun 2000 03:00:44 GMT
Subject: instantiate multiple automation objects
References: <3939d663.1392772@nntp.pipeline.com> <6ru_4.14017$Rq.1311147@pouncer.easynews.com>
Message-ID: <393b1787.3856102@nntp.pipeline.com>

Thanks Roger.  That did the trick.  You saved me hours of time and
hair-pulling.



On Sun, 04 Jun 2000 15:18:26 GMT, "Roger Upole" <rupole at compaq.net>
wrote:

>win32com.client.Dispatch always tries to connect to a running
>instance first, so if you've already done a Dispatch for that
>object, any subsequent call will return the same instance.
>However, you could use DispatchEx to get a second
>instance, as it apparently doesn't try to connect to a running
>instance first.
>           HTH
>                  Roger Upole
>
>
>"Yoon S. Choi" <ychoi at pipeline.com> wrote in message
>news:3939d663.1392772 at nntp.pipeline.com...
>> I am trying (without much luck) to create multiple instances of an
>> automation object.  I can create 1 instance correctly using the
>> function call win32com.client.Dispatch().  However, when I use the
>> same function to try and create another automation object, it just
>> refers to the first instance.
>>
>> Here is an example using Microsoft Word as the automation object:
>>
>> import win32com.client
>> w1 = win32com.client.Dispatch( "Word.Application")
>> w2 = win32com.client.Dispatch( "Word.Application")
>> w1.Visible = 1  # Microsoft Word becomes visible
>> w2.Visible = 1  # Nothing happens because w2 is referring to the same
>> object as w1
>>
>> How can I for example create another instance of Microsoft Word?  In
>> other words, w1 and w2 would refer to two separate automation objects
>> of the same type.
>> Can I do this without having to resort directly to the pythoncom
>> module and its associated CoCreateInstance and CoGetClassObject
>> functions?
>
>



From eviltofu at rocketmail.com  Wed Jun 28 21:01:56 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Thu, 29 Jun 2000 01:01:56 GMT
Subject: Module Name Problem
References: <eviltofu-C678D0.18293328062000@news.earthlink.net> <8F61CAE59gmcmhypernetcom@199.171.54.194>
Message-ID: <eviltofu-E613EC.21014728062000@news.earthlink.net>

In article <8F61CAE59gmcmhypernetcom at 199.171.54.194>, gmcm at hypernet.com 
(Gordon McMillan) wrote:

> Jerome Chan wrote:
> 
> >If I have a module called
> >
> >maxvu.os
> >
> >and I use
> >
> >import os
> >
> >Somehow, when I try to use the os calls or maxvu.os calls it gets 
> >confused. Is there a workaround besides renaming my module?
> 
> 
> Not if you want access to os within maxvu.
> 
> - Gordon

Will this be fixed in 1.6?


From jkraska1 at san.rr.com  Thu Jun 15 21:16:14 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 16 Jun 2000 01:16:14 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu>
Message-ID: <39498084.972E36D2@san.rr.com>

> > Python may be too slow for some tasks, but in the area
> > of programmer productivity, very little beats it, and
> > arguably the environments that do (e.g., lisp) have too
> > many other negatives associated with them.
> 
> Care to explain?

Very high learning curve. Syntactically heavy weight. High
current cost of maintenance. Continually deteriorating
support and availability of technical expertise. Expensive
license fees from Franz. Paucity of recent commercial publication,
indicating failure of advancement of the state of the art
vice recent trends.

> I think Common Lisp already has some good solutions to many issues
> confronting Python now:

Yeah. The Gods Of Python ought to take a look at the first
three of these, but if they do number four, they're all first
up against the wall when I become Supreme World Leader. :)
 
> - Optional type declaration;
> - Optimizing compiler (taking advantage of the type declaration);
> - Powerful macro (superseding the need for preprocessor);
> - Pseudo case insensitivity...

This actually irritates the shit out of me. One day I'm going
to have to get even with my coder friends who treat case in
their lisp files AsIfITDoesNTMatTEr.

> Even the syntax (parentheses) became an asset after I learnt to use
> Emacs effectively.

Arguably, it's a very bad sign when a language is so complex
that it requires an editor tuned to it in particular in order
to be effective in the language.

For example, try compiling any decent sized lisp files in ACL
without doing it from inside an emacs buffer, and watch the
nightmare begin. The interpreter will happily have aneurisms
without so much as indicating the line it was having trouble
with. Perhaps the lisp "gurus" I am learning from aren't so
guruish, but they insist that there isn't any straightforward
way to request the line number of a load/compile failure from
the interpreter.


C/


From jkraska1 at san.rr.com  Sun Jun  4 17:26:07 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 21:26:07 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl>
Message-ID: <393ACA13.FD2416F1@san.rr.com>

Thomas Wouters wrote:
> 
> On Sun, Jun 04, 2000 at 08:19:10PM +0000, Courageous wrote:
> 
> > 1. For slice operations, ilow/ihigh indices which undeflow
> > or overflow the list in question are reset to the edges
> > of the list. Why was it that PyExc_IndexError wasn't
> > thrown instead?
> 
> I believe this was done to avoid having to use a lot of len() and 'or''s
> inside indexes. If you want the first ten characters, or less, of a string,
> currently you do:

Ahhhh. That makes sense. Well that throws all my "better"
exception raising code out the window, now doesn't it? :)-


C/

> Actually, it's very intuitive, if you see the start and end index of the
> slice not as items, but as the gaps between the items.

Well, yeah, but [0:3] as "element zero through three, inclusive"
is about as intuitive as you can get. While I understand completely
how the current implementation works (reading the source code and
all, dontcha know), I still fail to see *why* this was chosen.
I'm presuming there must be some conditions where this particular
idiom was required or caused for nasty syntactic transformations
if it wasn't used?


C/


From eviltofu at rocketmail.com  Sat Jun  3 12:55:00 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 03 Jun 2000 16:55:00 GMT
Subject: Erm... MacOS X?
References: <eviltofu-1AEA18.02454303062000@news.earthlink.net> <usenet-6DC18E.10333803062000@news.visi.com>
Message-ID: <eviltofu-660E0A.12550003062000@news.earthlink.net>

In article <usenet-6DC18E.10333803062000 at news.visi.com>, James Felix 
Black <usenet at inter-slice.com> wrote:

> In article <eviltofu-1AEA18.02454303062000 at news.earthlink.net>, Jerome 
> Chan <eviltofu at rocketmail.com> wrote:
> 
> > Has anyone compiled Python for MacOS X?
> 
> Yes.  It builds right out of the box.  The only trap is if you build on 
> an HFS+ as opposed to UFS volume; then, you have to be careful about 
> case (HFS+ is case-insensitive but preserving.)
> 
> (jfb)

The problem with UFS is that the MacOS portion of MacOS X has problems 
with it as of DP 4. Someone should patch the configuration files and 
send in an update. I'm too new to autoconf to figure this out.


From lohgan at my-deja.com  Mon Jun 19 04:54:00 2000
From: lohgan at my-deja.com (lohgan at my-deja.com)
Date: Mon, 19 Jun 2000 08:54:00 GMT
Subject: win32gui DialogBox
Message-ID: <8ikn36$716$1@nnrp1.deja.com>

Hi all.  I'm attempting to open a dialog box on a click event from a
task bar icon.  I do not have much experience with Win32 programming
and am experiencing unexpected results.

I load the dialog template from a resource only DLL.  Then when I try
to show the dialog box, the function "DialogBox" blocks after handling
the message "WM_INITDIALOG".  The main and child (dialog) windows
freeze up as if the message loop is broken.  When I double click on the
task bar icon and the WM_QUIT message gets posted, all the blocked
DialogBox functions release control.

I hope this is something stupid like simply needing another message
loop somewhere.

Any help is appreciated.  Please CC me on responses.  Mail me
personally if you wish a copy of 'Resources.dll' and other files.

Thanks,
-Sean

---> application.py
from win32api import *
from win32gui import *
import win32con

class MainWindow:
    def __init__(self):
        self.message_map = {
            win32con.WM_DESTROY: self.OnDestroy,
            win32con.WM_USER+20 : self.OnTaskbarNotify,
        }

        # Register the Window class.
        wc = WNDCLASS()
        hinst = wc.hInstance = GetModuleHandle(None)
        wc.lpszClassName = "PythonTaskbarDemo"
        wc.style = win32con.CS_VREDRAW | win32con.CS_HREDRAW;
        wc.hCursor = LoadCursor( 0, win32con.IDC_ARROW )
        wc.hbrBackground = win32con.COLOR_WINDOW
        wc.lpfnWndProc = self.message_map # could also specify a
wndproc.
        classAtom = RegisterClass(wc)

        # Create the Window.
        style = win32con.WS_OVERLAPPED | win32con.WS_SYSMENU
        self.hwnd = CreateWindow( classAtom, "Taskbar Demo", style, \
                 0, 0, win32con.CW_USEDEFAULT, win32con.CW_USEDEFAULT, \
                 0, 0, hinst, None)
        UpdateWindow(self.hwnd)
        ShowWindow(self.hwnd, win32con.SW_SHOW)

        # Add the taskbar icon
        hicon = LoadIcon(0, win32con.IDI_APPLICATION)
        flags = NIF_ICON | NIF_MESSAGE | NIF_TIP
        nid = (self.hwnd, 0, flags, win32con.WM_USER+20, hicon, "Python
Demo")
        Shell_NotifyIcon(NIM_ADD, nid)

    def OnDestroy(self, hwnd, msg, wparam, lparam):
        nid = (self.hwnd, 0)
        Shell_NotifyIcon(NIM_DELETE, nid)
        PostQuitMessage(0) # Terminate the app.

    def OnTaskbarNotify(self, hwnd, msg, wparam, lparam):
        if lparam==win32con.WM_LBUTTONUP:
            print "You clicked me."
            try:
                import ui_preferences
                prefs = ui_preferences.UIPreferences(self.hwnd)
            except:
                import sys, traceback
                traceback.print_exc(file=sys.stdout)
            print "Should be loaded?"
        elif lparam==win32con.WM_LBUTTONDBLCLK:
            print "You double-clicked me - goodbye"
            PostQuitMessage(0)
        return 1

---> ui_preferences.py
from win32api import *
from win32gui import *
import win32con

RESOURCE_FILE = 'Resources.dll'
ID_DIALOG = 106
ID_ICON = 101

ID_USERNAME = 1002
ID_PASSWORD = 1003
ID_CLOSE = 1000

class UIPreferences:
    def __init__(self, parentHWND):
        message_map = {
            win32con.WM_INITDIALOG: self.InitDialog,
            win32con.WM_DESTROY: self.OnDestroy,
        }

        self.parentHWND = parentHWND
        self.res = LoadLibrary(RESOURCE_FILE)
        self.id_dlg = DialogBox(self.res, ID_DIALOG, self.parentHWND,
message_map)
        print "ID:", self.id_dlg
        print "LastError", GetLastError()

    def InitDialog(self, hwnd, msg, wparam, lparam):
        print "InitDialog"
        return 1

    def OnDestroy(self, hwnd, msg, wparam, lparam):
        return 1




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


From gmcm at hypernet.com  Wed Jun 14 09:35:08 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Jun 2000 13:35:08 GMT
Subject: Killing wxPython process in PythonWin
References: <8i7178$44c$1@nnrp1.deja.com>
Message-ID: <8F536925Fgmcmhypernetcom@199.171.54.194>

Mark Morrison wrote:

>When prototyping wxPython apps using PythonWin a crash sometimes leaves
>the wxApp class lying around (with main loop running). Next time I try
>to run the script, wxPython complains that only one wxApp can be
>running. Is there a simple way to kill the old running script without
>exiting PythonWin and killing the PythonWin process from the
>taskmanager?


I hate to tell you this, but except for very, very simple scripts, I never 
run anything from within Pythonwin or IDLE. If I do run from inside the 
IDE, (say, to use the debugger), I generally expect I'll have to exit and 
restart between runs. I've chased too many "bugs" (that were really fixed, 
but something wasn't reloaded) to trust anything but a fresh start. And I'm 
certainly not going to bother with the bizarre hacks people come up with 
(eg, to replace the "code" but not the data in "running" instances).

The cost to me is a ^G nnn (or ALT-G nnn in IDLE) to goto the line from the 
traceback. The value is complete certainty.

- Gordon



From rhaper at houston.rr.com  Fri Jun 16 23:55:52 2000
From: rhaper at houston.rr.com (Rod Haper)
Date: Sat, 17 Jun 2000 03:55:52 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <394AD545.AC791A2B@san.rr.com>
Message-ID: <394AF6C8.FFCBF182@houston.rr.com>

Courageous wrote:
> 
> Well, speaking of tabs, in vim (.vimrc), I have:
> 
> set tabstop=4
> set expandtab
> 
> This sets up vim to "on tab, take the cursor to the
> next appropriate tabstop, but to insert spaces instead
> of tabs when doing so". This is always just what I
> want.

I used to set the tabstop option to 4 (my preference also) in vim but it could really be confusing when I'd edit a file that someone else had written using a mixture of tab characters and spaces to format their indentation levels.  Then I found a much better solution by using the softtabstop option.  Now I use

set expandtab
set softtabstop=4
set shiftwidth=4

This leaves the tabstop option set at it's default 8 characters, keeps the indentation spacing consistent (and sane) for mixtures of tabs and spaces, and let's me use my preferred indentation spacing of 4 spaces.

Rod


From vigenereNOviSPAM at leland.stanford.edu.invalid  Fri Jun  2 10:09:50 2000
From: vigenereNOviSPAM at leland.stanford.edu.invalid (cheesecake)
Date: Fri, 02 Jun 2000 07:09:50 -0700
Subject: file.seek misbehaving?
Message-ID: <203da3d8.c492c8fd@usw-ex0103-019.remarq.com>

I've run into a problem that is perhaps not a Python-specific
bug. I was processing some log files using Python, when I
noticed some mysterious bugs in certain files. I tracked down
the problem to the seek method misbehaving.

Some of the files I worked with were files created in NT or
saved in NT using a text editor. The problem files were copied
over ftp from a unix machine to the NT machine on which I work.
None of the files that were saved in NT (ie, NT native files)
had any problems.

It seems that seek doesn't work correctly because it's not clear
how large the file really is. I copied all the contents of one
bad file into a new file, and the new file was over 5KB bigger
according to NT. The new file works just fine with the seek
operations. But when I read() both files into memory, they're
identical strings and have the content-length of the original
file.

Could the problem be that the differences between the NT file
system and the unix file system of the other computer are not
properly handled because I'm running on NT, so the Python
interpreter figures it should assume everything is NT?

Let me summarize the problem with some code:
fd1 = open("somefile.txt", "r")  # somefile.txt is a unix file
fd2 = open("someother.txt", "r") # someother.txt is NT
lines1 = []
lines2 = []
lines1[0] = fd1.readline()
lines2[0] = fd2.readline() # lines1[0] == lines2[0]
pos1 = fd1.tell()
pos2 = fd2.tell() # at this point pos1 == pos2, like it should be
lines1[1] = fd1.readline()
lines2[1] = fd2.readline() # lines1[1] == lines2[1]
fd1.seek(pos1)
fd2.seek(pos2) # pos1 == pos2 (still)
lines1[2] = fd1.readline()
lines2[2] = fd2.readline() # lines1[2] != lines2[2] !!!!?

What gives here? That's like having file.seek(file.tell())
moving you somewhere else.

Anyway, I'd like to find out if anyone else has had such
problems or knows what the solution is. I don't read this ng, so
please send replies to my email address at townsend at beehive.de.

TD

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



From dgoodger at bigfoot.com  Thu Jun 29 23:01:44 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Thu, 29 Jun 2000 23:01:44 -0400
Subject: How to concatenate external files ..?
In-Reply-To: <slm1odapjev11@corp.supernews.com>
References: <slm1odapjev11@corp.supernews.com>
Message-ID: <B58185D7.5994%dgoodger@bigfoot.com>

on 2000-06-29 04:20, Randolph MacKenzie (rmack at eznet.net) wrote:
> I want to combine all the little *.htm files in a directory into one big
> file.
...
> Now what  .. how do I iterate over the list and specify that each file is to
> be read and appended to "outfile" ?

If/when you get to the point where you want to do something to the file
contents, other than concatenate them together, take a look at the
"fileinput" module (docs in Python Library Reference).

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From sean at digitome.com  Mon Jun  5 12:05:32 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Mon, 05 Jun 2000 16:05:32 GMT
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee>
Message-ID: <393bc9f8.3018710@news.iol.ie>

On Fri, 02 Jun 2000 18:01:12 +0300, Kaimar Karu <kaimar at ut.ee> wrote:

>Hi,
>
>Where can I find some kind of helping material (tutorials etc) about how
>to use xml in python?
>I need some examples, just pure specification is not enough.
>
>tia,
>--
If you can wait a couple of weeks, my book "XML Processing with
Python" should be on the shelves, published by Prentice Hall.
ISBN 0130211192.

regards,



From darrell at dorb.com  Wed Jun 28 21:43:42 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 28 Jun 2000 18:43:42 -0700
Subject: Following a file, or cloning tail
References: <8jdj3t$or4$1@news1.xs4all.nl>
Message-ID: <0b4201bfe16b$78339560$6401a8c0@home>

If your on Windows and can wait for the file to complete writing, check out
http://www.dorb.com/darrell/win32WorkSvr/

The best solution might be to use popen if you can.

Trail might be a problem if the other application doesn't flush() the file
once in a while.
If it does you should be able to just call read() on an interval or use
select to avoid polling.
Select won't help you on Windows though, but Win32 calls will.

--Darrell Gallion


----- Original Message -----
From: "Boudewijn Rempt" <boud at rempt.xs4all.nl>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, June 28, 2000 12:19 PM
Subject: Following a file, or cloning tail


> I'm trying to write an application that follows the output of another
> app, analyzes it, and serves it up as html. Kind of filtering tail,
> really. Is there any accepted 'best' way of 'tailing' a file?
>
> I thought about having two threads, one doing the tailing, one
> doing the servering (instantiating new threads as needed), but I
> can't even get the tailing right!
>
> --
>
> Boudewijn Rempt  | http://www.valdyas.org
> --
> http://www.python.org/mailman/listinfo/python-list




From mwh21 at cam.ac.uk  Wed Jun 21 09:40:27 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 21 Jun 2000 14:40:27 +0100
Subject: 123.4+2=125.40000000000001
References: <3950A30D.DEE4CF4D@brb.midat.de>
Message-ID: <m31z1r5f04.fsf@atrus.jesus.cam.ac.uk>

Peter Koppatz <pkop at brb.midat.de> writes:

> >>> 123.4+2
> 125.40000000000001   # is this OK?
> >>> print 123.4+2
> 125.4
> >>>
> 
> On my calculator: 123.4+2 = 125.4 !?
> I am using Python 1.6a2

FWIW, on my Python:

>>> 123.4
123.40000000000001
>>> 0.4
0.40000000000000002

This isn't surprising, really, given that floats are represented in
base 2; floats are "really" rationals of the form M*(2**N) where M and
N are (bounded) integers, and 0.4 is 0.011001100110011001100110... in
binary.

Cheers,
M.

-- 
  There are 'infinite' number of developed artifacts and one cannot
  develop appreciation  for them all.  It would be all right to not
  understand something, but it would be imbecilic to put judgements
  on things one don't understand.               -- Xah, comp.lang.lisp


From jparkey at nospamrapidbs.com  Tue Jun  6 02:04:16 2000
From: jparkey at nospamrapidbs.com (John Parkey)
Date: Tue, 6 Jun 2000 07:04:16 +0100
Subject: Memory leak: assigning function object to instance variable
References: <8hh0pf$i7$1@phys-ma.sol.co.uk> <slrn8joise.ik.nascheme@cranky.arctrix.com>
Message-ID: <8hi4e2$afm$1@phys-ma.sol.co.uk>

Thanks for the explanations, and also to Mark Hammond and Tim Peters who
responded directly via the bug list.

In the short term I will get around this by del statements in the class
destructor.

jp


"Neil Schemenauer" <nascheme at enme.ucalgary.ca> wrote in message
news:slrn8joise.ik.nascheme at cranky.arctrix.com...
> John Parkey <jparkey at nospamrapidbs.com> wrote:
> >I have registered this bug, with code fragment below, but wondered if
> >anybody else was falling foul of memory leaks of this type.
>
> The latest version of my GC patch will prevent this leak.
> Hopefully it will be an optional feature in 1.6.
>
>     Neil
>
> --
> The internet: Learn what you know.  Share what you don't.




From embed at NOSPAM.geocities.com  Tue Jun 27 10:12:26 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 27 Jun 2000 10:12:26 -0400
Subject: Multithreading Embedded C/Python : Giving each thread its own sys.stdout 
References: <w6265.9118$227.218565@nnrp1.uunet.ca>
Message-ID: <GC265.9157$227.218433@nnrp1.uunet.ca>

Okay, here's a simpler twist on the previous EMAIL:

How do you do a deep copy of a module? Ie, if I have two namespaces, and I
import sys into both of them, I am getting two refcounts on the SAME object,
and modifying ONE will modify both. I really want a deep copy of a module.

Unfortunately, according to the documents, you can't deep-copy a module.
How about making something that looks like a copy of sys but isn't truly a
module.  Perhaps a class containing the same objects as the module sys would
have contained. Then the main thread can have the module sys, and the rest
of the threads can have a pseudo-sys, one that could tell them what thread
context and namespace they are in, as well as having it's own version of
sys.stdout.

Warren






From shouldbe at message.com  Tue Jun 27 11:35:49 2000
From: shouldbe at message.com (Arinté)
Date: Tue, 27 Jun 2000 10:35:49 -0500
Subject: Threading +  embedding = System Dump.
References: <8j7udf$15au$1@news2atm.raleigh.ibm.com> <8F5FCDB88gmcmhypernetcom@199.171.54.154>
Message-ID: <8jadre$k2c$1@news2atm.raleigh.ibm.com>

> That sounds very much like there is no multithreading.
>
> So why doesn't "diz[0].write(mx)" block? Or does it?
It doesn't need too.  The data is queued in a seperate subsystem.  My app
only cares if the subsystem couldn't receive the data, so the response is
quick no need to block.  This specific \x1b\x47 command tells when the
device actually finished, I don't always want to wait for that .

>
> Is the real problem here that you're using one of the PyRun_ calls when
you
> need to go down a level and use PyImport_ etc. so the C++ has some control
> over what's going on?
Seemed like I can only run a script once.  If I tried to run the same script
two times in a row the second time it did nothing, all the import calls
return successfully both times.  I don't Py_Finalize until the dll is
unloaded.

I really appreciate your help, but I think I will need to figure something
else out.  I don't know what else to say, that could further explain my
problems, short of posting my whole program.  Which is of moderate size (dll
<100K, app <200K), but no comments.  I didn't expect many others to look at
the code solo/self made project.  I know I still need comments =-(.

Why, is it that if use python Event.wait function, the app crashes, as
oppossed to popping up a messagebox and manually closing that box when I see
the message come thru?  Does wait shut down the thread the script is running
on in some sense?  I would try to do a notify from c++ on the python Event,
but it seems I can't get any access to python when it is in the wait state?

I am currently, tossing around the idea of python call a C++ function called
wait, then in the call back in calls the C++ function notify, and at that
point using a Windows Event I will allow the wait function to return.  I'd
much rather use python's wait, so I won't have to worry about recompiles,
but I don't know what to do to get this working.

thanx again.
Hopefully this is the last, I think I am trying to do too much.




From mwh21 at cam.ac.uk  Tue Jun 20 10:29:42 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jun 2000 15:29:42 +0100
Subject: TABs (was: Waffling between Python and Ruby)
References: <XFMail.000620085230.mikael@isy.liu.se>
Message-ID: <m3k8fk5stl.fsf@atrus.jesus.cam.ac.uk>

Mikael Olofsson <mikael at isy.liu.se> writes:

> On 20-Jun-00 Just van Rossum wrote:
>  >  (Due to historic narrowmindedness, a point is the same as a pixel under
>  >  MacOS, no matter what physical resolution the display has; I believe
>  >  Windows has (had?) the same problem, except that a pixel is 72/96th
>  >  points...)
> 
> Erm, isn't a pt as in 10 pt font the typographic unit 1/72 inch?

Well, sort of; the thing is the software doesn't know how big your
monitor is physically, so can't work out what's it's own dpi is.
You'd think a task as simple as "draw something this size" would be
pretty simple after fifty years of computing, but you'd be wrong...

Cheers,
M.
(who's spent longer wrestling with this than he'd have liked to)

-- 
  SCSI is not magic. There are fundamental technical reasons why it
  is necessary to sacrifice a young goat to your SCSI chain now and 
  then.                                                  -- John Woods


From no_spam at umich.edu  Mon Jun 12 12:46:54 2000
From: no_spam at umich.edu (Jeff)
Date: Mon, 12 Jun 2000 09:46:54 -0700
Subject: WIN32API GetLastError()
Message-ID: <8i3460$ppm@news.or.intel.com>

Hey, I see that when you import win32api, the function GetLastError is
exported.  When trying to use this, I can never get the (any) error code
other than 0.  I am suspecting that this means that the GetLastError it is
trying to "get" is an error from the pythonwin process and not from an API
call I make (example writing to a file when only read access was given, or
from an extension).  So my question is, what can GetLastError be used for,
and is it at all possible to get the windows error code from a bad
win32file.CreateFile??

jy




From piet at cs.uu.nl  Wed Jun 21 09:12:51 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 21 Jun 2000 15:12:51 +0200
Subject: 123.4+2=125.40000000000001
References: <3950A30D.DEE4CF4D@brb.midat.de> <Pine.GSO.4.20.0006211310100.17041-100000@luna.dsv.su.se>
Message-ID: <uem5rgoto.fsf@cs.uu.nl>

>>>>> Kalle Svensson <karl-sve at dsv.su.se> (KS) writes:

KS> On Wed, 21 Jun 2000, Peter Koppatz wrote:

>> On my calculator: 123.4+2 = 125.4 !?

KS> In fact, this probably isn't true. It's just that your calculator rounds
KS> the answer before handing it to you, just like print. Maybe. I should
KS> probably change my name to Search Deja... :)

I think calculators use decimal representation internally, so they don't
get this kind of error. Of course this introduces errors in other cases.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From moshez at math.huji.ac.il  Tue Jun  6 11:09:28 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 6 Jun 2000 18:09:28 +0300 (IDT)
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
In-Reply-To: <393D1155.D067A821@decrc.abb.de>
Message-ID: <Pine.GSO.4.10.10006061807370.26389-100000@sundial>

On Tue, 6 Jun 2000, Peter Mueller wrote:

> it looks like that I have the same problem. I use the pickler. When I import
> gtk/pygnome modules, the pickler can't read my pickled object (containing
> floats) anymore.
> This behaviour is new since redhat 6.2. On redhat 6.1 the same code with the
> same pickled object works without any problem. The python versions on both
> redhat 6.1/6.2 are the same. There must be a difference one of the modules.

I have a hunch this is related to locale. See if you can convince your
system that you live in the US (<wink>) -- fiddle environment variables,
etc. IF this solves the problem, please let us know.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From bjorn at roguewave.com  Tue Jun 27 12:35:03 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 27 Jun 2000 10:35:03 -0600
Subject: Passing variables between scripts
References: <8j745m$qqj$1@oslo-nntp.eunet.no>
Message-ID: <3958D7B7.BC3F664@roguewave.com>

Olav Viken wrote:
> 
> Can anyone tell me if there is a good way to pass a variable between two CGI
> scripts?

There are many variations of this, but you can do a lot with the
following model:

  form1 -> cgi1 -> form2 -> cgi2

meaning that your starting html page contains form1. When submitting the
form, you call cgi1 which during it's processing prints form2. When
form2 is submitted it calls cgi2.

If you only have simple values (and don't require a very secure system),
cg1 can print out a <input type="hidden" name="foo" value="%s"> tag
where you substitute the %s for the repr of a dictionary containing all
the values you want to pass.  You can then use eval on the value in cgi2
to resurrect the dictionary (as I said, not particularly safe, but easy
;-)

hth,
-- bjorn



From kens at sightreader.com  Fri Jun 16 01:12:16 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 22:12:16 -0700
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
Message-ID: <3949B730.B166F141@sightreader.com>

Warren Postma wrote:

> I was looking at the Ruby editor/IDE for windows and was disappointed with
> RubyWin, although it is definitely coming along. I noticed the smalltalk
> like immediate mode (Type code, hit ctrl-J, results inserted into the
> buffer) but did not see any debugging features. Are such features
> (debugging, syntax highlight) in the works for this? Is there going to be a
> cross-platform gui other than Tcl/TK for Ruby?  I love Python and I love the
> wxPython effort, but I also like a few things about Ruby better than python.
> They are:
>
> Chief Irritants In Python:
>
> 1.  After using Python seriously for 6 months, I must confess I still
> dislike the "whitespace as syntax".  I like the fact that Ruby uses block
> markers (begin, end) instead of tabs. This means cheifly to me that I don't
> need to worry about how my editor handles tabs and spaces.  Having something
> look correct on the screen, but not work because of misuse of tabs versus
> spaces, or incorrect indentation, is something that still continues to bug
> me.

I'm in the pro-whitespace club.  However the number one most obvious improvement
in python would be to simply make tab characters illegal.  This would neutralize
your complaint, which I have to agree with.

> 2. Python uses ":" to introduce a new scope, something I find extraneous,
> and more irritating than begin/end or Ruby's def and end for functions, for
> example.

A matter of taste.

> 3. Python is much slower than Ruby, in my informal tests.
>
> So here are the things I'm still wondering, and perhaps there is a "Python
> to Ruby FAQ" somewhere to answer it:
>
> 1. What is the Ruby equivalent of Python's ability to import a module and
> view its contents through code:
>
>             import somemodule
>             print dir(somemodule)
>
> 2. What is the Ruby equivalent of eval, and exec?
>
> 3. Is there a decent Language Reference guide similar to Pythons?
>
> 4. Is there a reference guide to writing extensions in C for Ruby similar to
> Pythons?
>
> I am by no means going to abandon Python, as it is more mature, it solves
> many problems, and Python has extremely advanced Windows integration (the
> windows extensions give more COM and DCOM and WinAPI functionality than
> Ruby) and a more thorough set of GUI toolkits, and a better set of database
> toolkits, and I can live with the whitespace-as-syntax as long as I have the
> right kind of editor. Python is truly a joy to work with. However,  I forsee
> room for both Python and Ruby in my repertoire in the future, and I see that
> Ruby will be a success as both a scripting and an application development
> language.
>
> You want a killer Ruby feature?  I suggest static typing and a Ruby to C
> converter that accelerates 2x or more over ruby.
>
> Ruby is a great language. If I had to make it even, by criticizing something
> in Ruby, it would be that I am not sold on some of the Perl idioms ($_
> anyone? Blech). But I like Ruby's syntax, and I don't have to use the perl
> headspace if I don't want to. :-)
>
> Warren

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/65d973c7/attachment.vcf>

From JohnH at PHM.GOV.AU  Wed Jun 28 00:29:03 2000
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Wed, 28 Jun 2000 14:29:03 +1000
Subject: Help with launching acrobat
Message-ID: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EB76@phm-exch.phm.gov.au>

>Hi!

>I am currently faced with a problem -

>My company has a windows product which needs to launch a PDF.

>Previously, we were putting the path to the acrobat executable into the
>autoexec.bat.

>This was fine, until some of our customers decided to run it in NT.

>My question is :-

>How can I query the windows registry for the Acrobat path?

>**OR**

>How can I use com to launch Acrobat reader?

If you can use wxPython it has an ActiveX control container in which you can
run Acrobat reader.
There is a demo called ActiveXWrapper_Acrobat.py which does exactly that.
It could be worth a look.

Jon





From bparsia at email.unc.edu  Thu Jun 15 19:57:56 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 15 Jun 2000 19:57:56 -0400
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <1ec8zim.1jmu7c7qqgdkwN%bparsia@email.unc.edu> <3948C878.E97ED299@uab.edu>
Message-ID: <1ec9y6t.8ui4sq1yhm958N%bparsia@email.unc.edu>

Shae Erisson <shapr at uab.edu> wrote:

> Bijan Parsia wrote:
[snip]
> 
> I'm looking for people to try VirtualPairProgramming (Python) with me.
> The best setup I've come up with so far is:
> SpeakFreely
> Emacs with make-frame-on-display or screen with -x
> 
> How are you doing VirtualPairProgramming?

IRC mostly :( I plan to use VNC too for Virutal Pair Squeaking.

IRC sucks because you have to paste code and/or descriptions constantly.

I've been working on some things that make IRC a little less painful,
because, after all, IRC is rather convenient :)

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From anders.eriksson at morateknikutveckling.se  Thu Jun 15 04:28:51 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 15 Jun 2000 10:28:51 +0200
Subject: Newbie : Add to PYTHONPATH
References: <8i83a7+kraa@eGroups.com>
Message-ID: <8u3hksgtiqk38lbbijrj9fmlcr34hcvt82@4ax.com>

On Wed, 14 Jun 2000 14:02:47 -0000, hohage at muenster.de wrote:

>Hello,
>this is an absolute newbiequestion.
>I'd like to know how to add a path to sys.path, but not temporarily.

This has just been discussed in a  thread called: "Where should i put
my HTMLgen directory?"

// Anders



From "dbroadwell" at spam.mindspring.com  Thu Jun  1 18:42:11 2000
From: "dbroadwell" at spam.mindspring.com (David Broadwell (remove spam.))
Date: Thu, 01 Jun 2000 18:42:11 -0400
Subject: Calling Function in DLL on Windows
References: <dThZ4.101952$fF5.3223396@news1.rdc1.il.home.com><3935F38E.FF0061DC@sightreader.com> <xSu2lLAMsgN5Ewy0@jessikat.demon.co.uk> <VmwZ4.102768$fF5.3267741@news1.rdc1.il.home.com>
Message-ID: <8h6ojo$2lq$1@slb6.atl.mindspring.net>

It is certianlly something that i at least want to see, so that i can
grab the dll for my web can and stream it to my localnet.. as a
project..

dsavitsk wrote:
> 
> > try Sam Rushing's windll/calldll stuff; should be findable at medusa
> >
> > Robin Becker
> 
> thanks for the suggestion
> 
> it looks like grabbing the calldll.pyd and windll.py files from sam rushing
> at www.nightmare.com (both in dynwin.zip) will allow something like...
> 
> >>> from dynwin.windll import *
> >>> # The DLL in qustion here is called
> >>> # SFWLink.DLL. It is a connection to
> >>> # a scanning engine for a NCS OMR Scanner
> >>> mod1 = module('SFWLink') # Load's DLL
> >>> mod1.LoadSFW('c:\\sfw\\', 'Param') # calls f(x)
> 
> This example is mostly from MHammond's Book where it is given about 1/2 of a
> page.  He says to look at Sam's site.  Unfortunatly the documentation is not
> really there for those of us who are new. MHammond also presents some
> information on passing values ByRef.
> 
> As soon as i get to work with the hardware in question again (next week)
> i'll put up a post on doing this, and how it works.

-- 
 --
	Got CODE?
 --


From mwh21 at cam.ac.uk  Thu Jun 22 16:37:32 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 2000 21:37:32 +0100
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu> <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk> <1ecjb1r.27wv4iz2rfkpN%bparsia@email.unc.edu>
Message-ID: <m34s6l4flf.fsf@atrus.jesus.cam.ac.uk>

bparsia at email.unc.edu (Bijan Parsia) writes:

> Michael Hudson <mwh21 at cam.ac.uk> wrote:
[snip]
> > Does Python 1.2 have the local variable optimisation? 
> 
> What do you mean?
> 
> [snip]

Certainly at some point there was no LOAD_FAST opcode; every name
lookup would first look into a local dict, then the global one, then
the builtins.  Now references to locals are just indexes into (C)
vectors; this is much quicker.

I don't know when this happened; it was quite a whiles back (before my
time by a long way), but so was Python 1.2!

Cheers,
M.

-- 
  languages shape the way we think, or don't.
                                        -- Erik Naggum, comp.lang.lisp


From wmcclain at salamander.com  Fri Jun  9 09:38:16 2000
From: wmcclain at salamander.com (Bill McClain)
Date: Fri, 09 Jun 2000 13:38:16 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <sk1sq867h5126@corp.supernews.com>

In <8F4E9C028stevemulozemailcomau at 139.130.250.4>, "stevemul at ozemail.com.au (Steve Mullarkey)" wrote:

> I have read in several places productivity claims of 5 to 10 times over
> 'C' and "C++". I would like to ask for some feedback from "C++"
> programmers who have moved to Python as to whether these estimates are
> realistic.

I've been using C++ for work for several years and use Python at home. I'm
trying to introduce Python at work.

What frustrates me about C++ is the enormous declarative overhead required
to get anything done. Every class requires me to instruct the compiler on
how to copy, assign, compare and reference objects, and how subclasses
are allowed to inherit from the parent. STL works well for data structures,
but the advanced features are very cumbersome.

Much of this burden simply vanishes with Python.

When my coworkers come to me for help, I always warn them: C++ is a 
difficult language to use _well_. The types of issues everyone
eventually encounters can be found at Herb Sutter's "Guru of the
Week": 

    http://www.peerdirect.com/resources/
    
It's not pretty.
    
-Bill


From walter111 at structurex.net  Mon Jun 26 13:47:04 2000
From: walter111 at structurex.net (walter hanagriff)
Date: Mon, 26 Jun 2000 12:47:04 -0500
Subject: trouble with tkinter, with source
Message-ID: <39579717.8DA35EB5@structurex.net>

this is just a framework i am using as a simplified test case for adding
tkinter to my main program
The first class makes the first frame of the program giving the user 3
options, whichever option he chooses opens up a second frame from the
second class, but they just keep piling up, i want it so that if an
options frame is there when they
pick another frame, the first options frame will disappear
i can easily delete the main frame from the Main class, but dont know
how to call the frames from the defs in the second class, whether i try
to call them from Main or Step2, so how do i make what i want happen
i think that best explains what i want, any help would be appreciated

------------------------------------------------------
class Main:
    # this function makes the first frame for the program
    def __init__(self):
        frame = Frame(root, width = 300, height = 400)
        frame.pack()
        Label(frame, text = "this program figures the area, perimeter,
and power of n \n"
        "____________________________________________").pack()

        self.v = IntVar()
        self.v.set("1")
        Radiobutton(frame, text = "1 = Find the power of a number",
variable = self.v, value = 1).pack(anchor = W)
        Radiobutton(frame, text = "2 = Find the area of a shape",
variable = self.v, value = 2).pack(anchor = W)
        Radiobutton(frame, text = "3 = Find the perimeter of a shape",
variable = self.v, value = 3).pack(anchor = W)

        next = Button(frame, text = "Next", padx = 10, command =
self.ifpick)
        next.pack(side = LEFT, padx = 5)
        close = Button(frame, text = "Close", padx = 10, command =
root.destroy)
        close.pack(side = LEFT, padx = 5)

    # this says which frame to open depending on which radiobutton above
is picked by user
    def ifpick(self):
        st = Step2()
        if self.v.get() == 2 or 3:  st.frame2.frame22.destroy
        if self.v.get() == 1:  st.frame2()
        if self.v.get() == 2:  st.frame3()
        if self.v.get() == 3:  st.frame4()

# Destroys frames not needed then displays all but the first frame in
the program
class Step2:
    # this frame is for the power of a number, radiobutton option v == 1

    def frame2(self):
        frame22 = Frame(root, width = 300, height = 410)
        frame22.pack()
        Label(frame22, text = "this frame is where you put information
to get power of n \n"
        "______________________________________________").pack()

    # this frame is for the perimeter of a shape, radiobutton option v
== 3
    def frame4(self):
        frame44 = Frame(root, width = 300, height = 400)
        frame44.pack()
        Label(frame44, text = "this frame is where you put information
to get the perimeter of a shape \n"
        "______________________________________________").pack()

if __name__== '__main__':
    root = Tk()
    root.title(title)
    start = Main()
    root.mainloop()




From stevena at neosynapse.net  Fri Jun 30 15:10:20 2000
From: stevena at neosynapse.net (Steven D. Arnold)
Date: Fri, 30 Jun 2000 15:10:20 -0400
Subject: Python MP3 Player
In-Reply-To: <aB575.2343$iN5.538590@ptah.visi.com>
References: <395CE50A.3BA16777@bioeng.ucsd.edu>
Message-ID: <4.3.1.2.20000630150810.01ff3df0@phear.dementian.com>

At 06:24 PM 6/30/2000 +0000, Grant Edwards wrote:
>In article <395CE50A.3BA16777 at bioeng.ucsd.edu>, Curtis Jensen wrote:
>
> >Does anyone know of a Python MP3 Player?
>
>Surely you're kidding?
>
>Do people really do real-time DSP type stuff in Python.

Why not, provided the most time-sensitive stuff were written in C or 
C++?  Surely most of the user interface, except perhaps visuals (a graphics 
display that moves in time with the music), could be beneficially written 
in Python.


steve


--
Steven D. Arnold           Que quiero sera         stevena at neosynapse.net
"If you have  built castles in the air, your work  need not be lost; that
is where they should be.  Now put the foundations under them!" -- Thoreau




From Dan at Grassi.org  Tue Jun 27 19:34:27 2000
From: Dan at Grassi.org (Dan Grassi)
Date: Tue, 27 Jun 2000 19:34:27 -0400
Subject: Need help: Python and Oracle on Linux x86
Message-ID: <B57EB243.44D%Dan@Grassi.org>

I am trying to use Python on a Linux system but need Oracle DB access.  I
have tried compiling the DCOracle interface with no success, there are
compile errors.

I am trying to replace the Perl that is used with python but the DB access
issue is a killer.  Can anyone help?

Thanks,

Dan



From stephen at cerebralmaelstrom.com  Tue Jun  6 02:29:34 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Tue, 06 Jun 2000 06:29:34 GMT
Subject: Local Databases with Python
Message-ID: <iT0%4.36661$WS3.441157@typhoon.we.rr.com>

    I'm writing my current project, i'm writting a program primarily in
Python, but am embedding it in C++ for a couple reasons. Primarily, I want
the native, fast and flexible GUI that I can get out of using a RAD tool
like C++Builder, but also, I am not aware of any easyish ways to have a
local database with Python, so will provide the data-access needed via BDE
or some other such tool in C++Builder.

    A majority of my actual functionality will be done in Python, and i'm
wondering if I really need C++Builder to get the database functionality I
want. Being Cross-platform doesn't concern me at the moment -- but
working -good- and -prettily- (heh) on Win98 is very important to me. Most
of the database interfaces I saw on the Vaults, deal with various
Client/Server SQL database engines that are available... but I don't need
that. There will only ever be one connection -- my program -- to this one
database.

    Are there any options for a local database which is accessible via
Python? Gadfly seems like one, except the idea of storing it all in memory
terrifies me. There may be a LOT of information in this database as time
goes on :)

Thanks,

--S




From kens at sightreader.com  Thu Jun 15 18:12:27 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 15:12:27 -0700
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com> <3949A6BE.81B500E1@colconsulting.com> <3949A423.13371179@sightreader.com> <394A1E28.480F7319@colconsulting.com>
Message-ID: <394954CB.F907B706@sightreader.com>

"Frank V. Castellucci" wrote:

> Ken Seehof wrote:
> >
> > "Frank V. Castellucci" wrote:
> >
> > > Ken Seehof wrote:
> > > >
> > > > "Frank V. Castellucci" wrote:
> > > >
> > > > > Courageous wrote:
> > > > > >[snip]
> > > > > > Shrug. If it weren't for execution speed, your C++ environment would
> > > > > > be toast. Admit it. :)-
> > > > >
> > > > > While the last line seems rather toungue in cheek, there are other
> > > > > things that C++ brings to the table such as encapsulation, access
> > > > > control and type strictness.
> > > >
> > > > All three of these items were intentially left out of python because python
> > > > would not benefit from them, though C++ does.
> > > >
> > > > "Frank V. Castellucci" might have written (but actually didn't):
> > > > > A fish has certain definite advantages over a duck.  For example a fish has
> > > > gills.
> > >
> > > I didn't so it came from elsewhere.
> >
> > I made it up.  That was a deliberate witty attempt to put words in your mouth :-)
>
> Well the words in my mouth did make me witty, when I tried it out on my
> son he said "Your a dimwit", so find something better for me please as I
> am loosing ground here.
>
> > > > Encapsulation, access control and type strictness are often erroneously
> > > > referred to as if they were features.  They aren't.  They are aspects of a
> > > > design philosophy.  To make a system less error prone, constrain the programmer
> > > > in order to prevent the programmer from writing bugs.  Of course, that's
> > > > impossible, since it is always possible to write bugs anyway.  Because C/C++
> > > > are extremely error-prone languages, they need stuff like encapsulation,
> > > > access, control and type strictness.  Because python is less error prone, it
> > > > doesn't need these "features".  Python is based on different philosophy: To
> > > > make a system less error prone, make it simple.
> > > >
> > > > You have to ask, "what is the actual value of these features?"  The answer is
> > > > that they prevent certain kinds of bugs from appearing.  Now ask, "what bugs
> > > > would a C++ programmer encounter if encapsulation, access, control and type
> > > > strictness were removed from C++."  Undoubtedly it would be a mess.  Get a
> > > > picture in your mind of these bugs.  Now ask a python programmer how often
> > > > these kinds of bugs come up.  Answer: "can't remember that ever happenning!"
> > > >
> > > > Look at the big picture.  Do encapsulation, access control and type strictness
> > > > really make C++ really less error prone than python?
> > >
> > > Somewhere in your conversation you went from the aspects of a particular
> > > language, while true they do reduce errors because the language is still
> > > low level enought to let you get in trouble, to implying that they were
> > > put in there to make a system less error prone. You then finish with
> > > justifying a position based on this posit.
> > >
> > > I disagree with the postulation.
> >
> > My point is that encapsulation, access control and type strictness are relevant in a
> > comparison between C++ and python if they produce a specific advantage in C++ over
> > python that would make C++ preferable for some task.  My assumption is that
> > encapsulation, access control and type strictness produce robustness in C++, which
> > would be relevant if C++ were more robust than python as a result, which it isn't.
>
> I have found that encapsulation and type strictness are useful in class
> libraries. If we consider invarient state in line with B. Meyer, I have
> some insurance that the inner workings will continue to do so without
> fear of change. And I know that the user can change this by changing the
> code, but that is the same everywhere no? I enjoy and employ development
> by contract, if you use the class then you agree to it's interface, and
> you agree to use it as such.
>
> But more to your point, I won't argue over robustness advantages because
> the argument goes both ways and therefore cancels itself out. In my
> dimwitted view anyway.
>
> > > Before these articles of disagreement, C had (and still does)
> > > structures. Systems built in C can be a mess, or not. That is based on
> > > the talent and experience writing the code. Python does not prevent you
> > > from writing "messy code", it just reduces the potential devastation
> > > caused by messy code. Python has the ability to hide data, you just use
> > > a different notation for the variable in the class. Python has the
> > > ability for type strictness, but you have to code it by hand as it is
> > > not a feature of the grammer.
> >
> > Python doesn't /prevent/ messy code, but all other things being equal, it is easier
> > to write python code that is less messy because it's a simpler language. Typical ways
> > to be messy in C++ include: inconsistent choice of integer types and other basic
> > types, poor indentation practices, excessive abstraction layers, class inheritance
> > messiness, preprocessor messiness, etc.  Python has fewer basic types, no choice
> > about indentation, less need for abstraction layers, cleaner inheritance system and
> > less need for inheritance, and no need for a preprocessor.
>
> I would say that the enumeration of what in C++ allows one to get messy
> are vast. I would also say that experience has a tendency to cancel
> these things out as "newbie" mistakes, or a poor design.
>
> Like with any language, Python can be messy. There are not less ways to
> be messy in Python, there are just different ways. As I am pretty much a
> neophyte in the Python ways (but trying to come up to speed), a review
> of any code I have written by someone like yourself would probably
> elicit the statement "Well, thats a wee bit messy lad".

I doubt that.  Your python code is probably much cleaner than most C++ code I've seen over
the years. :-)

Well it seems like there are less ways to be messy in python, but my knowledge of set
theory (enumeration of infinite sets, etc) is insufficient.  Both languages allow an
infinite number of ways to be messy, assuming unbounded memory capacity.  The same could be
said of assembly vs. C++.  There are just as many ways to be messy in C++ as there are in
assembly.

I've been coding C++ since it first arrived, and I consider my C++ code to be relatively
neat and robust.  And yet, my first major python project is less messy, required less
planning, and seems to be more robust than recent C++ projects of similar or smaller
scope.  Python code that I have downloaded has generally been of exceptional quality
(unlike the vast majority C++ source code, free or otherwise).  This shouldn't be
surprising.  High level languages are typically less messy than low level languages.
Otherwise we should all just go back to assembly (except when we need portability).

Unfortunately I don't have any quantatative data on the subject.  Does anyone reading this
know of any relevant studies.

> > Python is less messy /and/ messiness is less devastating in python.
> >
> > > My original response was to the (lite hearted) statement that the only
> > > thing C++ brings to the table is execution speed. I will posit that
> > > languages for a solution space should be used that appropriatley address
> > > the design requirements. It ain't all C++, and it ain't all Python.
> >
> > Agreed. So the question becomes: What are the relevant characteristics of the problem
> > spaces and corresponding characteristics of the candidate languages?  The best answer
> > I have so far is that python is a high level language and C++ is a low level
> > language, and that the two languages work well together.
>
> I would add to this that the protection the interpreter sandbox offers
> has value, although the 'C/C++' or other language extensions are not
> protected there.
>
> > > I also believe that the larger the project, the more diminished the
> > > return on this particular measurement. Complexity takes it's toll on any
> > > language.
> >
> > In my experience, complexity takes less of a toll on python because it is a higher
> > level language.  I find that when I write a module in python, I tend to make it
> > behave as a component (i.e. having minimum dependency issues).  More often than not,
> > these components can be separately posted on Parnassus and used (hopefully) in
> > unrelated projects.  Component programming is extremely good for making robust large
> > scale projects.
>
> I would be curious to know how large scale Python applications fare
> (defects per release, etc.) but I (in my newbie Python suit) am not yet
> aware of the really big projects (Zope? - ignorance disclaimers apply).
>
> > In C++, which I've been using for much longer, I tend not to componentize quite as
> > well.  As a result, my code tends to hang together on a large scale in python quite a
> > bit better than in C++.  In addition, the small scale advantages hold true for large
> > scales projects as well.  So IMHO, based on experience, pythons productivity
> > advantages relative to C++ tend to increase as the project gets larger and more
> > complex.  The advantages of C++ tend to be more significant on a small scale.
>
> Just remember you are unique, just like everyone else. <grin> (Now thats
> witty, although I can't claim it).
>
> But seriously, your habits are not neccessarily the standard are they?
> You could do a better job with your C++ work, if you elect to.

Actually I consider my C++ to be of very high quality.  It's just not as good as my python
code.  There are situations in C++ where one has to choose between different kinds of
messiness, but can't avoid messiness altogether, especially when integrating with multiple
external modules.  C++ lets you encapsulate messiness, but those extra layers are another
form of messiness.

> > Specific features, such as the requirement that one use semicolons, don't amount to
> > much unless they produce characteristics that differentiate between languages.  So
> > the real question is, what characteristics does C++ bring to the table, and for what
> > kind of jobs do these characteristics make C++ the better choice.  IMHO, C++ brings
> > speed and legacy to the table, so I use C++ for high speed components within a python
> > project, and for integration of legacy software.  Maybe it brings other things as
> > well, but I don't know what they would be.
>
> As I said earlier, I believe you can be more productive in Python
> (although I acutally need to be so first before I commit totally to that
> opinion). But I also know from experience that in addition to speed, C++
> can be used in a very clean way to take advantages - To increase the
> robustness of the deliverables. If done right, it reduces maintenace
> cost, as well as more resiliant to change. I am not saying Python isn't,
> just that C++ is also.

It is possible for a C++ programmer to use techniques specific to C++ in order to increase
robustness, reduce maintenance and be more resiliant to change than other C++ programs.  In
the ultimate case, a C++ program could be 100% robust, have very low maintenance cost, and
be extremely resiliant to change.  However, this is not likely to happen on a tight
schedule with a random team of programmers.  It is more likely to happen with python under
similar conditions.

> The net of all of this will be that there are different views for the
> two from different people.
>
> > The question of which language scales better is very important.  If C++ scales
> > better, python should perhaps be used primarily as a scripting language within C++
> > applications.  On the other hand if python scales better, C++ should primarily be
> > used for high speed components within python applications.
>
> I don't know if I posed it more as a question or a tenet of large
> systems. See my large system question earlier in this missive.
>
> --
> Frank V. Castellucci

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/e4a0d95b/attachment.vcf>

From ftian at cs.wisc.edu  Mon Jun 26 20:08:16 2000
From: ftian at cs.wisc.edu (Feng Tian)
Date: Mon, 26 Jun 2000 19:08:16 -0500
Subject: client side cookie with httplib
In-Reply-To: <8F5FCDEB5gmcmhypernetcom@199.171.54.194>
References: <p9R55.17984$Yr4.303276@news1.rdc1.tn.home.com> <8F5FCDEB5gmcmhypernetcom@199.171.54.194>
Message-ID: <Pine.GSO.3.96L.1000626190300.13567A-100000@nova11.cs.wisc.edu>

Hi,

Is there a way to use Cookie.py and httplib on clientside?
Basically, I want to find a python equivalent to perl

$cookie = new HTTP::Cookies;
$useragent = new LWP::UserAgent;
$useragent->cookie_jar($cookie)

Then, when I do POST, GET etc useraget accepts/send cookie
automatically.

Thanks.

Feng





From info at pythonware.com  Tue Jun 13 11:23:39 2000
From: info at pythonware.com (PythonWare)
Date: Tue, 13 Jun 2000 17:23:39 +0200
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 2000)
References: <01b901bfd547$24a4aef0$0a00a8c0@SCREAM2>
Message-ID: <00a101bfd54b$5a8089a0$0900a8c0@SPIFF>

Anders M Eriksson wrote:
> > >
> > >    The PythonWare py152 distribution is a small and self-contained
> > >    Python distribution for Windows.  This kit includes Python 1.5.2,
> > >    Tkinter/Tk 8.2.3, PIL 1.0, PST, and other useful extensions.
> > >
> >
> > It's a pity that this packages doesn't work on Windows NT!
>
> > First there is an error in the installation that causes it to be
> > aborted. Welcome.py refers to TCL8.1/TK8.1 but in the package you have
> > 8.2.3

This buglet has been fixed, but it looks like the correct version didn't
make it from the staging area...  fixed! (the fixed version is 1,633,504
bytes; the original release was a few bytes smaller).

(It doesn't really affect the installation itself, though.  All you're
missing is the final splash screen...)

> > Second: It will not register Python in the registry. which then causes
> > win32_all not to install.

It's done this way on purpose -- lots of people (including us) like to
be able to install things without messing up the registry.

But maybe we could add a "add me to registry" button to the final
screen...  maybe in the 1.6 release?

> I have tried to contact Pythonware but all mail bounces...

Temporary technical problems (we installed new servers over the
weekend).  Everything should be up and running again.

regards,
the pythonware crew




From herzog at online.de  Thu Jun 15 08:40:15 2000
From: herzog at online.de (Bernhard Herzog)
Date: 15 Jun 2000 14:40:15 +0200
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <20000614155102.E16232@freei.com> <8i9cbh$mev$1@news2.isis.unc.edu> <8i9cu3$mlj$1@news2.isis.unc.edu>
Message-ID: <m3k8frw21c.fsf@greebo.nodomain.de>

"Daehyok Shin" <sdhyok at email.unc.edu> writes:

> Maybe a little better, but not enough good.

What's missing?
 
> class A:
>  x = 1

What's this class variable good for if you override it in the
constructor anyway?

>  def __init__(self):
>    self.__dict__["x"]=1
> 
>  def __setattr__(self, name, value):
>   if not self.__dict__.has_key(name):
>    raise AttriubuteError
>   else:
>    self.__dict__[name] = value
> 
>  def __getattr__(self, name):
>   try:
>    return self.__dict__[name]
>   except KeyError:
>    raise AttriubuteError

This __getattr__ implementation does exactly what happens by default
anyway (modulo the misspelling of AttributeError) and is therefore
completely redundant. After all, if self.__dict__ contains name,
__getattr__ will not be called. That also means that in general you
don't have to have to use self.__dict__ to access instance variables in
__getattr__.

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


From pinard at iro.umontreal.ca  Tue Jun  6 11:34:29 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 11:34:29 -0400
Subject: [Human] Lanaguage translation
In-Reply-To: "Jason Cunliffe"'s message of "Sun, 4 Jun 2000 12:34:52 -0400"
References: <sjl1243u5ri150@corp.supernews.com>
Message-ID: <oqsnuqddru.fsf@titan.progiciels-bpi.ca>

"Jason Cunliffe" <jasonic at nomadicsltd.com> writes:

> Does anyone have any ideas for language translation using Python?

A lot, but none is simple!

> I welcome any comments from others about this topic and ways to acheive
> dynamic translation with Python.

Keep me posted about your progress, at least.  Make your sources available.
>From my viewpoint, automatic translation looks like an extremely difficult
task, but I would be extremely happy being proven wrong... :-)


P.S. - It reminds me Donald Knuth's `Art of Computer Programming', vol I.
The last sentence of the `Notes on the exercises', page xix, reads:

   3. [M50] Prove that when n is an integer, n > 2, the equation
   x^n + y^n = z^n has no solution in positive integers x, y, z.

With an earlier explanation for "50":

   A research problem which (to the author's knowledge at the tie of writing)
   has not been solved satisfactorily.  If the reader has found an answer to
   this problem, he is urged to write it up for publication; furthermore, the
   author of this book would appreciate hearing about the solution as soon as
   possible (provided it is correct)!

I found that pretty humorous at the time, as at the time Knuth wrote this,
there was an open contest, with a substantial money prize, to anyone who
would bring a solution to this problem.  The prize was later withdrawn,
a few years before a proof was found.  (So far that I heard about this.)


The parallel is that anyone solving the natural language translation
problem satisfactorily might, with some entrepreneurship, become rich now
and forever known!  So yes, please publish the Python code! :-) :-)

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





From olczyk at interaccess.com  Wed Jun  7 09:36:06 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 07 Jun 2000 13:36:06 GMT
Subject: Calling external programs: ioredirection
Message-ID: <393e4e6f.72239515@nntp.interaccess.com>

Sorry for being very brief and sctech, but my ISP news server is
flakey right now. I can post now but maybe not in ten minutes.

I want to to be able to call external programs and read their output.
For example: I might want to call grep and look at the results one by
one. 

I want to do this in order to do things similar to sh. For example:

wc -l *.h *.cpp|awk 'begin int i=0; {i+=$1 } end print i;}'

( I know the script is probably wrong, but you get the idea. )

Any idea of how?


From tismer at appliedbiometrics.com  Sun Jun 18 20:12:41 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Mon, 19 Jun 2000 02:12:41 +0200
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <394D6579.8B2FF840@appliedbiometrics.com>


William Dandreta wrote:
> 
> Let me preface my remarks with:
> 
> --I am using the Python 1.2 binary for DOS with the Python 1.5.2 library
> which causes some problems. I avioded some problems by importing functions
> directly from strop instead of string. Replace is not in strop so I used the
> one in string but I had to comment out (put an octothorpe at the start of
> each line) the function's descriptive string to get it to work.--

Are you sure that you are talking about Python 1.2 ???
Let me assume it is 1.5.2 - ok?

> A while ago I posted a message called Why is Python so slow? At that time
> this Python script took 90 minutes to run. By making some relatively minor
> changes, I was able to reduced the time to 6 minutes.
> 
> The biggest improvement came (about a factor of 6) when I changed the
> replace(x,y) function in string with
> joinfields(splitfields(x,y),''). Considering that this is exactly what the
> replace function in string does, I was quite surprised at the results.
> Essentially I reduced 3 nested function calls to 2.

This is a great improvement. But I see that you don't use
the most recent functionality - can it be that you missed
a couple of improvements?

1) both joinfields and splitfields are deprecated. Please use
   split and join instead.
2) Is the split(join(...)) pattern necessary here? If the pattern
   is a simple string, then you might be reasonably faster with
   a single string.replace(str, oldstr, newstr) call.

> The only thing that makes sense to me is that Python is spending 85% of the
> time processing the comment lines and 15% of the time doing useful work.
> There is a string.pyc so I would have assumed that it would not contain any
> comments or descriptive strings but apparently that's not true.

The time spent in comment lines is very small,and it is reasonably
smaller if you use Stackless Python. Anyway, if you are concerned
about multi-line comments, then try this out:
Put a """ line in front of a longer couple of comments,
and end it the same way.
I should wonder if execution time would change reasonably.

> I am not very familiar with interpretive languages, Python is the first one
> I have used, but it seems that comments can cause a serious perfomance hit
> because they have to be processed along with the code.

If you move all of your comment text into the first line of your
funcion, then it will become the DOC string, and it will cost
you not a single cycle.
Please let me know if this gives you a change of execution
speed. Also feel free to send me your source code for
inspection.

cheers - chris-the-optimizer

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



From quinn at xeno.ugcs.caltech.edu  Thu Jun  8 19:58:56 2000
From: quinn at xeno.ugcs.caltech.edu (Quinn Dunkan)
Date: 8 Jun 2000 23:58:56 GMT
Subject: toString()
References: <8hmspg$r7n$1@bob.news.rcn.net>
Message-ID: <slrn8k0cv9.qk.quinn@xeno.ugcs.caltech.edu>

On Wed, 07 Jun 2000 20:26:42 -0500, David Allen <s2mdalle at titan.vcu.edu> wrote:
>Sorry if this is a FAQ - I can't seem to find the answer anywhere.
>
>Is there such thing as a toString() for objects like java has that
>will be automatically called when an object is used "as a string"?
>
>I.e. when you create an object, __init__ gets called automatically,
>so if I say:
>
>x = Spam()
>print x
>
>can I put a method in Spam such that it will print out something
>useful based off of the class fields rather than the ugly data 
>definition?

Yes, it's __str__  See the Language Reference 3.3 "Special method names" for
all the double-underscore magic you can accomplish.


From tralfaz at pacbell.net  Fri Jun 23 14:08:15 2000
From: tralfaz at pacbell.net (Michael D. Marchionna)
Date: Fri, 23 Jun 2000 11:08:15 -0700
Subject: Successful Solaris 2.6 Compile???
References: <393F9E65.80E9C383@usa.alcatel.com> <20000608215318.A20700@xs4all.nl>
Message-ID: <3953A78F.6C578CD9@pacbell.net>

It generally compiles fine, but depending on what compiler you
are using a environment settings you can pull in the ucblib version
of dirent.h instead of the SYSV version.  One way to check is
to compile the problem source file with -E and redirect the
output to a file.  The look through the file to see which header files
are being included.  Generally if you have gcc installed you won't
run into this.  If you are using /usr/ucb/cc then you may have problems.

--MDM

Thomas Wouters wrote:

> On Thu, Jun 08, 2000 at 08:23:49AM -0500, Sara Fowler wrote:
>
> > Has anyone successfully compiled the Python Interpreter on a Solaris
> > 2.6?  If so, do you know where DIR is declared?
>
> DIR is a standard define in one of the standard UNIX header files, usually
> dirent.h (either directly or indirectly.) If you are missing dirent.h, you
> have a very broken system. Are you able to compile anything else ? Did you
> install the proper Solaris 'packages' ? (From what I recall, long ago, about
> SunOS installs and compilers, and gcc and the like, you have to be careful
> about what you install where.)
>
> --
> Thomas Wouters <thomas at xs4all.net>
>
> Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From sanner at scripps.edu  Wed Jun  7 12:16:30 2000
From: sanner at scripps.edu (Michel Sanner)
Date: Wed, 7 Jun 2000 09:16:30 -0700
Subject: Numeric on Dec Alpha
Message-ID: <1000607091630.ZM254763@noah.scripps.edu>

I built Numeric on a Dec Alpha under OSF1 V4.0. I built fine but qhen I ran it
I witnessed strange behavior.

a = Numeric.identity(4)
a.shape = (16,)

would raise an exception about the size of the array needing to remain the same
???

Using the debugger I found in arrayobject.c:2201

	if (PyArray_As1D(&shape, (char **)&dimensions, &n, PyArray_LONG) == -1)
return NULL;

After this call shape [0] is 4 BUT shape[1] is 0 !

I changed the code to
                  if (PyArray_As1D(&shape, (char **)&dimensions, &n,
PyArray_INT) == -1) return NULL;

and got the right result.

Did anyone else run into this kind of preblems ? what is the correct way to fix
that ?

thanks

-Michel

-- 

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

>>>>>>>>>> AREA CODE CHANGE <<<<<<<<< we are now 858 !!!!!!!

Michel F. Sanner Ph.D.                   The Scripps Research Institute
Assistant Professor			Department of Molecular Biology
					  10550 North Torrey Pines Road
Tel. (858) 784-2341				     La Jolla, CA 92037
Fax. (858) 784-2860
sanner at scripps.edu                        http://www.scripps.edu/sanner
-----------------------------------------------------------------------




From jkraska1 at san.rr.com  Wed Jun 28 00:05:21 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 28 Jun 2000 04:05:21 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com>
Message-ID: <39597A25.8BF553C@san.rr.com>

> On top of that, very likely that if SUN sues they win.

Ridiculous.




C/


From 12klat at sightreader.com  Tue Jun  6 22:58:14 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 19:58:14 -0700
Subject: Example of Python code (C code to Python code)
References: <Ych%4.758$0l2.6362@news2.tor.primus.ca>
Message-ID: <393DBA46.8369E288@sightreader.com>

Ack!  Somebody else beat me to it.  And for free!  I Hate Open Source!
 :-(

Robin Porter wrote:

> Could someone please give me an example of some Python code.  I am just new
> to this language and would appreciate if you could re-write this simple
> piece of C code in Python.
>
> Thank you,
>
> Robin
>
> #include "stdafx.h"
> #include <stdio.h>
> #include <math.h>
>
> float
> HoursWorked,PayRate,Overtime,IncomeTaxes,GrossPay,NetPay,TaxPayable,Overtime
> Hours;
>
> const float TaxRate=.23;
> const MinPay=300;
> const RegHours=40;
>
> main()
> {
>
> TaxPayable=0;
> Overtime=0;
> OvertimeHours=0;
>
>  printf("\nPlease enter the hours worked per week? ");
>  scanf("%f",&HoursWorked);
>
>  printf("\nPlease enter the rate of pay? ");
>  scanf("%f",&PayRate);
>
>  if (HoursWorked>RegHours)
>  {
>   OvertimeHours =HoursWorked-RegHours;
>   Overtime = (OvertimeHours)*(PayRate*1.5);
>  }
>
>  GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime;
>
>  if (GrossPay>MinPay)
>  {
>   TaxPayable = (GrossPay-MinPay)*TaxRate;
>  }
>
>  NetPay = (GrossPay-TaxPayable);
>
>  printf("\nThe gross pay is %f",GrossPay);
>  printf("\nThe net pay is %f",NetPay);
>
>  return 0;
> }

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/26319212/attachment.vcf>

From aahz at netcom.com  Thu Jun 15 09:19:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 13:19:34 GMT
Subject: Text processing for a non-programmer
References: <TdsX4.12$zT4.3106@news3.dion.ne.jp>
Message-ID: <8ial56$cof$1@slb6.atl.mindspring.net>

In article <TdsX4.12$zT4.3106 at news3.dion.ne.jp>,
S D G <colin at doco.net.this_part_is.invalid> wrote:
>
>I'm a tech writer with practically no programming experience, but 
>I'm interested in learning programming so as to be able to write 
>some text manipulation programs.  It seems like Python would be a 
>good language to start with :)
>
>As a starting project, I'd like to write something that will format 
>my usenet followups nicely.
>
>Something like: Leave all the headers alone Wrap unquoted lines at 
>the 65th column If quoted lines are longer than 80 characters, wrap 
>them
>
>Has anybody got a similar python script that I can use as a starting 
>point.  I learn best by hacking examples :)

Hmmm....  There's a C program called 'par' that does this, but what
you're asking to do is actually a pretty tricky problem.  I think you'd
be better off finding a simpler first project.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From tjg at exceptionalminds.com  Fri Jun  9 12:21:00 2000
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Fri, 9 Jun 2000 09:21:00 -0700
Subject: win32all-132 available
In-Reply-To: <rh705.459$cQ.7688@news-server.bigpond.net.au>; from mhammond@skippinet.com.au on Fri, Jun 09, 2000 at 02:36:07PM +0000
References: <rh705.459$cQ.7688@news-server.bigpond.net.au>
Message-ID: <20000609092100.A5243@reepicheep.avalongroup.net>

On Fri, Jun 09, 2000 at 02:36:07PM +0000, Mark Hammond wrote:
> Ive just made win32all-132 available.  Everyone who thinks it isnt
> terrible is welcome to upgrade :-)

Dang, stole my line...

Anyway, as a guy who rights python code--for the most part on Linux--thanks
for making the times when I work on Windows a little easier!

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
Red Hat Certified Engineer                           MIG #1433
Avalon Technology Group, Inc.           
>>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<



From thomas at xs4all.net  Thu Jun  8 06:26:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 8 Jun 2000 12:26:19 +0200
Subject: My wishlist for Python3K
In-Reply-To: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>; from hzhu@rocket.knowledgetrack.com on Thu, Jun 08, 2000 at 01:47:14AM +0000
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
Message-ID: <20000608122619.I469@xs4all.nl>

On Thu, Jun 08, 2000 at 01:47:14AM +0000, Huaiyu Zhu wrote:
> My wishlist for Python3K (This is not to start a flame war):

> 2. Augmentation operators: +=, *=, etc.  Dot operators .*, ./, etc (for
>    numerical computation to distinguish matrix and elementwise operations).

This needn't wait for Py3K, and in fact I'm writing code for this right now,
loosely based on Michael Hudsons' patch and with directions from Michael and
Tim. Given that I'm facing *another* day of almost non-stop meetings, and
hacking Python is a great way to relieve the built-up stress, I'll probably
post a working-but-unfinished patch this weekend or so.

I'm not sure if the patch will get accepted by Guido for 1.6, though, since
it lacks a JPython counterpart. I know some Java but nowhere near enough to
start a hack for something like this. Volunteers, anyone ? ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jkraska1 at san.rr.com  Fri Jun 23 02:32:04 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 23 Jun 2000 06:32:04 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <39529090.507147FD@zip.com.au> <slrn8l5leh.4pd.quinn@mark.ugcs.caltech.edu>
Message-ID: <395304F9.581D3FF6@san.rr.com>

> 4 seconds is too long, even for NT.

I use python on NT and its startup is nigh on instantaneous.
So something is awry here.

I have a 400 mhz PIII with 256 meg of ram. When I had 128 meg
of ram, everything was blur-fast then, too.

I wonder if this fellow is doing something like running
python with so many other applications running that the OS
has to swap stuff out? That's what it sounds like...

C/


From rhicks at nospam.rma.edu  Wed Jun  7 16:15:12 2000
From: rhicks at nospam.rma.edu (Opus)
Date: Wed, 07 Jun 2000 20:15:12 GMT
Subject: WxPython & MOTIF (Not GTK)
References: <8hlur3$8fg$1@nnrp1.deja.com>
Message-ID: <k3y%4.1$RB1.1185@news>

You would have to go to wxWindows.com or some such to see if there is a
wrapper...

<heath7952 at my-deja.com> wrote in message news:8hlur3$8fg$1 at nnrp1.deja.com...
: Dear All,
:
: I would like to use WxPython and MOTIF only (not GTK!). I currently use
: WxWindows MOTIF Only, and would like to do the same with WxPython.
:
: Is this possible. If so how.
:
: Heath.
:
:
: Sent via Deja.com http://www.deja.com/
: Before you buy.




From warlock at eskimo.com  Sun Jun 11 22:41:54 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Sun, 11 Jun 2000 19:41:54 -0700
Subject: the windows interpreter
References: <Pine.SOL.4.21.0006051507510.9521-100000@scooter.mdstud.chalmers.se>
Message-ID: <slrn8k8ind.id.warlock@troll.myth>

On Mon, 5 Jun 2000 15:23:06 +0200, 
 Tim Gahnstr?m, in the persona of <md9tim at mdstud.chalmers.se>,
 brought forth the following words...:

>I downloaded and installed the python 1.5 for windows interpreter.
>I can use some pyton comands at the prompt like
>
>>>> print "Hi"
>will causet the output 
>Hi
>I can also create simple files with idle and run them with F5.
>But when I try to use an example program like
>
>----
>#! /usr/local/bin/python
>import sys
>if '-h' in sys.argv or '--help' in 
>sys.argv or '--help' in sys.argv: print '''
>help.py--does nothing useful (yet)
>options: -h, -help, or --help-display this help
>Copyright (c) Jacek Artymiak, 2000 '''
>    sys.exit(0)
>else:
>    print 'I don't recognize this option' sys.exit(0)
>---
>
>Then it will not work at al, not even if I remove 
>the first "linux inspired" line
>I get the errormessage
>
>----
>Traceback (innermost last):
>  File "F:\pyton\Python\Tools\idle\ScriptBinding.py", line 131, in
>run_module_event
>    execfile(filename, mod.__dict__)
>  File "F:\pyton\Python\test.py", line 3
>     if '-h' in sys.argv or '--help' in
>                                        ^
> SyntaxError: invalid syntax
>---
>
>
>I would apreciate if anyone had a suggestion.
>
>Tim Gahnstr?m
>


The syntax is massively incorrect, please read the tutorial, it is very 
helpful and will cover this pretty early on. 

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



From ahopkins at ahopkins.dynacare.com  Tue Jun 20 13:06:23 2000
From: ahopkins at ahopkins.dynacare.com (Albert Hopkins)
Date: 20 Jun 2000 13:06:23 EDT
Subject: Changing directory ACLs
References: <394F96B3.34E31C68@darwin.cwru.edu>
Message-ID: <slrn8kv94e.44t.ahopkins@ahopkins.dynacare.com>

Check the win32security module.  I've never used it, but it appears to
have what you're looking for.


On Tue, 20 Jun 2000 12:07:15 -0400, Mike Warfe <jmwarfe at darwin.cwru.edu> wrote:
>Hello, I have been using Python to do administrative tasks on my NT PDC.
>
>Is there a Win32 module that I can use to set directory privileges, that
>
>has the functionality of cacls.exe using python? I am currently using
>cacls to set the privileges by doing a os.system command in my script.
>Thanks for the help in advance!
>
>
>Mike Warfe
>


-- 
                                                     Albert Hopkins
                                             Sr. Systems Specialist
                                                      Dynacare Inc.
                                              ahopkins at dynacare.com

--
Computers let you make More mistakes faster
than any other invention in human history,
with the possible exception of handguns and tequila


                                                    Mitch Radcliffe


From dworkin at ccs.neu.edu  Wed Jun 28 18:44:08 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 28 Jun 2000 18:44:08 -0400
Subject: Python design philosophy
In-Reply-To: Steve Juranich's message of "Wed, 28 Jun 2000 11:02:37 -0700"
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <vnd4s6dbf47.fsf@camelot-new.ccs.neu.edu>

Steve Juranich <sjuranic at condor.ee.washington.edu> writes:

> I was wondering why there really isn't such an idea as a "private"
> member of classes?

Because there isn't a good reason to have them in Python.

There are plenty of good reasons to have them in other languages, but
that isn't what the question was about.

> I understand that members can be hidden, but not really protected
> from the user.  I was wondering what the benefit of this would be?
> Wouldn't it open the possibility of security holes?

Non sequitur.  No matter what one's opinion is on private class
members, this has nothing to do with security.

> Also, what benefit is gained by allowing a user to add/delete
> members from existing classes?  Wouldn't it be more desirable to
> create his/her own derived class?

In situations where I want to extend an already-created class, I will
usually inherit from it, as that is often the cleanest (and most
desirable, as you put it) thing to do.

However, allowing modifications on the fly to existing classes is a
major feature of Python, and part of what makes it so usefully
dynamic.  Taking away the ability to do that would not make sense.

I note that you didn't actually include any reasons why you think this
behavior should change.  Are there any such reasons, or was this just
a troll?

-Justin

 





From mcalla at home.com  Thu Jun 29 17:44:44 2000
From: mcalla at home.com (Mike Callahan)
Date: Thu, 29 Jun 2000 21:44:44 GMT
Subject: Questions about Installer/Builder/Standalone
References: <p9R55.17984$Yr4.303276@news1.rdc1.tn.home.com> <8F5FCDEB5gmcmhypernetcom@199.171.54.194>
Message-ID: <grP65.21733$Yr4.417778@news1.rdc1.tn.home.com>

I tried this but it did not work:

1. I copied and pasted runTk.py to the top of my script.
2. I ran simple.py to generate a proper Install_xxx.exe
3. I ran the install_xxx.exe and got the tk folder, the xxx.exe and the
support files. The tk folder is at c:\xxx\tk
4. I moved everything into a new test folder on a different machine with the
tk folder under the test folder ie. c:\test\tk.
5. I double click on xxx.exe and nothing happens.

I thought that if I moved the tk folder to the same relative position,
directly under the main folder, and I ran runTk first that my app would use
the tk at c:\test\tk. Is installutils.installCArchive() doing more than
simply unpacking?

Mike C.


"Gordon McMillan" <gmcm at hypernet.com> wrote in message
news:8F5FCDEB5gmcmhypernetcom at 199.171.54.194...
> Mike Callahan wrote:
>
> >I am playing with Gordon McMillan's Installer 0.3f for Windows and have
> >gotten it to work using freeze, simple, and standalone with non-Tkinter
> >apps. Good job Gordon! I can make a Tkinter app using simple but I
> >cannot get it to work using standalone. I know that simple does some
> >magic to get the app to use the local Tk it installs and Gordon talks
> >about it in the docs. But I cannot figure out what the magic is.
>
> Two pieces of "magic". Look in the gen_install.py from Simple. There's a
> snippet of code there that unpacks the bundled up tcl/tk:
>
> installutils.installCArchive(os.path.join(idir,'TCLlib.pkg'), idir,
> 'lib/tcl')
>
> Basically, this is unpacking TCLlib.pkg into lib/tcl (relative to where
> you're installing). That happens once. Somehow you need to get
> InstallShield to do it for you.
>
> Then, each time you run, there's some code that runs first (which you'll
> find in projects/BuildTK/runTK.py). All it does is set environment
> variables so that _tkinter will find the private tcl/tk stuff installed
> above. If you cut&paste from runTK to the top of your main script, it
> should run fine.
>
> - Gordon




From ajm at enteract.com  Wed Jun 28 16:30:09 2000
From: ajm at enteract.com (Alan Miller)
Date: Wed, 28 Jun 2000 15:30:09 -0500
Subject: win32print - adding some APIs?
Message-ID: <MPG.13c414ba715d9c249896b9@news-enteract>

[forgive me, I'm a wordy SOB]
I've run into a semi-interesting situation where I need to tweak code 
that's completely ready to go to the printer - PCL in this case, but the 
basic problem could apply to just about anything.  The problem is that 
once I modify the data stream, I don't have a good generic way to get it 
to the printer - I can open up the printer share for writing (for shared 
printers, which covers most of what I'm addressing currently) and send it 
there, but it would be nice to have the ability to send data to a non-
shared printer as well.

What I'm wondering is what it would take to extend win32print?  As it 
currently stands it looks like it's pretty basic - OpenPrinter with a 
printer name to get a handle, GetPrinter to get some information about 
the printer, ClosePrinter to close the handle out again.  No way to deal 
with actual print jobs that I can see unless the handle is usable as a 
device context handle for printing with win32ui.

The things I see missing are StartDocPrinter and EndDocPrinter, along 
with WritePrinter to actually get data to it.  Is sending data directly 
to the printer or queue without going through the Windows drivers 
something that people have considered or would find useful?

If nobody else is interested I can maybe take a hack at the source, but 
my current C++ experience is pretty much limited to modifying existing 
code (appropriate here) and I'm not quite sure what the procedure would 
be for getting changes incorporated....

I do have a VB class that I put together to do this, but it seems silly 
to create an ActiveX server in VB so that I can use it from Python to get 
to the Windows APIs.  Inelegant, you might say.  On the other hand, I'd 
love to convert that into Python code if I can get to the APIs.

Any thoughts?

ajm


From tim_one at email.msn.com  Wed Jun 28 00:40:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 28 Jun 2000 00:40:35 -0400
Subject: float literal too large
In-Reply-To: <ulsntz2kra.fsf@runner.princeton.idaccr.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCMENFGGAA.tim_one@email.msn.com>

[posted & mailed]

> I just wrote a python program to read output produced by a Fortran
> program that somebody had written.  Much to my surprise I got the
> exception:
>
> float() literal too large.
>
> The literal in question was '0.23489798479873E-2325' (this was
> generated on a Cray computer).  I couldn't find anything in the
> documentation about this restriction.  Even more puzzling was the fact
> that eval() worked just fine (it gave the answer 0.0 which is
> perfectly acceptable to me, given that the platform on which I was
> running had less precision than the platform that generated it).  ...
> ...
> I'm running:
>
> Python 1.5.2 (#4, Nov 23 1999, 14:57:03)  [GCC 2.7.2.2] on sunos5

This inconsistency in 1.5.2 was inherited from the differing behaviors of
atof and strtod in your glibc.  It should be fixed for you in 1.6; e.g., but
under WinTel  (which also used to display the inconsistency in 1.5.2) 1.6a2:

C:\Python16>python
Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> float(0.23489798479873E-2325)
0.0
>>> eval("0.23489798479873E-2325")
0.0
>>> x = 0.23489798479873E-2325
>>> x
0.0
>>>





From jnastos at fenchel.math.uwaterloo.ca  Tue Jun  6 20:46:33 2000
From: jnastos at fenchel.math.uwaterloo.ca (Jim Nastos)
Date: Tue, 6 Jun 2000 20:46:33 -0400
Subject: Nth digit of PI
In-Reply-To: <393D999B.E1150BCB@home.com>
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <kk5rjski2pc4g4tor99tc2h5ursinprg80@4ax.com> <393D999B.E1150BCB@home.com>
Message-ID: <Pine.SOL.4.05.10006062044210.6934-100000@fenchel.math.uwaterloo.ca>

On Wed, 7 Jun 2000, Rajarshi Ray wrote:

> Don't you just hate presumptuous bastards like Ullrich who will jump
> your gun without even considering if you actually have a point?? Perhaps
> you should be glad you're a non-expert; the more expert you get, it
> seems the more of a pompous ass you become too.

  On the contrary, it seems to me that the more 'expert' one gets, the
more qualified they are in asking the right questions to probe 'pompous
asses.'
  You seemed to have totally missed his point, so stop criticizing someone
for 'jumping a gun' without considering if *he* 'actually [had] a point.'

Jeeeeezz....

  - Jim Nastos



From aahz at netcom.com  Fri Jun  2 16:42:42 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jun 2000 20:42:42 GMT
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de> <3938090C.9FB5D057@stud.ntnu.no>
Message-ID: <8h9682$v4h$1@nntp9.atl.mindspring.net>

In article <3938090C.9FB5D057 at stud.ntnu.no>,
Peter Schneider-Kamp  <peter at schneider-kamp.de> wrote:
>mlauer at trollinger-fe.rz.uni-frankfurt.de wrote:
>> 
>> But why ? How does the for loop knows
>> how much elements the sequence contains ?
>> I did not program the __len__ function,
>> so... how ?
>
>As far as I understand the for loop goes on until an IndexError is
>raised. Then the error is cleared and the for loop ends.

Let's be technically precise here: an exception is raised and caught by
the "for" handler; the for loop is then broken (precisely as if the
break statement was executed).  Like any other caught exception, the
exception is thrown away (not "error is cleared").
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From trentm at activestate.com  Sun Jun 18 20:11:16 2000
From: trentm at activestate.com (Trent Mick)
Date: Sun, 18 Jun 2000 17:11:16 -0700
Subject: command-line parser
In-Reply-To: <v1mqksou1vnlilrr2l4d7f8bfhqk1r29qi@4ax.com>
References: <kdgqkscjmqa1aqs78l6grsmkonfmjcpsb6@4ax.com> <20000618151125.A22384@activestate.com> <v1mqksou1vnlilrr2l4d7f8bfhqk1r29qi@4ax.com>
Message-ID: <20000618171116.A22632@activestate.com>

I didn't really look over all of this (no time!) but some more comments:

help() should probably be __str__. When I wrote something like this before I
made sure that the help() output kicked out a nicely formated usage line such
as:

  -v, --verbose       give verbose output

With code something like this:

	def help():
		tabs = (2, 20)
		sys.stdout.write('%s-%s, --%s%s%s' %\
			(' '*tabs[0], self.short_form, self.long_form, ' '*tabs[1],
			self.description))
	


On Sun, Jun 18, 2000 at 11:15:18PM +0000, Ben Wolfson wrote:
> [comments excised].
> 
> Some advice taken, and it's a lot more readable, or seems so to me.
> An instance of the Option class can take a callback that gets called on its
> value, so if a particular option's value needs to be an int, say, you could
> instantiate it with
> 
> Option(<short>, <long>, 1, val_cb = int)
> 
> If the callback raises an exception, the option's help() function prints
> its description
> 
> # arglist.py
> class Error(Exception):
>     pass
> 
> class Option:
>     def __init__(self, short, long=None, has_value=None, default_value =\
>                  None, val_cb = None, description = None):
>         self.short = short
>         self.long = long
>         self.has_val = has_value
>         self.val_cb = val_cb
>         self.value = default_value or None
>         self.description = description

Don't use 'short' and 'long'. These typically represent types. Confusing for
C programmers, and 'long' is a Python built-in. As well in function
declaration Guido suggests not having a space btwn then name and its default
value. This is all just nitpicking though. :)

>         
>     def get_keys(self):
>         return self.short, self.long
>     
>     def handle(self, val):
>         if self.has_val:
>             if val:
>                 self.value = val
>             elif self.value is None:
>                     raise Error, 'Option %s requires a value' % opt
>         else:
>             self.value = 1
>         if self.val_cb:
>             try:
>                 self.value = self.val_cb(self.value)
>             except:
>                 self.help()
>         del self.short, self.long, self.val_cb, self.has_val

I think it would be better *not* to catch the conversion error. Or if you do
then:

	raise Error, "conversion of option value %s by %s failed" %\
		(`self.value`, `self.val_cb`)

> 
>     def help(self):#rather simpleminded, yes?
>         if self.description:
>             print
>             print self.description
>    

See comment above


>         
> class Arguments:
>     def __init__(self, *opts):
>         self.__dict__['nonopts'] = []
>         self.__dict__['dict'] = {}

Whoa! Do you need to mess with self.__dict__ directly? Is this not the same
as self.nonopts = [], etc.?

>         for opt in opts:
>             short, long = opt.get_keys()
>             if short and long:
>                 self.dict[short] = self.dict[long] = opt
>             elif short:
>                 self.dict[short] = opt
>             else:
>                 self.dict[long] = opt
>

Use something like 'opt_handlers' or whatever rather than 'dict'.

>     def process(self, options):
>         for opt in options:
>             if opt[0] <> '-':
>                 self.nonopts.append(opt)
>                 options.remove(opt)
>         for opt,val in map(splitstrip, options):
>             self.dict[opt].handle(val)
> 
>     def __getattr__(self, key):
>         if not self.dict.has_key(key):
>             raise Error, 'No option %s' % key
>         return self.dict[key].value
> 
>     def get_nonopts(self):
>         return self.nonopts
>             
> 
> def splitstrip(opt):
>     return optsplit(optstrip(opt))
> 
> def optstrip(opt):
>     while opt[0] == '-':
>         opt = opt[1:]
>     return opt
> 
> def optsplit(opt):
>     if ':' in opt:
>         return opt.split(':',2)
>     elif '=' in opt:
>         return opt.split('=',2)
>     return opt, None
> -- 
> http://www.python.org/mailman/listinfo/python-list


Trent

-- 
Trent Mick
trentm at activestate.com



From sdm7g at virginia.edu  Wed Jun 28 11:56:46 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Wed, 28 Jun 2000 11:56:46 -0400 (EDT)
Subject: Darwin (Not Python, yet...) [OT]
In-Reply-To: <8jahjq$fd8$1@bob.news.rcn.net>
Message-ID: <Pine.A32.3.90.1000628114111.14339A-100000@elvis.med.Virginia.EDU>

On Tue, 27 Jun 2000, Jeffrey P Shell wrote:

> Find the .zip file (if Stuffit didn't destroy it) and use the 'unzip'
> command in Darwin to unpack it.
> 
> >   This zip file doesn't seem to be in gunzip format so I couldn't
> >   unzip it from Darwin to get the right line endings.
> 
> Hmm.  unzip worked in DP4.

I'm sure DP4 has all sorts of extra goodies -- Darwin isn't up to
the level of a good Linux distribution yet. There is the "Missing 
Link" project at <http://www.darwinlinux.com/>, which plans to get
a CD distribution of Darwin + XFree86 + lots of linux packages. 

Softrak on stepwise.com has been down for the last couple of days,
so I haven't been able to search for a zip there. I may have to 
try taking it to another machine to unzip and tar up. 
( The main complication in getting anything done is the fact that
  I can only have Darwin up for a small part of the day -- If I 
  had DP4 and could run some Mac apps, I could find more time to
  play with it! ) 


> 
> Does Darwin follow the MacOS X directory structure convention?  (ie,
> /Local/Users/Bob/
> /System/Library/Frameworks/...
> )?
> 

Yes. There even seem to be some dangling symbolic links and empty
directories from missing MacOS X applications and libraries! 

BTW: Did you post any of your problems to either of the Darwin 
mailing lists ( -users or -developers ) ? I've just signed up 
and I haven't had time to look thru the archives yet. I would 
think this must not be a python specific problem. 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

"There's a new programming language called C+++. It's 50% better than C++, 
         but every time you compile, your modem hangs up."





From Rod.Haper at sage.nelesautomation.com  Tue Jun 20 14:07:24 2000
From: Rod.Haper at sage.nelesautomation.com (Rod Haper)
Date: Tue, 20 Jun 2000 13:07:24 -0500
Subject: A few python questions
References: <slrn8ktmmj.2md.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <394FB2DC.7DC9D02D@sage.nelesautomation.com>

Suchandra Thapa wrote:
> 
>     The first problem I've encountered is trying to determine
> the size of a file.  The file type doesn't seem to have any
> function that returns the length of a file.  I've been seeking
> to the end of the file and then using tell.  However this seems
> kind of inelegant, is there a way to get the file size of a file
> without doing this or using the stat function from the os module?


os.stat(path) returns a stat tuple.  The seventh element is the file
size in bytes.



>     Finally, is there some documentation for the python interface
> to postgresql?  The rpm package that comes with redhat 6.2 doesn't
> have any documentation.


The PyGreSQL package ftp://ftp.druid.net/pub/distrib/PyGreSQL.tgz
(Latest released version) has a tutorial directory.


-- 
+----------------------------------------------------------------------+
| Rod Haper, Senior Systems Analyst      |  Neles Automation           |
| Phone: 713-346-0652 - direct           |  SCADA Solutions, Inc.      |
|        713-939-0393 - fax              |  7000 Hollister             |
| Rod.Haper at sage.nelesautomation.com     |  Houston, Texas 77040, USA  |
+----------------------------------------------------------------------+


From fiona at sitegnome.com  Mon Jun 26 22:14:06 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Tue, 27 Jun 2000 12:14:06 +1000
Subject: The automagic of Tkinter
References: <20000626135336.A8180@bork> <8j7p5e$qia$1@nnrp2.deja.com>
Message-ID: <39580DEE.EFCAA22@sitegnome.com>

> Your class Knop inherits from Button which in
> turn inherits from Widget which inherits from
> BaseWidget.

....

> What does knop mean anyway?

In dutch: 

knop == button

:-)


From ben at marcin.org  Sat Jun 24 06:32:45 2000
From: ben at marcin.org (Marcin Benke)
Date: 24 Jun 2000 10:32:45 GMT
Subject: SAP R/3 and Python?
References: <C59F2D5004BDD311BEC500508B6FF2AD08E816@dewdfx16> <39538247.28CFCB57@lemburg.com> <ZFU45.219719$MB.3967328@news6.giganews.com>
Message-ID: <slrn8l93hb.1jn.ben@trawa.apcon.com.pl>

Christopher Browne <cbbrowne at news.hex.net> pisze:
>>Could it be that SAP (or a partner) already has such a Python
>>interface available ? I wouldn't want to reinvent the wheel
>>here and I assume this to be a highly non-trivial and time
>>consuming task.
>
>I'd suggest taking a look at:
><http://www.mcs.net/~garth/SAP/>
>
>There are RFC interfaces both for TCL and for Python; it ought not to
>be _overly_ nasty to adapt those to Python.

There are interfaces for Perl and Tcl there. Alas, no Python one.
OTOH, the other day I tried to construct a Python RFC interface using SWIG
(on Linux). It run, but was hardly usable as the parameter passing 
conventions in Python and SDK are highly incompatible, so some bits
of additional C/C++ code are needed. But I guess that a week of work 
should do to get something usable (provided one knows the RFC SDK)


Cheers,
	Marcin Benke
------------------------------------------------------------------------
YMMV. RTFM, of course, but sometimes when you RTFM, you just don't FTFA.


From pete at visionart.com  Wed Jun 14 15:57:44 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 14 Jun 2000 12:57:44 -0700
Subject: somebody stop me
References: <3947D4F3.1D315713@visionart.com>
Message-ID: <3947E3B8.EC9FEA18@visionart.com>

Pete Shinners wrote: 
> i have this routine which is running pretty quick
> on my machine. i may be getting overly greedy, but
> if i could shave off more time i'd be pleased
> (btw, this is with numeric python)
> (btw2, WIDTH and SAMPLES are defined elsewhere as global)
> 
> def pumpmeup_func(img):
>     rndpos = RandomArray.uniform(0, WIDTH, SAMPLES).astype(Int)
>     rndcol = RandomArray.uniform(100, 255, SAMPLES).astype(Int)
>     for x, color in map(None, rndpos, rndcol):
>         img[x,-3:-1] = color
> 
> - creating a list/array of the slices and then use numpy
>   to assign the random colors to that?  (like to see this!)
> 
> ...from .045 seconds per call, down to around .013


ok, i'm replying to myself. (ps, not crazy) but i was able
to crunch some more time off function call. the trick was
indeed the "list of slices". here's my latest version, down
from the previous optimized .013 to .009

def pumpmeup_func(img):
    rndpos = RandomArray.uniform(1, WIDTH-2, 100).astype(Int)
    rndcol = RandomArray.uniform(100, 255, 100).astype(Int)
    carved = array(map(lambda x,f=img: f[x,-3:], rndpos))
    carved[:,0] = rndcol
    carved[:,1] = rndcol
  
urgh, how do i clean up those last two lines into one?
i have a feeling the array(map(lambda...))) is doing
some extraneous work. most of my previous optimization ideas
are still open, can anyone drill this down another notch? 

woohoo, i'm my own best guru   :]


From heiland at ncsa.uiuc.edu  Tue Jun  6 10:09:49 2000
From: heiland at ncsa.uiuc.edu (Randy Heiland)
Date: Tue, 06 Jun 2000 09:09:49 -0500
Subject: PyRun_SimpleString error handling
Message-ID: <393D062D.8413AC36@ncsa.uiuc.edu>

I'm trying to figure out how to detect/return an error when using
PyRun_SimpleString.

Here's my simple pgm:

#include <Python.h>
#include <import.h>
#include <graminit.h>
#include <pythonrun.h>

void main()
{
  char save_error_type[1024], save_error_info[1024];
  PyObject *errobj, *errdata, *errtraceback, *pystring;
  int retval;

  Py_Initialize();
  retval = PyRun_SimpleString("print foo");
  printf("retval (from PyRun_SimpleString) = %d\n",retval);

  PyErr_Fetch (&errobj, &errdata, &errtraceback);

  if (errobj != NULL) printf("errobj != NULL\n");
  else  printf("whoops -- errobj == NULL\n");
}

You guessed it - I get the "whoops", the errobj is NULL.

Can someone enlighten me?

--Randy



From speed at ?.com  Mon Jun 19 18:07:17 2000
From: speed at ?.com (Jake Speed)
Date: Mon, 19 Jun 2000 22:07:17 GMT
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news and  links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net> <394E96D2.5787F2A3@gmx.de>
Message-ID: <SKT6CLIE7F158@CORP.SUPERNEWS.COM>

Michael.Scharf at gmx.de (Michael Scharf) wrote in
<394E96D2.5787F2A3 at gmx.de>:

>"Andrew M. Kuchling" wrote:
>>     In the same thread, David Bolen and Huaiyu Zhu write a script to
>>     verify that a class implements all the methods of a template
>>     class:
>>         http://x53.deja.com/getdoc.xp?AN=635109618
>>
>
>Trying to extract the code or even read the code in the daja.com
>representation seems impossible. Either I am an old-fashioned ascii
>formatting fetishist or just to stupid to figure out, how to display
>a deja article in simple plain ascii. Looking at the source of
>this article: the entire script is put into one line.....

It's at the bottom of the article,
 'View Original Usenet Format'

http://www.deja.com/getdoc.xp?AN=635109618&fmt=text

-Speed!


From jerome.quelin at insalien.org  Fri Jun 30 15:46:30 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 21:46:30 +0200
Subject: [Tkinter] Entry widget and textvariable option?
Message-ID: <962394390.813366342@news.libertysurf.fr>

Hi all,

I can't manage to use the textvariable option in an Entry widget. And
I can't find docs on this... And since we can't take references to
variables in Python, I don't see how to use this feature?

When running the following code:

>>> from Tkinter import *
>>> Entry(width=10, textvariable=spam).pack()

spam is to be defined, since Python interpolates it in the previous
statement. And that is definitely _not_ what I want Python to do! I
want the content of the Entry to go into the variable spam... In perl,
I would use '\$spam' to take a reference to the variable, and in Tcl
I'd use 'spam' to indicate a variable name, without interpolating
it. But what should I do in Python? 'spam' interpolates the content of
the spam variable, and we can't take variable references in Python...
Is it that we can't use the textvariable option in Python/Tkinter?

Any help welcome.

Thanks,
Jerome
--
jerome.quelin at insalien.org


From claird at starbase.neosoft.com  Sun Jun  4 18:33:31 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 4 Jun 2000 17:33:31 -0500
Subject: [Python-Dev] What's that sound?
References: <Pine.LNX.4.10.10005310928270.30220-100000@nebula.lyra.org> <4.3.1.2.20000604002128.0165a620@phear.dementian.com>
Message-ID: <703A4A2E8F200AC1.17365361774558AF.FF51B284ABF8FCA8@lp.airnews.net>

In article <4.3.1.2.20000604002128.0165a620 at phear.dementian.com>,
Steven D. Arnold <stevena at permanent.cc> wrote:
>At 09:14 PM 5/31/2000 -0500, Cameron Laird wrote:
>>5.  Bruce is starting to release
>>     chapters of *Thinking in Python*.
>
>Wow, where are these?  I've got Bruce's `Thinking in C++' and love it, even 
>though I don't particularly like C++. ;-)  I'm sure the combination of 
>Bruce's writing and a much more pleasant language will be fantastic!
			.
			.
			.
<URL:http://www.mindview.net/Python/ThinkingInPython.html>
-- 

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


From thomas at xs4all.net  Tue Jun 13 06:05:20 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 12:05:20 +0200
Subject: pipes and popen2
In-Reply-To: <slrn8kbqug.pd0.stianh@localhost.localdomain>; from stianh@ifi.uio.no on Tue, Jun 13, 2000 at 08:08:47AM +0000
References: <slrn8kbqug.pd0.stianh@localhost.localdomain>
Message-ID: <20000613120519.T20700@xs4all.nl>

On Tue, Jun 13, 2000 at 08:08:47AM +0000, Stian Husemoen wrote:

> I was a bit bored this weekend, and passed time coding python.

If only you had spent a little time reading the newsgroup as well ! ;-)

> In one snippet I wanted to make a UI to ispell. The idea was to create a
> bidirectional pipe, keep that pipe open, and feed ispell with words to
> spell. Like in this pseudo-code:

> 	stdout, stdin = popen2.popen2('ispell -a')
> 	
> 	for word in wordlist:
> 		# spell this
> 		stdin.write('pyhton')
> 
> 		# result
> 		print stdin.readline()
> 
> Problem is popen2 wont give me any output until I close stdin. Is there any
> way to both read and write to a program while keeping the pipe and
> process open? I can't seem to find any postings or documentation on this
> subject.

Curious, as I replied something very similar to someone else, only yesterday
;-) The solution is to set the buffersize of the pipe, by passing the
optional 2nd argument to popen2:

stdout, stdin = popen2.popen2('ispell -a', 1) # or something similarly small

So that the read*() methods don't stop to wait for more input.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From aahz at netcom.com  Mon Jun 12 17:17:42 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jun 2000 21:17:42 GMT
Subject: Printing exceptions
References: <8i2p3i$3pd$1@nnrp1.deja.com>
Message-ID: <8i3k1m$cjk$1@slb0.atl.mindspring.net>

In article <8i2p3i$3pd$1 at nnrp1.deja.com>, Arint?  <jamarijr at my-deja.com> wrote:
>My embedded python app sets an exception string, python can easily
>print this error using
>print sys.exc_value
>but, when I try to post it in my MessageBox
>pos.alert(sys.exc_value)
>the application crashes, I even tried putting the failing part in a
>try..catch block and it still crashed (What the...)
>Of course
>pos.alert("testing this messagebox") works fine

Do you get any kind of error message when your application crashes?

I'm suspecting this probably has something to do with the slightly weird
reference counting with exceptions.  In addition, you're supposed to use
sys.exc_info() because it's thread-safe.

Here's some code that will extract a safe string from the exception that
you should then be able to pass into your MessageBox:

import sys, traceback
try:
  ....
except:
  s = apply ( traceback.format_exception, sys.exc_info() )
  pos.alert(s)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From hgg9140 at skewer.ca.boeing.com  Tue Jun 20 11:09:48 2000
From: hgg9140 at skewer.ca.boeing.com (Harry G. George)
Date: Tue, 20 Jun 2000 15:09:48 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no>
Message-ID: <xqxwvjkie2r.fsf@skewer.ca.boeing.com>

Is it ready for prime time?  It is good enough to do useful, serious
glueware tasks and cgi's in-house.  I can't speak to heavy-duty,
externally-visible stuff.

As noted by others, see 
  http://www.opensource.org/

Additional comments:

We have recently approved a company-wide "toolkit" of OpenSource
libraries and apps.  The cost-benefit argument was that a common base
of tools simplifies infrastructure, cross-training, and in-house
consulting.  Further, by authorizing the toolkit as a unified package,
we get order-of-magnitude reduction in authorization paperwork and we
get reduced variation in processes.

Since we already had Perl and Java bundles (external to the OpenSource
toolkit), I added Python (with a good selection of add-ons) to the new
toolkit.  Initially mgmt was worried we were adding yet another
language to a complex mix (they just recently grappled with C-->Java
and shell scripts-->perl).  However, a few projects done "on spec"
made it obvious python was a significant player.  High-visibility
glueware tasks (e.g., XML interfaces among engineering tools) made the
case.

For the toolkit as a whole, we originally targeted UNIX boxes, but it
was such a success mgmt is considering extending to cover WinNT.  That
is, the same libraries and tools would be available on all our
development platforms.  With that of course goes python.

We did *not* push for Linux at this time.  Linux is a hot-button here.
If it is in your shop as well, concentrate on the financial case for
OpenSource.  Once that is established, Linux is simply a
faster/better/cheaper way to provide the tools.

One more point: It was crucial to make clear that we were
supplementing existing infrastructure (read "political turf"), not
attacking it head-on.  We will simply demonstrate ability to do
more/better/faster, and adjust the next budget cycle on that basis.


thomas at cintra.no (Thomas Weholt) writes:

> 
> Hi,
> 
> I want to convince alot of my co-workers that OpenSource software,
> especially Python, is ready for big business. ( is it ?! )
> 
> How do I sell an OpenSource solution like Python to hardheaded
> closedsource-fanatics? I know they care about money, so that would be
> my best angle. If anybody has any good examples of companies using
> Python, or similar technology, with profit, that would be great. 
> 
> To be more exact :
> 
> My mission is to swap expensive, oversized databases solutions like MS
> SQL 7.0 with PostgreSQL 7.0, crappy inconsistent script-languages like
> VBScript with Python and MS SiteServer / IIS with Zope, Linux with
> Windows  2000/ NT 4.0. I argue that the load on the system will never
> be to great for PostgreSQL to handle as supposed to MS SQL-server,
> VBScript, it sucks and they agree to that, Linux instead of Windows
> will save lots of cash, so will the "reduced" cost of PostgreSQL
> instead of MS SQL 7.0. 
> 
> The scenario above is normal in my experience, and I`d appreciate any
> feedback from people with similar ideas. Related Case-studies would
> also help alot.
> 
> The thing is that these people don`t thrust OpenSource. They laugh at
> script-languages in general, and the idea of sharing code is even more
> stupid. 
> 
> The growth of OpenSource, Python, Linux etc. depends on how well it
> can be a suitable substitution for ClosedSource-alternatives. A more
> agressive form of propaganda, showing off successful solutions, WITH
> benchmarks, results of stress-tests etc. is vital. Right now there are
> alot of talk about how great Python are, but few actual test of how
> well Python performes under pressure, and not just Python alone, but
> in a more complex environmet, like in Zope or using Python as
> interface to a RDBMS.
> 
> I need selling points for OpenSource, the idea and philosophy,
> examples of successful technology and eh .. moral support. I feel like
> I`m in the lions den right now. ( " Think of a happy place. Calm blue
> ocean, calm blue ocean ..." )
> 
> I truly love Python and would like to see it get the credit it
> deserves.
> 
> Best regards,
> Thomas Weholt

-- 
Harry George                E-mail:  harry.g.george at boeing.com 
The Boeing Company          Renton:  (425) 237-6915 
P. O. Box 3707  OY-89       Everett: (425) 266-3149 
Seattle, WA 98124-2207      Page:    (425) 631-8803


From karl-sve at dsv.su.se  Wed Jun 21 07:35:14 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Wed, 21 Jun 2000 13:35:14 +0200
Subject: 123.4+2=125.40000000000001
In-Reply-To: <3950A30D.DEE4CF4D@brb.midat.de>
References: <3950A30D.DEE4CF4D@brb.midat.de>
Message-ID: <Pine.GSO.4.20.0006211310100.17041-100000@luna.dsv.su.se>

On Wed, 21 Jun 2000, Peter Koppatz wrote:

> >>> 123.4+2
> 125.40000000000001   # is this OK?
> >>> print 123.4+2
> 125.4
> >>>

Short answer: This is the way it is.

Long answer:
It has to do with

1) Representation of floating point numbers internally
2) The difference between __repr__ and __str__

Point 1 ist that floating point numbers are not exact. Lots of people can
explain this better than me. Search deja, the answer is out there!

Point 2 is that print uses the __str__ method of an object to get the
printable string, but the interactive interpreter uses __repr__. __str__
does the rounding, __repr__ doesn't. Again, many others probably explain
this better than me. I could even be wrong, strange as it might seem!
Once again, search deja [1].

> On my calculator: 123.4+2 = 125.4 !?

In fact, this probably isn't true. It's just that your calculator rounds
the answer before handing it to you, just like print. Maybe. I should
probably change my name to Search Deja... :)

Also note that there has been a lot of discussion on this topic here, and
that the behavior might change in a later version of python.

HTH,
  Kalle Svensson

[1] You might answer "No, you get eye cancer from looking at Deja."
    To this I have no reply. Just get a good health insurance. :)



From cut_me_out at hotmail.com  Sat Jun 10 12:26:40 2000
From: cut_me_out at hotmail.com (Alex)
Date: 10 Jun 2000 12:26:40 -0400
Subject: namespace question
References: <eviltofu-8E2F45.12073510062000@news.earthlink.net>
Message-ID: <etdn1ktec3j.fsf@w20-575-109.mit.edu>

> If I do the following:
> 
> import socket
> 
> socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> 
> how do I call a function from the socket module later on?

Well, the best answer might be, "don't do that."  I would call the
socket 'socket_' to prevent the clash.  You could also give the
'socket' module another name, e.g.

import socket
socket_ = socket

socket = socket_.socket_ (blah)

...but I would be inclined to avoid that, as people would expect the
module to be called 'socket', and calling it something else would
proabably confuse them.  It would confuse me for a bit, at any rate.

Alex.


From alex at magenta.com  Fri Jun 30 06:57:39 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 30 Jun 2000 12:57:39 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr>
Message-ID: <8jhum101vku@news2.newsguy.com>

Jerome Quelin <jerome.quelin at insalien.org> wrote in message
news:962352947.103057778 at news.libertysurf.fr...
> Hi all,
>
> I was wondering how to pass variable-length keyworded arguments list
> to my base-class constructor? Look at the example code below:

Doesn't apply work for __init__ just as for any other function?


Alex






From rod at inf.bme.hu  Tue Jun 13 04:16:50 2000
From: rod at inf.bme.hu (Gabor Papp)
Date: Tue, 13 Jun 2000 10:16:50 +0200
Subject: where is PyOpenGL?
References: <39409F0C.8935C54F@inf.bme.hu>
Message-ID: <3945EDF2.F99E23BD@inf.bme.hu>

thank you for your help.
Gabor Papp


From DOUGS at oceanic.com  Thu Jun 15 18:03:50 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Thu, 15 Jun 2000 12:03:50 -1000
Subject: locate
Message-ID: <8457258D741DD411BD3D0050DA62365907A21F@huina.oceanic.com>

I'm assuming you mean character based user interface rather that graphical.
This is very platform dependent.  If on *nix there is the curses module.  On
Win platforms there's some curses work alikes and other solutions.

If you are on a unix or linux platform there are several resources I can
point you towards that can get you started.  Let me know.

-Doug-

> -----Original Message-----
> From: A[r]TA [mailto:arta at NOSPAMx-stream.nl]
> Sent: Thursday, June 15, 2000 8:26 AM
> To: python-list at python.org
> Subject: locate
> 
> 
> Hi,
> 
> I'm wondering if there's a method to do something like LOCATE 
> in BASIC.
> Example:
> 
> LOCATE  4,13
> PRINT "COOL!"
> 
> This prints the message COOL! on the screen on y, x location 4,13.
> 
> Can somebody tell me, if this is possible in Python?
> 
> A[r]TA
> 
> --
> -Life can kick your ass, but all you got to do is stand up 
> and move on...
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From hzhu at rocket.knowledgetrack.com  Fri Jun  9 19:38:43 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 09 Jun 2000 23:38:43 GMT
Subject: MatPy-0.2.7 released
Message-ID: <slrn8k305t.10m.hzhu@rocket.knowledgetrack.com>

I've just released MatPy-0.2.7.  You can find all the related links (home
page, project page, download, CVS, docs in html, ps, pdf, ...) starting from

	http://MatPy.sourceforge.net/

Apart from many small additions and improvements, there is an interactive
console contributed by Henning Schroeder that simulates Matlab start up
(example follows).

Enjoy!

Huaiyu <hzhu at users.sourceforge.net>

==================================================================
Here's an example:
$ matpy.py
Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- (<MatPy-console>)
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

    Command completion bound to tab.  You can change it to C-g by typing
    readline.parse_and_bind('C-g: complete')
	    
You are working in MatPy mode! Type help to get started
>>> help
Type 
    help(function)   to get more information on a special function 
    lookfor(name)    to search the help for name
    functions        to see a list of available functions
	
    log_on()         to start logging
    log_off()        to stop logging
    dump()           to show the history
    load('filename') to load filename
    save('filename') to save filename
			    
    CTRL-D to exit
				
It is recommed to read the included documentation so that you understand
the way MatPy works.

>>> functions
Matrix_c, Matrix_r, Mrange, abs, acos, acosm, approx_real, arange,
array2str, array2strC, asin, asinm, atan, atanm, ceil, cos, cosh, coshm,
cosm, cov, cumprod, cumsum, diag, diff, efunc, eig, exp, expm, eye, floor,
log, log10, logm, long, longC, max, mean, median, mfunc, mfuncC, mfunc_p,
min, mnorm, norm, ones, prod, ptp, rand, short, shortC, sign, sin, sinc,
sinh, sinhm, sinm, solve, sort, sqrt, sqrtm, std, sum, sums, svd, tan, tanh,
tanhm, tanm, to_Matrix, to_array, to_number, trapz, wait, zeros
>>> a = rand((2,3))
>>> print sin(a)*5 - exp(a)
[ -0.815   1.46    0.872 
  -0.894   1.47    0.0668 ]



From klaus.bruns at teleatlas.com  Wed Jun  7 05:09:34 2000
From: klaus.bruns at teleatlas.com (Klaus Bruns)
Date: Wed, 07 Jun 2000 11:09:34 +0200
Subject: Tkinter: disable a item in a Menubar
Message-ID: <393E114E.1F59AD44@teleatlas.com>

Hi all,

I want to change the state of an item in a Menubar, but I don't now how
to do.
For example: in the Menu 'File' there is a funktion to open and save a
file. After using the 'open' funktion this funktion is disable till
using the 'save' funktion.

I use Pmw for creating the Menubar:

import Pmw
from Tkinter import *

class mymenu:
   def __init__(root):
	Pmw.initialise(root)
	self.balloon=Pmw.Balloon(root)
	self.menuBar=Pmw.MenuBar(root,hull_relief=RAISED,hull_borderwidth=1,
                                 balloon=self.balloon)
        self.menuBar.pack(fill=X)
        self.menuBar.addmenu('FILE','open and save files')
	self.menuBar.addmenuitem('FILE','command','open a file',label='open',
				  command=self.open_file)
	self.menuBar.addmenuitem('FILE','command',save a file',label='save',
				  command=self.save_file)
	
   def open_file(self):
	file  = open(filename)
	## here I want to change the state of 'open' to DISABLED

   def save_file(self):
	## do something to save the file
	## and change the state of 'save' to DISABLED and 'open' to Normal

root = TK()
mymenu(root)
root.mainloop()

Can anyone help me?

	Klaus


From perly at news.xnet.com  Thu Jun 15 20:40:34 2000
From: perly at news.xnet.com (Pat Szuta)
Date: 16 Jun 2000 00:40:34 GMT
Subject: Python segfault
Message-ID: <slrn8kits2.ggk.perly@typhoon.xnet.com>

I don't know if anyone has seen this,  but python segfaults when given a
rather large input.

[pat at perly pat]$ perl -e "print '2+2+' x 10000" > input
[pat at perly pat]$ python < input
Segmentation fault (core dumped)
[pat at perly pat]$ 

#0  0x80631ab in PyNode_AddChild ()
(gdb) where
#0  0x80631ab in PyNode_AddChild ()
#1  0x80633d0 in PyParser_Delete ()
#2  0x806353f in PyParser_AddToken ()
#3  0x804fe9a in PyParser_ParseFile ()
#4  0x804fdb6 in PyParser_ParseFile ()
#5  0x805f91f in PyParser_SimpleParseFile ()
#6  0x805f7a9 in PyRun_File ()
#7  0x805efbd in PyRun_SimpleFile ()
#8  0x805ec9d in PyRun_AnyFile ()
#9  0x804fc02 in Py_Main ()
#10 0x804f746 in main ()
#11 0x400669cb in __libc_start_main (main=0x804f738 <main>, argc=1, 
    argv=0xbffffac4, init=0x804ed84 <_init>, fini=0x8084c3c <_fini>, 
    rtld_fini=0x4000ae60 <_dl_fini>, stack_end=0xbffffabc)
    at ../sysdeps/generic/libc-start.c:92
(gdb) 

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


From claird at starbase.neosoft.com  Thu Jun  1 10:18:45 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 1 Jun 2000 09:18:45 -0500
Subject: Using python (or Tcl) on the web
References: <MPG.1365adcfa8470eef9896aa@news> <L3iZ4.154057$681.2782248@news-east.usenetserver.com> <8h4eeu$bpg$1@saltmine.radix.net> <gCjZ4.155377$681.2810806@news-east.usenetserver.com>
Message-ID: <696FCF6078F44C05.0A1D8D07896635A0.39C501803FD07528@lp.airnews.net>

In article <gCjZ4.155377$681.2810806 at news-east.usenetserver.com>,
J.M. Ivler <ivler at basecamp1.netquest.net> wrote:
>In comp.infosystems.www.misc Cary O'Brien <cobrien at radix.net> wrote:
>> Well, they've got mod_dtcl and aolserver.
>
>Excellent products both. There is also cgi.tcl, another excellent piece of
>tcl support.
>
>> Look, this is mature technology. Can't everyone just get along?
>
>We do. I hope. I was just pulling CL's leg. :-)
>
>> (Plus you forgot embedded perl - embperl).
>
>Hey, don't let the book get you thinking I do Perl, that was a demand by
>PTR/PH. They also wanted fastCGI, but I talked them out of that. :-)
>
>jmi

... and for those trying to keep score, this thread
has actually only touched on about half the script-
ing-savvy Web servers for these languages.  Notice
no one has yet mentioned NeoWebScript, Velocigen,
tclhttpd, YAWP, PMZ, Zebra, ...

My compliments to JM for standing up to PH; the book
does its job well without the complication fastCGI
would represent.
-- 

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


From jkraska1 at san.rr.com  Tue Jun 13 00:51:38 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 13 Jun 2000 04:51:38 GMT
Subject: Threads -- play well with others? If not, what instead?
References: <Pine.GSO.4.10.10006130725470.20303-100000@sundial>
Message-ID: <3945BE7F.336F3C2A@san.rr.com>

> Yes. Use event based programming. Tcl has support for fileevents -- events
> that get called when a file is ready to read from.

In the unix environment, this translates to select on a file
descriptor block, where activity on any of the file descriptors
(which includes pipes, sockets, and files) results in activation
of the thread, which otherwise goes to sleep until there is
something to do, ensuring that the application does not eat
cpu time. Certainly, there is similar functionality on windows,
probably implemented in a similarly general way.



C/


From dwhite2 at seas.upenn.edu  Fri Jun 23 16:27:58 2000
From: dwhite2 at seas.upenn.edu (dave white)
Date: Fri, 23 Jun 2000 16:27:58 -0400
Subject: writelines() or write()
Message-ID: <3953C7B9.E32E8599@seas.upenn.edu>

When I use writelines() and my strings have \r characters at the end
somehow my output file only contains the last line.  What's the correct
way to write multiple lines to a file?

strings = ['one \r', 'two \r', 'three \r}
outfile.writelines(strings)
---
outfile looks like this:
'three '
---
What I want to have happen of course is:
'one '
'two '
'three '

thanks,
david


From cbiegay at objectmentor.com  Fri Jun 30 12:46:49 2000
From: cbiegay at objectmentor.com (Chris Biegay)
Date: Fri, 30 Jun 2000 16:46:49 GMT
Subject: Uploading binary files
Message-ID: <Z9475.7$qm5.709@ord-read.news.verio.net>

I'm trying to upload files through an HTML form, but it only works
for text files.  When I try to send a jpeg or something the process just
hangs, and I can't even change the script afterward since the webserver
won't release the file!  Any ideas?  This is basically how I'm doing it now:

form = cgi.FieldStorage()
formItem = form[inputName]

if formItem.file:
    print "Received file [" + formItem.filename + "]"
    data = formItem.file.read()
    ...




From redaelli at inc.it  Sat Jun  3 16:41:09 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sat, 03 Jun 2000 22:41:09 +0200
Subject: float does not parse '0.000000E+00'
Message-ID: <39396D65.D2196E66@inc.it>

1st of all, cheers to everyone!

I need to parse numbers coming from a file (made by a fortran program)
the problem comes when parsing 
When I made some tests on the shell they went fine, like this:

=======================
>>> a = "0.000000E+00"
>>> float(a)
0.0
>>> 
============================

but if I run a program that does x=float(x) where x = "0.000000E+00 "
(being a string) here's the output. 

===========================
[paolo at aldebaran Telaio]$ ./pyplot.py
x(<type 'string'>)='0.000000E+00' y(<type 'string'>)='0.000000E+00'
Traceback (innermost last):
  File "./pyplot.py", line 198, in ?
    t = Telaio()
  File "./pyplot.py", line 172, in __init__
    x = float(x)
===========================
I looked into Python manuals adn found  that a float can't looks like an
octal. But an octal does NOT have the dot!

Thanks very much in advance for your attention,
	Paolo Redaelli


From wlfraed at ix.netcom.com  Sat Jun  3 00:43:20 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Fri, 02 Jun 2000 21:43:20 -0700
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <CC59868A0A76D311B1F50090277B321C1F338A@VRTMAIL>
Message-ID: <gd2hjsoet6itkl5f6vfpbpbk0aucq5f703@4ax.com>

On Fri, 2 Jun 2000 19:25:42 -0400 , Mike Fletcher
<mfletch at tpresence.com> declaimed the following in comp.lang.python:

> 
> 	module m dynamically receives a reference to an object o that in one
> instance has an attribute p (it also, in that instance has an attribute P)
> --> run-time system needs to know when m was written (thus whether it is
> intending the first or second casing), as well as the renaming map within
> object o (gack)
>
	Oh gad... Persistent objects in shelves and similar...

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From wjdandreta at worldnet.att.net  Sun Jun 18 23:30:14 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Mon, 19 Jun 2000 03:30:14 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <01bfd93c$66c03180$2ace8ad1@daves-dell> <oO935.12054$Xx5.557215@bgtnsc06-news.ops.worldnet.att.net> <01bfd96b$f7ddcea0$85a38ad1@daves-dell>
Message-ID: <atg35.9524$C44.540890@bgtnsc05-news.ops.worldnet.att.net>

Hi David,

Thanks for the sample copde. I will test it out.

Bill

David C. Ullrich wrote in message <01bfd96b$f7ddcea0$85a38ad1 at daves-dell>...

>...On the machine I
>tested this on calling testre() seems to take about
>six seconds while testfind() seems to be less than
>three.





From thomas at xs4all.net  Wed Jun 14 12:45:08 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 14 Jun 2000 18:45:08 +0200
Subject: tuple creation in C extensions
In-Reply-To: <etdya48z0m4.fsf@w20-575-109.mit.edu>; from cut_me_out@hotmail.com on Wed, Jun 14, 2000 at 12:30:27PM -0400
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu>
Message-ID: <20000614184508.B26436@xs4all.nl>

On Wed, Jun 14, 2000 at 12:30:27PM -0400, Alex wrote:

> I was thinking it ought to be possible to write a script that would pull
> the signatures of the API functions from the docs, and write some C code
> to determine their refence count effects.  You might be able to get a
> long way to the sort of table we need that way.

The docs already have that:

   PyObject* PyList_GetItem (PyObject *list, int index)

   Return value: Borrowed reference.

          Returns the object at position pos in the list pointed to by p.
          If pos is out of bounds, returns NULL and sets an IndexError
          exception.

   PyObject* PyList_New (int len)

   Return value: New reference.

          Returns a new list of length len on success, or NULL on
          failure.

   PyObject* PyList_GetSlice (PyObject *list, int low, int high)

   Return value: New reference.

          Returns a list of the objects in list containing the objects
          between low and high. Returns NULL and sets an exception if
          unsuccessful. Analogous to list[low:high].

And:

   int PyList_SetItem (PyObject *list, int index, PyObject *item)
          Sets the item at index index in list to item. Note: This
          function `steals'' a reference to item.


-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From olipt at mayo.edu  Sat Jun 10 01:36:06 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Sat, 10 Jun 2000 00:36:06 -0500
Subject: Reloading on the fly: was: Re: Python Productivity over C++
In-Reply-To: <3941C013.F1F6DE85@Lugoj.Com>
References: <20000609185450.91797.qmail@hotmail.com> <8hrjg3$dn2$2@news1.xs4all.nl> <3941C013.F1F6DE85@Lugoj.Com>
Message-ID: <Pine.LNX.4.10.10006100025040.6036-100000@us2.mayo.edu>

While on the subject of reloading...

I'm not sure if you covered this already, but what if I have an extension
module written in C that is already loaded in memory.

If I change the extension module and recompile it.  Is there a way to get
python to reload the new extension module.  I've tried to do it but always
get a segmentation fault.

-Travis




From bjorn at roguewave.com  Fri Jun  9 13:05:45 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 09 Jun 2000 11:05:45 -0600
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available 
 now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news>
	 <8hlrkn$5t2$1@nnrp1.deja.com>
	 <1ebvwau.af3wl21301xdcN%bparsia@email.unc.edu> <87em67fpp6.fsf@ev.netlab.co.jp>
Message-ID: <394123E9.F3293D77@roguewave.com>

Yukihiro Matsumoto wrote:
> 
> bparsia at email.unc.edu (Bijan Parsia) writes:
> 
> |Are Ruby's *classes*, themselves, instances of classes (aka
> |metaclasses)? Like, oh, let's see, *Smalltalk*!
> 
> I know it's off-topic, so in sort answer:
> 
>   yes
>                                                         matz.

:-)

I've looked at Ruby a couple of times, and when it comes to language
features I like it a lot! (e.g. the iterator concept looks miles easier
to work with than the Python idiom for non-sequential collections -- and
the alternative uses like mutex.lock <block> virtually makes me salivate
<wink>)

The reason I haven't done any programming in it yet, however, is that
the English language documentation is not yet good enough.  I.e. I get a
feeling that it is not complete (vs. the Japanese docs), and I'm afraid
that if I got stuck, gramatical issues would prevent me from solving my
own problems...  I'll contine to come back to it though.

-- bjorn



From petersc at stud.ntnu.no  Tue Jun 13 02:41:08 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 13 Jun 2000 08:41:08 +0200
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu>
Message-ID: <3945D784.FD062D7@stud.ntnu.no>

Roy Katz wrote:
> 
[nice anecdotal hint for Python productivity]
> At the same time, C was faster than (straight) Python by a factor of
> ten.  Whatever, I now have three implementations of a virtual CPU to play
> around with :)

Would you mind putting the code somewhere (on a site or something).
I'd like to take a look to count number of lines/evaluate
how expressive the languages are, if I can read and extend
the code easily (let's say I add two CPU commands, I measure
how long I need to understand enough of the code to be able to
do it and then how long I need to do it) etc.

of-course-I'd-like-to-play-with-it-too-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From mrfx at etang.com  Sun Jun 11 16:01:02 2000
From: mrfx at etang.com (mrfx)
Date: Mon, 12 Jun 2000 04:01:02 +0800
Subject: How to display UNICODE text in idle?
Message-ID: <8i255t$1r0j$2@news.cz.js.cn>

I want to write a python program in IDLE, but idle cannot display
other language text. How to do it?








From stephen at cerebralmaelstrom.com  Thu Jun  1 01:34:52 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 01 Jun 2000 05:34:52 GMT
Subject: ANNOUNCEMENT: PythonWorks 1.0 now available (may 31, 2000)
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com>
Message-ID: <0CmZ4.30723$WS3.355848@typhoon.we.rr.com>

    Hey, that looks kind of nice, a professional IDE (?) for Python.. I just
have one little question, though....

    Three hundred and ninety five dollars? Are you kidding?

    Is this product up to the quality level of, say, any of the RAD
environments from Borland, who's standard-editions are about $100? Or how
about Visual Basic (Professional $279 -- I don't see a "standard" on
shop.microsoft.com, but the Learning edition is also about $100) [ MS VC++'s
standard is about $100 I think, also, but i'm not including it in this
because I have a hard time calilng it a 'RAD' environment... that
splitting-of-a-window thing it makes you do is irritating, I want to be able
to draw a whole form :P ]

    Are your features comparable to any of the above mentioned products, in
intent or _completion_? I'm not flaming you or anything like that, i'm just
rather flabbergasted. I've shelled out a hundred bucks to both MSoft and
Borland for their C++ environments(Granted, I was getting the 'Pro'
environments at a student discount, but still), and have done similar for
other such products.. But I can't even fathom the possibility of spending
twice that much on something which isn't BETTER then the aforementioned
products, let alone four times as much.

--S, who's still in shock.


PythonWare <info at pythonware.com> wrote in message
news:00de01bfcaf3$6236de90$0500a8c0 at secret.pythonware.com...
it's time for a major release from the labs:

    PythonWorks is a rapid application development tool, designed for
    efficient development in Python.  PythonWorks includes editors,
    a debugger, a user-interface design tool, project browser, deploy-
    ment tools, integrated on-line help, and a scriptable environment.

    To learn more, visit the PythonWorks product site:

        http://www.pythonware.com/products/works

    If you order now, you get:

    - PythonWorks 1.0 for Windows 9X/NT/2000 (more platforms 3Q00)
    - a personal license for PythonWorks 1.X on all available platforms
    - free upgrades throughout the entire 1.X series
    - support via mail

    For more information, please visit the product site, or mail us
    at info at pythonware.com.

regards,
the pythonworks team

    "Secret Labs -- makers of fine pythonware since 1997."






From kpmurphy at my-deja.com  Wed Jun 28 14:17:33 2000
From: kpmurphy at my-deja.com (kpmurphy at my-deja.com)
Date: Wed, 28 Jun 2000 18:17:33 GMT
Subject: entry widget and rotor module
Message-ID: <8jdff7$mc0$1@nnrp1.deja.com>

can anyone help out... what's going on below?

anna% python
Python 1.5.2 (#12, Jun 26 2000, 13:56:56)  [GCC 2.95.2 19991024
(release)] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import rotor
>>> rt = rotor.newrotor('qwerty', 20)
>>> rt.encrypt('boat')
'.\314q\217'
>>> foo = rt.encrypt('boat')
>>> type(foo)
<type 'string'>
>>> foo
'.\314q\217'
>>> print foo
.?q
>>>

...it appears that the string foo has the right data in it, but prints
out weird characters.  Thanks,

keith


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


From robin at jessikat.demon.co.uk  Tue Jun 27 19:02:53 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 28 Jun 2000 00:02:53 +0100
Subject: Controlling Acrobat Reader with COM
References: <200006262159.AA3991339306@mail.mamey.com>
 <s7khls0in7rrfbqnqdo4cha6chh9qqhstr@4ax.com>
Message-ID: <Q38VarAdKTW5Ewcx@jessikat.demon.co.uk>

In article <s7khls0in7rrfbqnqdo4cha6chh9qqhstr at 4ax.com>, Sam <?@?.?>
writes
...
>>Andres Corrada
>Try typing these strings directly into the interpreter.
>
>>>> print 'c:\test\readme.pdf'
>c:     est
>eadme.pdf
>>>> print "c:\test\readme.pdf"
>c:     est
>eadme.pdf
>>>> print 'c:\\test\\readme.pdf'
>c:\test\readme.pdf
>>>> print 'c:/test/readme.pdf'
>c:/test/readme.pdf
I tried already with correctly escaped filenames and always got the same
error ie 'c:\\Python\\Installs\\test.pdf' got the same error as 
'gnarg.pdf' (and I don't have any docs called gnarg.pdf :)).
-- 
Robin Becker


From invalid.address at 127.0.0.1  Fri Jun  9 13:40:35 2000
From: invalid.address at 127.0.0.1 (David)
Date: Fri, 09 Jun 2000 17:40:35 GMT
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <39432bc2.54792472@news.telus.net>

On Thu, 8 Jun 2000 23:18:26 -0700, "Lawrence Kesteloot" <lk at veriomail.com>
wrote:

>If one of the goals of Python 3000 is to make it easy for new
>programmers to learn, how about removing the distinction
>between = and ==?  The only kinds of examples I can think of
>where this is useful is something like:

How about fixing the error codes to be more newbie-friendly?  When a = is
used in place of ==, the error code should use clear language and suggest a
solution.
 
I think newbies would learn extremely quickly that = != ==.



From petersc at stud.ntnu.no  Wed Jun  7 02:35:09 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 07 Jun 2000 08:35:09 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.10.10006060938200.4711-100000@sundial> <memo.20000606232050.11341B@digitig.compulink.co.uk>
Message-ID: <393DED1D.C7803A6C@stud.ntnu.no>

Tim Rowe wrote:
> 
> god doesn't need to shout...

I am sorry I have to disagree, Sir, but god DOES shout. For
anecdotal evidence please go to your local video retailer
and buy a copy of the catholic documentation film Dogma.

thinking-alanis-was-a-good-choice-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From redaelli at inc.it  Sun Jun  4 05:58:45 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 11:58:45 +0200
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <003b01bfcdde$0cd2f950$6401a8c0@home>
Message-ID: <393A2855.49389A4A@inc.it>

Darrell Gallion wrote:
> 
> Too much work to decode what your doing.
Well, you're right! Too much code! Thanks for the tip.
I'm posting some more things about it
-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/


From mrfx at etang.com  Sun Jun 11 15:55:37 2000
From: mrfx at etang.com (mrfx)
Date: Mon, 12 Jun 2000 03:55:37 +0800
Subject: How to display UNICODE text in idle?
Message-ID: <8i255k$1r0j$1@news.cz.js.cn>

I want to write a python program in IDLE, but idle cannot display
other language text. How to do it?






From db3l at fitlinxx.com  Mon Jun 12 18:52:25 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Jun 2000 18:52:25 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au> <8F51A5DF3gmcmhypernetcom@199.171.54.194>
Message-ID: <ur9a2ms0m.fsf@ctwd0143.fitlinxx.com>

gmcm at hypernet.com (Gordon McMillan) writes:

> Simply that in Python, inheritance doesn't necessarily have anything to do 
> with runtime behavior. (You can, of course, introduce that dependency, but 
> you'd be well advised not to.)

(...)

> That is, it's the interface that matters (not the derivation), and you 
> don't even have to get formal about that.

For me, this hit home in a small but practical case of overriding
sys.stdout with an alternate "file" object.  The requirement is that
it satisfy the interface (in this case, just a "write" method) and not
a particular class hierarchy.

This makes it extremely simple, and generally informal, to build a
class that fits the requirement (and/or augment an existing class that
may be deep in your own hierarchy) and use that as a replacement
sys.stdout object.  As long as you meet the interface requirements
(dynamically at run time) you're set.

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


From aahz at netcom.com  Mon Jun 26 21:43:29 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 01:43:29 GMT
Subject: stopping threads
References: <3957C1DC.E7C77E2E@sec.noaa.gov> <8j8id2$v3h$1@nntp9.atl.mindspring.net> <etdg0q0hqvt.fsf@w20-575-109.mit.edu>
Message-ID: <8j90s1$7vj$1@slb7.atl.mindspring.net>

[BTW, please keep attributions; they make it easier to follow threads]

In article <etdg0q0hqvt.fsf at w20-575-109.mit.edu>,
Alex  <cut_me_out at hotmail.com> wrote:
>Attribution restored for Aahz:
>>
>> I'm currently reading up on Java threads, and while such a function
>> exists, the authors recommend very strongly against doing so.
>> Instead, your thread should watch for some condition and return/exit
>> from its main function/method when that condition becomes true.
>
>Do they say why?

I don't remember off-hand, but assume it's essentially the same reason
you avoid using "kill -9" (or the equivalent) to end actual processes.
Too much chance of improper cleanup, if nothing else.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From adams_s at lab.eng.usyd.edu.au  Mon Jun 12 00:55:03 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Mon, 12 Jun 2000 14:55:03 +1000
Subject: Newbie: displaying binary numbers
Message-ID: <8i1q49$2nk3$1@news1.wire.net.au>

Hi all,

I'm trying to display integers as binary numbers, I know there are the calls
for making hexadecimal and octal strings, but couldn't find anything for
binary?

are there any modules, or operations not in the docs?

all I need is to convert a number to a binary with n bits.

thanks, Steven




From jkraska1 at san.rr.com  Wed Jun 21 17:04:26 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 21 Jun 2000 21:04:26 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <394F0C4D.3911FF90@commsecure.com.au> <394FE68A.D3AB1FF8@sightreader.com>
Message-ID: <39512E1C.4B160E72@san.rr.com>

> > > I want to convince alot of my co-workers that OpenSource software,
> > > especially Python, is ready for big business. ( is it ?! )

Yes. In my opinion, it's easier to make fault-tolerant programs
in python than it is C++.

> > > How do I sell an OpenSource solution like Python to hardheaded
> > > closedsource-fanatics?

Find every angle from which closed-source is promoted, and show
them that Python beats them at it (hint, you won't be able to do
this, entirely, but you can try). :)-

> > > The thing is that these people don`t thrust OpenSource. They laugh at
> > > script-languages in general, ...

Script languages have a well-deserved reputation for being slow.
But perhaps it was your mistake for selling python as a "scripting
language". I'd call it, "a modern object oriented rapid prototyping
system" if it were me.

One of Python's great features is the relative ease with which you
can build external C connections to python. Or... the opposite. 
You could have a C/C++ core making occasional use of python scripts.

Furthermore, the python source core is VERY well written. Having
direct availability to the source allows you to investigate python
itself for bugs, right in the debugger of your choice.

A while back I mentioned a desire for a priority queue, and
a fellow c.l.p'er hammered one out in python for me within 2 days.
When was the last time one of your "closed source fanatics" had
the company which was supporting them assign a coder to craft
a solution for them immediately, without hesitation?

Never, I bet.

I mean, dang. There's people here on this group that, if you
allege a bug in the python interpreter, will compile a DEBUG
executible and trace through to verify the problem within
hours of you mentioning it.

Can you say the same of those closed source solutions? No?

Cat got your tongue? :)-



C/


From ge at nowhere.none  Mon Jun 26 17:02:45 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 26 Jun 2000 21:02:45 GMT
Subject: SMTP with Python 1.6
References: <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com> <8j7n8d$4u64$1@www.univie.ac.at> <VqK55.1715$iN5.345609@ptah.visi.com> <8j7vnb$1kue$1@www.univie.ac.at>
Message-ID: <VxP55.1764$iN5.361073@ptah.visi.com>

In article <8j7vnb$1kue$1 at www.univie.ac.at>, Matthias Huening wrote:
>
>> Gosh, on all my computers, "localhost" is the address of a valid
>> SMTP-server.  Are your computers all broken?
>
>Because of this line in Colin's script:
>
>> #!c:\python\python.exe

Good point.  I missed that.

>I assumed that he is on a Windows-computer. And that makes it unlikely that
>localhost is a SMTP-server. (No, I am not going to discuss the question
>whether Windows computers are broken by definition...)

;)

-- 
Grant Edwards                   grante             Yow!  ... I have read the
                                  at               INSTRUCTIONS...
                               visi.com            


From parkw at better.net  Tue Jun 13 17:21:09 2000
From: parkw at better.net (William Park)
Date: Tue, 13 Jun 2000 17:21:09 -0400
Subject: help formatting number
In-Reply-To: <8i645i$lht$1@nslave2.tin.it>; from mtbros@tin.it on Tue, Jun 13, 2000 at 10:01:15PM +0200
References: <8i645i$lht$1@nslave2.tin.it>
Message-ID: <20000613172109.A6362@better.net>

On Tue, Jun 13, 2000 at 10:01:15PM +0200, Mario wrote:
> Hi..
> this is a simple question but I can't find a simple answer... so help me.
> 
> I have an integer variable and I want it to become a string variable
> formatted so that there is a dot between each three digits of the original
> number. An example:
> var =3456789
> It must become:
> "3.456.789"
> Ideas ?
>  Thanks
>         Mario

def f(b):
    c = ''
    b = str(b)
    while (len(b)>3):
	c = '.' + b[-3:] + c
	b = b[:-3]
    return b + c

>>> f(str(3456789))
'3.456.789'

--William



From perly at news.xnet.com  Tue Jun 20 17:49:09 2000
From: perly at news.xnet.com (Pat Szuta)
Date: 20 Jun 2000 21:49:09 GMT
Subject: Access violation when no memory
References: <olD35.82$_%.4481@newsfeed.avtel.net>
Message-ID: <slrn8kvpml.flp.perly@typhoon.xnet.com>

I think that this might be a problem with the python GC.  Basically you
create circular lists,  and because their reference count never goes down
to zero, the GC never frees them.  For example:
ython 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> while 1 == 1:
...     x = [1]
...     y = [2]
...     x[0] = y
...     y[0] = x
... 

This will eat up your memory like there is no tomorrow.  I think that the
problem that you're describing my be similar.

On Mon, 19 Jun 2000 22:20:46 -0700, Phil Mayes <nospam at bitbucket.com> wrote:
>The following program progressively allocates all memory:
>    size = 1 << 20
>    list = [None,]
>    while size:
>        try:
>            mem = [None] * size
>            mem[0] = list
>            list = mem
>        except:
>            size = size / 2
>
>It fails with an access violation.  (Turn off virtual memory
>to provoke it.)  This is not a theoretical situation for me:
>I have a beta release in the field that is crashing like this.
>The immediate cause is this:
>
>1. we are handling a MemoryError
>2. the following call cannot allocate a traceback object
>     ceval.c line 1800:    PyErr_Fetch(&exc, &val, &tb);
>3. the NULL pointer at tb is pushed here:
>     ceval.c line 1816:    PUSH(tb);
>     ceval.c line 1817:    PUSH(val);
>     ceval.c line 1818:    PUSH(exc);
>4. later it is popped here and its ref count is decremented:
>     ceval.c line 682:    case POP_TOP:
>     ceval.c line 683:        v = POP();
>     ceval.c line 684:        Py_DECREF(v);
>     ceval.c line 685:        continue;
>
>This is with 1.5.2 but the crash still happens with 1.6a2.
>At this stage I am way out of my depth, and  the only way I
>see to fix this is to change Py_DECREF to Py_XDECREF, which
>is a (minor?) inefficiency. Can anyone recommend a better way?
>
>And should I report this at http://www.python.org/python-bugs?
>I do not have access to the current CVS tree.
>
>TIA, Phil Mayes
>--
>pmayes AT olivebr DOT com
>
>
>


-- 
-------------------------------------------------------------------------------
Pat Szuta						www.sourcelight.com
Software Engineer					"Guiding you to
Sourcelight Technologies				 great choices"(TM)
(847)-475-1000 ext. 209


From ionescu at pik-potsdam.de  Mon Jun 26 04:02:50 2000
From: ionescu at pik-potsdam.de (Cezar Ionescu)
Date: Mon, 26 Jun 2000 10:02:50 +0200
Subject: Can anyone solve this?
References: <8j737m$t5a$1@nnrp1.deja.com>
Message-ID: <39570E2A.BBF53AB1@pik-potsdam.de>

[Rene]

[...snip]
> I need the code to give every possible combination of the lists in
> inputlist.
[...snip again]
> 
> Anyone able to solve this?

Uhhh, I *think* so:

def list_comb(inputlist):
    outputlist = [[]]
    for list in inputlist:
        newoutputlist = []
        for incomplete in outputlist:
            for element in list:
                newoutputlist.append(incomplete + [element])
        outputlist = newoutputlist
    return outputlist

if __name__ == "__main__":
    a = [1, 2, 3]
    b = ["c", 32]
    c = [1, 2]
    inputlist = [a, b, c]
    print list_comb(inputlist)

This prints the results you wanted. However, it probably won't do what
you want for inputlist with one element, or with no elements, so please
take this as "a first approximation" only.

Best,
Cezar.


From speed at ?.com  Fri Jun 16 17:00:05 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 16 Jun 2000 21:00:05 GMT
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151553280.8145-100000@fep132.fep.ru> <3949028F.3B512AF9@darmstadt.gmd.de> <394A27FE.3CA5D365@darmstadt.gmd.de>
Message-ID: <skl5alf6is440@corp.supernews.com>

shecter at darmstadt.gmd.de (Robb Shecter) wrote in 
<394A27FE.3CA5D365 at darmstadt.gmd.de>:


>>Surely this level is higher than HTTP.
>
>I disagree.  I think that an HTTP request can be seen as a black box
>with these "knobs", or inputs:
>
>Destination URL (host, port, path)
>A set of parameters
>A set of headers
>A method (GET, POST ... I'm not interested in the others)
>
>...and then, this black box should have a big "send" button on it,
>which results in the usual python 3-tuple popping out.  All of these
>items are specified in the RFC.


Technically the 'set of parameters' isn't -- RFC's 2396 and 2616 don't 
make any reference to the format of the URL query segment, and 2616 
doesn't make any restriction on the POST content type.  

GETting 'path?name1=value1&name2=value2' is a convention of web 
browsers that submit HTML forms, as is POSTing this:

   Content-Type: application/x-www-form-urlencoded

   name1=value1&name2=value2

but it's not inherent to the HTTP spec.

Well, 2616 does refer to 1867 - 'Content-Type: multipart/form-data',
but in that case the 'add_parameter(name, value)' method won't be
sufficient; you might need (name, filename, mimetype, filedata).
 
-Speed!





From bwarsaw at python.org  Tue Jun  6 14:40:28 2000
From: bwarsaw at python.org (Barry A. Warsaw)
Date: Tue, 6 Jun 2000 14:40:28 -0400 (EDT)
Subject: "syncmail" in CVS repository
References: <20000606145723586.AAA450.401@hermes.cinetic.de>
Message-ID: <14653.17820.796934.529321@anthem.python.org>

Thanks!  I'll take a look at this when I get a chance and commit it to
the master copy (which happens to be in the Mailman project's
CVSROOT).

-Barry



From mfletch at tpresence.com  Mon Jun 12 02:19:31 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 12 Jun 2000 02:19:31 -0400
Subject: Newbie: displaying binary numbers
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33AE@VRTMAIL>

Hmm, here's a naive implementation which will work for positive numbers:

>>> def bits( integer ):
... 	result = ''
... 	while integer:
... 		result = str(integer & 1) + result
... 		integer = integer >> 1
... 	return result
... 

See also:

http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=621175578&fmt=text

HTH,
Mike

-----Original Message-----
From: Steven Adams [mailto:adams_s at lab.eng.usyd.edu.au]
Sent: Monday, June 12, 2000 12:55 AM
To: python-list at python.org
Subject: Newbie: displaying binary numbers


Hi all,

I'm trying to display integers as binary numbers, I know there are the calls
for making hexadecimal and octal strings, but couldn't find anything for
binary?

are there any modules, or operations not in the docs?

all I need is to convert a number to a binary with n bits.

thanks, Steven


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



From donn at u.washington.edu  Wed Jun  7 11:22:22 2000
From: donn at u.washington.edu (Donn Cave)
Date: 7 Jun 2000 15:22:22 GMT
Subject: Calling external programs: ioredirection
References: <393e4e6f.72239515@nntp.interaccess.com>
Message-ID: <8hlpbe$2npu$1@nntp6.u.washington.edu>

Quoth olczyk at interaccess.com (Thaddeus L. Olczyk):
| Sorry for being very brief and sctech, but my ISP news server is
| flakey right now. I can post now but maybe not in ten minutes.
|
| I want to to be able to call external programs and read their output.
| For example: I might want to call grep and look at the results one by
| one. 
|
| I want to do this in order to do things similar to sh. For example:
|
| wc -l *.h *.cpp|awk 'begin int i=0; {i+=$1 } end print i;}'

As already suggested, os.popen() is good for that.  You'll give
it a shell command as its first parameter, and it can include
any kind of shell expression like |, etc., because it actually
does run the shell to interpret the command.  (So it's a big
potential exploit if any external data goes into the command.)

In general you can do about anything the shell can do, from
Python.  The problems start when you try to use pipes to do
things that the shell can't do.

	Donn Cave, donn at u.washington.edu


From benji_york at cal-na.com  Tue Jun  6 10:18:47 2000
From: benji_york at cal-na.com (benji_york at cal-na.com)
Date: Tue, 06 Jun 2000 14:18:47 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <393CCEE6.E5C1C267@marconicomms.com>
Message-ID: <8hj17n$rj$1@nnrp1.deja.com>

I've been following this thread on and off, so I'd bet someone else has
already expoused this point of view.  My initial reaction to the case-
insensitivity question was: "Great! Now I can impose my naming
conventions on other modules, and even the language itself."

I'm one of those who likes classes, member functions and stand-alone
functions to start with an upper-case letter and variables and member
variables a lower-case letter.

I'm not saying this is enough of a reason to change the language, but I
definitely see an advantage to not mixing naming conventions in my
modules just because I want to use someone else?s module that doesn't
name the same way I do.  In a production environment this can be quite
annoying (or it has been for me).

Example of what I would like to be able to do:

from Module import Thing

foo = Thing
for key in dict.Keys():
    if anotherDict.HasKey(key):
        print foo.NotTheOriginalCase(key)

Perhaps there is a better way to get the same result without being case-
insensitive.  A "case-mapping" directive or such, but that seems to
over-complicate things.

Along the same lines; identifiers could also ignore imbedded (not
leading or trailing) underscores, so I could use dict.HasKey() instead
of dict.has_key().

I think both of these suggestions have /some/ merit. After all,
identifiers carry meaning, so why not ignore those parts of the
identifier that carry little or no meaning to the reader.  If the
underscore in "has_key" is meant to separate words, why not let me
separate the words like "HasKey"?  This has the added advantage of
removing ambiguous situations where there are two names with the
same "human" meaning, but different "machine" meanings.

Some have said that this would make it harder for people just learning
Python, but in training the people where I work (long-time COBOL and
Fortran guys) I've found that the differences between dict.has_key()
and HTMLgen.TD() and pickle.dumps() causes some confusion.

Benji York


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


From bmcd at es.co.nz  Fri Jun 16 21:31:38 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 17 Jun 2000 01:31:38 GMT
Subject: numpy bug?
References: <slrn8klh53.56e.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8kllct.120.bmcd@ripley.sulaco.net>

On Sat, 17 Jun 2000 00:18:34 GMT, Huaiyu Zhu <hzhu at knowledgetrack.com> wrote:
>>>> from Numeric import *
>>>> a = arange(16)
>>>> (a-5)/4
>array([-1, -1,  0,  0,  0,  0,  0,  0,  0,  1,  1,  1,  1,  2,  2,  2])

Numerical Python is rounding towards zero.

>I believe this is not correct. 

I agree, this behaviour is inconsistent with "stock" Python. In general, the
choice of rounding direction varies between languages or implementations, but
fortunately Python fixes this. See below.

> And the stock python gives correct result:
>
>>>> for i in a: print -i/4
>... 
>0
>-1
>-1
>-1
>-1
>-2
>-2
>-2
>-2
>-3
>-3
>-3
>-3
>-4
>-4
>-4
>
>Any enlightenments?

"Stock" Python is rounding towards negative infinity, as required by the
language definiton (see "Binary arithmetic operations").

The C (1989) standard sidesteps this issue and doesn't specify a direction of
rounding, and merely states that the identity (a/b)*b+a%b==a holds. If either a
or b is negative, not even the signs of the divisor or remainder are specified.
This vagueness is (I assume) to support the various hardware implementations.

The new C (1999) standard (or at least the C9X draft I have) requires rounding
towards zero, fixing the vagueness of the above standard [or at least breaking
it consistently :-)   ].

GNU C rounds towards zero (on all platforms).

It seems that Numerical Python has left the rounding direction to the
underlying C implementation. As a C programmer, I'm used to not relying on the
rounding direction, but need not be inflicted on Numerical Python programmers.

This should be fixed.

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From alan.gauld at gssec.bt.co.uk  Thu Jun 29 08:15:03 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Thu, 29 Jun 2000 13:15:03 +0100
Subject: Any other good tutorials?
References: <sl2i51h0e7f149@corp.supernews.com>
Message-ID: <395B3DC7.D83FB4CE@gssec.bt.co.uk>

SugarKing wrote:
>      Are there any other good tutorials besides Guido van Rossum, I don't
> think he's very explainatory. I like full explainations:)

Mine is aimed at non programmers and tries to get them to the 
point where the official tut makes sense. It has a lot more 
explanation than the others mentioned so far.

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

Available as a zip from the site and prints out at 
around 90 pages.

The book version will hopefully be out late this year...

Alan g.

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


From ray at commsecure.com.au  Tue Jun 20 08:31:17 2000
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Tue, 20 Jun 2000 12:31:17 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no>
Message-ID: <394F0C4D.3911FF90@commsecure.com.au>

Thomas Weholt wrote:
> 
> Hi,
> 
> I want to convince alot of my co-workers that OpenSource software,
> especially Python, is ready for big business. ( is it ?! )
> 
> How do I sell an OpenSource solution like Python to hardheaded
> closedsource-fanatics? I know they care about money, so that would be
> my best angle. If anybody has any good examples of companies using
> Python, or similar technology, with profit, that would be great.

Having been involved in several Python based Online share trading systems for
banks I have had quite a deal of experience in these sorts of "debates".
The interesting thing that emerged under any serious questioning is
how appalling their experiences have been with commercially bought
and supported software. They all look towards the OpenSource software
with great disdain, and argue that support is non-existent, however
when you ask about software they have paid for, they frequently
tell of documented "features" that they expected to use during the
project that never worked. And this is over many years, they then
address this by buying another package which purports to solve
these issues. At this point when you mention that the great support
that they should expect in commercial software doesn't seem to amount
to much, they start making excuses for the manufacturer ... really
weird!

By way of example, we managed to use Python/Linux/MySQL with a very
conservative international bank and after a year of 7x growth, no
problems have emerged with the solution. The bank initially developed
with MS SQLserver, but it never worked so we swapped it out to a MySQL
system in the last few weeks of the project (since we were using Python,
we had enough time up our sleeves to help their side of the
project too!)
> 
> To be more exact :
> 
> My mission is to swap expensive, oversized databases solutions like MS
> SQL 7.0 with PostgreSQL 7.0, crappy inconsistent script-languages like
> VBScript with Python and MS SiteServer / IIS with Zope, Linux with
> Windows  2000/ NT 4.0. I argue that the load on the system will never
> be to great for PostgreSQL to handle as supposed to MS SQL-server,
> VBScript, it sucks and they agree to that, Linux instead of Windows
> will save lots of cash, so will the "reduced" cost of PostgreSQL
> instead of MS SQL 7.0.
> 
> The scenario above is normal in my experience, and I`d appreciate any
> feedback from people with similar ideas. Related Case-studies would
> also help alot.
> 
> The thing is that these people don`t thrust OpenSource. They laugh at
> script-languages in general, and the idea of sharing code is even more
> stupid.

An alternative is to get your Visa in order and move to a new employer
that knows a good thing ... we are always looking ...

They can laugh at Python all they like, I prefer it that way, but in
a competitive environment we have been able to produce highly maintainable
mission critical system in an order of magnitude less time with
fewer errors in a very compressed development timeframe. I consider
it to be a very serious competitive advantage, so caught in two minds
as to whether educating companies such as these as to the merits of
Python and OpenSource tools.

> 
> The growth of OpenSource, Python, Linux etc. depends on how well it
> can be a suitable substitution for ClosedSource-alternatives. A more
> agressive form of propaganda, showing off successful solutions, WITH
> benchmarks, results of stress-tests etc. is vital. Right now there are
> alot of talk about how great Python are, but few actual test of how
> well Python performes under pressure, and not just Python alone, but
> in a more complex environment, like in Zope or using Python as
> interface to a RDBMS.
> 
> I need selling points for OpenSource, the idea and philosophy,
> examples of successful technology and eh .. moral support. I feel like
> I`m in the lions den right now. ( " Think of a happy place. Calm blue
> ocean, calm blue ocean ..." )
> 
> I truly love Python and would like to see it get the credit it
> deserves.
> 
> Best regards,
> Thomas Weholt

See www.hsbcinvestdirect.com.au and www.quicktrade.com.au for examples
of full Python share trading systems, and a few more to follow soon ...
Total lines of code including charting and full failover redundant servers
(done at the application level) 3200, simple enough for even a bank to
maintain.


From Jaime.Villela at fnc.fujitsu.com  Tue Jun 20 15:45:18 2000
From: Jaime.Villela at fnc.fujitsu.com (Jaime Villela)
Date: Tue, 20 Jun 2000 14:45:18 -0500
Subject: Can't Compile 1.6
Message-ID: <C42D90E3FCDD4132.725BB8D5EF6124B6.DE7AEEE25F005000@lp.airnews.net>

I'm trying to compile the latest version of Python and I'm getting the
following error (Solaris 5.6):

gcc  -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H -c ./posixmodule.c
In file included from ./posixmodule.c:3031:
/usr/include/sys/statvfs.h:38: parse error before `fsblkcnt_t'
/usr/include/sys/statvfs.h:38: warning: no semicolon at end of struct or
union
/usr/include/sys/statvfs.h:39: warning: data definition has no type or
storage class
/usr/include/sys/statvfs.h:40: parse error before `f_bavail'
/usr/include/sys/statvfs.h:40: warning: data definition has no type or
storage class
/usr/include/sys/statvfs.h:41: parse error before `f_files'
/usr/include/sys/statvfs.h:41: warning: data definition has no type or
storage class
/usr/include/sys/statvfs.h:42: parse error before `f_ffree'
/usr/include/sys/statvfs.h:42: warning: data definition has no type or
storage class
/usr/include/sys/statvfs.h:43: parse error before `f_favail'
/usr/include/sys/statvfs.h:43: warning: data definition has no type or
storage class
/usr/include/sys/statvfs.h:51: parse error before `}'
/usr/include/sys/statvfs.h:51: warning: data definition has no type or
storage class
./posixmodule.c: In function `posix_fstatvfs':
./posixmodule.c:3044: storage size of `st' isn't known
./posixmodule.c: In function `posix_statvfs':
./posixmodule.c:3096: storage size of `st' isn't known
*** Error code 1
make: Fatal error: Command failed for target `posixmodule.o'
Current working directory /scratch/downloads/Python-1.5.2/Modules
*** Error code 1
make: Fatal error: Command failed for target `Modules'

What does this mean and how do I fix it?  I've looked at the file in
question (statvfs.h) but I don't see anything wrong with it.  Even if I
did, I don't have permission to change it.
-- 
Jaime Villela
Fujitsu Network Communications
Jaime.Villela at fnc.fujitsu.com
(972) 479-2769


From mwh21 at cam.ac.uk  Tue Jun 20 19:15:40 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 21 Jun 2000 00:15:40 +0100
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net> <j8y35.1071$ul.67488@news1.rdc2.tx.home.com> <394EB843.EEB3B68F@uniserve.com> <m3g0q856w6.fsf@atrus.jesus.cam.ac.uk> <394FF41E.E4FE1BCD@uniserve.com>
Message-ID: <m3bt0w54gz.fsf@atrus.jesus.cam.ac.uk>

Bob van der Poel <bvdpoel at uniserve.com> writes:

> Michael Hudson wrote:
> > 
> > Bob van der Poel <bvdpoel at uniserve.com> writes:
> > 
> > > Yes, but I'm using __name to keep the variable private/static to a
> > > particular instance of the Goption().
> > 
> > This is a bit of a wild guess but are you suffering from
> > instance/class confusion?
> > 
> > Cheers,
> > M.
> 
> More than likely.... Perhaps I should have made my example and
> reason more clear.... I have a simple (tkinter) program which does some
> datafile searches using either grep or agrep. I thought it'd be nice to
> be able to change the options fo grep/agrep so I wrote a little function
> to do this. Problem with this is that I ended up with 2 nearly identical
> functions (one for grep, another for agrep); and I found that I needed
> some global variables to store the current and default settings.
> 
> so, being a newbie to the wonderful world of classes, objects and other
> things I don't understand I figured that I'd do this as a class. To
> create the class I do:
>         grep_opts = Goption('-ih', "Grep")
> 
> which creates a class instance. I can (hopefully) now do the following:
> 
>         grep_opts.get() to get the current setting
>         grep_opts.change() to call the gui to change the current
> setting.
> 
> So, here's the simple thingie I came up with. Note, that this uses the
> dialog class from tkinter to create the dialog window (and this is where
> my problems come in.

Hmm.  I know next to nothing about Tkinter, but here's how I'd do
something a bit like this.  Spread "IMHO"s liberally around what
follows :-)

Firstly, don't bother nesting the class MyDialog inside Goption.  It's
just not worth the hassle.

class MyDialog(tkSimpleDialog.Dialog):
    def __init__(self,gopt,**kw):
        self.gopt = gopt
        # this is warty:
        apply(tkSimpleDialog.Dialog.__init__,(self,),kw)

    def body(self, master):
        Label(master, text="%s Options:" % self.gopt.searchName).grid(row=0)
        self.e1 = Entry(master)
        self.e1.insert(0, self.gopt.optSetting)
        self.e2 = Button(master, text='Default',
                         command=self.default)
        self.e1.grid(row=0, column=1)
        self.e2.grid(row=0, column=2)
        return self.e1 # initial focus

    def default(self):
        self.e1.delete(0,END)
        self.e1.insert(0, self.gopt.optDefault)

    def apply(self):
        self.result = self.e1.get()

class Goption:
   def __init__(self, title, setting):
        self.searchName = title
        self.optDefault = self.optSetting = setting
   def get(self):
        return self.optSetting
   def change(self):
        a = MyDialog(parent=root,gopt=self,
                     title="FindSong %s Options" %self.searchName)
        if a.result:
            self.optSetting = a.result
        
> Am I on the right track...or is my concept of how to do this just plain
> dumb?

I think you're on the right track, but perhaps not quite sure how to
get down it.

I might be able to help more if I knew how Tkinter worked...
 
> Am I missing something really obvious on how to get the values I
> need?

I don't think so.  I hope this helped!

Cheers,
M.

-- 
  Indeed, when I design my killer language, the identifiers "foo" and
  "bar" will be reserved words, never used, and not even mentioned in
  the reference manual. Any program using one will simply dump core
  without comment. Multitudes will rejoice. -- Tim Peters, 29 Apr 1998


From drew.csillag at starmedia.net  Fri Jun  9 13:07:25 2000
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Fri, 9 Jun 2000 13:07:25 -0400 (EDT)
Subject: Python Productivity over C++
In-Reply-To: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <Pine.LNX.4.10.10006091252070.10098-100000@naxos.skunk.org>


On Fri, 9 Jun 2000, Steve Mullarkey wrote:

> I have been a 'C' programmer for the past 15 years and a "C++"
> programmer for the past 3 years. I'm not a genius but would modestly (?)
> class myself as reasonably competent.
> 
> In the past week I have become aware of Python. I've bought a book and
> read this newsgroup every day.
> 
> I have read in several places productivity claims of 5 to 10 times over
> 'C' and "C++". I would like to ask for some feedback from "C++"
> programmers who have moved to Python as to whether these estimates are
> realistic.
> 
> In "C++" I have :-
> 
> 1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
> Profilers, Programmers Editor including Class Browser, etc..

Python's tools are very good (and don't need them a lot of the time
anyhow).  The main thing missing is a GUI builder (but glade and PyGlade
are getting there quickly).  Memory leak tools just don't apply (Python
has reference counted garbage collection).  Programmers editors are
numerous and one is included (IDLE).

> 2. I use the STL (Standard Template Library) which gives powerful string
> and container classes.

There is no STL library for Python since you don't need them, things like
mappings, lists, strings, etc. are built in.  

> 3. I use incremental compiling and linking. This reduces wait times to
> very small amounts.

Python - minimal compile time (happens automatically, so no fighting with
project defs or makefiles, dependancy tracking, etc.).
 
> 4. I can also use background compiling and linking.

C++ link time always takes forever, just sometimes forever is shorter
occasionally.  There are may stories of people starting a full compile and
going to lunch since it would be that long before the compile finished...
These stories are true (lived it myself) when dealing with a reasonably
large project.

> 5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
> etc..

All available for python too (most included in the standard dist).
Python's standard library is almost a good enough reason in and of itself
to use it.  Not to mention that the language rocks.... :)

> 
> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated. I am genuinely interested
> to ask "C++" programmers :-
> 
> 1. What productivity increase do you achieve ?

Better than 10x.
 
> 2. How long did you use Python before you achieved increased
> productivity ?

Well, since I was doing Python on the side, it was about a month or so.  I
was doing C++ for about 3-4 years before that.


IMNSHO, after writing Python for a month, I didn't want to do anything in
C++.  Luckily, for my current employer, I program in Python almost
exclusively (some C here and there too).  Basically, what I've discovered
is that:
    1) I write the application in Python
    2) if it's too slow, I try tweaking the slow section in python.
    3) if it's still too slow, I write a C extension.

Generally, too slow doesn't happen often and I rarely get to step 3.

My main beefs with C++ are this:
    * If you want to write dynamic programs, you spend a lot of time
      fighting the compiler, convincing it you are correct.
    * Even when not writing dynamic programs, you still spend a lot of
      time telling the compiler "I know better than you, bow to me" -- 
      by using casts.
    * C++ compilers bite when it commes to debugging templates, some are
      better than others, but they all suck.
    * Memory management sucks.  With Python, it doesn't enter the picture.
    * Can you say "buffer overflow"?
    * The language is so big that very few people know the *whole*
      language and you wind up at any job with 1 or 2 language lawyers who
      write code that nobody else understands.

Python makes all of the above much easier.

Just my $.02

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






From pduffin at hursley.ibm.com  Tue Jun 20 12:20:36 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Tue, 20 Jun 2000 17:20:36 +0100
Subject: Conversion from proprietary satellite control script language
References: <20000616084700.29919.qmail@web2006.mail.yahoo.com>
Message-ID: <394F99D4.C28E013D@hursley.ibm.com>

TJ Judson wrote:
> 
> I'm afraid my only other answer is to use a language that supports
> "goto", like Tcl.
> 

Tcl does not support goto.

I think that it should be possible to analyse all the possible paths
through your satellite scripting language so that you create a graph
of flow, in those cases where you see a goto you *simply* replace
it with something else which does the job.

e.g. Pseudo C code.

   if (x) {
     goto end;
   }

   if (y) {
	:
   }

end:
   if (z) {
	:
   }

Replace with

   _goto_end_ = 0;
   if (x) {
     _goto_end_ = 1;
   }

   if (_goto_end_) {
      if (y) {
	:
      }
   }

   if (z) {
	:
   }

However whatever process you use you will have to exhaustively test your
resulting scripts.


From embed at NOSPAM.geocities.com  Thu Jun  8 10:25:32 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Thu, 8 Jun 2000 10:25:32 -0400
Subject: Stealing from Ruby to Pay Python. SDBM?
Message-ID: <h6O%4.752$227.21053@nnrp1.uunet.ca>

Has anyone ever built the SDBM module as a Python module. Attached find the
source from Ruby's SDBM module, a simple dbhash (21k of C code) replacement.
How comparable is this to the BSDDB's dbhash?

Warren Postma


begin 666 sdbm.zip
M4$L#!!0````(`+D,#2<B6UZ\G@,``#,)```&````<V1B;2YHC5;;;MM&$'VF
M`?_#P 5:T;!N0= X$5K4CM1&B&0;DAZ2!H6P)(?60B27V%W65F[?WME=4B9I
MJ8@>1.CLG+F<F1VJ?WYZ`N>@HB"%+F3F\2#DMLL2OD78,+7!""*F6< 40L(#
MR>3.4 at P0@<C@#F7WBHQG3"J1_:)@O,M8RD-X1VR>W0-+[H7D>I.J'EQ/5S"\
MA,[P]:M+OV<=L4)OA'P#XO,?&3X6JK>C!(I>R%QFFNE"O8&\"!+R&8F4\:P'
MYJQ_>O(3C[,(8V^]'%_/U^_6A-!/GF$-V6,POIZ]7T[_AI>#U[\^H7<E.AR\
M>%E#KZ:+^=4'0@>7GN?U*4\9<&W*MT4[4O?&I5&2EG>S%9&\X<!14O;(TR(E
M!1+Q0&JI/.%:68KG.9-82&"@2*<$@6<*I>;DO^YU/%W\.?FP\LYZ$9=G]13_
M*O&<W9^90O4N1SHDT601:OA"87BF at 6CQR// ?2 at H`1AJ0:7$G,)&J$+)<P)<
M:H9#+IL<`O"X>9RP>]6P=XWKHY1D9X\O0"%"@*2%(R:"IH,T"C+1I/+/""*V
M>7,K"P1[W2PI+$PS&B2")%(FA$-6I '*FOTF96I[T-R,.)C3FG60;',M#WM/
M1+BU3:-9U5O</4E /XZQC-TS3A6J57S%L7IK`1)9U'^@^X,U&AVV16OP2#%C
M4L2.$VZ8+(%/Y>3^,SK86U<?V<4-_:@3Q^(Y!@4T-HV #O at T/A"P-8#/HGZC
MNSH?->_N?+T8W][,/GJ#QV'=%2TGNXQ Y)A9N;HB2W;-*T3LZ>UDL2#RBX/D
M:?\6W*P:WNE)WRW&0G.ZLCN:D% *56V=TJO9FFL9F6B=*/ ]KV,>W=_MM,//
MM9S]%LE&.LJQF?KUZBTI3)!)8EI>D_8;?'\BVLI$$6[*4AI.S)W*1-V%0=KY
MF=O?M#*(?V3'V&Z?1V;ZJXUCKO#(])$4+E+;29I]E)D#Z(8FI"NZ$]KBX*)'
MG?5ZN1J_7:]]^/IU#\Z7X]ME+<=<"BTZN0\Y89 at H/'#4,?:813RNM=/)1;?*
M5JEH`K'54ROCS7*R6'F#UO1-[F97;R?>L.;/OC"I8)0Q"RM79:'$@7,;QPZF
M2ZSCM+HP)/OE^Z,]XU_!HZK50E6U=*RCNIV3T!K&J*G/=<,+=URW-QUQS<<$
M-?ZPN=7GD/7^T2J at GAB7RNZZ'RBB6HO/3/<Z"[U!>4S>7&+>EO?_9+8;S3+M
M\G_>F#*VFQ[SDJ[^3-CX_P%02P,$% ````@`Y&A0)S/K_&5%````D@````8`
M``!D97!E;F2++TY)RM7+MU*(!S.2%50T,E**4C*+-/6+2I,J]3* `B7Y!6"!
MY/R\M,QTL!!,34IJ6F9>:K%>!B]79EYF"<@@,$V!.0!02P,$"@``````N0P-
M)TV V;XK````*P````H```!E>'1C;VYF+G)B<F5Q=6ER92 G;6MM9B<-"@T*
M8W)E871E7VUA:V5F:6QE*")S9&)M(BD-"E!+`P04````" "Y# TGK>'J(E()
M``"D,@``!@```&EN:70N8^T:^V_;-OIG%^C_P/F&0DZ5QDD?:Z,F at Y=Z788D
M[3GNMB(7"(I$QUQDR9#HK%XO__N1_"B2DF5)=I);N[N at 2*./_-X/?GQL;:SX
M\_#!PP<(I<'%A$2$/O'1)D"^[<WH.$YVT<2C?Z)O!>B-1_$NVG[UZM56]^76
M]E/4?;[[]+O=YUT8]Q/,)@3(H[OHQX2 at 8V^.T'>H^W+WZ;/=G6?HY].A0 8&
M!_%TGI#+,47608?#GV^R7R_1Q]D5&9,D9N at TG4UB&O/YM7H4?K8XTC](Y(>S
M`*-V,KN8/QFW\T"NM `JT.N1']'PR7C?A.$DB6()&P5XA#Z<]MV#7]]^.#UD
M(*3FI?-TRYB+HX",.,-?>D<?^L at _??/#L<._4YK,?(H8\\"C'OK,C8$8&8I0
M0-R4_(D=`#$$M,% ;":#W$ADCQ(?7<<D>/C #^,4!R[7P^H\?" I)1=NXI$4
M6^P//)A%E$QP/TGBQ$9MP$!<%C0B(6YW.&5A%J89B3!ZBRD;M.*+WVTNXK2#
M/O]+BL.D==FP>RH4@"EY922*(S'("%D<L+D/6J"]/=3MH)S8?.[- at F:C!&.8
M( @"N8+A-OB0(]0N9]<18>T*=OD!:6#.!1 at X>2&$RY at 4`C]UXRF.+"^Y]&W$
M?E_;Z"KTTK2C'<?')$WP]@:?EX,(%$<Y"8#<!3:ZGL1!WN?"X14J9WP-1"#(
M?.*8YF AD/I>Y#)QTIP&[>WMMHT>@0"/A 0=[I[M#@_(%O]&S%DO7KQP6JVM
M#<2BPYN%%%U[X0PCGEZ<Q0W\A\,4"W8GAT?N>PNHY0AM;@.9!--9$J&(A'S^
MFQ]0%%.$/Y&4EM(T*)Q\.-XY/!E*XDYN[L$8^U?NJ3?"+#0MKE-'64&$'>HZ
MVB:"X#Z/1$8=AH6?A9<'I\/!X<E;(+*Y/Z4L:]ZY at S>_#O[]SCT8]'M#&YD2
M<(+?\)!:B=;M:;P[.?JHJ13H"+LI5?>X^3N9Z?_);,\P6CPRYJD[\DA8PL0P
M,/S!`HO))4K L7>%LQH at HMHN%(&NK=+7SK)+NL(H!4A'N1K at Q0^B)>,KI8:X
M7IJBD.)L4F<Q&3Y79M+*I6V%RB89!Y4UJ&"4;E%SZ:^EJH\P]<<@\!6>,YF+
M)LC at VA1,G=F$0VU(Y_I:8U0E,^.&\RFV at +J-ABY$4:88 at S\)6"@QI;+XD@**
M"#-G2;<O3 MQI#@NK$O:?A!+RHIR`"J5S"6P$@](SM'THI at F!14^Y+F3MSV?
M>I-S"DL>ZK$"S'U-$S?"?QATI%5!K<J5A40!_H3SA5D'<<.%)1?E`&+R.)H&
M439D<&80)KR7S+G0.X*S6 at WC!%EDK^L@\EKP1>3Q8VD/0)G.TK'%\&RT$'M<
MK#-RWEDH'%G%%R(MM42 0TSQ?SV,>1'$_BS!UK-,\"\\LF\?V[(M(!0G'HT3
M=VJQI9H!Y at 2'02:ALRP)-''3;XHYL% RF_6\5=67:EJ(KT at X:!>6>-5BY&GO
M*[5R\,W-TK2M6472,1G1ZE4D'W,K19RFF,45HP`!O308UP at -1EP%!DE2RKZY
MS3H.TJ[[)HO at 8E? )S"IJB.KU/$F_U2D1DF-S-C:.CDZFNM2-#:F"VL>3==C
MUH?$?L;%,&]E`08M7#+Z>KW.!T3E7N[XS.R.&1P1_J2F0 95.KZUAE];:SA5
ME*?!L'\ZM%1-JO)N1PJ?A79%5:HL0 :>68$$718^K9LUZHD?8B]I&%DK!=7=
M18X"F97Z[D*JQB-2BY6=LHXS2'2-D[^DNNNAL9>.(1OX7R*<=>'\G\[ZS"1>
MBJG%_\KLI]I!9R6?8\\?NU./U*2?7CY&L\CWPA#,S]LC)G8266U%J-VQ49?]
M*UU/9#^<LGX*(]>R!- V?G? SWDA9U,646SUL3A]X?2<I JJY36:4Q@=NKW!
MH/?1;/H&`B+&H=%$K]&.:N-UKO62RRS/^%PTF:4476!T=J4ZZO-B%V:H"7&2
M9\8<>]8]+X-NGQ=6[+IM+=@&W!'3,2[9#PBPX<NLI\66<IJ<9*/,U+"_RHM2
M4S at 2/ T]?R51BN7?N5\)P1^QWB.IU7%!T(7B]/G6E4]'7I;X>\*[$')+MW=.
M]0ZC9'$S:Q&CQ&H%+RLDNC0V+6KF76[0C)--V71DCJG2#G#-8I at 76K>G<EY1
M9#G!%%D5S;)IIL at K'V@%RQJ!I<N/6MN-<I %D,T#Q!WTWQ_U#OJPT,B[DY]Z
MO_1=/GAPU.L/^H,!"P.=+3A)8/72]R>BAQ#7*CRP^N_[@V.Q756'E]UL`:EN
M(Z V+VXN<XFFTF)IKC$C7]+Q?31UPJ;\N :9IX"W.YF$/3+?(DL%#T^&.S\>
M_I:?4=EIR(IV-^W at X\?:\&511YR"0S)Y2>FRFSD%3Z9T[DZ_*J^\- at XNRJS>
MNAN3"^;"[(6N6=VO$).[<?MXHR4G\O at HJ]=,T6Q6!AJQS,&5/N(KGN at J[JT'
M!\ 7V4>;>]DF/?&*[:YA8B[5/>3!7VK;*NM=E>Q"UK=>_9:A-#Z;W]I6G at 6M
M<=SW-]L$-CCPT;Y=Q;N<RGVM#T#,2^:W<21#1^;=C+H_O+,44C<X[ ^[J2^7
MF1NT76IN4>B;&OPVIRU?MM5K%X5:G]0L#JOX9,SV'E>P.JQ[R;:2DP#P==P4
M+_<2GR WMX4KDW6Z(.X#U00MWZ:K=>$>O' MS]6*7KCC'>@7V88I1X(&Y2J@
M1X]DXSS!$W\R-7*PS#!F9D*?G)UL+\9)OB&I#Q<:N][_B^C*150W#PU7.9L1
M8C](_#2MP at W*,#QF/(P(+3[2%,]!P6KP\-+U^5,JJ\WA;:&%_^[B=^S3C(TX
ME!9/3=U)',Q"; D:8NJD'\TF./$NS)=OFG1*HLL0TSAR)YB.XR##;/,G9NWL
M-0>\<[3Y8S&G,05FG48$"FCB/91"%%_R;+T2[>R\G7M[8J-Z5F?G>UI"?@AD
MHYU:)/DRAR&BW%.=1LJQN<1?%QD.F)3$\%EJ&R\D7IKAH38/RK8F4,M(')I\
MKQC),Y0F7N [)8VG3[*;8<+ZE\<7H,8$6-[FT47Y:X at L[G+6$IXW&PJ3?S1!
M at D98H<%G$T3Q&$4'+O]J$NUP%JWPX+,YHDM&!5P&:2*N.+XU$II]U:.UX5)6
MH<%G`SRX+%%X\%FO95O>Y2A$^2TQ:]-:5%^=,K*A;F)=.75M7!$W>6R9,_7X
MZ_*MX5F#S5L7A<L_]-WI?P!02P,$% ````@`EDG(*/%,L-BS`@``V 4```@`
M``!-86ME9FEL9:53;6O;,!#^'(/_ at V"&VJ&VF[Z,X5%H&KMMAMN4Q"L=ZPA^
M41HOLA4LAZ9?]MMW)SF)R[(7F#](Y[M']]R='DUN at C DY\1-\M(5<UW3M7?P
MD4D=5S7A,R)>14T+DO)REC^OJKC.>4D$37%W"&+QC*C2+*\@S\!SJU7R:O><
M4^?4I>O:%5E2Z-K#?3^Z^4-<UVJ^W)-"U^99M=?O!Y/('XXA@*<'`S!2)LVK
ML'\](00<MC^\O1^-(V(/#5,16-)62<$N.>//'(!WT7F/V*,UD?%E16?YVG+S
M,F6KC *%'_HJ[SF!O]"?W/3'@2])B1WZLHCQY\LOT^'=).J'X?2N?QM &$O&
MF,J(DUZ)RF4\C:%8NJ;I=!O9TNH:RQ/5M&&V,)8+?KB at Y>*Y!5"FY1KF+_R6
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M-&:JOH[<T)>7HL8NX/R%4A!T7I%9S3D3Q*;DX HH/:\`<AB%,+LPTP?K`%M0
M+]#:2NJ-;R>D-^Y&:];?^)JZ3&3[>O3MD$BC!\;1V=G9(:FK%555R$'NY7#;
M%[.Q/8"BEK "P]P\9)F@>>)0$&\GEFC848IRWVK:DD]"UQI1>HTZ4[)3CI36
MO*5\]6"D:X/)X%I+*IPY7@>*VE/B_H\\/P%02P,$% ````@`:II4)_IGM[DT
M````-P````@```!-04Y)1D535/-U]/-T<PT.X>6*+TY)RM5+YN5*22U(S4OA
MY4JM*$G.STO3*TKBY<K,RRP!R8'59/!R`0!02P,$% ````@`ETG(*%A=;C E
M`0``)P(```@```!S9&)M+F5X<&U0RTZ#4!0<:I6:&&UTZX(%)I at TI+HS+MIJ
MFTA*H:%T36C!AJ18`OB(&S_$/_!;_ Y_`\^]/$JC$V8X=^;<`SFZT,#W9^_F
M0@!:X!!4WW-3%Q5&1(_X59P/B'UZ5,]?/*_D6=G7(_Y@%WVT[]B;?:/TA!K_
M1^(M0M5;KZ$]!:G#3L 1^0VBB'/2/5Y)I$U>R:3[O%*X,G:0_ZO([YW at D,]N
M\>'^6Z3 at K"T6HU"T9M+I)%C&FV3SF$J*=2GIFC%&?[D)(S7P,JF)+&,WZBL2
M^(SDW7!#'[BNG'@5O;!<J3EY5Z?FO)HQ[;9;.;+1)5SAMG(8CK%%`S'I=C-.
MK72",')V#&TR-2W;&8YF]Y8VM4VK#(RYKO]-\<'B/+0?YL;8&0[L`7X!4$L#
M!!0````(`+D,#2=NT+>J:1@``!]3```'````7W-D8FTN8^P\:W/;.)*?E2K_
M!R2[R4BQ+$OQS#BQ8L\Y8V62VKS*3F9G+N=R421E\4R16CZB*)/<;[]^@0!(
M^C'/NJTZIR*+0*/1:/03#7K[_L8M=5_EP72AME2"OU9I=K'EQ=%%J.9>/@\#
M%7B%-_7R4,71-/.R-0W!AD"EB7H39EN''D"_\+(\3;[*U=$Z\1:1KY[!\"@Y
M5UY\GF91,5_D`_7D^5LU>JBZHT>[#WL#PN25Q3S-]E3ZZ3^2\&.9#]9 03GP
M/2:M\(HRWU/+<AH#SB!=>%%"`ZG;3[-096E91$F88\OVQJV-6W^+9DD0SH#@
MI-BXM7W?GWL9+?(L\_,H>'^J]M6=OS\/]M09M@[\_@<U&CQ0HT>/'FT/'VZ/
M=M3PF[V=W;UOAFKA%9_4Y.-2_?W.&+'_+4R":,:S)'Y<!J&Z0TCF=^PF/TUF
MT3DU(@6_E<U%F1 +DT MTPP`HC at JUK#L)"^RTB]R]3Y)"Y6$7A:O59BDY?G\
M]'JV,I> 1< V]>3GMY.3Y__9Z3P4WB'KGAW^.#E[]^KYR=NCLV?6PAX#07D1
M#.8'+BMHT).3HZ\?&,0GD\D_SDXF;SLO\+-J[RS"11X6W;SO]Y->9_HIS%)X
M4/"]LWT?9 I6LIJ'B?)5E"OL)8*MT?YRW<U'_?P!C?=3?'S05] $2%S(Q=*&
MM)YL\F5_H at 7Q."DTVY=>MLA5=^[->PVNP>*.WKTY(9*]>.6M<Q6$1>@7*BB7
M(*I>$>8VW>K)X='+R0D/. ^3, ,(Y:E%F.?>>:AF:89"D1?*!R'8V[C5Z724
M at H<$-WB!&I:EZ4)YLR+,U,F;%V]?'OZD<IBJX'ED$8LH]X58V96CR9-W/Q at Z
M at G!:GG<_]CK+#-1CICXB)^(\;$*T[?#S'\XF!O*'R=NSDV>OC]]VEWT5]3K=
M;K=,\N@\"8.>^:I(_^[WNLM>[WTWZMU_<*H>/U8/>VI3=:\"@_[1:<_LJ'KS
MSIX/=KQW-0)4=+<;R()!!P<P>U]=.W?+\)SH<1G69,.R]SXZO9)P#813"$XC
MCL:.++TH`RN"VXFF$$P@;!K(Q2PJL$LML[1(NUTFOX^=O=ZX at OV01H$"V]D*
M"P:G7, at O>Q U@(A>/LB&9G*",/X5T/[\8HDR[T"WDI!<A.L;K1$4H8FRK_3O
M.$W.:9#(L6AO?17E\NI5M!C_QV -HY3MH=T&5EP::<*7)T>O3VP0/48P6F/7
M^7:Q7H9Y'2>T([4V5FUP':!9%(>"FZ6TZIWY21%3ER5BC\%,IMD:FFWWIEWH
MZ[,GSU\='O]L1%FW=(;M_ BS+-'\8!R3-Y/CEP8!/78FA]]_/SEI94 !ANF<
M,%0+/3L#1_3]V5F-R=#5]$0TYZMW+UZ8*?')I5>L9?@1K&GBQ;DQF30\+Q/S
M('O'L"0JM$9+'AIXP9JOO"P0K):0 at 4P'TZC0(G;TY*4CGA9D?F-(4E4C^Y= 
M&IV"A3BP=7S@;,C5-/$9Y:M66N;AK(QAC)^E%1>%[5,O0$<#INYC;Q LBTSM
M[]->J,^?%;7ET2?8236T3&7X$0.B;E2$BUZ'@C9ZY@;"`H9;'G"X-31*86.Z
MP12-*WQN'<QB[SQ7G_<5K.'L^>O)\7'/]N&OGSX]>W/X0S>=S6 (,4S1`ZSK
MS9,7_X"HR 4^>G[<"GRD at 2L^8C\P);_(*=K\!?SYT/IYT5?6T\A]W'G1=\%W
MG?Z1^[BS6P/?=?I'[N/.;@U\U^D?N8\[NS7P7:=_Y#[N[-; =YW^D?NXLUL#
MWW7Z1^[CCG[4X+M._\A]W*D>!7S7Z1^YCSOFD<%W38.RH/EQQWHD\%VK05G0
MNV" OI"JL.HE91RCW*(XH!(`,'>3 at H'DH*RGRS#IH at 7O*Q+>OEJD`<IX%IY#
MX VA'WLE!!E;K:BW-*#>B,.A#06P:J<)@^G8;F.T0036<!&V]("%:?0 at _H26
MT(EFBLBV-?PV4=G#4#8+BQ)L)]E-6/_D^:L?#X$!8G]HQ+ at R*4D(T5:^]'R.
MBHM5"L9PR0$S8D0ZM*GI)( ./$8L;$-MQ-A-FD!AGTY^>MLS+:#ONN6!(;TK
M"P=D"R^.4]\*9"&GZ.EEM2[FU>N7DY>7+69:1G&@BCF3KAS:S:1 '&0,7?P%
MR8RT]VD(Q*EZ%<A\V0<98X#-"J4!5S at 2%,V!,)DT])5F";,CF"((RN(R"Y>&
ME at K<EDM$/,O"L KBE)Z=!84819+VI2[I-\"^<:LNE759M,7^!M+.23,=*J@`
M/VT)P'7K/10A0/^2SK"Q]YMD`#9:?H*I]D;[:CAVV^?H)UK:I_$%^<Q&!P3%
MID/DS O^NX3D$=QQQAQ1>0IB!PO]Y_'K5R]^5M/03T'VU/'1/X_[BL?D*@O_
M5489Z-MT#="0;X/:74 P,5"'<8YZ5ZBT)$VG.(F#&T!/FGE\A)@'6IS)!M#4
M]R!*Y&F)67KA5>__5-WJ,X B2>,&`L9."&SNH2OG+K-XM)J5CN5 +1#^KS),
M?! H/#NA#<=^D"0*CVG4\YE:P1 OB@%HO9J'&8"729"J\$.8K0L\O>HKV6G<
M4;U2-Z;O5%&&#HW')B+4LK5U`$([`_+)P+=*.Z2E:FO4HUC!#$.*]3!C%]J'
MV?;36JVB**N 90$?IG&4ST&^/RH,+I-R,0VS:@=YWAFRJYJ\K^Z1JI#\#^UY
MZ&@&<>^I,(+Y,N4I, : 7I]E]15(20K!TPKZH3<'CH)KX\ QH<,+ at D7U!RPP
M'06R((:@?X0_MVCK"$E34KGN;2)KD!=G% :J[]10[0$KQ at 88L#&PTPJ+YU8'
M`9 D49T<B+%M@)\NIIA@$_C82N,M at 3/#OHX3>^/+8)%@@3UR8)&+>,AS'[(-
MWP_S7!YEL;;QA,<O]"GH_3C-PVJ'&!\!-/I1ZJC_2]-23VTC7;-<9*X1F5AK
MC=&.1(QA_44$'6VH921==X^384;*`G[%2BY?1/L*O at BY%&8)O;.P\.<PN*_ 
M6+83S5$9=+<N``,8S%]P^%4AC([JC">1=(PFEV2&D-A8Y'3%$28$8BP"5*4S
MSBRT5#K8IH4&81P6X5^Q4M8A',M'Z0&>U")YK8,PR9<QU_'%6+S;<H[4X M-
MH><@I#K]7 88%J*U(].!]JZR&(@2!"Z\J*2H7^5\QCP`>_5)=0]S41R)GA;8
MLLJB(K0&VV1IQ9?TU9!G;9M>ONX:BKBV[C$!UW8W+]*LVMR^^N#%8OROWVC^
M#B/&;IA$`H"!.W@)UG]!0RDK;HO9LK]62,AO[2/Y<C"PB>3S=[/CX!QXMR-(
M#=(4W- at YY.'\/<>XK<=9`P8S54WEN^_L((4F at FW#1='W`XB!GQ^_//SIVJ6U
MR'(7>09PCDS;?C*B.(.]<@KQQ#+V,#!AS:7%P)AM\8-1QC%6E.7%`$,4<)&P
M$BP"%!RA0%8"0C%PI%R"(XF.)F]>''X_8:$48WJ98HVK:*8Z$NVPE:;]YP/1
MJ]41V:+#'6?%$$@AV7/O0RBE*2IA]+&/&H$;% .X:[DM9]O.K,@]F99 at Y(R*
M-@#9[D!<IEQ"G9[>H at GC% Y'@-?BCF=E/(!0$@+J0F\21RX84;;9';,,8;H<
MO-?91WHLIO[_B(WB8,0)160IO\YRZ2*4G"!N684K2#"(PP56UI!O*039P..8
M&$C#2&>EF at LA(W1Y!3B]98&28C"A>NL"6!%A3C,-H8UQG$<?<()RJ7?#'&QN
MW+I$:EIMJ6T.+7N)YI.ZDG"U1*90,%*LHN2]</MTW, at 0" ;"61N$L:1Q4$ "
M8JE0I\IQ,4*M-K.:ZCY,##TXHV.^Z<P8X_I]S1R6,-!"8XI8QI'[?IEE(0S1
MS+?E5HHIF*8 at AU9]*TW=5*.>424O""#@SU4Z(Z1(FH,0F82Q.IG(>P8-YGW=
M!E((SZGTV+V#X_;4W3CXK^0.D;#L6;.2R*-)\]:8GWF!\I!PS/>V?<\'0O)Y
M.9O%6$;%18.A]0M+5U/D%F3#7..-LCTT6%AI0M^2JO,4/^WU]*UI03P9519^
MB-(RQ_R&LTPL1%?#,.M!">]#"P"!Y,%L;!R;&T"&'FQE'S*DQ1+KWS250P)-
M`:;+$T'/"U20.$TOZ(: M3[8#,K70_0@]C3:D;#QHNJ$8I_BY<K$&ZQZ(#=Y
MVI=Z,QI&LH3L?!WN:45K2+VV*4\1W3R-)0T<(K<77@!HT/(C1)=Z@;I7K]_*
MAN9LL5B.1$M DNJF$D_2R3).7AV1"$GF]0DY7TO/5G-,AKO:NI)4@>L7Y%88
M>N4D$CAH"UPWOS2Q8W<1KVU[R91RNJ;7M;FOAXB9-04EIDCTIZ8R0O+O\"&_
MS8FX*W$S;6T6[508CRF-+7$R9F)#%6]<M at C:JE;R6VAW)FEXOJ'E=V]+C8W\
M`2(`98''7F. MIYA**%-(WY(TI4;D$D,(\NN0A,K9A*D1;96WCG>:AH,=%C!
M=)#&\VZ#.N at JWRHMXX (H.$08(/YF(:^5^:D^GA<)VC$*MLFH&\",QINHE)0
MNBC!@S Z]QJH)^_>VH&:L4B"CH:_!WV'," ,3L7TE 49$X0#!Q6D*R0"C]C*
M+&3S"O85)K-B/L-X$)\'N"33L*E8E"Y5 at .]@Q%[E/:K.S_LM[N4R]?[SPRT1
MNB]*3-#6%GIJRZ.Q5(%I%WM,D1 &0Z$78]Z%0IT/U&'+Y9^^CFG9(<B%H5E4
M!:T#E7MK<&IID$3G\XKQ^@H!7TL"ZD1?>8$/^GRM;4\CE$#8O7XDX5D^(#?]
M]=?V'0EW[79T2 >U:1RG*PH%5VEU at X\]SQ16?:$0!PH#9DE1FN3(G>0K6!7$
M at 52%3Y7G^VF9@)/KTITZ"!UZLCCG* A3*>!#]U>>7EU_X*/-0N$A9\3'#K5@
MW_C8\"J9'+:(HD at BJNQ-!?'*$R77@ ]UBU-^H!:G[F!V6.X4=/DXKG$4AWU_
M"ONOF%^'IR!-YE;?-$H\,+9%%H$AR\"N93E7-&I9 at IW<H]UI)[OUW.27C5I$
M/@<K-MZHG[($K:VX!6.IP9$Q),;/K>]B/JC[,>VXG&7?NZ=OEH at OTUXLL,TJ
M&51C2^6X>X27\7"6S4W7H"(A$H_C6(C'`_!1=S1Z3:IEO?6ZC.G=ETL0B/^4
M1^ RH;V>$HSK;A;-Q#1._8OJ](3J`BCV:W15<F^)W5A:A'ML-3$B]G+(!A2G
M`C"/\J88M8^&=]4<Z,1"LE.T(HINFVP+'(#Z19EZ"B%GPXQ6.L_+!5[=O(-1
MZQTQTQRR<KD"J:=H?H$E.2J3L&;*L4X]W[K61%SGN7#4%3[KUU@*-[:C*$FN
MZUGT59%G,Z1U;(F1Z$J2H(GH06DBKT&T5R4**SXB1:[II2OA-]!+D4A'+ZG#
M;]'**!-J?9%EM6T5@[!<CX+K0^M1%:UKXX5](D)<HFK-)[B at IJ\2(>!-MHU'
M:=2X;4=7;YNA`A=2+<NH<Y2YFT"4Z$TPVV"F?.^KNWK64S0>9#78$MW53.I5
M]M?>MOS_M^VOVK;+]@O#X4LW3+,">P[V+:_2TS2)DX$L55R&( 9+Y]1*?P/3
M&E'U#9CF!M5UBR%&6U]TW,)O<I("CW3*DU1G)7RJ`G0'NDI=1:C<C=E9-9CB
M.^MR at 1L[2,QC12.7ACMVG08>\4"J.QZ+Z)E(:W.3Q )IWM>WH1TO84 Y`<(-
MH/()AFFWK6!*<XFG-.?@4K?/RH13-RJ<A)D^*:+W=])$R<V0\W PJ'+6BBN8
M@ X&<ILB3G.^,;),\PCC=A[.N2K?:UO)`5-5`@"!<;-!)\[LB'16;D74E<-3
MB%EH?=?E=0Q=%T YSJ"$H]+ RGN9<149-_*D^QJQP7N='S607S8N/V>O5X&W
M[_,!G$^\T]'3O^W+7,B<K3C\$,8W?*4+%_\'O]+%L\B]^FMOR?^N6\NT?ZP7
M?%M<-7(1U(U+7O,P[S8TKF0OY9 ;UDD'X)W-K2M_-CEF3=+.9Y6HSV at DTMD,
MOH"P\!?=0H NNLWF%YK3&KW%_^#GH".WV#Z[=+43*(@ZZJDZ5A/X=TB_#]MQ
MM&/2Q*C'^"2D?.9+1]5/*S6;-705)C3'RL;Q69JN7%<=$['2]V*_C#U3$IO-
M(%K*J4*!5UWPS,/;4SH;XMM:-#*=J3#!3#8'L4K2]\/3GGY;KF^ATH6,R:LC
M:*E\6X5;5],YST+!'*C7Z %6$=[>XGJH-M]),>?)>$*8$4E8?Y7+9*808(=_
MS9-0ZQKGN%%;:UPN=EM at ID8;KL9IS,$&%>I^Q/?$Y*FG>#X`1+&$CB[>'3;O
M;@%X?TC7Z8:8!KL=20^R8NN,GCBXC^2 / $B/&)$"\<71FE*<LF4MVYR`E[O
MW##Y-:&[&W":*\$=MNFS8R?4X>L+^WK==L#--[9TW9E87A6<7;9?=E7D-^S!
MG\]N,6PA5K7TL2;GTCINH(VP[HX@@54T) :WI20!^P8C^TH'37V#@;;/O+='
M5-&5:GSGP]"4I"NB"*U!C1C[\@H14S5<30R!$3%F0#LQ#VK$R"5@>LN CD8U
M1>[08;^^#8BKGKR9>-81IBODJ"$[44-V2)IL>;NY&"$'+Y&A_7TW-6A>RNM&
M,%"\J9BB]EV_#IG>'#[=`+ZYU$15+;O:NP;%40]LQF7#9%)FSA<)2BYY91&X
M+Q=T?NWF7,EK34-#2$!5U;U[&YSJ.LRL@[;S%A'PJDP`U4R>,,GA'2H7;;ZB
M7(R;6=3OLU" $SKP$]\>V5+5);8&"U XZ-H8P!XT)?%2L1$;R,-&+0:OQ*!]
MY)H]&%=E<^WBABP::S 1-O9)K7!F;YEKC;,:?4WLMRA[TU&XZO\'ZO?P#U%L
MJ[([`V,9F()9-#.ACXH],*D4Z1#H^V@`05*$V!+<LE-,:,GLBO7%,03.)I at 3
MHSG&]_1FOUJDD/=B98("R"!=88XL<1K?"0PHL^'"'J1=LZ(*# E0`TDCGV&8
MR1<\XWL^MAY"GAO!_^14^P+D&K6PP/W"'+&VC6HKIDG>P<F+2@:[M/H1"+'(
M*DAMW9XA[EX;HCSSK[6=+CU*?4KENCZ at A?&\^T[PM%4N,7Z"L E@[;M"S#]D
M]39L;4[\UISH+)JVF2.YAOY095/^GL*[IT]-XO?R]8]/5.?^UA8S"+X`@4P?
M,GK!\0TQV5U5G*9T-6H!,S[DK3 at X4#NRMDZ^B at I_CMWW5)?[-1H4(S7<DYM=
M>("!1(RY?7>/GPSDMWMV_S>-_J^=_IU&_P.G?V3WVQ5M7 ]O'A<"R,K+"^A\
M+6CR$D9.B._A`K6@"SSKXX;V%9DG@>_H.M=B:XL/4FKL'3??O'8#($E^(DCF
M/ZIR6>VWX*U+?RT\BNK^+*+P"#Y0MGB)D9S-R7%-#0.6;UP4* 1;'&-I\U,_
ML<Q#+X,=Y\O+8?VDDI5:ACKK$Z#,2[#ZK-7=@1]*18KM7-O-R!8+BI?1/M4<
M0,/@UY2;78,#!4C&-PH-V45:.=:-/0>=GZ(?&(UY\?@+DR[K,!_(0*/5YAPA
M'JNB&B5_-(;6WVJD>LP7XTWT?M**Q-%?9MBL$M=PW):Z-:]>YEPRL7>AN@<Z
MEFNJN2F7.$&1'6C7V=V:U[D;0//XI7MM];I=`7B>S<UOH)D8ZEY]< N at U#V\
M"H(XTJB-M at 4-XTHNDA1T%?0=@T\\Y;(EPPJO@,#&9@,:58%=$5[Q='9NT(9M
M9$&U)@;#IN.Q+]O6[[$NTX at 2XNYT;=\#)8'AN@;=V:FL7^U6>M=Z80&<# _[
MCBZ<[BD=`#IWU=MD6Q:58,+[0$1\PQQLW WDSO'=8%O.-;J.K/3>#T_5-F2R
MR&@YC[,@@)H*HA4`"-4`IOPH#//GH7_!O,J])"K6>WS>RN=H]AD:H,-;=5,`
M3!<AU3W$?GIYFGC36"I%&+;I6$SL+AMA at P":TRR0ERJY<N+K/KKUNJ"_61:=
MIUE:YFVV6-<)<&FNXC>LZ.\_G+DBZM9OR.#K,1+G;;N'6)?$Y(D5]#2,R at VT
MLCWU.B#5`X+J`ECUT%B4CQ:(UE2M?I/U"@&W)+QR^DU'3F66?_-Z#*A]N.1;
MZF!-5M[:5&CF,JN49T1V3>$AC==)NHB\&/]"'=ZOP@(@O:]$A41\X2-.5SGG
M)Z0:$'<LZ-(]L@;4;>%E%Z!N:TBJXAAK&VKF@<V;AD4A)]_%7-[U)H[2+:)9
MF?A8:QRH8WZY"LW@&JN0F%QE47Y!R0^^'+^GOOWFFT>/.DGD2TA%/]#X<!=^
MAQ\`.4\VT.48]+!RE8WL98YY91PFO4;TDU.ET(WRPZ2NM]4?%./72)2^3N>D
M%R;!>'9X\NQ[<G&(?W,3W HM`>U8I7$PC:5SK7D&@K1E&IQH4'<]TV at D&D2+
M233XL8+\=N]_AZ+ %$.!":H"8PP%1J@*#)$58.]K0&H;1%\#UL0'>@?:>0"%
MG(8&).Q ]9R1J:DF) PMS'U @6B-N68U#Y*C`5!+`0(4`!0````(`+D,#2<B
M6UZ\G@,``#,)```&``````````$`( "V at 0````!S9&)M+FA02P$"% `4````
M" #D:% G,^O\944```"2````!@`````````!`" `MH'"`P``9&5P96YD4$L!
M`A0`"@``````N0P-)TV V;XK````*P````H``````````0`@`+:!*P0``&5X
M=&-O;F8N<F)02P$"% `4````" "Y# TGK>'J(E()``"D,@``!@`````````!
M`" `MH%^! ``:6YI="YC4$L!`A0`% ````@`EDG(*/%,L-BS`@``V 4```@`
M`````````0`@`+:!] T``$UA:V5F:6QE4$L!`A0`% ````@`:II4)_IGM[DT
M````-P````@``````````0`@`+:!S1 ``$U!3DE&15-44$L!`A0`% ````@`
METG(*%A=;C E`0``)P(```@````````````@`+:!)Q$``'-D8FTN97AP4$L!
M`A0`% ````@`N0P-)V[0MZII& ``'U,```<``````````0`@`+:!<A(``%]S
;9&)M+F-02P4&``````@`" "K`0```"L`````
`
end



From wlav at atlas03.cern.ch  Fri Jun  9 13:30:19 2000
From: wlav at atlas03.cern.ch (Wim Lavrijsen)
Date: 9 Jun 2000 17:30:19 GMT
Subject: Problem with script and typing
References: <394123ad.89325968@nntp.interaccess.com>
Message-ID: <8hr9jb$14g$1@sunnews.cern.ch>

olczyk at interaccess.com (Thaddeus L. Olczyk) writes:

I'm not quite sure what your code is supposed to do, so I'm just
guessing...

>project =[]

>class Project:
>   def IsNull(self):
>      return self.name==""
>      project.append(self)

I presume you meant 'global project.append( self )'?

>def FindProject(x):
>    for project in projects:
>        if project.name == x:
>            return x
>        null=Project("")
>        projects.remove(null)

I don't know what your trying to do here ... where does 'projects' come
from? It's never assigned an object.

>def is_project(s):
>    if s[len(s)-3:] == 'prj':
>        project=FindProject(s)
>## error here
>        if project.IsNull():
>            project=Project(s)
>        project.Scan()

Again ... 'global project'. Also, should this be a member of Project?

>----------
>error message AttributeError: 'None' object has no attribute 'IsNull'

And rightfully so! Sorry, but if you loop with project over projects
in FindProject to find a Project in project, someone is really going
to dislike you. :)

>2) Is there a better way of handling null objects? ( Most languages I 
>    know have objects which are explicitly null )

Yep, use None:

>>> project = None
>>> if project: print "there's a project!"
...
>>> if not project: print "no project!"
...
no project!

Best regards,
      Wim Lavrijsen

PS: you should use spaces instead of tabs.



From mfletch at tpresence.com  Sat Jun 10 02:11:06 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sat, 10 Jun 2000 02:11:06 -0400
Subject: mxTextTools
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33AB@VRTMAIL>

sets -> [a-zA-Z] in regex/bnf concepts, that is, match any element in the
set

	AllInSet ( "test" ) in "setrmgeneric" --> 3 
	# matches s,e,t from the set of allowed characters

sWordStart -> passes if there are some number of characters (including zero
characters), before the start of the word for which we are searching.

	sWordStart ( "test" ) in "esttest" --> 3
	sWordStart ( "test" ) in "test" --> 0
	WordStart ( "test" ) in "esttest" --> 3
	WordStart ( "test" ) in "test" --> FAIL

Hope this helps,
Mike


-----Original Message-----
From: Stephen Hansen [mailto:stephen at cerebralmaelstrom.com]
Sent: Saturday, June 10, 2000 1:57 AM
To: python-list at python.org
Subject: mxTextTools


Okay, i'm learning mxTextTools on top of the dozen other things i'm doing at
once, at the moment, and have two questions:

With the commands that deal with 'sets', i'm assuming they're talking about
sets returned via the 'set()' function and its similar friends? But what
exactly IS a character set? Why use one instead of a string?

Secondly, 'sWordStart' -- I don't get the description of what it does. It
says it uses a Search Object to perform the required action, which is
fater.. but zero matching characters are allowed.. So, it only passes if
there is NOT a match?

Thanks

--S


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



From sdhyok at email.unc.edu  Fri Jun 16 13:46:58 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Fri, 16 Jun 2000 13:46:58 -0400
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <8ia4tv$e19$1@nnrp1.deja.com>
Message-ID: <8idr2g$fvi$1@news2.isis.unc.edu>

How can I get the DCI module?

Peter

"Fred Gansevles" <gansevle at cs.utwente.nl> wrote in message
news:8ia4tv$e19$1 at nnrp1.deja.com...
> In article <8i8ujt$hop$1 at news2.isis.unc.edu>,
>   "Daehyok Shin" <sdhyok at email.unc.edu> wrote:
> > How can I prevent illegal definition of a variable to an object?
> > For instance,
> >
> > class A:
> >     __init__(self):
> >             self.x = 1
> >
> >     __setattr__(self, name, value):
> >             if name == "x":
> >                     self.x = value
> >             else:
> >                 raise AttributeException
> >
> >     __getattr__(self, name):
> >             if name == "x":
> >                 return self.x
> >             else:
> >                 raise AttributeException
> >
> > >>> a = A()
> > >>> a.x = 10
> > >>> a.y = 20 # I want to raise an AttributeException.
> >
> > Peter
> >
> >
>
> Some time ago I posted my DCI module. This module also has a attribute
> module wich can, among other atribute-related stuff, do this for you.
>
> For instance,
> >>> import dci.attr
> >>> class A:
> ...     def __init__(self):
> ...         dci.attr.init(self) # initialize the attribute mechanism
> ...         dci.attr.declare(self, 'x')
> ...
> >>>
> >>> a = A()
> >>> a.x = 10
> >>> a.y = 20
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "dci/attr.py",
> line 52, in __setattr__
>     raise AttributeError, "Undeclared attribute: %s" % `name`
> AttributeError: Undeclared attribute: 'y'
>
> --
> -----------------------------------------------------------------------
> ----          Linux/Windows-NT/IntraNetware Administrator          ----
> -- Whenever it sounds simple, I've probably missed something! ... Me --
> -----------------------------------------------------------------------
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From BgPorter at NOacmSPAM.org  Tue Jun 13 16:39:20 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Tue, 13 Jun 2000 20:39:20 GMT
Subject: no SyntaxError?
References: <Pine.LNX.4.10.10006131359020.17071-100000@localhost.localdomain>
Message-ID: <YZw15.7$lU5.825@news1.rdc1.nj.home.com>

"Robert W. Bill" <rbill at digisprings.com> wrote in message
news:Pine.LNX.4.10.10006131359020.17071-100000 at localhost.localdomain...
> However, the following snippet generates "0 1 whoops" when it seems like
> it should also generate "SyntaxError: invalid syntax" because there is no
> if before the else.
>
> def test1():
>     for i in range(2):
>         print i,
>     else:
>         print "whoops"
>
> test1()
>
> Why is it that I do not get a "SyntaxError" in the second example?
Loops in Python may have an optional else: clause that's only executed if
the loop runs to completion. It's used like this:

def find(list, value):
    for i in list:
        if i == value:
            print 'Found it!'
            break
    else:
        print 'couldn't find %d' % value





From johngrayson at home.com  Fri Jun 30 13:25:47 2000
From: johngrayson at home.com (John Grayson)
Date: Fri, 30 Jun 2000 17:25:47 GMT
Subject: Tkinter focus problem
References: <395CC8DE.AA663D4@ieee.org>
Message-ID: <8jil65$hs6$1@nnrp1.deja.com>

In article <395CC8DE.AA663D4 at ieee.org>,
  Didier WILLAME <didier.willame at ieee.org> wrote:
> This is a multi-part message in MIME format.
> --------------EAB9DEA81EB868EA42827E00
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
>
> Hello all,
>
> I've slight problems with focus in Tkinter, running under win NT.
>
> With the following script I'm able to create 'nodes' (left button) in
a
> canvas. The color of a 'node' toggles if the mouse is putting on.
> Moreover, I would like display "Hello World!" when I press on the key
> '1' and, when the mouse is putting on this 'node'.
>
> But I'm unable to bind a key event (line 25), although I can run the
> script with a button event (line 26 instead of line 25).
> I should probably give the focus to the 'node' object inside the
> entryNode function (line 57).
>
> Thank you for your time,
> Didier
>

Well, you need to give the canvas focus. However, I don't think that
your original code is adequate, since I'm sure you will want to know
*which* node is active.

I've restructured your code a bit, storing the current item that
has been entered, giving it a new highlight when '1' is pressed.

Should get you going again...

   John Grayson    (Python and Tkinter Programming)


# ----------
from Tkinter import *

class Graph( Frame ):

 # ----------
  def __init__( self, master=None ):
    Frame.__init__( self, master )
    self.pack()
    self.createCanvas()
    self.activeNode = None

  # ----------
  def createCanvas( self ):
     self.editingArea = Canvas( self )

     # --- To create a node with the left button.
     self.editingArea.bind( '<Button-1>',   self.mkNode )
     self.editingArea.bind( '<KeyPress-1>', self.activateNode )
     self.editingArea.bind( '<Button-2>',   self.activateNode )

     # --- To toggle the node color when passing with the mouse.
     self.editingArea.tag_bind( 'node', '<Any-Enter>', self.entryNode )
     self.editingArea.tag_bind( 'node', '<Any-Leave>', self.leaveNode )

     self.editingArea.pack(side=TOP)
     self.editingArea.focus_set()

  # ----------
  def mkNode( self, event ):
    """Create a new node at (x,y)"""

    x = event.x
    y = event.y

    new = self.editingArea.create_oval( x-5, y-5, x+5, y+5 )
    self.editingArea.itemconfigure( new, tag='node', outline='black',
                                         fill='red' )

  # ----------
  def entryNode( self, event ):
    self.editingArea.itemconfigure( CURRENT, fill='blue' )
    self.activeNode = self.editingArea.find_withtag( CURRENT )

  # ----------
  def leaveNode( self, event ):
    self.editingArea.itemconfigure( CURRENT, fill= 'red' )
    self.activeNode = None

  # ----------
  def srcNode( self, event=None ):
    print 'Hello World!'

  def activateNode( self, event ):
      if self.activeNode:
	self.srcNode()
	self.editingArea.itemconfigure( self.activeNode, fill='green')

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

graph = Graph()
try:
  graph.mainloop()
except KeyboardInterrupt:
  graph.quit()


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


From olczyk at interaccess.com  Mon Jun 12 12:22:10 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 12 Jun 2000 16:22:10 GMT
Subject: Testing for a file in python?
Message-ID: <39450724.231994953@nntp.interaccess.com>

How would you go about testing if a certain file exists in python?
In most scripting languages you have the -f test:

if -f filename do
   whatever_you_were_going_to_do
   fi
else do
   print "Error" file not found
   fi

stat doesn't work, it aborts if the file does not exist.


From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 28 09:32:51 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 28 Jun 2000 06:32:51 -0700
Subject: long lines, long string
References: <j9tcj8.i0b.ln@iiab.soluma.de>
Message-ID: <0b9419d8.107f512e@usw-ex0105-038.remarq.com>

Hi Holger,

You can extend python code over lines in a few ways. Firstly
with a \ or with one half of a ( or { or [ and the other half
when you complete it.

You can extend strings over lines with triple speech marks

i.e.

spam=1
eggs=1
moreSpam=1
if spam and \
   eggs and \
   moreSpam:
    print 'chicken'

aBigTuple=('Hello','Holger',
          1,2,3,
          5,6,7)
aBigString="""A String can
stretch lots of lines
if you put triple speech marks
"""

print aBigTuple
print aBigString

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From kens at sightreader.com  Thu Jun 15 23:32:17 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:32:17 -0700
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <3948D25D.7C83CD6@micron.net>
Message-ID: <39499FC1.8C31F7B5@sightreader.com>

LOC (lines of code) is also a measure of programmer output.  The idea is
that you evaluate your programmers by the sheer quantity of code they
crank out.  The one who cranks out the most lines of code per week is
clearly the best programmer (yeah right!).  Anyway, it is sometimes used
as a rough extimate of productivity.

Peter Immarco wrote:

> Courageous,
>
> Tried Lisp for over 8 months, never did learn to like it.  I recognize
> it's strong points but its just "not for me".  What's the LOC acronym
> stand for?
>
> Peter.
>
> Courageous wrote:
>
> > Re: python vs c++
> >
> > > However my C++ friends are badgering the crap out of me...
> >
> > Most of your C++ friends would be picking their tongues
> > up off the floor if they saw the LOC of python programs
> > in comparison to their C++ counterparts.
> >
> > Python may be too slow for some tasks, but in the area
> > of programmer productivity, very little beats it, and
> > arguably the environments that do (e.g., lisp) have too
> > many other negatives associated with them.
> >
> > C/

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/1b288b4c/attachment.vcf>

From herzog at online.de  Thu Jun 29 16:06:40 2000
From: herzog at online.de (Bernhard Herzog)
Date: 29 Jun 2000 22:06:40 +0200
Subject: stderr and stdout
References: <qzI65.16987$_b3.366473@newsread1.prod.itd.earthlink.net> <m3itusjswb.fsf@greebo.nodomain.de> <8jg591$n9b$1@nnrp1.deja.com>
Message-ID: <m3bt0kjlpr.fsf@greebo.nodomain.de>

Arint? <jamarijr at my-deja.com> writes:

> Ok, I took care of that stuff do you know of anyway to keep the stdout
> and stderr in sync.

Here's a simple python solution:

import sys, random
    
class Collector:

    def __init__(self):
        self.strings = []

    def write(self, tag, str):
        self.strings.append((tag, str))

    def dump(self):
        for item in self.strings:
            sys.__stdout__.write("%s:%s" % item)

_collector = Collector()

class TaggingOStream:

    def __init__(self, tag):
        self.tag = tag

    def write(self, str):
        _collector.write(self.tag, str)

sys.stdout = TaggingOStream('stdout')
sys.stderr = TaggingOStream('stderr')

def main():
    # Generate some random messages in stderr and stdout
    for i in range(20):
        stream = random.choice((sys.stdout, sys.stderr))
        stream.write("Random message %g\n" % random.random())

main()

_collector.dump()


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


From gmcm at hypernet.com  Fri Jun 30 20:34:39 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 01 Jul 2000 00:34:39 GMT
Subject: command-line history in win
References: <395c304e.1224185@news.student.lu.se> <8jhm03$qnm$1@nnrp1.deja.com> <8jig08$dk7$1@nnrp1.deja.com>
Message-ID: <8F63D360Egmcmhypernetcom@199.171.54.154>

david_ullrich at my-deja.com wrote:


>  David Currie <davec999 at my-deja.com> wrote:

>> It works for me (Python 1.5.2)...

>     Keen. Maybe I'm the only one who doesn't know:
>_What_ works for you, exactly?

Perhaps David is on NT, where it just works (not only for DOS cmds, but if 
you start Python, up-arrow will take you thru the history for your last 
Python session in that DOS box). This is all built in to NT.

Sorry, can't help with Win9x.

- Gordon


From sdhyok at email.unc.edu  Wed Jun 14 21:53:05 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 21:53:05 -0400
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <20000614155102.E16232@freei.com> <8i9cbh$mev$1@news2.isis.unc.edu>
Message-ID: <8i9cu3$mlj$1@news2.isis.unc.edu>

Maybe a little better, but not enough good.

class A:
 x = 1

 def __init__(self):
   self.__dict__["x"]=1

 def __setattr__(self, name, value):
  if not self.__dict__.has_key(name):
   raise AttriubuteError
  else:
   self.__dict__[name] = value

 def __getattr__(self, name):
  try:
   return self.__dict__[name]
  except KeyError:
   raise AttriubuteError

"Daehyok Shin" <sdhyok at email.unc.edu> wrote in message
news:8i9cbh$mev$1 at news2.isis.unc.edu...
> Even though I corrected minor errors, it does not work.
> Refer to the following comments.
> >> class A:
> > >     def __init__(self):
> > >             self.x = 1  # this code calls __setattr__
>                                   # when I change it into self.__x = 1, it
> raised AttributeError in __setattr__
> > >
> > >     def __setattr__(self, name, value):
> > >             if name == "x":
> > >                     self.x = value # this code calls __setattr__
> recursively
> > >             else:
> > >                 raise AttributeError
> > >
> > >     def __getattr__(self, name):
> > >             if name == "x":
> > >                 return self.x
> > >             else:
> > >                 raise AttributeError
>
> I found one solution, but expect better one.
>
> -------------- Solution, but not good ------------------
> class A:
>  def __init__(self):
>    self.__dict__["x"]=1
>
>  def __setattr__(self, name, value):
>   if not self.__dict__.has_key(name):
>    raise AttriubuteError
>   else:
>    self.__dict__[name] = value
>
>  def __getattr__(self, name):
>   if not self.__dict__.has_key(name):
>    raise AttriubuteError
>   else:
>    return self.__dict__[name]
>
> "Sean Blakey" <sblakey at freei.net> wrote in message
> news:20000614155102.E16232 at freei.com...
> > On Wed, Jun 14, 2000 at 05:48:45PM -0400, Daehyok Shin wrote:
> > > How can I prevent illegal definition of a variable to an object?
> > > For instance,
> > >
> > > class A:
> > >     __init__(self):
> > >             self.x = 1
> > >
> > >     __setattr__(self, name, value):
> > >             if name == "x":
> > >                     self.x = value
> > >             else:
> > >                 raise AttributeException
> > >
> > >     __getattr__(self, name):
> > >             if name == "x":
> > >                 return self.x
> > >             else:
> > >                 raise AttributeException
> > >
> > > >>> a = A()
> > > >>> a.x = 10
> > > >>> a.y = 20 # I want to raise an AttributeException.
> > >
> > > Peter
> > >
> > >
> > > --
> > > http://www.python.org/mailman/listinfo/python-list
> >
> > The code you posted seems to do what you want, provided you
> > a)add the 'def' keyword to your function definitions
> >     and
> > b) define AttributeException (perhaps you meant AttributeError?)
> >
> > --
> > Sean Blakey, sblakey at freei.com
> > Software Developer, FreeInternet.com
> > (253)796-6500x1025
> > "Uncle Cosmo ... why do they call this a word processor?"
> > "It's simple, Skyler ... you've seen what food processors do to food,
> > right?"
> > -- MacNelley, "Shoe"
> >
>
>




From cbiegay at objectmentor.com  Fri Jun 30 11:17:07 2000
From: cbiegay at objectmentor.com (Chris Biegay)
Date: Fri, 30 Jun 2000 15:17:07 GMT
Subject: Uploading files through HTML
Message-ID: <TR275.5$qm5.707@ord-read.news.verio.net>

I'm trying to upload a binary file through an HTML form, but it only works
for files smaller than a few kb.  I tried changing the buffer size in cgi.py
but it hasn't helped.  Any ideas?  This basically how I'm doing it now:

form = cgi.FieldStorage()
formItem = form[inputName]

if formItem.file:
    print "Received file [" + formItem.filename + "]"
    data = formItem.file.read()
    ...




From thomas at xs4all.net  Tue Jun 20 08:32:56 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 20 Jun 2000 14:32:56 +0200
Subject: Why is Python so slow ?- revisited.
In-Reply-To: <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu>; from bparsia@email.unc.edu on Mon, Jun 19, 2000 at 08:47:09PM -0400
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl> <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu>
Message-ID: <20000620143255.F26436@xs4all.nl>

On Mon, Jun 19, 2000 at 08:47:09PM -0400, Bijan Parsia wrote:
> Thomas Wouters <thomas at xs4all.net> wrote:

[ Python's lack of an instruction/function cache ]

> > Indeed. Such a cache would have to be invalidated whenever the function
> > object or one of the objects used to lookup the function, was altered (or
> > rather, 'touched'.) The dynamic nature of Python makes this quite
> > complicated, and probably quite expensive.

> I'd love to hear more about why this would be the case. (Mere
> "dynamicness" can't be it: Smalltalk is a s dynamic and most
> implementations make quite effective use of various sorts of cache. So,
> if your contention is correct, there must be some *special* aspect of
> Python's dynamic nature that makes it' *especially* resistent. Note that
> I am *very* skeptical about the existence of such a special aspect, but
> I am quite willing to be enlightened! Merely being "more dynamic" or "so
> very dynamic" which is what I usual read, doesn't cut any mustard.

I know next to nothing about Smalltalk, let alone how it optimizes, but I
can tell you why it would be complicated in Python ;) Lets take a simple
example:

def spam(x, y):
    if x in range(1, y):
        return 1
    if x in range(y, 1):
        return 1
    return 0

Might not seem very useful, but it's just an example ;) This can easily be
optimized, right ? We lookup 'range', it's a builtin, we store the address,
and call it two times. Well, no. We dont *know* range a is a builtin, it
could be a global that points elsewhere. If the 'spam' function contained a
'from module import *' or 'exec' statement, it could even be a local.

So, we have to check if range is still the same function. Which means we
have to look it up and compare identity -- and we've already lost any
advantage the cache could give us. Or we could place a marker on the
__builtin__ module, and invalidate the cached entry of 'range' whenver
'range' is changed. But then we also have to add a marker to the global
namespace, because something can *add* a range there.

And don't think it can't happen: 'range()' could be a function that modifies
the global or builtin namespace. Or it could be a function that deletes
references to a few objects (not that uncommon an operation, really.) And
those objects could then be destroyed (no more references), and then those
objects could trigger destruction of other objects, and one of those objects
could call a function on some module or other that modifies the original
global or builtin namespace. 

Another example, the simple case of:

def spam(egg):
    egg.spam()
    egg.spam()

If egg or a parent class of egg does not define a spam() method but does
define a __getattr__, this could return two entirely different things. So,
in caching egg.spam, we need to make sure it's been located 'statically'.
(This can be read as 'without __getattr__', or as 'analyzed __getattr__ and
determined each call with the same arguments would return the same object')

Also, we need a way to invalidate the cached entry of egg.spam if any of egg or
any of its parent up to the one defining 'spam', have something done to
their 'spam' attribute. And that means than whenever you change an attribute
on a class or instance, you need to traverse the entire tree of subclasses
and instances (up to one that 'masks' the change) looking for cached entries
and invalidating them.

And of course, if you do something like:

  egg.ham.viking.spam.plate.drop()

you have to consider *each lookup* cached ;-P I don't think lookups are
*that* expensive. I'm not sure what the penalty for a failed lookup is, I
seem to recall something about a successful lookup being relatively cheap,
compared to a failed one, and a 2nd lookup is not likely to fail ;)

> Indeed, as my bet is that the majority of the time, in the majority of
> code, things are quite "hands off" (i.e., no touching :)), I would be
> surprised that this would prove a problem. In inner loops, indeed, it
> could be quite effective, and would be no different than *manually*
> copying the function to the local--a move, I hope you agree, can provide
> a substantial speedup.

I've measured about 10% speed increase (according to pybench) in method
calls, by storing each method in the instance-dict after it was created. It
also seemed to have the same effect on the TryExcept test, but that's
probably because it uses a lot of methods.  However, at the same time, some
other tests decreased in performance about as much, ending up actually 0.76%
slower than before ;)

The speedup can probably be a lot more, if you do (compile time) analysis of
howmany times a given method is called, and caching only those. Or, of
course, if you *manually* cache the function to the instance-dict or the
local namespace (even better -- one lookup less).

> I confess to not being an expert in these matters, but I don't know I
> understand why the *whole* cache would have to be invalidated if one
> function object were "touched". Er..in mere other words...More please!

I was a bit careless (again ;) in my wording. I meant to say that the
individual cached entry has to be invalidated.

> > What's more, currently 'bound methods' (methods which have a 'self' to which
> > they apply) are created on demand -- the unbound method exists on the class,
> > but it has to be 'bound' to an instance before it can be used. This is done
> > each time you access the method, though CPython is being smart and you have
> > to try hard to actually see that ;-)

> Hmm. Ok. I think. Maybe. Why it has to be do each time one accesses the
> *instance*'s method I find a tad confusing. I mean ok, you change the
> class def, you'll want to update the instances. But surely it's ok to
> make class updates more expensive than method access?

Yes, but this works, and caching & invalidation doesn't :-) It would be no
little ammount of code to add it, even if you discard the ideas about
invalidation (say, if you add the new code under a -OOO option, and tell
people not to f*** around with base classes after instances have been
created, which I doubt Guido or any of the other core developers would
allow.)

> Ewww! :) But *why* do this? walk hasn't changed. It's surely
> determinable that it hasn't changed over the course of x,y=(cleese.walk,
> cleese.walk), yes?

Yes, by comparing identity of the retrieved objects, but then it's already
too late :-) "a,b=c.walk,c.walk" gets compiled roughly in this:

          6 LOAD_FAST                0 (c)
          9 LOAD_ATTR                1 (walk)
         12 LOAD_FAST                0 (c)
         15 LOAD_ATTR                1 (walk)
         18 BUILD_TUPLE              2
         21 UNPACK_TUPLE             2
         24 STORE_FAST               1 (a)
         27 STORE_FAST               2 (b)

As you see, different bytecodes. The bytecodes have trouble seeing what
previous bytecodes were doing, so you'd have to either recognize this in the
compiler and generate alternative bytecodes, or rewrite the bytecode
interpreter to do caching on *all* lookups (of a certain type.) And see
above about that problem ;)

So, to the casual reader it might seem obvious that Cleese's walk doesn't
change, but the compiler doesn't know *anything* about Cleese, and though
the interpreter does know everything about Cleese, it doesn't know that
Cleese has already been queried about his walk method. And adding the
knowledge to either (or better (for optimization), both) is a major project.

Hopefully, now that a couple of the Big Brains can devote their energy to
the furtherance of Python full-time, they can devote some of that in this
direction ;) But I can imagine there are more important things than this
potential 10% (or so) speedup.

So, now for the DISCLAIMER: I'm just a hobbyist. I don't know *that* much
about Python (yet ;), I've only been mucking around in its internals for a
couple of weeks. I also have no experience with compilers, cache-mechanisms
or optimizations other than a few good books on the subject (parts of the
Dragon book, O'Reilly's High Performance Computing, and books about the
internals of computers/compilers in general, like those of Tanenbaum and v/d
Linden.) So, please, do not take my words for unbendable truth, for even if
they are true, they're likely flexible beyond belief ;)

laywer-like-ly yr's,
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From donn at u.washington.edu  Mon Jun 26 12:11:47 2000
From: donn at u.washington.edu (Donn Cave)
Date: 26 Jun 2000 16:11:47 GMT
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi>
Message-ID: <8j7vc3$6b6q$1@nntp6.u.washington.edu>

Quoth ge at nowhere.none (Grant Edwards):
| In article <kuzoobue0g.fsf at lasipalatsi.fi>, Erno Kuusela wrote:
|>>>>>> "Tony" == Tony Keating <keating at mech.uq.edu.au> writes:
|>
|>    Tony> start the program, then send various strings to the program,
|>    Tony> read some output from the program then send some more
|>    Tony> strings to the program.
|>
|> there are a couple of expect modules for python (so named because
|> there is an old tcl-based language called expect for doing just this).
|> parnassus lists 3 modules (<URL:http://www.vex.net/parnassus/
|> apyllo.py?find=expect>), i haven't
|> used any of them however...
|
| Or you can roll your own using the popen2 module.

OK, but check to see if your application will work reliably on
a pipe, before you put too much work a pipe solution.  I append
a short sample program that works with the shell as a remote
application, but deadlocks with awk.  The awk behavior is what
I would expect from most programs, so normally you're sunk.

	Donn Cave, donn at u.washington.edu
----------------------------------------
import popen2

# rfd, wfd = popen2.popen2(('/bin/sh', '-c', 'echo hi!; read ans; echo $ans'), 0)
rfd, wfd = popen2.popen2(('/bin/awk', 'BEGIN {print "hi!"} {print "bye"; exit}'), 0)
hi = rfd.readline()
# ^ Deadlock here.  To fix, move write+flush up before readline, which lets
# awk exit - and flush its output - before we need to read it.
print 'application says', repr(hi)
wfd.write('OK.\n')
wfd.flush()
hi = rfd.readline()
print 'application responds', repr(hi)


From neilh at scintilla.org  Thu Jun 22 05:04:41 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 22 Jun 2000 09:04:41 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se>
Message-ID: <JEk45.262$Sn2.1571@news-server.bigpond.net.au>

> However, on the NT system, the very same command takes 4.0 s.

   One thing to try:
set path=
python -c "print 'hello'"

   Although you will now have to give a full path to run Python.

   Neil




From jparkey at nospamrapidbs.com  Tue Jun 20 05:32:14 2000
From: jparkey at nospamrapidbs.com (John Parkey)
Date: Tue, 20 Jun 2000 10:32:14 +0100
Subject: Newbie : Add to PYTHONPATH
References: <8i83a7+kraa@eGroups.com>
Message-ID: <8indsf$6qm$1@phys-ma.sol.co.uk>

Try

http://www.python.org/windows/python/registry.html

jp

<hohage at muenster.de> wrote in message news:8i83a7+kraa at eGroups.com...
> Hello,
> this is an absolute newbiequestion.
> I'd like to know how to add a path to sys.path, but not temporarily.
> Python is running on Win 98.
>  Thanks.
>
>




From kern at caltech.edu  Wed Jun 21 19:47:12 2000
From: kern at caltech.edu (Robert Kern)
Date: Wed, 21 Jun 2000 19:47:12 -0400
Subject: speedup numpy ufuncs
References: <395001D6.BD6D9728@visionart.com>
Message-ID: <39515400.C51C0123@caltech.edu>

Pete Shinners wrote:
> 
> i've been doing some simplish image processing
> stuff with numpy and a thought occured to me about
> a possible speedup.
> 
> if i have 4 or 5 arrays i want to add together,
> the fastest way i've found to do this is by
> calling the add ufunc many times, specifying an
> output argument.
> (this gets rid of extra temporary arrays)
> 
> array1 = array2 = array3 = array4 = arange(100)
> add(array1, array2, array1)
> add(array1, array3, array1)
> add(array1, array4, array1)
> 
> i was wondering if there was a way to add these all
> in one pass?
> addmany(array1, array2, array3, array4, array1)

This is where you get to the really cool part of ufuncs.  ufuncs have
methods.  You want the "reduce" method.

new_array = add.reduce([array1, array2, array3, array4])

You make a single new array.  I don't believe there is a way to put the
results into an old array.

[snip]

--
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From hellan at acm.org  Thu Jun  8 05:33:58 2000
From: hellan at acm.org (Jon K Hellan)
Date: 08 Jun 2000 11:33:58 +0200
Subject: GUI editor
References: <393f4568$0$25329@wodc7nh0.news.uu.net> <vskujs867ghjso3aqi2r38n5mei0bg4lhq@4ax.com>
Message-ID: <87zoowseih.fsf@parus.parus.no>

Anders M Eriksson <anders.eriksson at morateknikutveckling.se> writes:

> On Thu, 8 Jun 2000 01:59:37 -0000, "Blu Dragoon"
> <eddthompson at ssi.parlorcity.com> wrote:
> 
> >Is there a gui editor written in python for tkinter?
> >The reason I ask is that I use python on 6 different platforms.
> >And tkinter seems to be the most prevailant of the gui's.
> >Or should I go ahead a take the PIL? and is there any gui editor
> >for PIL too.
> >
> 
> <opinion>
> This is the MAJOR lack of python! I sys that the first one that comes
> out with a IDE/GUI editor which works has set the standard of which
> GUI that will be used
> </opinion>
> 
> AFAIK, there is no GUI editor for TKinter. There are some rumour about
> Pythonware's 'PythonWorks' it will contain a GUI editor in some
> version...

Didn't they go Gold recently?

Apart from that, I don't know of any. But for somebody who might end up in
a rainy place during summer vacation, it might be a good idea to bring
along a laptop and sources for glade and libglade. Not to extend glade
to tkinter, but to see how hard it is to come up with something
similar.

Jon K?re


From fuessler at solutec.de  Wed Jun 28 08:07:27 2000
From: fuessler at solutec.de (Holger M. Füßler)
Date: Wed, 28 Jun 2000 14:07:27 +0200
Subject: long lines, long string
Message-ID: <j9tcj8.i0b.ln@iiab.soluma.de>

Hallo!

Is there a possibility to put long lines of python-source-code in multiple
lines (for better clarity)?
How can I accomplish the same with strings?

Greetings

Holger




From jkraska1 at san.rr.com  Sun Jun  4 01:22:45 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 05:22:45 GMT
Subject: Um? *boggle* Silly question about listobject.c
References: <000e01bfcddf$d2e09d40$1d2d153f@tim>
Message-ID: <3939E84B.8AE86E66@san.rr.com>

> Critical?  No more than that your experience with more-recent C caused you
> to make the similar "critical oversight" of suggesting "long" is adequate.
> Guido is on record as favoring changing PyObject_VAR_HEAD to declare ob_size
> of type size_t, but that won't happen in time for 1.6.  In the meantime, if
> you really need lists with more than 2 billion elements,...

*GACK*!

After running through a quick bit-shifting test, I discovered
to my chagrine that I was thinking for some dumb reason that
int was 16 bits. This is what I get for working on programming
problems all day long during my leisure time. I heard a rumor
that programming destroys brain cells more quickly than liquor.
Is it true? :)-

*bonk self on head with rubber mallet*


C/


From morten at src.no  Tue Jun 27 10:51:46 2000
From: morten at src.no (Morten W. Petersen)
Date: Tue, 27 Jun 2000 16:51:46 +0200
Subject: Algoritm that'll find usable account names
Message-ID: <Pine.LNX.4.21.0006271644130.14053-100000@lade.src.no>

Does anybody have any snippets of code that'll morph names ?

I.e., if some user, at my service, registered with the name John Doe
(ta-da) and subsequently got the name John.Doe at sweetservice.com, and some
other user, with the same name registered at the same service, could
get a different variant, for example JohnDoe at sweetservice.com ?

-Morten
---------------------------------------------------------------------

 How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From stephen at cerebralmaelstrom.com  Thu Jun  1 18:10:51 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 01 Jun 2000 22:10:51 GMT
Subject: search for a string?
References: <8h6211$ark$1@news-int.gatech.edu>
Message-ID: <LbBZ4.31403$WS3.364156@typhoon.we.rr.com>

    Depends on how you have the document opened/loaded...

    If its all in one string you could just use:
string.find(Document,Substring)
    [Document.find(Substring) in py1.6]

    If its in a list, a la readlines? Hm.
string.find(string.join(Document),Substring)
    ["".join(Document).find(Substring) in py1.6]

    Etcetc..you could cycle through a file or do whatever you wanted to..

    As for inserting the contents of another file, well, something like:

    OldString = File.read() + OldString

    I perfer, though: OldString = "%s %s" % (File.read(),OldString)

--S
Shengquan Liang <gt2558b at prism.gatech.edu> wrote in message
news:8h6211$ark$1 at news-int.gatech.edu...
>
> how do i search for a string in a document
>
> and inseart content of another file before
> that string?
>
>
>




From angband at blueyonder.co.uk  Sun Jun  4 07:26:27 2000
From: angband at blueyonder.co.uk (Bablos)
Date: Sun, 04 Jun 2000 11:26:27 GMT
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk> <slrn8jkej0.igk.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <393A4B95.MD-1.4.4.angband@blueyonder.co.uk>

On 4 Jun 2000 11:16:47 GMT, Remco Gerlich wrote;
> > Is there a way that I can get Python to print this all on the same
> > line, so that it looks as if the line is being updated constantly, as
> > opposed to having a ridiculously long list of 'uploaded...' lines?
> > 
> > Hope someone can help.
> 
> Print, sys.stdout.write and so on all just push bytes into stdout. Any
> tricks you could use to do this are very platform dependent. However, maybe
> printing a lot of ^Hs to clear a line will work; see what

I've managed to get it to work by adding '\v', a vertical tab to the
beginning of each line.  I've tried adding lots of backspaces, but it
doesn't solve my problem.  I've tried adding a '\r' to the beginning
of the line, and suppressing the newline at the end, but under the
Amiga the printing output is then buffered and flushed for some
reason, so I get the required effect, but due to the buffering, I only
see about 1 in 15 messages.

> Nice to see people from the Angband world here, btw :-)

Oh, I'm flexing my Python wings this weekend.  Now that the potential
of Kang has disappeared, I'm doing a bit of programming as a stepping
stone before wading waist deep into Zang's scripting ;o)


Bablos.


From bjorn at roguewave.com  Fri Jun  9 15:56:59 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 09 Jun 2000 13:56:59 -0600
Subject: Why is Python a memory hog?
References: <Ta%%4.1757$iy.138198@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <39414C0B.2B655203@roguewave.com>

William Dandreta wrote:
> 
> I downloaded Python 1.2 to run under DOS without a memory extender.  The
> interpreter seems to take up almost all of the 640K. When I try to run all
> but the tinyest programs I get an out of memory error.
> 
> Is there anything I can do to improve this situation?

Sure. Get yourself an OS and a current version of Python.

-b



From hschroeder at gmx.net  Mon Jun 12 18:12:02 2000
From: hschroeder at gmx.net (Henning Schroeder)
Date: Tue, 13 Jun 2000 00:12:02 +0200
Subject: SMTP via pipes?
Message-ID: <i7n3i8.lk.ln@odin.henning-schroeder.de>

Hello,
I have some users who don't have a local SMTP server running although 
sendmail is available. All mail has to be sent via pipes. When you
call sendmail with the -bs parameter you can talk to sendmail like you are
connected to a smtp-server. So thought I write the below attached
wrapper. Unfortunately it doesn't run :-(
Can anybody help or give any hints how pipes behave and why it doesn't
work that way? Your help is very appreciated.

Henning



# Module pipesmtp.py

import smtplib
from popen2 import popen2

SENDMAIL = "/usr/lib/sendmail -bs"


class SocketDummy:

    def __init__(self, file):
	self.file = file
	self.send = self.file.write
	self.close = self.file.close

	
class FileDummy:
   
    def __init__(self, input, output):
	self.ifile = input
	self.ofile = output
	self.read = self.ifile.read
	self.readline = self.ifile.readline
	self.write = self.ofile.write

    def close(self):
	self.ifile.close()
	self.ofile.close()

	
class SMTP(smtplib.SMTP):
    
    def connect(self, host="doesnt_matter", port=0):
	r, w = popen2(SENDMAIL)
	self.file = FileDummy(r, w)
	self.sock = SocketDummy(self.file)
	(code, msg) = self.getreply()
        if self.debuglevel > 0 : print "connect:", msg
        return (code, msg)

    

if __name__ == "__main__":
    s = SMTP()
    print s.connect()  # this works
    print s.help()     # here it hangs
    print s.quit()


From dan at netrics.com  Fri Jun 16 14:19:55 2000
From: dan at netrics.com (Dan Gindikin)
Date: Fri, 16 Jun 2000 18:19:55 GMT
Subject: performance benefits from byte code files?
Message-ID: <8idr3p$33h$1@nnrp1.deja.com>

I was wondering what the performance benefit is from having a .pyc file
around.

Here an example that motivates this quiestion. Lets say I have a
relatively large script.py CGI script. Since it is always executed
directly, a script.pyc file is never generated. Is it then beneficial
to write a tiny wrapper that does nothing but
   import script
   script.run()
because then script.pyc does get generated?

--
Dan


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


From thomas.heller at ion-tof.com  Thu Jun 15 09:18:24 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 15 Jun 2000 15:18:24 +0200
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com> <dCL15.4673$cQ.38267@news-server.bigpond.net.au> <8i8m11$bni$1@nnrp1.deja.com> <8_W15.4949$cQ.40252@news-server.bigpond.net.au>
Message-ID: <042601bfd6cc$2fe59230$4500a8c0@thomasnb>

[Neil]
>    I'd like to see the Python download smaller rather than larger so I'm
not
> in favour of adding large chunks of the Win32 extensions - just the bits I
> like, OK :) Lua manages a full script interpreter in 160K, and less than
80K
> if compressed with UPX, so Python looks bloated to me.
>
>    All that is really needed is for Sam Rushing's calldll/windll to be
> included in the standard distribution. Then any required OS calls could be
> implemented in Python rather than requiring an extension written in C. I
> know it opens Python up to the world of untyped, unchecked hardware
pointers
> but the increase in capability is large.
This would not help for the popen case, because on win95/98 (not on NT)
an additional executable helper program is needed.

>
>    Why are Unix specific modules like commands included in the Win32
> installation? It just leads to the confusion which caused this thread.
This may be the best solution:
Remove unix-specific modules (and those that do not work on windows) from
the core,
and use optional packages if these are needed.

Thomas




From kens at sightreader.com  Tue Jun 13 18:55:10 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 13 Jun 2000 15:55:10 -0700
Subject: Python Productivity over C++
References: <m130pw3-000wcDC@swing.co.at>
Message-ID: <3946BBCE.D0AF2339@sightreader.com>



Christian Tanzer wrote:

> "Frank V. Castellucci" <frankc at colconsulting.com> wrote :
>
> > > > 2. There is the type information in standard conforming compilers.
> > >
> > > Nowadays, there is.
> >
> > Agreed, but the language had the ability for you to write your own, or
> > use someone else's library which provided the same.
>
> Sure it provides it. But you have to implement it. And that costs
> time.
>
> The decisive difference is:
>
> - in C++ you have to add RTTI code to every class in the class
>   hierarchy -- it is simply impossible to factor the behavior into the
>   root of the hierarchy.
>
> - In Python, you can implement the entire mechanism needed in the root
>   of the hierarchy.

> Of course, nobody forces you to do that, and you might end up with
> exactly the same design as in C++.

What are you referring to?  In python, you get more introspection capabilities
than RTTI without writing any mechanism at all!  Arbitrary python objects can,
for example, be serialized with absolutely no serialization coding at all even
at the root of the heirarchy (provided that the objects are "well behaved").

> <snip>
> --
> Christian Tanzer                                         tanzer at swing.co.at
> Glasauergasse 32                                       Tel: +43 1 876 62 36
> A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000613/3e53dea6/attachment.vcf>

From n at n.com  Tue Jun 13 23:31:03 2000
From: n at n.com (Arinté)
Date: Wed, 14 Jun 2000 03:31:03 GMT
Subject: Threading tips.
Message-ID: <X%C15.25733$hp4.604260@newsread1.prod.itd.earthlink.net>

Its me again with questions about my embedded python app.  My scripts are
primarily to talk to devices.  If a script is running and the device does
something, then I want the app to call the py script callback if one was
set.  Do I need to block the Python threads?  What other steps do I need to
take?

thanks for the primer.




From ivanlan at home.com  Tue Jun  6 13:44:54 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Tue, 06 Jun 2000 11:44:54 -0600
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <001101bfcfde$0a6a6f20$48a2143f@tim>
Message-ID: <393D3896.CE02FA96@home.com>

Hi All-

Tim Peters wrote:
> 
> [Michael Hudson]
> > ...
> > Tangentially, does anybody know of any implementations of JavaScript
> > (not necessarily connected to a browser) that let you type interactive
> > commands into a listener and genereally muck about with the language?
> 
> Unsure about NS, but under IE5 you can type JavaScript expressions directly
> into the address bar (as if entering a URL), but prefixed with
> "javascript:".  For example, type this in:
> 
> javascript: 1+2
> 
> then hit ENTER; IE5 echoes back
> 
> javascript:%201+2
> 
> in the address bar and displays
> 
> 3
> 
> in the main window. 

Netscape echoes back the 

javascript: 1+2

line in the address bar, and does indeed display '3' in the main window.

> Unfortunately(?), vrbl assignments do not persist, so
> you're really limited to a single line of expressions:
> 
> javascript: x=3; y=4; x*x+y*y
> 
> So the answer to your question is "no" <wink>.  However, as with the Python
> shell, it's a convenient way to see what a line of JS does.
> 
> not-sure-this-new-fangled-web-stuff-will-ever-catch-on-ly y'rs  - tim
> 

<funny-thing-for-a-'bot-to-say>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From mwh21 at cam.ac.uk  Wed Jun 14 20:02:33 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 15 Jun 2000 01:02:33 +0100
Subject: SV: Python Productivity over C++
References: <Pine.GSO.4.10.10006142339350.3119-100000@sundial>
Message-ID: <m33dmf3j6u.fsf@atrus.jesus.cam.ac.uk>

Moshe Zadka <moshez at math.huji.ac.il> writes:

> On Wed, 14 Jun 2000, Eric Lee Green wrote:
> 
> > Michal Vitecek wrote:
> > >  this is highly debatable - properly written C/C++ source is pretty well
> > >  readable and self-documenting. similarly, badly written python source
> > >  makes it as unreadable as in other languages - there's not a point in
> > >  trying to prevent people from writing (un)readable source code.
> > 
> > It's harder to write unreadable Python source because you can't write code
> > like, hmm,
> > 
> > opendir(FOO,\"%s\");@files=readdir(FOO);foreach (@files) { if
> > (/^[^\\.].*\\.txt$/) { print \"%s/$_\\n\";};};closedir(FOO);
> 
> import os,re
> for _ in os.path.listdir("%s"): if re.match("^[^.].*\.txt$': print(
> "%%s/%(_)s" % vars())

But that doesn't work:

>>> for _ in os.path.listdir("%s"): if re.match("^[^.].*\.txt$': print(
  File "<stdin>", line 1
    for _ in os.path.listdir("%s"): if re.match("^[^.].*\.txt$': print(
                                     ^
SyntaxError: invalid syntax
 
> writing-obfuscated-python-isn't-that-easy-ly y'rs, Z.

Indeed.

or-was-that-your-point-ly y'rs
Michael

-- 
  SCSI is not magic. There are fundamental technical reasons why it
  is necessary to sacrifice a young goat to your SCSI chain now and 
  then.                                                  -- John Woods


From donn at u.washington.edu  Thu Jun  8 13:26:02 2000
From: donn at u.washington.edu (Donn Cave)
Date: 8 Jun 2000 17:26:02 GMT
Subject: How to tell if I can open a file or not
References: <Pine.GSO.4.10.10006081936380.15963-100000@sundial>
Message-ID: <8hokva$2rqa$1@nntp6.u.washington.edu>

Quoth Moshe Zadka <moshez at math.huji.ac.il>:
| On Thu, 8 Jun 2000, Skip Hollowell wrote:
|> Is there a way I can check the attrbutes of the file before I open it?  This
|> is easy enough in c or perl, but have yet to find an equivalent in Python.
|
| Try os.stat and the stat module, though exception handling is probably
| better.

One reason it's better than the C way:  sooner or later, you will check
whether a file can be opened, go ahead to open it on the strength of this
check, and find that it can't be opened anyway.  Your program crashes,
the air traffic control system goes out to lunch and hundreds die.

After you check, there's a period of time before you actually open
the file.  It's a brief window, but it's real, and many things can 
happen in that time if the system is busy, including changes to the
file's permission bits or even whether it exists.  Might as well
just open the file, and if it works then you have a green light.

	Donn Cave, donn at u.washington.edu


From borlak at home.com  Mon Jun 19 18:48:46 2000
From: borlak at home.com (Michael Morrison)
Date: Mon, 19 Jun 2000 22:48:46 GMT
Subject: Binary Input: read how?
Message-ID: <irx35.917$ul.59581@news1.rdc2.tx.home.com>

Here's the basic problem: I can read a binary file using open(...,'rb') but
when I assign it to a object (b = file.read()) it stores it as an escape
character or \300 - whatever range.

I want to test this byte of binary data by comparing it's bits.  b[0] & 1 ..
b[0] & 2
for x in range(30):
    if b[0] & 2**x:
        print 1,
    else
        print 0,

This doesn't work.  Tried all sorts of things.... Can't get it to work :(

What am I missing here?

thanks




From richard_chamberlain at ntlworld.com  Tue Jun 13 12:35:26 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 13 Jun 2000 17:35:26 +0100
Subject: Two newbie questions
References: <20000613165659.A7828@basajaun>
Message-ID: <Mgt15.508$9K.10789@news6-win.server.ntlworld.com>

or,

import string
st=string.join(l,'')

will do it.

if you do your random thing interactively and don't assign to i you'll
notice you get an exception so you'll have to play around based on that.

I'm sure I'll be corrected but I guess the range your trying to use
(-maxint-1 to maxint) is not an int it's a long so it doesn't work.

random.randint(0,i-1)

does work though.


Richard

Arkaitz <ma9aba at bath.ac.uk> wrote in message
news:20000613165659.A7828 at basajaun...
> Hi all,
>
> I'm learning Python, as it seems everybody around here :-), and I found
two
> problems that I can't find the solution to:
> 1. As the strings are inmutable, if I want to modify the order of the
characters
> in the string(e.g. sort them), what I do is:
>
> st = "asdfasfasdf"
> l = []
> for c in st:
>     l.append(c)
>
> Then I process the list, and the problem is that I have no idea of how I
can get
> the string back, any help?
> 2. The second question is about random numbers, how can I get a random int
between the minimum and maximum integers of the machine?
> The next code doesn't worK
> import sys, random
>
> i = random.randint(-sys.maxint-1, sys.maxint)
>
> Thanks in advance,
>
> Arkaitz.
>




From rupole at compaq.net  Wed Jun  7 17:49:00 2000
From: rupole at compaq.net (Roger Upole)
Date: Wed, 07 Jun 2000 21:49:00 GMT
Subject: Win32net.NetServerEnum Unicode problem
References: <8hlnd2$lrm$1@nnrp2.deja.com>
Message-ID: <erz%4.2026$wp.166067@pouncer.easynews.com>

It looks like you are missing the domain name, which should
be right after SV_TYPE_ALL.
        Roger Upole

<jcashman55 at my-deja.com> wrote in message
news:8hlnd2$lrm$1 at nnrp2.deja.com...
> Having trouble with the following line:
>
>         entries, total, resume= \
> win32net.NetServerEnum('Dopey', \
> 102,win32netcon.SV_TYPE_ALL, 0, 4096)
>
> When I run code that contains this line, I get:
>
> TypeError: The object can not be converted to a
> Unicode object
>
> I have run some of the the other win32net calls
> without a problem.  Anyone have an idea?
>
> Thanks.
>
> Jim
>
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From thomas at xs4all.net  Mon Jun 12 18:53:40 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 00:53:40 +0200
Subject: SMTP via pipes?
In-Reply-To: <i7n3i8.lk.ln@odin.henning-schroeder.de>; from hschroeder@gmx.net on Tue, Jun 13, 2000 at 12:12:02AM +0200
References: <i7n3i8.lk.ln@odin.henning-schroeder.de>
Message-ID: <20000613005339.Q20700@xs4all.nl>

On Tue, Jun 13, 2000 at 12:12:02AM +0200, Henning Schroeder wrote:

> I have some users who don't have a local SMTP server running although 
> sendmail is available. All mail has to be sent via pipes. When you
> call sendmail with the -bs parameter you can talk to sendmail like you are
> connected to a smtp-server. So thought I write the below attached
> wrapper. Unfortunately it doesn't run :-(
> Can anybody help or give any hints how pipes behave and why it doesn't
> work that way? Your help is very appreciated.

Setting the buffersize passed to the popen2 call rediculously low (like 1)
will probably solve your immediate problem, but I'm not sure if it's the
ideal solution. It depends on exactly how you want to send mail, but
usually, starting 'sendmail -t' in a one-way pipe and passing it a message
is easier and more convenient than using -bs.

At least, if you just have a rfc822.Message or similar object, you can do:

sendmail = os.popen("/usr/lib/sendmail -t", "w")
sendmail.write(str(msg))

Of course, this is more of a 'fire and forget' solution: you don't have
extensive error checking on your mail. If you want something more elaborate,
including splitting up the list of recipients into sendmail-accepted
numbers, take a look at Sendmail.py in Mailman.

But if you just want this to work, 

> 	r, w = popen2(SENDMAIL)
 	r, w = popen2(SENDMAIL, 1)

should do it ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jerome.quelin at insalien.org  Fri Jun 30 11:31:13 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 17:31:13 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8F635367Cgmcmhypernetcom@199.171.54.154>
Message-ID: <962379073.1230449645@news.libertysurf.fr>

gmcm at hypernet.com (Gordon McMillan) wrote:
>>But how to pass this dictionnary to the parent class constructor?
>class F(Frame):
>  def __init__(self, *args, **kwargs):
>    apply(Frame.__init__, (self,)+args, kwargs)
>Generally you have to muck with the kwargs dict before passing it to the 
>base class __init__.
That is exactly what I want to know: what this mucking consists of?

Jerome
--
jerome.quelin at insalien.org


From matt at nightrealms.com  Tue Jun 27 23:16:37 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Tue, 27 Jun 2000 19:16:37 -0800
Subject: Guesstimate for 1.6 beta timeline?
Message-ID: <Cid65.2526$h8.109632@news-west.usenetserver.com>

The Python home page says that Python 1.6 (final) will be released
around June 1, which is currently about 25 days ago, and it still
looks like Python is in alpha.  Does anyone have some guesstimate
as to when beta will be out?

Thanks in advance.





From paulb at infercor.no  Mon Jun 19 09:01:56 2000
From: paulb at infercor.no (Paul Boddie)
Date: Mon, 19 Jun 2000 15:01:56 +0200
Subject: ANNOUNCE: jdbi-0.2
References: <m266r8v1i9.fsf@iscariot.local.net>
Message-ID: <394E19C4.91A2F344@infercor.no>

Sverre Stoltenberg wrote:
> 
> JDBI is a jpython module that can connect to any database that supports
> jdbc. I'm trying to make it DB-API spec v2.0 compatible.
> 
> http://www.terminator.net/jdbi/

Excellent! I know that a number of people have been looking for this kind of
portability between the JPython and CPython environments. Now I must try and
find some time to take a closer look...

Paul


From wjdandreta at worldnet.att.net  Sun Jun 18 11:04:56 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 18 Jun 2000 15:04:56 GMT
Subject: Regular expressions vs find?
Message-ID: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net>

I recently read a message that suggested that using regular expressions
might be faster than the find function. I've heard the term regular
expression used from time to time but I don't know what it means. Could
someone give me a simple explanation?

Bill




From tjjudson at yahoo.com  Fri Jun 16 04:47:00 2000
From: tjjudson at yahoo.com (TJ Judson)
Date: Fri, 16 Jun 2000 01:47:00 -0700 (PDT)
Subject: Conversion from proprietary satellite control script language
Message-ID: <20000616084700.29919.qmail@web2006.mail.yahoo.com>

I've culled through all of the discussions on the "goto" statement.?
I believe the summary is:

1. It doesn't exist, but it would not be hard to add (The necessary
opcodes are there -- well, the one you need: JUMP_ABSOLUTE --
mwh21 at cam.ac.uk)

2. There are a few "good" uses for it:  (jump to cleanup code, for
use by code generators, as a performance optimization -- John
Mitchell)

I have a new and much better reason :-) to add a goto statement... To
support conversions from other, existing scripts/languages.

I need to convert an extensive set of scripts from a proprietary
satellite control language to some new, open source scripting
language.  Having looked at them, I think Python is superior to the
others.  I can handle everything except for the label/goto construct.
 This is used infrequently, but enough that it poses a major problem.
 I need to convert millions of dollars of existing scripts, so
complete automation is the only answer.  It's cost prohibitive to
rework the logic of the original scripts.  As I said, I can handle
everything except for the goto.

I'm afraid my only other answer is to use a language that supports
"goto", like Tcl.

Question/Plea:
--------------
If it is easy to add, then wouldn't this be a good feature to add for
the sole purpose of easing (promoting) the conversion to Python?  I
think it would be neat if Python were being used to fly satellites. 
There are many "old" scripting languages out there and most of them
have a goto/label construct.

It could be added as a standard feature with a strong warning
regarding its use.  Or it could be provided only as a patch...

If the powers that be decide "too bad, you're screwed", does anyone
have any pointers on how I might add this myself?  I have built my
own interpretor before, so I have some hope of being able to figure
it out on my own, but I haven't yet poked around the Python
internals.  I really wouldn't want to have my own version.  I hope
the powers that be see the argument that it would be better to add a
less than perfect construct into the language, for the sole purpose
of gaining more widespread use.  With more widespread use, the
language will ultimately grow and become better than it otherwise
could be.  Providing this "necessary evil" opens up the possibility
of many more people being able to adopt Python in favor of their old
languages.

(Sorry for drumming up what is a well worn subject, but I'm
desperate)

Thanks,
-TJ Judson
judson at iname.com



__________________________________________________
Do You Yahoo!?
Send instant messages with Yahoo! Messenger.
http://im.yahoo.com/



From wjdandreta at worldnet.att.net  Wed Jun 21 12:16:34 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 16:16:34 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <8iqin6$muc$1@newshost.accu.uu.nl>
Message-ID: <CT545.13338$C44.784749@bgtnsc05-news.ops.worldnet.att.net>

Hi Martijn,

This suggestion is the most appealing to me for what I am doing because I
always know exactly how many fields there are.

Now if there were just some way to similarly name a slice...

Bill

>In many cases I'd use tuple/list unpacking for this:
>
>date, address, name, whatever = string.split(line, ",")
>
>If you don't know how many fields you're going to get, this can be
>more tricky. But imagine you know the first 4:
>
>date, address, name, whatever = string.split(line, ",")[:4]
>
>Of course changing a field is harder here. Generally I use the idiom of
>constructing new lists (or dictionaries or objects) in that case,
>though. After all, if you're really going to manipulate your fields
>it makes sense to wrap them up in objects in any case.
>
>Regards,
>
>Martijn







From claird at starbase.neosoft.com  Thu Jun 29 17:14:45 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 29 Jun 2000 16:14:45 -0500
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org> <8F62AF73Egmcmhypernetcom@199.171.54.194>
Message-ID: <36D5F4DFCA9BCA46.B5AE15F2F19F393C.28F9B766F10B5E23@lp.airnews.net>

In article <8F62AF73Egmcmhypernetcom at 199.171.54.194>,
Gordon McMillan <gmcm at hypernet.com> wrote:
			.
			.
			.
>Um, you said you had to be portable. Not only is tail not standard on 
>Windows, but os.popen is flaky there too (depending on circumstances too 
>painful to enumerate).
I think I'm going to start growling every time
I hear that.  To be precise:  popen* flakiness
is one of the chief discouragements in my use
of Python (which is, I recognize, slightly
skewed from other's).
>
>However, tail (usually - some implementations differ) does almost what the 
>following code does. Note that even this isn't portable, because Windows 
>doesn't have ino's to check. I have the code this is extracted from 7x24 
Guys, guys:  there are enough seriously hard
problems in the world.  Don't make this one.

Win* FSs don't have i-nodes.  So a portable
formulation of the requirements shouldn't
involve i-node.  That's OK.  Just poll for
either readability or file size.  While I
confess I haven't verified that Py-built-ins
for these are properly portable, they REALLY
need to be if they aren't already, and I'm
willing to write test drivers and demonstra-
tions if these remain in doubt for one more
round of c.l.p dialogue.
			.
			.
			.
-- 

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


From mwh21 at cam.ac.uk  Fri Jun  9 18:33:36 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 23:33:36 +0100
Subject: Gnome Applets and Python
References: <39416F2B.96AD2322@oreilly.com>
Message-ID: <m3bt1av60v.fsf@atrus.jesus.cam.ac.uk>

"Stephen R. Figgins" <fig at oreilly.com> writes:

> Any good suggestions on Gnome Applets and Python information?

Use the example that comes with gnome-python.  Read the docs for the C
API.  Change, test, refine, test.  This seems to work for me...  I'll
post my code to my starship page sometime soon...

Cheers,
M.

-- 
  languages shape the way we think, or don't.
                                        -- Erik Naggum, comp.lang.lisp


From answer4u at earthlink.net  Fri Jun 16 21:17:50 2000
From: answer4u at earthlink.net (Money At Home)
Date: Fri, 16 Jun 2000 18:17:50 -0700
Subject: How You Can Make $350.24 A Week Or More In Your Spare Time...
Message-ID: <200006170125.SAA14367@avocet.prod.itd.earthlink.net>

How You Can Make $350.24 A Week Or More In Your Spare Time...

Are you looking for a REAL money making opportunity?

Don't waste another minute on those get-rich-quick
programs, you can work for legitimate companies for
real paychecks right at home.

Choose from simple jobs such  as:

*Word Processing
*Data Entry
*Customer Service/Support
*Clerical
*Claims Processing
*Sales
*Telemarketing
*Surveyors

Plus other simple tasks for regular pay.

These are real money making work at home jobs offered
by legitimate companies that hire people like yourself
to work as independent contractors.

Begin immediately, only requirements are that you be over
18 years of age and a U.S. resident.

Request a free listing of current work at home jobs
right now by sending an email to bizgold2k at earthlink.net
with "Work At Home" in the subject heading.

Best Wishes,

Work At Home Center




From just at letterror.com  Thu Jun 22 11:57:30 2000
From: just at letterror.com (Just van Rossum)
Date: Thu, 22 Jun 2000 16:57:30 +0100
Subject: Am I the only one to receive the messages several times????
In-Reply-To: <7BD10B680501D411B9DF009027E06F32011AD2@EXCHANGE>
Message-ID: <l03102802b577e7881f31@[193.78.237.153]>

At 2:24 PM +0200 22-06-2000, Max M?ller Rasmussen wrote:
>This starting to be a drag.

Yep. It seems only messages that come from usenet are duplicated. And they
just keep on coming...

>Furthermore I received a message from BeOpen about one of my messages that
>was to long for their listserver.
>
>Perhaps a BeOpen list is mailing to the python list, and the Python list is
>mailing back to them???? This should cause a loop and make messages appear
>several times.

I got one, too. Barry W. told me:

"Let's just say there was a local configuration error"

Whatever that means ;-)

Just





From 12klat at sightreader.com  Thu Jun  1 20:25:31 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Thu, 01 Jun 2000 17:25:31 -0700
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and 
 others" is this possible?
References: <393692FB.8610EBCA@dallas.net> <007901bfcbf2$c3b4e680$1906a8c0@fc.fenx.com>
Message-ID: <3936FEFB.BA248CA3@sightreader.com>

On NT I get:

>>> mylist = os.popen('dir').readlines()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
OSError: (0, 'Error')

(same thing with any command)

popen (command[, mode[, bufsize]])
    Open a pipe to or from command. The return value is an open
    file object connected to the pipe, which can be read or written
    depending on whether mode is 'r' (default) or 'w'. The bufsize
    argument has the same meaning as the corresponding argument
    to the built-in open() function. The exit status of the command
    (encoded in the format specified for wait()) is available as the
    return value of the close() method of the file object, except that
    when the exit status is zero (termination without errors), None is
    returned. Availability: Unix, Windows.

- Ken Seehof

Emile van Sebille wrote:

> Jim, Take a look at os.popen, as in: import osmylist =
> os.popen('locate myname').readlines()for eachfile in mylist: do
> something with eachfile HTH,
> Emile van Sebille
> emile at fenx.com
> -------------------
> ----- Original Message -----
> From: Jim
> To: python-list at python.orgSent: Thursday, June 01, 2000 9:44
> AMSubject: Python to call commands, e.g. "find , perl scripts , nmap ,
> and others" is this possible?
>
>
> Dear list,
>
> I'm new to Python, and I'm sorry if I'm asking about anything that has
> been
> already discussed.
>
> I would like to use Python to call commands, e.g. "find , perl
> scripts  , nmap , and others" is this possible?
>
> I do not need GUI. All I need is to link the out put and work with it.
> Is perl or shell scripts my only
> answer?
>
> jim
>
> --
> Proprietary & Confidential.  The information transmitted is intended only for the person or entity to whom it is addressed and may
> contain confidential and/or privileged material.  Any review, retransmission, dissemination, or other use of, or taking of any
> action in reliance upon, this information by persons or entities other than the intended recipient is prohibited.  If you received this in error, please
> contact the sender and delete the material from any computer.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000601/6f2c88c8/attachment.html>

From db3l at fitlinxx.com  Wed Jun 14 21:45:58 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 21:45:58 -0400
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com>
Message-ID: <ug0qflns9.fsf@ctwd0143.fitlinxx.com>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> Now only if somebody could enhance that to check inherited methods too (the
> "more general case").  That is, if a method is not found, go recursively
> into the superclasses.  The following example is not correct at the moment:

Ah, good point - try this variant - I switched to getattr() which
takes the inheritance into account (you could also write your own
recursive function that used Instance.__bases__ to look in the
superclasses).

Note that the object returned by getattr() is a method, whereas the
raw dictionary object is just a function, so there's a small change in
the type comparison.  I'd be interested if anyone more familiar with
the introspective/metaclass features of Python might have a cleaner
way to identify the class member functions:

    def verify_class (Reference, Instance):

	import types

	print "Checking %s for conformity with interface %s" % \
	      (Instance.__name__,Reference.__name__)

	for curmethod in Reference.__dict__.keys():
	    if type(Reference.__dict__[curmethod]) == types.FunctionType:
		try:
		    if type(getattr(Instance,curmethod)) != types.MethodType:
			raise TypeError
		except AttributeError, TypeError:
		    print "Method %s not supported" % curmethod


which for your test cases yields:

>>> verify_class(MyInterface,MyBase)
Checking PostBase for conformity with interface PostInterface
Method method3 not supported
Method method4 not supported
>>> verify_class(MyInterface,MyDerived)
Checking PostDerived for conformity with interface PostInterface
Method method4 not supported

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


From n at n.com  Fri Jun 30 10:26:59 2000
From: n at n.com (Arinté)
Date: Fri, 30 Jun 2000 09:26:59 -0500
Subject: Is this a leaker?
References: <GH075.17918$NP5.455672@newsread2.prod.itd.earthlink.net> <0S075.17928$NP5.457944@newsread2.prod.itd.earthlink.net> <slrn8lp6ul.u8u.thomas@centurion.xs4all.nl>
Message-ID: <8ji745$1d5u$1@news2atm.raleigh.ibm.com>

> This depends on how you document your function :) If you document it as
> returning a borrowed reference, the caller should not DECREF the returned
> value. If you document it as returning a new reference, it should take
care
> of the DECREF itself. In this case, it doesn't make sense to return a
> borrowed reference: If you do a DECREF on the fresly-created
> Py_BuildValue()-object, it'll end up destroyed !

I was speaking of returns to python scripts.  All of the 'return
Py_BuildValue()' return to python scripts.  I don't have to decref in py
script files do I?





From gansevle at cs.utwente.nl  Wed Jun 28 07:01:20 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Wed, 28 Jun 2000 11:01:20 GMT
Subject: SystemError: bad argument to internal function
Message-ID: <8jcltq$1li$1@nnrp1.deja.com>

 Hi,

sometimes, not reproducable, I get a

    SystemError: bad argument to internal function

The location of the error varies and most of the time it doesn't occur.
The only thing common to all occurences of the SystemError is that it
happens whet I call 'tree.get(name)' (see below)

Can anybody help me with this?
The app is a rather critical app, so I'm worried about this.

here is the latest traceback.

Traceback (innermost last):
  File "E:\SAMBA\session1.py", line 105, in main
    logon (user, server)
  File "E:\SAMBA\session1.py", line 53, in logon
    sess.start (status)
  File "E:\SAMBA\smb\profile.py", line 1346, in start
    dir.dump ()
  File "E:\SAMBA\smb\profile.py", line 127, in dump
    self._dump (self.__data, self.__path)
  File "E:\SAMBA\smb\profile.py", line 1060, in _dump
    value, src = tree.get (name)
SystemError: bad argument to internal function

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From aahz at netcom.com  Wed Jun  7 10:01:59 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 14:01:59 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <8hiq8d$ngm$1@newshost.accu.uu.nl> <slrn8jr8qm.b4q.wtanksle@dolphin.openprojects.net> <8hlk0t$h4q$1@newshost.accu.uu.nl>
Message-ID: <8hlkkn$gu3$1@slb2.atl.mindspring.net>

In article <8hlk0t$h4q$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>> On 6 Jun 2000 12:19:25 GMT, Martijn Faassen wrote:
>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>>> On 5 Jun 2000 12:01:04 GMT, Martijn Faassen wrote:
>>>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:

May I suggest that the two of you take it to e-mail?  Or should we send
the Spanish Inquisition?  ;-)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From daniel.lundqvist at vogon.se  Wed Jun 28 08:32:45 2000
From: daniel.lundqvist at vogon.se (daniel.lundqvist at vogon.se)
Date: Wed, 28 Jun 2000 12:32:45 GMT
Subject: buildtime flags
Message-ID: <8jcr9d$5hm$1@nnrp1.deja.com>

Hi,

Is there a way to get the link and compile flags which were used to
build python runtime?

Kind regards,
Daniel Lundqvist


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


From bjorn at roguewave.com  Fri Jun  9 19:00:36 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 09 Jun 2000 17:00:36 -0600
Subject: I need a lesson - howfast.py Revisited
References: <8hrphd$6m7$1@ionews.ionet.net>
Message-ID: <39417714.1A36B548@roguewave.com>

To create an executable, you'll need Installer from Gordon's Starship
pages (http://starship.python.net/crew/gmcm/distribute.html)

The command you're looking for is:

  python c:\where\you\installed\installer\Freeze.py myscript.py

(there are variations on this for different purposes of course, but this
should get you going ;-)

keep-hitting-those-walls'ly y'rs
-- bjorn

"F. G. Brown" wrote:
> 
> Greetings!
> First off I want to thank everyone for their responses to my post a few days
> back (2000 06 07).  I was able to learn much from all of your suggestions.
> At the same time I can see that it's going to take alot of effort on my
> part, to be proficient in using Python, so you might be hearing from me now
> and then when I hit a wall.  If anyone has a suggested challenge for me to
> work on next, I'll give it a try.
> Thanks again, Bjorn, Brent, Scott, David, Brett, & Moshe.
> Regards,
> Fred
> 
> Below is what I finally came up with using everyone suggestions.
> I next am going to see if I can make it an executable file in a DOS
> and Windows environment.  I seem to remember reading that I
> can download a program for doing that somewhere.
> Time to go hunting!
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> !!!!!!!!!!!!!!!!
> 
> howfast.py
> 
> decide = 'y'
> while decide == 'y':
>      distance = input("What is the distance you traveled in miles?  ")
>      time = input("How many seconds did it require?  ")
>      speed = distance*3600.00/time
>           print "You were going %d mph!" %speed
>      if speed > 75:
>           print "Hey you better slow down!"
>      decide = raw_input ("Would you like to do another? (y)es or (n)o")
>      if decide == "y":
>           print 'Here we go!'
> else:
>           print 'See ya!'
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From aahz at netcom.com  Tue Jun 27 10:07:30 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 14:07:30 GMT
Subject: threading module
References: <3958265D.906DECF5@yahoo.com>
Message-ID: <8jacf2$i4g$1@slb6.atl.mindspring.net>

In article <3958265D.906DECF5 at yahoo.com>,
Al-Amerrho H. Amerin <al_amerin at yahoo.com> wrote:
>
>I was using the threading module's activeCount() method when I noticed
>that when you import it, it spawns a thread, MainThread. And this
>becomes a problem when I call threading.activeCount(). It returns the
>number of threads I spawned + 1.

Nope, it doesn't spawn a thread.  That thread existed when you started
Python; what threading does is *register* that thread with itself.

>Python 1.6a2 (#9, Jun 22 2000, 00:09:08)  [GCC 2.95.2 19991024
>(release)] on linux2
>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>
>>>> import threading
>>>> threading.activeCount()
>1
>>>>
>
>I hope someone can enlighten me as to the reason why activeCount()
>includes the MainThread and why this shouldn't be fixed so
>activeCount() will only return the number of threads that a caller
>spawns.

Because the main thread is a thread just as much as any other thread.
For example, you could exit the main thread and your other threads would
continue running.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From 12klat at sightreader.com  Tue Jun  6 20:05:38 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Wed,  7 Jun 2000 00:05:38 -0000
Subject: No subject
Message-ID: <393D9CF2.DE8B0897@sightreader.com>

no content


From noone at do.not.use  Sat Jun  3 11:55:14 2000
From: noone at do.not.use (Russell Turpin)
Date: Sat, 03 Jun 2000 10:55:14 -0500
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com>
Message-ID: <39392A62.BE04911@do.not.use>

Christian Tismer wrote:
> Please note that there is an alternative to using doubly linked
> lists ..

My miscommunication. When I said deque, I didn't mean linked
lists, but just a vector that is allocated with spare space 
at both ends, at least whenever elements are added to or 
removed from the start. There's really no reason to copy or
reallocate a vector, every time the first element is 
deleted or inserted.  I think vector is the clear choice
over linked list as a default implementation; but I still 
want to be able to add and remove from both ends efficiently. 

Beating-a-dead-horsing-ly yrs,
Russell


From hodnicak at bigfoot.com  Tue Jun  6 05:19:10 2000
From: hodnicak at bigfoot.com (radoslav hodnicak)
Date: Tue, 6 Jun 2000 11:19:10 +0200
Subject: Nympy for win
Message-ID: <Pine.LNX.4.21.0006061109170.30720-100000@chelin.chtf.stuba.sk>

Hello.

I would like to use the Numpy package with my windoze/python1.5.2 system,
but it needs distutils and distutils need m$ visual c++ which I don't own.
Is here anybody who has it compiled and could send me the
binaries? Thanks.

Regards, radoslav

ps: I don't know if any distutils-developer reads this, but how about
adding support for free c compilers for win, like lcc-win32?
(http://www.cs.virginia.edu/~lcc-win32/)



From hzhu at knowledgetrack.com  Fri Jun 16 20:18:34 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Sat, 17 Jun 2000 00:18:34 GMT
Subject: numpy bug?
Message-ID: <slrn8klh53.56e.hzhu@rocket.knowledgetrack.com>

>>> from Numeric import *
>>> a = arange(16)
>>> (a-5)/4
array([-1, -1,  0,  0,  0,  0,  0,  0,  0,  1,  1,  1,  1,  2,  2,  2])


I believe this is not correct.  And the stock python gives correct result:

>>> for i in a: print -i/4
... 
0
-1
-1
-1
-1
-2
-2
-2
-2
-3
-3
-3
-3
-4
-4
-4

Any enlightenments?


-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From s2mdalle at titan.vcu.edu  Tue Jun 20 21:41:03 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Tue, 20 Jun 2000 20:41:03 -0500
Subject: Python sockets/select
Message-ID: <8ip6kt$1lc$1@bob.news.rcn.net>

I've been programming with sockets in python recently, and it's all
been going well and it's relatively easy with a few exceptions....

When I select() sockets for reading, writing, and error conditions,
there are sockets (usually connected to clients who have already
quit or died unexpectedly) that don't show up in the error queue,
but continually flag select() as being readable.  When I read them,
of course, '' comes back.

For some reason, sometimes '' comes in for valid clients, so I 
can't just disconnect a client because I get that from them.  Is
there a way to detect which clients are alive and which aren't?
What causes a client to send '' when I didn't tell it to, according
to select()?  (The clients are also written in python)

Any help would be appreciated.
-- 
David Allen
http://opop.nols.com/



From gwhite at wendigo.bio.dfo.ca  Sat Jun 10 10:55:14 2000
From: gwhite at wendigo.bio.dfo.ca (George N. White)
Date: Sat, 10 Jun 2000 14:55:14 GMT
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
In-Reply-To: <mtya4ha05p.fsf@astron.berkeley.edu>
References: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD> <14647.58023.493846.225899@buffalo.fnal.gov> <m3wvk1hm2u.fsf@chinon.cnrs-orleans.fr> <vmebt1dms3s.fsf@buffalo.fnal.gov> <m3ln0hh2ix.fsf@chinon.cnrs-orleans.fr> <mtya4ha05p.fsf@astron.berkeley.edu>
Message-ID: <Pine.SGI.4.21.0006100937230.10317-100000@wendigo.bio.dfo.ca>

On 7 Jun 2000, Johann Hibschman wrote:

> Konrad Hinsen writes:
> 
> > Charles G Waldman <cgw at fnal.gov> writes:
> >> 
> >> What do you think the consistent behavior should be?  Should
> >> math.exp(-10000) return 0, or raise some exception?
> 
> > I have yet to encounter a case where returning 0 would cause a problem,
> > so I'd vote for that solution. But raising an execption systematically
> > would be fine as well, one could then catch it and return 0 if that's
> > what is wanted.
> 
> Unsignalled underflows are very bad.  Most of my work conceptually
> fits on a log-log scale, so the difference between 0 and 10^-100 is
> immense.  If something underflows, I want to know that at the point of
> underflow, not at the point where I try to take the log of it.
> 
> -- 
> Johann Hibschman                           johann at physics.berkeley.edu

There are a couple issues in this thread:

1)  cross-platform consistency

2)  treatment of exceptions: underflow, overflow, invalid, ...

For 1), Sun's freely distributable math library should eliminate the
effects of differences in vendor-supplied math libraries

For 2), support for reliable reporting of f.p. exceptions when they occur
conflicts with techniques (e.g., combined multiply+add: y=a*x+b) used to
get better f.p. performance in new processor designs.  Since the
marketplace favors speed over all else, the conflict will likely be
resolved in favor of performance.  This is reflected in the lack of
support for f.p. exceptions in language standards.  It is much faster and
easier to simply store a 0, +/-Inf, or NaN and keep going.  Until we have
benchmarks that explicity test underflow and overflow behaviour, vendors
are going to focus on performance for the simplest cases.  In practical
terms, users need to a) write code that doesn't rely on exception handlers
and b) push for support in language stds. for features needed to work with
signed 0, NaN's and Inf's in (e.g.: don't optimize away expressions like
'x != x').

As it happens, just this week I encountered a subroutine that was
returning NaN's from a calculation of the form: y*log(y) where y=exp(-x)
and x was large enough to cause underflow, giving 0*(-Inf)=NaN.  I tried
trapping the underflow, but there were _thousands_ of log(y) expressions
used in a way that did not cause any difficulty before I reached the one
that was returning a NaN.  In this case, trapping underflows gave so many
false positives that it didn't help identify the real problem area in the
code.

Perhaps a better approach for Johann's specific example would be to use a
special non-underflow function, exp_p(-x), that calls an error routine
when x exceeds some threshold, or add a test before exp(-x) is called.

--
George N. White III <gnw3 at acm.org> Bedford Institute of Oceanography



From hat at se-46.wpa.wtb.tue.nl  Tue Jun 27 04:21:29 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 27 Jun 2000 08:21:29 GMT
Subject: boolean true and false values.
References: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl> <memo.20000626221037.43263C@digitig.compulink.co.uk>
Message-ID: <slrn8lgsjq.us.hat@se-46.wpa.wtb.tue.nl>

On Mon, 26 Jun 2000 22:10 +0100 (BST), Tim Rowe <digitig at cix.co.uk> wrote:
>In article <slrn8lf3q9.us.hat at se-46.wpa.wtb.tue.nl>, 
>hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:
>
>> I understand that now.
>> However, I am not quite sure that that is what one wants.
>
>Maybe it isn't; then another language may be more appropriate. When it 
>/is/ what one wants then Python may be your choice.

:-)
Changing Python is not my goal, I am only trying to really understand the
differences between C-like declarations, and the Python approach.

Digging around by finding counter-arguments is one way to find the core reason
for the difference in approach. Once you have this on the table, you can
make a good judgement about the design rationale for the c-like approach as
well as the python approach.

With a good language, it should be really hard to find a valid argument for
which the language design rationale does not provide a reason.
If one does find one, that argument is a weak point in the language design, which
probably means that some more thought about that part of the design should be put in.


Until now, I like the basic idea of Python of variables not having a type, but
I don't really see the advantage.
That means that if I would design the Python language, I would like to change
something (although I am not quite sure what, changing anything at this
fundamental level is likely to break just about any existing program).
However, I am not one of the language designers of Python. The best I can hope
for is that one or more of them is reading this discussion, and takes it into
account for the next version.

>Folks don't seem to be able to get away from the notion of a "perfect" 
>language, ideal for all applications. I don't believe it exists, and if 

I am not talking about ideal for everything.
Python is supposed to be aimed at 'the masses' in the next project of Guido. In
that context, enforcing something like type-safety of variables in order to
have better support for invariants seems ok to me.
For example, Perl is aimed at experienced programmers. They want a fast hack
language, not a cleanly designed language aimed at learning how to program.
Consequently, I consider type-safety in Perl as quite not important (and that
is an understatement).

>you "improve" Python for some you may well break it for others. Surely the 
>competent programmer has a number of languages under his/her belt?

I am aware of that problem. For that reason alone, it is not very likely that
any change in Python will result in this discussion. However, in my opinion it
is a step forward to have weaknesses in the language exposed. That gives the
language designers (and not only those of Python) the chance to take it into
account, and users better equipment to judge usability of the languages.

>first language I was taught was FORTRAN and would consider INTERCAL to be 

Hmm, I read the Intercal manual, and am anxious to write something in it,
although the right program hasn't yet arrived. 
Fortran is also on the list of languages to read about.
[ please don't tell any one about my habit of learning eeky languages :-) ]

>> In other words, variables are in reality quite tightly connected to 
>> their
>> values because a meaning is given to the variable.
>> For example, pointing the variable to a value of another type does not 
>> fit in
>> the idea of invariants. For example, by executing 'total="kjerg"', the
>> invariant is not longer true, since "kjerg" is not the sum of elements 
>> (it is
>> not even a number).
>
>But that's /exactly/ right! If you execute 'total="kjerg"' then your 
>invariant is indeed violated!

So, why is it then that Python allows this ?
For a language aimed at professionals, I understand (although I wouldn't
introduce it), but for a clean language such as Python ?

>> Thus, to a good programmer, the dynamic nature of variables is not 
>> needed. To
>> bad programmers, it is a tool which can be used quite abusively.
>
>As can any tool, of course. The dynamic typing is indeed never /needed/ 
>(probably one of the things Dijkstra proved!), but it's often very /very/ 
>convenient indeed. If it's likely to be a problem to you -- or to the 

Hmm, why is it convenient then ?
Without declarations you can simply pick another identifier when you need a new
variable, rather than re-use existing identifiers, which may cause all kind of
bugs and/or hamper readability.


Albert
---
Look ma, windows without Windows !!


From pduffin at hursley.ibm.com  Tue Jun 27 06:51:45 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Tue, 27 Jun 2000 11:51:45 +0100
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu>
Message-ID: <39588741.FB7EB5FE@hursley.ibm.com>

Donn Cave wrote:
> 
> Quoth ge at nowhere.none (Grant Edwards):
> | In article <kuzoobue0g.fsf at lasipalatsi.fi>, Erno Kuusela wrote:
> |>>>>>> "Tony" == Tony Keating <keating at mech.uq.edu.au> writes:
> |>
> |>    Tony> start the program, then send various strings to the program,
> |>    Tony> read some output from the program then send some more
> |>    Tony> strings to the program.
> |>
> |> there are a couple of expect modules for python (so named because
> |> there is an old tcl-based language called expect for doing just this).
> |> parnassus lists 3 modules (<URL:http://www.vex.net/parnassus/
> |> apyllo.py?find=expect>), i haven't
> |> used any of them however...
> |
> | Or you can roll your own using the popen2 module.
> 

I think that you will find that Expect offers a lot more than a simple 
pipe to the executable.

> OK, but check to see if your application will work reliably on
> a pipe, before you put too much work a pipe solution.  I append
> a short sample program that works with the shell as a remote
> application, but deadlocks with awk.  The awk behavior is what
> I would expect from most programs, so normally you're sunk.
> 

The main problem is that most interactive programs rely on the C library's
setting of the buffering of the stdout IO channels. Basically when 
stdout is connected to a tty it has line buffering but when connected to 
a pipe, or a socket, or any other type of channel the buffering is normally 
set to about 4K.

If you have access to the source of the interactive program then you can
simply add a call to setlinebuf for stdout; stderr is always line 
buffered. If not then the only solution is to use something like Expect
(actually there is nothing like it) which uses pseudo ttys to ensure
correct behaviour and also has some really nice buffer handling.

Obviously you could do all this and the GUI in Tcl/Tk but as you are
asking this about Python then you may want to look into Python versions
of Expect. If they are not satisfactory then you could probably write
a little Tcl/Expect wrapper around the application which forces the
correct buffering and then control that using Python.


From al at findacorp.com  Sun Jun 11 14:01:38 2000
From: al at findacorp.com (Alan Alper)
Date: Sun, 11 Jun 2000 14:01:38 -0400
Subject: HELP with HTTPLIB/URLLIB
Message-ID: <sk7kt5ajh51169@corp.supernews.com>

I need to duplicate the action of a browser POSTing form data to a server.
GET is easy, but I don't know how to do POST. Any and all help, preferably
with examples would be welcome.  Essentially, I'm trying to write a small
agent to fetch classified ad data from several newspaper websites for me so
that I don't need to visit the sites manually.  I have no clue how to do a
POST.

Thanx in advance.

Al




From clive at pisquaredoversix.force9.co.uk  Fri Jun 16 11:50:22 2000
From: clive at pisquaredoversix.force9.co.uk (Clive Tooth)
Date: Fri, 16 Jun 2000 16:50:22 +0100
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie>
Message-ID: <8idie3$l4i$1@mail.pl.unisys.com>

Russell Wallace wrote in message <394889C3.1ED5 at esatclear.ie>...

>Robert Israel wrote:
>> Plouffe has another algorithm that can find a digit in an arbitrary
>> base.  This takes C*n^3*log(n)^3 time for the n'th digit, so it is not
>> faster than other methods that would compute the first n digits, but it
>> does take very little memory.
>
>Correct me if I'm wrong, but I thought memory was the limiting factor?
>i.e. that a good algorithm on typical hardware can fill all the
>available memory with digits of pi in a few days or so?

You are wrong. The discussion was about methods of calculating digit N of pi
(in some base) without computing digits 1 thru N-1 . This can be done in
about O(log(N)) bits of memory.

--
Clive Tooth
http://www.pisquaredoversix.force9.co.uk/
End of document





From ted.drain at jpl.nasa.gov  Tue Jun 20 11:33:21 2000
From: ted.drain at jpl.nasa.gov (Ted Drain)
Date: Tue, 20 Jun 2000 08:33:21 -0700
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <ted.drain-2006000833210001@mjordan.jpl.nasa.gov>

The one thing that I really hate about the white space blocking is when
you have to reformat because of a code change.  This is something that has
hit me quite a few times.  

I'll write some code, and as I start debugging/prototyping/etc etc, I find
that I have to enclose some existing code in a control block (if,etc). 
With C/C++, I just put the code in brace, tell xemacs to re-indent the
block, and everything's good.  With Python, I have to hand edit everything
because the editor can't figure out how to do the indentation after I've
inserted the control block.  Once you get a few nested indentation blocks,
this gets very confusing and leads to bugs that are hard to find (code
that did work suddenly doesn't).

I love Python, but I'd really prefer to have open/close brackets for the blocks.

Ted


In article <394ac7ea.609024015 at nntp.interaccess.com>,
olczyk at interaccess.com wrote:

>After using Python for a while I've come to the conclusion that
>using whitespaces to determine the begining and ending of blocks
>is bad for three reasons.
>
>Before I list those three reasons let me make two points. People say
>that the indentation requirements don't matter because you generally
>indent the same way. That may be true, but do you think it is a good
>idea to make conventions requirements ( remember that conventions
>start flame wars )? The second is that modern editors with good
>indentation rules ( like emac's python mode ) make indentation based
>coding easy. The thing is that by doing this you are abrogating
>control flow to the editor. When it was the editor indenting code
>which was well defined then there was no problem. By having
>the code depend on indentation, and the editor do the indentation
>for you, you give the editor responsiblity in how the code will flow.
>
>Now on to my points.
>1) The first is rather stylistic and may not appeal to most, but I
>have often found it useful. When debugging code I often put in
>code to help me figure out whats happening. I also avoid indenting
>such code so that it easy to remember where it is later when I want to
>take it out. An example ( assuming I use begin and end to denote
>blocks ):
>
>def Average():
>begin 
>      sum=0
>print "number of numbers=",len(numbers) #illegal
>i=0
>      for n in  numbers:
>      begin
>print "n=", n #illegal
>            sum=sum+n
>print "sum=",sum #illegal
> i=i+1                     #illegal
>if sum>MaxNumber*I:#illegal
>begin                     #illegal    
>  print "sum is too large"#illegal
>end                        #illegal
>     end
>     if len(numbers)!=0:
>     begin
>        return sum/len(numbers) 
>     end
>     return 0
>end
>
>You can immediately see code that you used for debugging and remove
>it. However right now writing code that way makes it illegal.
>
>2) You can easily insert a tab introducing a bug.  Consider the
>following code ( similar to a bug I recently fixed ):
>
>def FindInvoice(idnum): 
>      for invoice in invoices:
>            if invoice.id=idnum:
>              return invoice
>            return None # should be indented one less tab.
>
>....
>
>for lines in file.readline():
>     idnum=lines[:6]
>     invoice=FindInvoice(idnum)
>     if not invoice:
>              invoice=InvoiceClass(idnum)
>         invoices.append(invoice)
>....
>Initially the line in error was indented correctly, but someplace
>along the line the extra tab was introduced. A lot of effort then had
>to go in to finding this bug. With begin...end or braces this is less
>likely to happen.
>
>3) Finally. Python is a language where it is almost easy to write
>extensible programs. To illistrate what I mean consider this example
>of a python program I recently wrote:
>There is a C++ project which involves many project files and workspace
>files ( the project files are roughly equivalent to makefiles, the
>workspace files primarily are used to couple projects and define
>dependencies ). Version 1 simply read all the project files and 
>generated a call to compile each one with all different configuartions
>(debug , release, alpha, beta etc..) . Since this built everything
>even if it was already built, and since this had to be run several
>times ( sometimes compiles would fail because a project that the
>present project was dependent on wasn't compiled ), I made version
>two. Version 2 read the workspace file and organised the projects
>into workspaces. Version 3 passed the string to a second script so
>that the second script could figure out if this was something that
>should be built ( for example whether a standard Win32 build or a
>WinCE build should be done ). Version 4 tracked dependencies more
>carefully ( it did a topological sort and orgainsied the build
>according to it ). Since the core stuff should be built first, I had
>to find out a way to figure out which was core. I used a definition of
>the core stuff is the stuff with the longest chain of dependencies.
>That was Version 5.
>
>As you can see I started with a simple program and made it more
>complex as I went on.  That's what I mean by extendible. Python is
>almost good at it. Almost. But that damned code indentation dependence
>backs its ugly head. In the process of extending code, I have to move
>blocks of code around. Sometimes put it in a loop, sometimes add an if
>clause to the top. Sometimes extract code into a subroutine. That
>means that I have to reindent the block. So as I reindent I have to
>try to remember how the indenting went. Something I stink at. 
>Emacs also doesn't reindent correctly. So I have to go through the
>code, and try to figure out how it works. This distracts from my 
>rewrite ( this also applies to refactorings ). I would rather think
>about how new code is supposed to work rather than how old code
>is supposed to work.
>
>Anyway I hope this causes TPTB to think about how blocks should be
>handled in Pyhton 3000.

-- 
Ted Drain   Jet Propulsion Laboratory    Ted.Drain at jpl.nasa.gov


From vtbl.moorepe at memo.volvo.se  Thu Jun 29 06:57:36 2000
From: vtbl.moorepe at memo.volvo.se (PM)
Date: Thu, 29 Jun 2000 10:57:36 +0000
Subject: Hi, my name is not Mrs so-called Gumblenose
Message-ID: <395B2B88.CC3041C1@memo.volvo.se>

My head hurts.

If I move to Python from Java from RPG from C++ from C from Basic from
6502, am I really, really, really, really stupid? I was not very good at
-any- of them, but I'm as persistent as a crap dance record on Radio 1.

Will this group answer all questions, solve life's problems? 
Can Python kick large bottoms with decent database access - not just SQL
not ODBC, could it go to DB2/400?

What I liked about good old DOS/basic was the ability to knock-things up
and try things out, now that's what has piqued my interest in Python,
but is it true.... I'll probably waste some time trying to find out, but
I'm not 14 years old. Wish I was - and could still drive my Audi.

--PM


From adams_s at lab.eng.usyd.edu.au  Fri Jun  9 00:42:24 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Fri, 9 Jun 2000 14:42:24 +1000
Subject: Newbie Q: CGI
References: <8hn0c1$1a22$1@news1.wire.net.au> <m3snuobxuv.fsf@underlevel.underlevel.net>
Message-ID: <8hps9f$30eq$1@news1.wire.net.au>

Thanks Jordan, I'd already looked through socket (for another project) but I
didn't think I'd need (or want) to use socket,
all the IP was needed for was part of a simple HTML message board/discussion
system I was writing, and I neede the IP of who was sending the message.

thanks anyways :-)

Steven - annoyed that Perl seemed to provide an easier way to do it...


Jordan Katz <webmaster at underlevel.net> wrote in message
news:m3snuobxuv.fsf at underlevel.underlevel.net...
> "Steven Adams" <adams_s at lab.eng.usyd.edu.au> writes:
>
> > Hi,
> >
> > How do you get the IP number from a request for a web page?
gethostbyname()
> > and gethostbyaddress() weren't mentioned anywhere in the cgi module docs
> > (that I could see, did I miss it?)
>
> The low-level socket module has what you need.  You can read all about
> it at http://www.python.org/doc/current/lib/module-socket.html
> --
> Jordan Katz <webmaster at underlevel.net>




From perl_monkey at my-deja.com  Fri Jun  9 14:08:01 2000
From: perl_monkey at my-deja.com (perl_monkey at my-deja.com)
Date: Fri, 09 Jun 2000 18:08:01 GMT
Subject: Porting CPython
Message-ID: <8hrbph$a4n$1@nnrp1.deja.com>

Hello,

First, I don't mean to start a flamewar with my nick, it was created
long ago on Deja.  :)

I am currently porting CPython to IBM's OS/390 Unix System Services.  I
wanted to point out a few issues that I've had regarding this port, and
ask for some advice from anybody currently involved in ports or anyone
who has ported python in the past.  OS/390 is a mainframe platform, and
Unix System Services is a fully functioning (albeit intensely weird)
UNIX running on top of the mainframe OS in an LPAR.

If you respond to this, please respond to s2mdalle at titan.vcu.edu (David
Allen)

- SIGMAX in Modules/signalmodule.c

  SIGMAX is not defined in any headers I am able to find on my system,
so I looked through signalmodule.c attempting to determine what it was
used for.  As it turns out, it's part of another macro, NSIG which as
far as I can tell ONLY is good for bounds checking on signal numbers.  I
checked /usr/include/signal.h, found that the highest numbered signal I
could see was 28, so I #define SIGMAX 28 in signalmodule.c

- Modules/pwdmodule.c

This is what I'm currently having problems with.  As it turns out, the
struct passwd on this machine is defined as this: (from
/usr/include/pwd.h):

       struct passwd {
             char  *pw_name;
             uid_t  pw_uid;
             gid_t  pw_gid;
             char  *pw_dir;
             char  *pw_shell;
       };

which of course is missing pw_gecos, and pw_passwd that are usually
found there.  Now, if I can get somebody to definitively say that it's
OK to replace references to somepasswdstruct->pw_gecos with (char *)NULL
or some such thing, then that's what I'll do.  As it stands, any
solution I can come up with may break some really obscure things down
the line, and I'd rather not deal with that.

(BTW, the system security on OS/390 machines is done by a separate
package on the MVS side of things, so there isn't a passwd file on this
UNIX.  And trust me, that's only the tip of the iceberg with regard to
the weirdness of this UNIX)

- Excluding Modules

What is the build procedure for omitting certain built in C modules?
(I"m talking about the stuff in Modules/)

Any help on any of these issues would be appreciated.  Also, when/if I
finish this, I'd be willing to diff the source and put it into the main
python branch if that's something people would be interested in.
There's quite a few more details to porting this that I'm leaving out,
(silly things like the way the compiler works with flag ordering, macros
that have to be defined to get to socket headers, things like that) but
that I'd be willing to share if anyone is interested.

As I said earlier, please post to this thread or reply to
s2mdalle at titan.vcu.edu in response.  This is a throwaway Deja account
that I'm using because I don't have access to anything else at the
moment.

-- David Allen
-- s2mdalle at titan.vcu.edu



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


From timr at probo.com  Sat Jun 24 01:20:11 2000
From: timr at probo.com (Tim Roberts)
Date: Fri, 23 Jun 2000 22:20:11 -0700
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <8F5B6C0A8gmcmhypernetcom@199.171.54.154> <slrn8l7bqu.cmq.andyl@azaal.dircon.co.uk>
Message-ID: <a4h8lso56cjl0004363kfj0gmhflt8bmq6@4ax.com>

andyl at azaal.dircon.co.uk (Andy Leighton) wrote:
>
>-- 
>Andy Leighton => andyl at azaal.dircon.co.uk
>"... January is your third most common month for madness" - _Sarah Canary_

  "Sarah Canary?"

  "No, semi-carnally!"

(This is on-topic for this newsgroup.  Take it away, Eric the orchestra
leader!)
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From jcm at bigskytel.com  Wed Jun  7 20:28:53 2000
From: jcm at bigskytel.com (David Porter)
Date: Wed, 7 Jun 2000 18:28:53 -0600
Subject: numpy installation woes
In-Reply-To: <8hmmko$qd3$1@nnrp1.deja.com>; from long_way_home@my-deja.com on Wed, Jun 07, 2000 at 11:42:19PM +0000
References: <8hmmko$qd3$1@nnrp1.deja.com>
Message-ID: <20000607182853.B22543@bigskytel.com>

* long_way_home at my-deja.com <long_way_home at my-deja.com>:
> Hi.
> 
> I am new to pyhton and have tried to install numpy, but the darn thing
> doesn't recognise it's existence.  I have also installed MatPy and that
> works fine.  When I try to import it I get the message
> >>> from Numeric import *
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: No module named Numeric
> 
> although the directory is there.  

What happens if you start python from the directory that numpy is in and
then import Numeric? If that works, then you need to put the directory on
your path. See site.py in the standard library for how this works. [ Quick
explanation: if you have numpy at /usr/lib/python/site-packages/numpy, then
create a file named numpy.pth in /usr/lib/python/site-packages with the word
numpy as its contents. ]


  david.

-- 
A world comfortable disease.  Progress is not a comfortable disease.
  -- e.e.cummings in a blender



From pduffin at hursley.ibm.com  Tue Jun 27 10:59:00 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Tue, 27 Jun 2000 15:59:00 +0100
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr>
Message-ID: <3958C134.7AAEF37B@hursley.ibm.com>

Alexandre Ferrieux wrote:
> 
> Paul Duffin wrote:
> >
> > [detailed explanation of stdio-buffering/ttys/pipes]
> >
> > you may want to look into Python versions
> > of Expect. If they are not satisfactory then you could probably write
> > a little Tcl/Expect wrapper around the application which forces the
> > correct buffering and then control that using Python.
> 
> Not withstanding Tcl ;-) honesty commands to also mention the
> existence of the 'pty' unix package, which is roughly "Expect minus
> Tcl", and especially its 'nobuf' tool:
> 
>         nobuf cmd [args...]
> 
> starts cmd as a child, strongly held between ptys, so that the overall
> command behaves properly (ie in unbuffered or line-buffered mode) betwen
> pipes. If all you want is that, Expect is surely an overkill.
> 

Where can I get the 'pty' package from ?

> But ptys are no rocket science anyway. The nobuf above should be doable
> in pure Python within 20 loc...
> 

While ptys are not rocket science they are also not particularly portable,
Expect handles all of this for you. Windows is also a real problem as it
does not have ptys and needs to be approached in a completely different 
way.


From piet at cs.uu.nl  Fri Jun 23 05:10:53 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 23 Jun 2000 11:10:53 +0200
Subject: running pyton on NT command lines
References: <8ithrb$cj1$1@nnrp1.deja.com>
Message-ID: <uitv0sqxu.fsf@cs.uu.nl>

>>>>> ghost_man at my-deja.com (gm) writes:

gm> I have written several scripts that I want to run at some set time.  I
gm> have been trying to get them to fire with the AT command line
gm> scheduler, but keep getting errors.  I have tried several different
gm> variations to get it to run even putting the python executable in a dos
gm> batch file and trying to run the dos batch file.  I also have the
gm> python registered in my Windows 2000 environment path.

gm> my command line looks something like this

gm> at 16:30 "myscript.py"

gm> it fails everytime.  I have also tried putting in a complete path to
gm> the python.exe.

Have you tried "start myscript.py"?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From dmpettyp at somanetworks.com  Thu Jun 22 13:58:19 2000
From: dmpettyp at somanetworks.com (Dave Pettypiece)
Date: Thu, 22 Jun 2000 13:58:19 -0400
Subject: minimal python install
Message-ID: <395253BB.C5021893@somanetworks.com>

can anyone refer me to a resource that describes to how make a *minimal*
python installation? i want to put it on an embedded device, and would
like to make it as small as possible.

thanks,
dave



From timr at probo.com  Thu Jun  8 00:35:16 2000
From: timr at probo.com (Tim Roberts)
Date: Wed, 07 Jun 2000 21:35:16 -0700
Subject: SQL Variable Substitution Question
References: <8hmacu$ia9$1@news-int.gatech.edu>
Message-ID: <uf8ujs0siug4cfsacqcuvrl37gimb82bel@4ax.com>

liang at cad.gatech.edu (Shengquan Liang) wrote:
>
>say i have yyyy = 1000
>
>in the documentation of DC Oracle, it says that
>i shoul d use 'select * from attt'
>		'where bbbb = :yyyy'
>
>to make it work.

No, SQL doesn't have any idea what your Python variables are.  To use the
":" feature, you have to have set an _SQL_ variable called "yyyy".  How you
do that varies with the backend.

In your case, it's probably easier to build the complete string in Python
and pass that to the database:

  'select * from attt where bbbb = %d' % yyyy

>BUT, when i tried it this way, the 
>iDC oracle reponded by:
>
>  oci.err:(1008,'ORA-1008: not all variables bound\012')
>
>what's the problem with that?

You haven't bound the SQL variable yyyy to any value.  As I said, the
method for doing that varies with the backend.

--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From matthias.huening at univie.ac.at  Thu Jun 29 06:15:53 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Thu, 29 Jun 2000 12:15:53 +0200
Subject: ODBC
Message-ID: <8jf7nm$n48$1@www.univie.ac.at>

Hi,

I use the ODBC module that comes with Marc Hammond's winall.
Everything works fine; I open a connection, do something, close the
connection, like this:

import odbc

conn = odbc.odbc(DSN)
cur = conn.cursor()
sql = "SELECT col FROM table"
cur.execute(sql)
x = cur.fetchall()
cur.close()
conn.close()

The problem: when I try to open another connection in the same script
(after I closed the first one), Python crashes. Any ideas?

Matthias

--
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/dt/





From jeremy at beopen.com  Tue Jun 13 11:14:10 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 13 Jun 2000 11:14:10 -0400
Subject: TypeError: unhashable type
References: <393d1c1a_2@news1.prserv.net> <20000606181546.H469@xs4all.nl> <F8476B9163F138C3.8DBC9B4CCE38EBEA.9FFEEE60294594AF@lp.airnews.net> <olg05.706$cQ.9336@news-server.bigpond.net.au>
Message-ID: <C955E8835231EBB8.94EB70832E26D3DB.A8CB195364633A9C@lp.airnews.net>

"Mark Hammond" <mhammond at skippinet.com.au> writes:

> "Jeremy Hylton" <jeremy at beopen.com> wrote in message
> news:F8476B9163F138C3.8DBC9B4CCE38EBEA.9FFEEE60294594AF at lp.airnews.net...
> 
> > It seems that "unhashable type" is a pretty obscure error message.  If
> > you understand hashes, types, and dictionary implementation, it's
> > fine.  If not, you're left scratch your head.
> 
> For the casual user, this generally means "dictionary keys must be
> immutable".
> 
> > It would be helpful to produce an error message like:
> > "dictionary key must be hashable type, not <type>"
> 
> This message isnt a real lot better.  Often I _know_ what the type is -
> its just "unhashable" - which still doesnt clue the user in.  The word
> "hashable" is a bigger problem than not printing the type.

I was hoping the new error message would provide a few more clues by
mentioning "hashable" and "dictionary key."  If the documentation for
dictionaries defined hashable and explained its significant, then a
user would have an easier time figuring out what went wrong.

I agree that "hashable" is puzzling, but I think it is doubly so when
you don't know its related to dictionary keys.

Printing the type of the object is probably of secondary importance.

> I think the situation could still be improved without the type name.  But
> for the life of me I can't think how.
> 
> "dictionary keys must be immutable" isnt really any better - WTF is
> "immutable"??

Yeah, immutable isn't any better.  In fact, it's bad because you can
write a mutable instance that has __hash__ and __cmp__ methods and use
it as a dictionary key.

> "dictionary keys must be objects unable to be changed" - WTF - I can
> change strings!!?  Or maybe this is trying to tell me that once I add a
> key, it can't be removed?
> 
> Maybe something like:
> "dictionary keys must be immutable (such as tuples, strings or ints, but
> not objects such as lists)" - *gag*.
> 

The right goal is probably a slightly clearer error message and
documentation tied specifically to that error message.  This matches a
general strategy I use when some piece of software spits out an
obscure error message.  I type the exact message into Google,
DejaNews, or AltaVista and see what comes up.  More often than not, I
find someone has already answered my question.

Jeremy


From jblazi at netsurf.de  Thu Jun 15 05:45:38 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Thu, 15 Jun 2000 11:45:38 +0200
Subject: initializing a class
Message-ID: <3948a566_5@goliath.newsfeeds.com>

I need several constructors for a class. So would this pseudo code be in
order:

class x:
  ...

class y:
  ...

class z:
  def __init__(self,a)
      if isinstance(a,x):
         z.t=x.t1+x.t2
      elif isinstance(a,y):
         return z(y.x)
      else:
         ...

When the constructor for class z is called, its argument is checked. If the
argument is class x, then the instance is constructed directly. Otherwise,
when the argument is class y then the constructor is called again with a
class y argument.

J.B.




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


From cgfandrich at my-deja.com  Mon Jun 12 13:07:45 2000
From: cgfandrich at my-deja.com (cgfandrich at my-deja.com)
Date: Mon, 12 Jun 2000 17:07:45 GMT
Subject: Embedded Python and Restricted Execution
Message-ID: <8i35cl$ch8$1@nnrp1.deja.com>

I'm embedding Python in an application and I'd like to be able to run
Python code (including callable PyObjects) in a restricted environment -
 much like rexec - but I'd like to control the environment in C/C++.
I've tried the following and had some success:

1. Create a new module.
2. Add a "__builtins__" dictionary to the module.
3. Copy references from existing "__builtins__" to new "__builtins__"
(except for "__import__" and "open" - use my functions for those 2)
4. Use the new "__builtins__" dictionary whenever I want to run code in
restricted environment.

When importing modules, I've also had to make sure that the module
being imported gets the correct "__builtins__" dictionary.  The only
problem is ... if I have 2 modules "test1" and "test2" that use the
same Python source code, I can't run them in the different
environments.  It seems like they both run with
whichever "__builtins__" was used to import the first module.

Any ideas?  Should I be doing this differently?

-chris
//====================================================================
static PyMethodDef restricted_methods[] =
{
    {"__import__", restricted___import__, 1},
    {"open", restricted_open, 1},
    {NULL,NULL}
};

void InitRestricted()
{
    PyObject *SecureModule;
    SecureModule = PyImport_AddModule("__restricted__");
    PyObject *SecureDictionary = PyModule_GetDict(SecureModule);
    PyObject *SecureBuiltins=PyDict_New();
    PyDict_SetItemString
(SecureDictionary,"__builtins__",SecureBuiltins);

    PyObject *MainBuiltins = PyEval_GetBuiltins();

    if(MainBuiltins && SecureBuiltins)
    {
        PyObject *MainKeys = PyDict_Keys(MainBuiltins);
        int i;
        // __import__
        value = PyCFunction_New(&restricted_methods[0], NULL);
        if (value != NULL)
            PyDict_SetItemString(SecureBuiltins, "__import__", value);
        Py_DECREF(value);
        // open
        value = PyCFunction_New(&restricted_methods[1], NULL);
        if (value != NULL)
            PyDict_SetItemString(SecureBuiltins, "open", value);
        Py_DECREF(value);
        for(i=0;i<PyList_Size(MainKeys);i++)
        {
            PyObject *key = PyList_GetItem(MainKeys,i);
            char *keyname=PyString_AsString(key);
            PyObject *value = PyDict_GetItemString
(MainBuiltins,keyname);
            if((strcmpi(keyname,"__import__")!=0) && (strcmpi
(keyname,"open")!=0))
                PyDict_SetItemString(SecureBuiltins,keyname,value);
        }
    }
}

void RunScript(char *Script,const BOOL Secure,PyObject *RetObject)
{
    PyObject *module;
    PyObject *dict;
    if(Secure)
        module = PyImport_AddModule("__restricted__");
    else
        module = PyImport_AddModule("__main__");
    dict = PyModule_GetDict(module);
    RetObject = PyRun_String(Script,Py_eval_input,dict,dict);
}



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


From nascheme at enme.ucalgary.ca  Thu Jun  1 15:30:24 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Thu, 01 Jun 2000 19:30:24 GMT
Subject: garbage collection and Tkinter
References: <200006011808.OAA26643@smtp3.andrew.cmu.edu>
Message-ID: <slrn8jdefe.a2.nascheme@cranky.arctrix.com>

Brian E Gallew <geek+ at cmu.edu> wrote:
>... is there a relatively easy way to examine the garbage
>collector to see why it is or is not collecting things?

The standard reference counting that Python uses cannot detect
reference cycles.  You could compile a new interpreter with
Py_TRACE_REFS defined.  The interpreter can then show you which
objects are still alive when it is shutting down.  You could also
grab my garbage collection patch from:

    http://www.enme.ucalgary.ca/~nascheme/python/gc.html

It allows reference cycles to be debugged quite easily.

    Neil

-- 
145 = 1! + 4! + 5!


From frankc at colconsulting.com  Fri Jun  9 19:19:36 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 09 Jun 2000 19:19:36 -0400
Subject: "Announce": Python Snippets site on hold
References: <200006092242.e59Mg9K29454@hera.cuci.nl>
Message-ID: <39417B88.1B17E56E@colconsulting.com>

Hans Nowak wrote:
> 
> Hi,
> 
> As some of you might have noticed, I decided to stop working on my snippets
> site. The main reason is that I just don't have the time anymore to scan
> the newsgroup for interesting code, much less collecting, editing and
> adding it.
> 
> However:
> 1. The site will stay online for a while; after all, it contains 390
> snippets, so it may as well stay as a reference point for users. (That is,
> if Blake Winton permits it... it's his server.)
> 2. There is such a thing as Python FAQTS (the Python knowledge base); the
> idea is roughly similar to that of the snippets site. I offered them my
> snippets (published and unpublished); it's up to them if you will actually
> see them in the knowledge base.
> 
> I'd like to thank everyone who has provided me with code and support during
> the time that the site existed.
> 
> Blessed be,
> 
> --Hans Nowak (zephyrfalcon at hvision.nl)
> Homepage: http://www.hvision.nl/~ivnowa/newsite/
> Python questions? See http://tor.dhs.org/~zephyrfalcon/snippets
> You call me a masterless man. You are wrong. I am my own master.

Consider starting a SourceForge project which is just that. They even
have a Snippets Collection function. It is free and hosted.

I am not employed by SourceForge.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From gmcm at hypernet.com  Tue Jun 27 16:25:41 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 27 Jun 2000 20:25:41 GMT
Subject: unexpected behaviour in class variables
References: <8j83ml$2s7$1@slb3.atl.mindspring.net> <8F5FCA994gmcmhypernetcom@199.171.54.194> <8j8vjr$v63$1@nntp9.atl.mindspring.net>
Message-ID: <8F60A1C6Bgmcmhypernetcom@199.171.54.154>

Andrew Dalke wrote: 

>Gordon McMillan wrote:

>>An attribute found in the instance dict will never get wrapped. Only
>>stuff found in the class __dict__ (or in a base class __dict__)
>>qualifies for wrapping.
>
>
>But the documentation at http://python.org/doc/current/ref/types.html
>says (with some spacing changes to enforce better alignment)
>
>: User-defined method objects are created in two ways: when getting
>:  an attribute  of a class          that is a user-defined function
>:  object, 
>: or when getting
>:  an attributes of a class instance that is a user-defined function
>:  object. 
>
>Isn't an "instance dict" in your terminology the same as "attributes of
>a class instance" in the quoted documentation?  

Terminology sherminology!

>>> def eggs():
...  pass
...
>>> class A:
...  pass
...
>>> a = A()
>>> a.eggs = eggs
>>> type(a.eggs)
<type 'function'>
>>>

It ain't wrapped. Beat up Fred, he needs it <wink>.

>>Why the wrapping? To hide away "self" - so it's implicit to the method
>>object but explicit to the function object.
>>
>>Make sense now?
>
>
>No.  Why does it only occur for user-defined function objects added to
>the class namespace and not C-based functions or callable objects?  That
>limits how I design my code if I want the ability to change between a
>C-based implementation, a Python-based one or to a function object,
>which is exactly the problem I have.  (I replaced a call to a C
>extension library to a Python based one used for regression tests.)

I'm not really understanding.

>>> class A:
...  pass
>>> A.sin = math.sin
>>> def eggs(x):
...  print "eggs"
...
>>> A.eggs = eggs
>>> a = A()
>>> a.eggs()
eggs
>>> a.sin(0)
0.0

Functionally, the difference is the need to provide an arg to the user-
defined function that will take the value of "self". Or, perhaps, the 
inability to use one in a C extension. Maybe your original post spelled out 
some brilliantly deduced logical consequences of this apparently innocuous 
difference. If so, they slid right past me <wink>.

- Gordon


From 12klat at sightreader.com  Thu Jun  1 04:00:48 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Thu, 01 Jun 2000 01:00:48 -0700
Subject: Dictionary size/speed limit?
References: <f3lZ4.557$r7.287786@news1.rdc2.tx.home.com>
Message-ID: <39361830.AC9431BF@sightreader.com>

Michael Morrison wrote:

> I'm comtemplating a new program I'll be creating, and I plan on having a
> dictionary with around 20000-50000 entries.
>
> This dictionary will be searched many thousands of times itself, all at
> once.  ie. read a file with 20000 keys, find value in dictionary for each
> key.  (The dictionary size and file size have no relationship, and actually,
> the file size will probably be larger than the dictionary in some cases)
>
> So, is using the regular Python dictionary just an insane idea? :)  Should I
> create something in C to do this for me, or are there modules to do this
> sort of thing?
>
> Thanks,
> borlak

You don't have to decide yet.  Write it with python dictionaries.  Rewrite it
later in C if necessary.





From tom-main at REMOVEME.home.com  Mon Jun  5 11:55:08 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Mon, 05 Jun 2000 15:55:08 GMT
Subject: Future of PythonWin?
Message-ID: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com>

I see that Mark Hammond has gone to ActiveState.

I'm interested to hear that they will be integrating Python into Visual
Studio 7, but what does this all mean for the future of the PythonWin?

Will PythonWin (the IDE and the MFC wrapper library) continue to be actively
and publicly developed?

Tom Malcolmson.




From mrfx at etang.com  Tue Jun  6 16:42:41 2000
From: mrfx at etang.com (mrfx)
Date: Wed, 7 Jun 2000 04:42:41 +0800
Subject: Is there any gui builder like glade for wxwindows?
Message-ID: <8hl1p3$272v$1@news.cz.js.cn>

Glade is a good gui builder for GTK,
is there any gui builder like glade for wxwindows?





From olczyk at interaccess.com  Thu Jun  8 17:26:07 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Thu, 08 Jun 2000 21:26:07 GMT
Subject: Using python with Corba
References: <eieJ4.8408$H7.704991@brie.direct.ca> <871z4awiz0.fsf@parus.parus.no>
Message-ID: <39410efe.18494828@nntp.interaccess.com>

On 13 Apr 2000 09:36:19 +0200, Jon K Hellan <hellan at acm.org> wrote:

>"Fanny Yeung" <fannyy at idmail.com> writes:
>
>> Hi everybody,
>> 
>> I am new python and I am particular interest in use python with corba. After
>> some research, I find out omniOrb provide interface 'omniOrbpy'. Any one
>> know where I can find it ?
>
>http://www.uk.research.att.com/omniORB/omniORB.html
>
>> Also, besides omniOrb2, is there any other Corba that Python can use ?
>
>Fnorb is almost pure Python.
>
>ILU does a lot of languages.
>
>ORBit is extremely fast.
>
>> Anyone know TAO can be used in Python ?
>
>Don't think so. But a Python ORB can interoperate with TAO.
>
>Regards
>
>Jon Kare

They are supposed to but (sigh ) most orbs have many parts of the
standard unimplemented yet. Do you know if Fnorv,ILU, OmniORb
or ORBit (didn't know ORbit supported python) can interoperate with
TAO?


From aahz at netcom.com  Mon Jun 19 15:43:17 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jun 2000 19:43:17 GMT
Subject: string parameter replacement question
References: <Pine.LNX.4.10.10006191432420.3161-100000@worf.mcnc.org>
Message-ID: <8ilt4l$fbi$1@slb7.atl.mindspring.net>

In article <Pine.LNX.4.10.10006191432420.3161-100000 at worf.mcnc.org>,
oliver  <fwang2 at yahoo.com> wrote:
>
>I have a function defined to accept one string parameter, inside this
>funtion, I need to call another function to take this string as part of
>the parameter, I am confused how to make the replacement happen. See
>following code:
>
>def fun1(str):
>  fun2("input=str")

Just drop the quotes:

def fun1(str):
  fun2(input=str)

In fact, you should just be able to do:
  fun2(str)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From db3l at fitlinxx.com  Fri Jun 30 19:30:53 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 30 Jun 2000 19:30:53 -0400
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org> <8F62AF73Egmcmhypernetcom@199.171.54.194> <36D5F4DFCA9BCA46.B5AE15F2F19F393C.28F9B766F10B5E23@lp.airnews.net> <8F62C20E2gmcmhypernetcom@199.171.54.154>
Message-ID: <uk8f63fwy.fsf@ctwd0143.fitlinxx.com>

gmcm at hypernet.com (Gordon McMillan) writes:

> Well, os.popen works fine between console processes on the NT family of 
> Windows. However, Windows has problems with 16 bit DOS apps and Win95 has 
> an MS acknowledged problem with any of this stuff. The win32 extensions fix 
> most of these problems, (I think Bill Tutt even added a workaround for the 
> Win95 bug), and I believe that 1.6 will use the win32 versions if they are 
> installed (at least, this was discussed seriously at some point).

If they do, then I hope that the extensions get augmented to return
the result code of the child process on close(), since that's really
the only thing holding me up from using them (for now I just use NT
and ensure I have a console allocated to my process).

I guess I'm just glad that my requirements to use Windows stops with
NT, since I also think Python might be a non-starter for a lot of my
administration apps if popen() didn't work reliably.  Of course, MS is
far more to blame for this than Python :-)

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


From anders.eriksson at morateknikutveckling.se  Tue Jun 13 10:43:01 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 13 Jun 2000 16:43:01 +0200
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 2000)
References: <002901bfcef1$d8fa3660$0500a8c0@secret.pythonware.com>
Message-ID: <bkhckscbhq9e9i18ijci51eh4r82f3qaqk@4ax.com>

On Mon, 5 Jun 2000 15:27:49 +0200, "PythonWare" <info at pythonware.com>
wrote:

>
>    The PythonWare py152 distribution is a small and self-contained
>    Python distribution for Windows.  This kit includes Python 1.5.2,
>    Tkinter/Tk 8.2.3, PIL 1.0, PST, and other useful extensions.
>

It's a pity that this packages doesn't work on Windows NT! 

First there is an error in the installation that causes it to be
aborted. Welcome.py refers to TCL8.1/TK8.1 but in the package you have
8.2.3 

Second: It will not register Python in the registry. which then causes
win32_all not to install.

// Anders
-- 
I have tried to contact Pythonware but all mail bounces...


From derek at ooc.com.au  Thu Jun 15 15:40:42 2000
From: derek at ooc.com.au (Derek Thomson)
Date: Fri, 16 Jun 2000 05:40:42 +1000
Subject: Python Productivity over C++ (Encapsulation)
References: <20000615183521.59984.qmail@hotmail.com>
Message-ID: <3949313A.D04687B3@ooc.com.au>

Hung Jung Lu wrote:
> 
> --- In python-list at egroups.com, Ken Seehof <kens at s...> wrote:
> >"Frank V. Castellucci" wrote:
> > > Courageous wrote:
> > > While the last line seems rather toungue in cheek, there are other
> > > things that C++ brings to the table such as encapsulation, access
> > > control and type strictness.
> >Encapsulation, access control and type strictness are often erroneously
> >referred to as if they were features.  They aren't.  They are aspects of a
> >design philosophy.  To make a system less error prone, constrain the
> >programmer
> >in order to prevent the programmer from writing bugs.  ...
> >You have to ask, "what is the actual value of these features?"  The answer
> >is
> >that they prevent certain kinds of bugs from appearing.
> 
> Type strictness is fine. It has all to do with performance: small, fast,
> efficient machine code. Encapsulation is a totally different story.
> 
> One of the biggest lies in modern computer language theory is that
> encapsulation is there for a good technical reason.

Yes, it is there for a good technical reason. 

>
> [Delete ranting]
> [Summary: Lack of "private" keyword in small but nice programming language >  brings love, peace and unlimited rice pudding to all.
>  Evil corporations crumble to dust instantly to be replaced by nice,
>  clean socialist re-education camps]
> 
> I just don't see how those little white lies can perpetuate after Visual
> Studio 7.0 incorporates Visual Python. Sleepyheads will wake up, and the
> corporate tower will start to collapse. Not a bad thing. :)

Then why does Python support encapsulation that is just as strict as C++? Guido
van Rossum - exposed as a traitor to the people's revolution all along!

I can easily subvert encapsulation in many C++ classes simply with the line
"#define private public" before I include any header files. Perhaps this means
that C++ will in fact lead to the overthrow of evil and oppression in the world?

Also, remember that Java's encapsulation isn't exactly inviolable. It can't be.
Think about it.

The only real way to implement the kind of encapsulation that you are talking
about is to run the objects on a different server, and specify access in CORBA
IDL, or a well defined socket protocol. And there are certainly very good
technical reasons (that don't involve actual world domination by shadowy evil
overlords) for doing this.

Regards,
Derek.


From urner at alumni.princeton.edu  Fri Jun 30 20:29:48 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 30 Jun 2000 17:29:48 -0700
Subject: PowerPoint slide presentation (Math + Python)
Message-ID: <uneqlsoqo59b38g19k8lus9gmb30obf9as@4ax.com>

Greetings educators, Python programmers --

For those of you with PowerPoint and/or the requisite browser 
plug-ins, I've tossed a slide show on the web at:

ftp://ftp.teleport.com/pub/users/pdx4d/images/synergeo.ppt

This is my standard 'Beyond Flatland' curriculum, wherein
I mix spatial geometry, virtual worlds, computer programming
(these days Python), and synergetics.

Given this is just an outline of a talk, there's not a 
lot of discussion (few words).  To present this material
effectively, you'd need to study the background readings
e.g. http://www.inetarena.com/~pdx4d/ocn/cp4e.html (see
'Numeracy + Computer Literacy' series).

Uses some of those special effects PowerPoint is all about.  
Apologies if this is distracting, but the business culture 
in which I operate kinda likes these bells and whistles.

Feedback welcome.

Kirby
4D Solutions
Portland, Oregon
http://www.teleport.com/~pdx4d/


PS:  use mouse clicks to get text bullets to appear, 
and to move from slide to slide (11 slides in all).

PPS:  also notified educators at
http://mathforum.com/epigone/math.teaching.technology
of this free download




From moshez at math.huji.ac.il  Thu Jun  8 12:39:20 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 8 Jun 2000 19:39:20 +0300 (IDT)
Subject: How to tell if I can open a file or not
In-Reply-To: <8ho4to0m8n@enews2.newsguy.com>
Message-ID: <Pine.GSO.4.10.10006081936380.15963-100000@sundial>

On Thu, 8 Jun 2000, Skip Hollowell wrote:

> I have a indexer that I am writing that goest through a user-specified
> directory and all subdirectories and indexes all the .htm* file contained
> therein.  This works great until I get to a file that someone has chmod-ed
> so that I can't read it.  Then I bomb after the open().

Let's be exact: you get an exception when open() fails. This is a Good
Thing(TM): you won't be able to forget to check if open() fails. The
downside, of course, is that you have to handle the exception. I suggest
that in your main loop, just put a try/except for IOError and os.error. If
all this exception handling stuff went over your head, sit down and read
the fine Python tutorial -- it covers exceptions quite nicely.

> Is there a way I can check the attrbutes of the file before I open it?  This
> is easy enough in c or perl, but have yet to find an equivalent in Python.

Try os.stat and the stat module, though exception handling is probably
better.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From dieter at handshake.de  Sat Jun 17 04:36:24 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 17 Jun 2000 10:36:24 +0200
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com>
Message-ID: <x7bt10d7qv.fsf@lindm.dm>

mjackson at wc.eso.mc.xerox.com (Mark Jackson) writes on 14 Jun 2000 16:23:30 GMT:
> "Warren Postma" <embed at NOSPAM.geocities.com> writes:
> > Having something
> > look correct on the screen, but not work because of misuse of tabs versus
> > spaces, or incorrect indentation, is something that still continues to bug
> > me.
> Indulge my curiosity - how many times has this actually happened in
> your six months of using Python?
It happens regularly, when I (working under linux with default
tabwidth=8) and a collegue (working unter MacOS with default
tabwidth=4) exchange code.
Of cause, we could know ... but sometimes forget ...


Dieter


From krussell4 at videon.home.com  Tue Jun 13 12:13:55 2000
From: krussell4 at videon.home.com (Kevin Russell)
Date: Tue, 13 Jun 2000 16:13:55 GMT
Subject: unsettling tkinter bug -- or is it just me?
References: <39458AFC.8AD5B850@tricity.wsu.edu>
Message-ID: <39465DD4.EB3C3CAF@videon.home.com>

Bob Lewis wrote:

> Consider the following code:
> --------
> from Tkinter import *
>
> root = Tk()
> img = PhotoImage(file='bobl.gif')
>
> def func():
>     cnvs = Canvas(root, width=img.width(), height=img.height())
>     cnvs.create_image(0,0, image=img, anchor=NW)
>     cnvs.pack(side=TOP)
>     Button(root, text='Quit', command=root.quit).pack()
>
> func()
> root.mainloop()
> --------
> This runs just fine: It creates a root toplevel with "bobl.gif" displayed
> in it and a "Quit" button.
>
> Now we make one wafer-thin change: Move the the "img =" inside "func()":
> --------
> from Tkinter import *
>
> root = Tk()
>
> def func():
>     img = PhotoImage(file='bobl.gif')
>     cnvs = Canvas(root, width=img.width(), height=img.height())
>     cnvs.create_image(0,0, image=img, anchor=NW)
>     cnvs.pack(side=TOP)
>     Button(root, text='Quit', command=root.quit).pack()
>
> func()
> root.mainloop()
> --------
> And now, while the canvas has the right dimensions, the image does
> not appear.  Okay, so "img" went from a global to local scope, but
> that shouldn't have affected execution, right?  Have I missed something?
>
> This is on a Red Hat Linux system running versions 1.5.2-13 of both
> python and tkinter.
>
>         - Bob Lewis
>           bobl at tricity.wsu.edu

I've noticed this ... um ... feature myself, but have never quite got
around to asking c.l.p about it.  It bit me when I tried to create a bunch
of  buttons with images, using a for-loop, like:

buttonlist = [('1.gif', SaveFunction), ('2.gif', OpenFunction), ...... ]
for (imagefile, command) in buttonlist:
    img = PhotoImage(file=imagefile)
    Button(toolbar, image=img, command=command).pack()

[Remembering it off the top of my head, so it may have mistakes
in it.]

Anyway, once the img variable gets reassigned, the PhotoImage
object that it used to refer to won't show up anymore.

It seems that you have to keep an active Python reference to a
PhotoImage -- the reference you'd expect its parent Tk object
to keep to it isn't good enough.  (Maybe the garbage collector is
over-eager.  Who knows?)

I got around it by making a global Python list of PhotoImage
objects, and just appending to it every time I make a new one.
Ugly hack, but it works.  If anyone has a more elegant solution,
I'd appreciate hearing it.

-- Kevin





From hack at watson.ibm.com  Fri Jun 16 14:47:07 2000
From: hack at watson.ibm.com (hack)
Date: 16 Jun 2000 18:47:07 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <394a5109.80394991@10.1.1.28> <394A5795.F4DE98FB@san.rr.com>
Message-ID: <8idsnb$j08$1@news.btv.ibm.com>

In article <394A5795.F4DE98FB at san.rr.com>,
Courageous  <jkraska1 at san.rr.com> wrote:
>
>> Suppose, however, that you could get digit 'n' of pi directly (or e,
>> or any other inifinte, nonrepeating sequence).  All I have to do is
>> communicate which digit of pi the message key starts at, and I can
>> generate the rest of the key.
>
>This is a collosally bad idea, IMO.
>
>C/
 
It works fine if the number of bits in 'n' is large enough, e.g. 16K bits.
Your key would then be the 16K bits of pi starting around bit 10^5000 (to
encode a message of 16K bits).  Of course, you haven't gained much:  you
might as well use 'n' itself!

Michel.


From a at a.a.a  Sun Jun 25 21:27:36 2000
From: a at a.a.a (Douglas Taylor)
Date: Mon, 26 Jun 2000 01:27:36 GMT
Subject: TCP/IP ports and RPC in python
Message-ID: <cky55.92$yS5.440413@news0.optus.net.au>

Surely there is someone out there who has written something already...

I just don't get it. I can do port programming, I understand that much at
least.

What I want to write is N-Tier -- I would implement this solution using COM,
but I need to run it on MACs too.

Has anyone out there got a small code snippet they would be willing to
share?

I mean, client and server RPC + serialisation equivalents.

I want to stop using Java. Won't someone help me?




From thomas at cintra.no  Wed Jun  7 08:40:34 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 07 Jun 2000 12:40:34 GMT
Subject: What is Python3000??
Message-ID: <393e426f.77267284@news.online.no>

Ok, not that intelligent Subject-description, but ...

I want to know more about Python3000. How is it going to be different
from Python as we know it? More features, new syntax ?? 

Thomas


From Vladimir.Marangozov at inrialpes.fr  Thu Jun 22 10:29:49 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Thu, 22 Jun 2000 16:29:49 +0200
Subject: Access violation when no memory
References: <olD35.82$_%.4481@newsfeed.avtel.net> <slrn8kvpml.flp.perly@typhoon.xnet.com> <o9e45.18$FG1.1680@newsfeed.avtel.net>
Message-ID: <395222DD.353DDD8F@inrialpes.fr>

Phil Mayes wrote:
> 
> No, I understand that - my program is explicitly designed to consume
> all memory by progressively consuming smaller and smaller blocks,
> throwing exceptions at each size.

Whatever your task is, I strongly doubt that consuming all the memory
of your machine is really what you want to do. Not from Python, at least.

The net effect of your program is zeroing the memory the OS is kind enough
to allocate for Python. It's not really Python's fault in such extreme
situations. Tim Peters reported (and one could try it easily with a
malloc(BIG) call) that Linux overbooks it's memory, that is, the malloc
succeeds, but accessing everything subsequently segfaults because the
requested mem size is not physically present (RAM + swap included).

So this boils down on how the OS handles such extreme cases. Python
usually does not trap OS signals, etc. and it cannot handle OS-level
mem overflows. All it does is trusting C malloc and checking the return
values for success or failure.

> My point is that eventually Python /crashes/ by attempting to write to
> low memory instead of failing with a MemoryError.

You get what you asked for <wink>. See above why.

> 
> So I need a ceval.c expert to vet my change.
>

Forget about it <wink>. Nobody is convinced that checking for NULL the
top of stack after popping its value would solve any problem. Your example
crashes on my combo in the builtin dict methods (& elsewhere). This is
really low-level OS stuff, and Python has not builtin emergency procedures
for these cases. I don't know why you want to eat all your memory, but
I'd suggest changing the approach or write a C module that deals with
the memory issues you're looking for.

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


From moshez at math.huji.ac.il  Sat Jun  3 06:37:21 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 3 Jun 2000 13:37:21 +0300 (IDT)
Subject: mixed solution: unicase (unique allowed case) (was: Re:
 Case-sensitivity: )
In-Reply-To: <87n1l3u7q1.fsf@deneb.cygnus.argh.org>
Message-ID: <Pine.GSO.4.10.10006031335390.17165-100000@sundial>

On 2 Jun 2000, Florian Weimer wrote:

> There's the convention to use "Foo" for classes and "foo" for
> instances.  I don't think people want to give up that.

OK, I've heard this one once too many times. 

What's wrong with:

>>> class Foo:
...     pass
... 
>>> aFoo = Foo()

Certainly much clearer?
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From claird at starbase.neosoft.com  Wed Jun  7 19:23:22 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 7 Jun 2000 18:23:22 -0500
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com> <8hluop$mlo$1@news1.xs4all.nl> <393e9acf@nntp.server.uni-frankfurt.de>
Message-ID: <1B9570FDE8DD9D42.387E1354F5B7C769.B82389FFF2B684D4@lp.airnews.net>

In article <393e9acf at nntp.server.uni-frankfurt.de>,
 <mlauer at trollinger-fe.rz.uni-frankfurt.de> wrote:
>Boudewijn Rempt (boud at rempt.xs4all.nl) wrote:
>> Tom <tom-main at removeme.home.com> wrote:
>
>> > I agree that MFC seems to be on the way out, but there are two things that
>> > prevent me (and some other MFC developers) from switching:
>> > 1) The CodeTools & CodeGuru sites contains TONS of peer reviewed code (in
>> > the form of messages at the end of every article) that is free, even for
>> > commercial use.  80% of this code is in the form of MFC extensions.
>> > 2) What's next?
>> > - STL?  Yes, but that's not a GUI.
>> > - ATL & WTL?  Too MS only.
>> > - wxWindows?  Only has a tiny fraction of the number of users & code that
>> > the other options have.
>> > - TCL/TK?  I've heard this is slow, and it can't create native windows' apps
>> > (users want the windows look & feel).
>> > - Java?  Again, too proprietary.
>
>> There's another option, of course. You could use Qt (and PyQt), which
>> works on Windows and Unix/X11, offers a great GUI and a good design.
>> And there are tons of programs using it you can look at & learn from.
>> (Not to mention the great documentation.)
>
>Not to forget PyGnome and PyGTK, which runs quite good on Unix/X11 and
>is evolving on Windows, BeOS and MacOS.
			.
			.
			.
... and more, which I'm slowly documenting in
<URL:http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>.
-- 

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


From greg at cosc.canterbury.ac.nz  Mon Jun 26 21:50:05 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 27 Jun 2000 13:50:05 +1200
Subject: apply for objects
References: <Pine.GSO.4.20.0006220932160.9460-100000@keg.cs.vu.nl>
Message-ID: <3958084D.42CFD928@cosc.canterbury.ac.nz>

Sander Hahn wrote:
> 
> >>> class C:
>         def m(x, y):
>                 print str(x) + str(y)
> 
> 
> >>> c = C()
> 
> >>> apply(c.m, ('hello', ' bye now'), {})

You didn't give your method a self argument:

class C:
    def m(self, x, y):
        print str(x) + str(y)

You will then be able to use apply() on it quite successfully.

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


From jkraska1 at san.rr.com  Wed Jun 14 00:13:15 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 14 Jun 2000 04:13:15 GMT
Subject: Python + libnet? Link Level Networking?
References: <3946eb3b.2033142@news.va.mediaone.net>
Message-ID: <39470705.7BA6988B@san.rr.com>

> least at the network layer, right? (I'm asking, don't know).  But I
> need to write packets (okay I'm playing with Cisco Discovery Protocol)
> which are a CDP packet inside of a SNAP frame on Ethernet.  Any ideas?

This really sounds like an embedded-C job to me;
if you have the python source release, you can
find an object called xxobject.c which is actually
only a skeletal framework that you can borrow in
order to jumpstart external C module creation.

Anyway, hang on, and perhaps someone will produce
the answer you are looking for... :)-


C/


From jmccarty at sun1307.ssd.usa.alcatel.com  Wed Jun  7 13:45:20 2000
From: jmccarty at sun1307.ssd.usa.alcatel.com (Mike Mccarty Sr)
Date: 7 Jun 2000 17:45:20 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <pcon1kylmzg.fsf@math.ntnu.no>
Message-ID: <8hm1ng$p94$1@relay1.dsccc.com>

In article <pcon1kylmzg.fsf at math.ntnu.no>,
Harald Hanche-Olsen  <hanche at math.ntnu.no> wrote:
)+ Rajarshi Ray <rajarshi at home.com>:
)
)| Besides the gee-whiz aspect of such algorithms, do they have any
)| value?
)
)They appear to have considerable entertainment value.  And, for some
)reason I could never fathom, every new supercomputer ever made is
)supposed to crank out ever more digits of pi to prove their worth.

[snip]

The computation of PI to a large number of digits is an excellent
*test* of the computer. Has nothing to do with computing PI per se.
-- 
----
char *p="char *p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
I don't speak for Alcatel      <- They make me say that.


From fiona at sitegnome.com  Mon Jun 26 00:19:22 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 26 Jun 2000 04:19:22 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 26th, 2000
Message-ID: <20000626041922.24062.qmail@synop.com>

Greetings,

Below are the entries I've either entered or edited within 
http://python.faqts.com today.

cheers, Fiona


## New Entries #################################################


-------------------------------------------------------------
Is there any way, other than using the curses library, to check if a key has been pressed, and if so, which one?
Is there any way to prevent curses.initscr() from messing w/ normal prints?
http://www.faqts.com/knowledge-base/view.phtml/aid/3988
-------------------------------------------------------------
Fiona Czuczman
Erno Kuusela

Yep, you can do it with the termios module (or with system() & stty) if
you dare.

Here is some code for one-key-at-a-time input that I had laying
around (might be borrowed from somewhere):

import termios, TERMIOS, sys, os

def getkey():
        fd = sys.stdin.fileno()
        old = termios.tcgetattr(fd)
        new = termios.tcgetattr(fd)
        new[3] = new[3] & ~TERMIOS.ICANON & ~TERMIOS.ECHO
        new[6][TERMIOS.VMIN] = 1
        new[6][TERMIOS.VTIME] = 0
        termios.tcsetattr(fd, TERMIOS.TCSANOW, new)
        c = None
        try:
                c = os.read(fd, 1)
        finally:
                termios.tcsetattr(fd, TERMIOS.TCSAFLUSH, old)
        return c

if __name__ == '__main__':
        print 'type something'
        s = ''
        while 1:
                c = getkey()
                if c == '\n':
                        break
                print 'got', c
                s = s + c

        print s

However this doesn't do it in a non-blocking way. You should
probably check for data with the select module prior to the read()
call. Set the timeout to 0 and it returns immediately iirc.
(it's also possible to set stdin to non-blocking mode, or use the
FIONREAD ioctl, or ...)


-------------------------------------------------------------
Can I initiate script B at the end of script A?
http://www.faqts.com/knowledge-base/view.phtml/aid/3990
-------------------------------------------------------------
Fiona Czuczman
David Bolen

Try the os.exec* functions in the os module.  It sounds pretty much
like what you want to do.  The exec* functions replace the current
running executable with another executable - only one process slot is
involved, and the original executing code does not exist following the
exec call.

Note however that since you'll be replacing your current process, you
actually need to exec python giving it the new script as an argument,
rather than just directly-exec()ing the script.

You can use sys.executable to determine the python that is executing
the current script and just exec() that with an argument of the new
script.

You don't note your platform, but I believe this is documented for
both Windows and Unix.


-------------------------------------------------------------
Is there a quick way to see if a string is a valid date?
http://www.faqts.com/knowledge-base/view.phtml/aid/3991
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

Marc Lemburg's mxDateTime module (do a search at www.vex.net/parnassus)
returns a RangeError if you create a date that doesn't exist. I've 
created a checkDate method below which returns 0 or 1.

mxDateTime is an excellent module if you're going to be doing anything
complicated with dates or times.

import DateTime

def checkDate(y,m,d):
    "year,month,day"
    try:
        checkDate=DateTime.DateTime(y,m,d)
    except DateTime.RangeError:
        return 0
    else:
        return 1
def testDates(y,m,d):
    print '%d/%d/%d' %(y,m,d)
    if checkDate(y,m,d):
        print 'is ok'
    else:
        print 'is not valid'

def main():
    testDates(2000,3,10)
    testDates(2000,13,3)

if __name__=='__main__':
    main()


-------------------------------------------------------------
What are the contents of .pyo file?
When is a .pyo file generated?
What is a .pyo file used for?
http://www.faqts.com/knowledge-base/view.phtml/aid/3993
-------------------------------------------------------------
Fiona Czuczman
Kalle Svensson

It's optimized byte code, generated by importing a module or using
py_compile.compile() in a python process run with the -O command line
option. The advantages of the .pyo over the .pyc file is just that it's
optimized and thus runs faster. I guess... :)


-------------------------------------------------------------
I can send mail via the smtplib but how do I attach files to the email?
http://www.faqts.com/knowledge-base/view.phtml/aid/3992
-------------------------------------------------------------
Fiona Czuczman
J?rgen Hermann, Grant Edwards

Just did this today (you're lucky), you need a combination of cStringIO
(create file objects from your text, optional), MimeWriter (create the 
MIME msg), mimetools (encode the parts, f.x. in base64), and smtplib 
(send the mail).

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

Take a look at <ftp://ftp.visi.com/users/grante/stuff/mfetch.py>.

There's a class called SmtpWriter that has a "Message" method that 
probably does what you want...

def Message(self,sender="", subject="", recipients=[], body="", 
attachments=[]):
    if self.__debugLevel > 2:
        sys.stderr.write("SmtpWriter: Building RFC822 message From: %s; 
Subject: %s; (Length=%d with %d
attachments)\n" % 
                          (sender, subject, len(body),len(attachments)))
[...]

The MimeWriter library module is where most of the magic happens...


## Edited Entries ##############################################


-------------------------------------------------------------
I'm new to Python and would appreciate advice on what book(s) I should be looking at?
http://www.faqts.com/knowledge-base/view.phtml/aid/2795
-------------------------------------------------------------
Fiona Czuczman
Chris Lada, Snoopy :-)), Paul Winkler, Dana Booth, Frank V. Castellucci,richard_chamberlain

There a lots,

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

for a list.  The manning and oreilly books have sample chapters online 
you could look at.

A range of books that should be useful when starting out with Python:

- "Quick Python" by Harms, and McDonald.

- If you already program, David Beasly's "Python Essential Reference" is 
very good. Another plus is that the book is high quality. It covers 
1.5.2, is tight fast text with no little fluffy animal or snakes 
writhing through the text, just down to the bone reading.

Snoopy :-)) wrote:

If you are New to Programming, then In addition to the Tutorials found 
on Python's Home Page, you should get as the 1st. choice "Teach Yourself 
Python in 24hrs".

BTY: "Learning Python" is also a very good book, but as a Newbie I found 
it quite overwhelming, to the point that I almost stopped learning 
Python. On the other hand I am very happy that I bought the "Teach 
Yourself Python in 24hrs". I find it considerable easier to comprehend 
the concepts, etc.

You can read the book online at the following:
                        http://www.pauahtun.org/TYPython/

Paul Winkler wrote:

OTOH, I got "Learning Python" and I've found that after reading just
the first half of the book and doing maybe 1/4 of the exercises, I
had a very good grasp of the language, including classes and
exceptions (both new ideas to me). From there the next step is to
skim the second half looking for useful bits, or poke around in the
library reference manual.

But then, it did take more than 24 hours. :)






From db3l at fitlinxx.com  Fri Jun 16 20:01:53 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 16 Jun 2000 20:01:53 -0400
Subject: Frozen socket.
References: <Pine.GHP.4.21.0006161522100.3877-100000@don.et.byu.edu>
Message-ID: <uln05qioe.fsf@ctwd0143.fitlinxx.com>

"C. Porter Bassett" <porter at et.byu.edu> writes:

> I have looked around, but I haven't been able to figure out why a program
> would choke on a send() command.  I don't even know if it is because of
> something I am doing on my end or something happening on the server
> end.  If it is my fault, how do I fix it?  If it is because of something
> happening on the server side, how do I compensate?

Under default conditions, send() may block if there is no room to
locally buffer the information you want to send.  It's not that it has
"choked" - just that it's waiting for room to become available.  You
can set a socket to non-blocking mode, but then you have to handle a
failure to send() all of your data and keep retrying until it is
accepted.

In most cases, this occurs because you have gotten far ahead of the
receiver - enough so to back up the TCP window size across the network
and your internal socket buffers.  As long as the systems are still
actually sending data, it will eventually catch up and you can keep
sending.

However, the fact that you say it happens regardless of how quickly
you send out the strings makes me wonder if the other side is really
receiving them at all.  Are you absolutely positive that you are
performing the initial communication properly so that the receiving
side is really receiving your data?  It may be just that it takes you
about that many strings to fill up your internal queues while the
remote system does not receive any data.

I suppose another possibility is something in the data is causing a
problem for the remote system such that after receiving that initial
amount it becomes busy and fails to continue receiving.

Do you have any other applications that perform this function that you
might be able to compare/contrast with - having some sort of control
or pre-existing behavior to compare against might give you some hint
as to what is going on.  But send() blocking is normally just a
symptom of something larger happening, and less likely to be a problem
in the actual send() call itself.

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


From jepler.lnk at lnk.ispi.net  Tue Jun  6 07:50:54 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Tue, 06 Jun 2000 11:50:54 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com>
Message-ID: <slrn8jnc3n.779.jepler.lnk@potty.housenet>

>Thomas Wouters wrote:
>> Actually, it's very intuitive, if you see the start and end index of the
>> slice not as items, but as the gaps between the items.
>
On Sun, 04 Jun 2000 21:26:07 GMT, Courageous
 <jkraska1 at san.rr.com> wrote:
>Well, yeah, but [0:3] as "element zero through three, inclusive"
>is about as intuitive as you can get. While I understand completely
>how the current implementation works (reading the source code and
>all, dontcha know), I still fail to see *why* this was chosen.
>I'm presuming there must be some conditions where this particular
>idiom was required or caused for nasty syntactic transformations
>if it wasn't used?

If it wasn't written in this way how would you express the use of a slice
to insert items:
	>>> s = [1,2,3,4,5]
	>>> s[3:3] = ["hi there mom"]
	>>> print s
	[1, 2, 3, 'hi there mom', 4, 5]

Jeff


From bvdpoel at uniserve.com  Mon Jun 26 13:33:04 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Mon, 26 Jun 2000 10:33:04 -0700
Subject: tkinter font selector?
References: <39564255.65F7C155@uniserve.com> <029c54f6.0e097830@usw-ex0101-005.remarq.com>
Message-ID: <395793D0.F62CFA3F@uniserve.com>

Thanks. I looked at the tk stuff which gets the list of families. What
I'm wondering is if there is a way for tk to get the possible
sizes/weights which are available for a family? I guess one could just
parse the output from xlsfonts and play with that...



Richard Chamberlain wrote:
> 
> Hi Bob,
> 
> In tkFont there is a method called families which returns a list
> of installed fonts. I've created a little script below which
> pops them in a combobox and changes the font of a label
> according to the item selected. It would be fairly easy to to
> add size and style (but I'll let you do that ;) ).
> 
> from Tkinter import *
> from tkFont import families
> import Pmw
> root = Tk()
> Pmw.initialise()
> 
> choice = None
> comboentries = families(root)
> lblTest = Label(root, text='Test Font',padx=20, pady=10)
> lblTest.pack(expand=1, fill=BOTH, padx=8, pady=8)
> 
> def selectFont(entry):
>     lblTest.configure(font=(entry,10))
> combobox = Pmw.ComboBox(root, label_text='Font:', labelpos='wn',
>                         listbox_width=24, dropdown=1,
>                         scrolledlist_items=comboentries,
>                         selectioncommand=selectFont)
> combobox.pack(fill=BOTH, expand=1, padx=8, pady=8)
> 
> combobox.selectitem(comboentries[0])
> 
> root.mainloop()
> 

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


From paul at prescod.net  Tue Jun  6 16:54:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 06 Jun 2000 15:54:52 -0500
Subject: indirect references
References: <8hha1n$phl$1@nnrp1.deja.com>
Message-ID: <393D651C.7626546C@prescod.net>

shaleh at debian.org wrote:
> 
> How do I do this in python:
> 
> foo = 'string'
> bar = foo
> 
> somehow assign to bar and change the value of foo.

There are hundreds of ways to do something like this but the best one
depends on what problem you are trying to solve. The most
straight-forward is to use a list of characters rather than a string
object, but that is probably not the best way.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Alabama's constitution is 100 years old, 300 pages long and has more
than
600 amendments. Highlights include "Amendment 393: Amendment of
Amendment
No.  351", "Validation of Laws Regulating Court Costs in Randolph
County",
"Miscegenation laws", "Bingo Games in Russell County", "Suppression
of dueling".  - http://www.legislature.state.al.us/ALISHome.html



From redaelli at inc.it  Sun Jun  4 05:30:49 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 11:30:49 +0200
Subject: float does not parse '0.000000E+00'
References: <39397EFF.E1B76E6E@inc.it> <8hbvv0$40h$1@news.wrc.xerox.com>
Message-ID: <393A21C9.E955293F@inc.it>

Mark Jackson wrote:

> Have you checked to see if you might have rebound float in that namespace?
What do you mean? I does not know the meaning of "rebound" in python
slang, although I've used it for almost 2 years...
10x in advance for your attention

-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/


From aahz at netcom.com  Thu Jun 22 13:02:58 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jun 2000 17:02:58 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <03bf34ab.667a04d2@usw-ex0105-040.remarq.com>
Message-ID: <8itgs2$4f0$1@slb6.atl.mindspring.net>

In article <03bf34ab.667a04d2 at usw-ex0105-040.remarq.com>,
RickZ  <rzantowNOrzSPAM at usa.net.invalid> wrote:
>
>I'd like to get my employer to add Python to its "approved" list of
>development tools, too. But which Python? It's an extra handicap to
>say, "Well, the language now is changing constantly in ways that will
>potentially break what we develop now. And I am not at all sure what it
>will look like at this time next year. But it's a great language!"

There were very few breaks from 1.5.1 to 1.5.2; the total time from
1.5.1 to 1.6 will be around three years.  Consider what other software
upgrades in that time caused breakage.  Note, too, that the majority of
code-breaking changes were already deprecated before 1.6 development
started.

I don't think this is a big issue, in the end.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From mhammond at skippinet.com.au  Tue Jun 13 14:48:39 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 13 Jun 2000 18:48:39 GMT
Subject: PythonWin and collections
References: <8hv12v$ol1$1@nnrp1.deja.com> <8i3g8d$kev$1@nnrp1.deja.com> <8i3qkm$rsh$1@nnrp1.deja.com>
Message-ID: <8i5vm2$bhr$1@nnrp1.deja.com>

In article <8i3qkm$rsh$1 at nnrp1.deja.com>,
  barryhart at my-deja.com wrote:
> Mark,
>
> Thanks for the tip. Is this build publicly available anywhere? I was
> looking at
> http://starship.python.net/crew/mhammond/win32/Downloads.html and
build

It appears it is only on the "old" starship.  I will update the "new"
starship ASAP (but that may not be for a few days, Im afraid)

Mark.


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


From bram at sara.nl  Wed Jun 21 05:07:41 2000
From: bram at sara.nl (Bram Stolk)
Date: Wed, 21 Jun 2000 11:07:41 +0200
Subject: CWI's Python and SGI's Performer
Message-ID: <395085DD.2429526D@sara.nl>

pfGreetings(),


I am currently evaluating the viability of embedding SGI's OpenGL|Performer
into CWI's Python. Or in other words, extend python with Performer 
functionality.

However, I would really like to know if someone has trodden this path before,
so I do not have to re-invent the wheel.

Basically, I want to do matrix math and basic operations on geoset, geodes,
etc. At later stage support for pfChannel also, maybe.

SWIG looks like a good tool to wrap the pf stuff for python use, but
has it been done before?

many thanks,

   Bram Stolk


-- 
------------------------------------------------------------------------------
 Bram Stolk, VR Specialist.
 SARA Academic Computing Services Amsterdam, PO Box 94613, 1090 GP  AMSTERDAM
 email: bram at sara.nl   Phone +31-20-5923059  Fax +31-20-6683167

 "I heard if you play the NT-4.0-CD backwards, you get a satanic message."
 "Thats nothing, if you play it forward, it installs NT-4.0"
------------------------------------------------------------------------------


From arun_rajappa at my-deja.com  Mon Jun  5 10:34:24 2000
From: arun_rajappa at my-deja.com (arun_rajappa at my-deja.com)
Date: Mon, 05 Jun 2000 14:34:24 GMT
Subject: Newbie Questions
Message-ID: <8hgdpf$2fn$1@nnrp1.deja.com>

hi,

i've been working with python for the last week and have gone through
the tutorial by guido. i have a few questions, and hope i will find
help and enlightenment here... ~~ :-)

a) Does Python support multi-threading ? What are the thread and
threading modules for ?

b) Can anyone give me a simple, "classy" example of implementing
polymorphism in Python ?

c) What is wxPython ? How does it compare with Tkinter ??

Thanks in advance,
arun rajappa


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


From rapto at my-deja.com  Thu Jun 15 13:22:11 2000
From: rapto at my-deja.com (rapto at my-deja.com)
Date: Thu, 15 Jun 2000 17:22:11 GMT
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain> <8hmmvu$ets$2@newshost.accu.uu.nl>
Message-ID: <8ib3bd$4pa$1@nnrp1.deja.com>

No way, did you ever see Monty Python's films translated (dubbed)? They
are just not funny.

Now, if I imagine them dubbed into Dutch, I might commit suicide.


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


From wjdandreta at worldnet.att.net  Fri Jun 23 03:43:15 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 23 Jun 2000 07:43:15 GMT
Subject: How to convert a string to hex?
References: <3952EC0D.D5CF9066@esec.com.au>
Message-ID: <nyE45.19061$Xx5.1052772@bgtnsc06-news.ops.worldnet.att.net>

Sam,

I asked the same question earlier. Here is the response:


Yes, everything except that trailing H anyway.  I'd use
string.atoi('c44d', 16), and I guess in the end that is
about the same thing as int('c44d', 16)

Donn Cave, donn at oz.net




From r2d2 at mao.acc.umu.se  Sun Jun  4 06:52:09 2000
From: r2d2 at mao.acc.umu.se (Niklas Frykholm)
Date: 4 Jun 2000 10:52:09 GMT
Subject: ANN: metahtmllib 0.5
Message-ID: <slrn8jkd6p.dr1.r2d2@mao.acc.umu.se>

metahtmllib 0.5

    by Niklas Frykholm
    released under GNU public license

metahtmllib is a Python library for facilitating the transformation of
structured text documents written in HTML-like languages. metahtmllib lets
you separate content (the source document) and form (a set of
transformation rules) so that the same source document can be used to
render HTML, LaTeX, text files, etc, in many different styles. You can
change the layout of any number of documents by only changing the style
rules. The transformation rules are written as Python methods. For
example, a rule for coloring text marked with <HILITE>...</HILITE> red
would look like this:

def hilite(self, text, attr):
    return "<FONT COLOR=#ff0000>%s</FONT>" % text

Some of the things which are easy to achieve with this library are for
example: automatic insertion of <P>...</P> tags, automatic insertion of
HEIGHT and WIDTH tags for images, generation of LaTeX and HTML from the
same source document, automatic gathering of indices and tables of
content, etc.

Extensive documentation with helpful examples is provided. Since this is
the first version of the module released to the public,
comments are very welcome.

Download at:

http://www.acc.umu.se/~r2d2/files/python/metahtmllib.html

// Niklas



From robin at jessikat.demon.co.uk  Sat Jun  3 12:42:33 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 3 Jun 2000 17:42:33 +0100
Subject: test
Message-ID: <B3tRKTA5VTO5Ew+Z@jessikat.demon.co.uk>

ignore
-- 
Robin Becker


From paul at prescod.net  Wed Jun  7 11:01:13 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 07 Jun 2000 10:01:13 -0500
Subject: Network byte ordering question ...
References: <CC59868A0A76D311B1F50090277B321C1F3397@VRTMAIL>
Message-ID: <393E63B9.4F68A406@prescod.net>

Mike Fletcher wrote:
> 
> ...
> 
> Don't know if there are any _current_ Waterloovian Pythonistas, the two I
> knew about (Paul and myself) having both graduated a while ago... and not
> even an alumni newsletter to show for it in my case ;) .  Enjoy,
> Mike

Also: Ka Ping Yee, Reuben Sumner and Blake Winton. It's clearly a
conspiracy. Hope I didn't miss anyone!
-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Alabama's constitution is 100 years old, 300 pages long and has more
than 600 amendments. Highlights include "Amendment 393: Amendment of
Amendment No.  351", "Validation of Laws Regulating Court Costs in
Randolph County", "Miscegenation laws", "Bingo Games in Russell County",
"Suppression of dueling".  -
http://www.legislature.state.al.us/ALISHome.html



From phil at geog.ubc.ca  Sun Jun 11 13:46:13 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 11 Jun 2000 10:46:13 -0700
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <m33dml6p34.fsf@brant.geog.ubc.ca> <110620000926542025%pecora@anvil.nrl.navy.mil>
Message-ID: <m3g0qk9km2.fsf@brant.geog.ubc.ca>

"Louis M. Pecora" <pecora at anvil.nrl.navy.mil> writes:

> In article <m33dml6p34.fsf at brant.geog.ubc.ca>, Phil Austin
> <phil at geog.ubc.ca> wrote:
> > If you're able to use the CXX extension, the sequence is:
> > 
> > Tuple t(2);
> > t[0]=Int(a);
> > t[1]=Int(b);
> > 
> > If the conversion fails, Int will throw an exception that
> > can be caught on the python side.
> 
> Phil,
> 
> What about reference counting for these CXX examples?

That's handled by the destructors for t, a, and b.  Here's
an example of a histogram extension I wrote recently that
takes a NumPy array and returns a dictionary with the bins
and counts.  I've also appended a session showing some
typical mistakes; CXX gives you quite  a bit of error checking
with minimal work

__________hist.cc_____________
#include"Python.h"
#include"arrayobject.h"
#include "CXX_Objects.h"
#include "CXX_Array.h"
#include "hist.hh"

using namespace Py;
using namespace std;

void inithist()
{
  static hist_module *hist = new hist_module;
}

static string hist_cc_ID=("$Id: hist.cc,v 1.3 2000/06/11 16:30:36 phil Exp phil $");


hist_module::hist_module(): ExtensionModule<hist_module>( "hist" )
{
  import_array();
  add_keyword_method("hist", &hist_module::ex_hist, 
		     "hist(data,max=max,min=min,numbins=numbins) -> dictionary");
  initialize( "documentation for the hist module" );

  Dict d( moduleDictionary() );
  d["rcsid"]=String(hist_cc_ID);
}

Object hist_module::ex_hist(const Tuple &args, const Dict &kws)
{
  args.verify_length(1);
  Array data(args[0]);
  if (data.species() != PyArray_FLOAT){
    throw ValueError("expecting float data");
  }
  float* dataPtr = (float*) data.to_C();
  int numbins = Int(kws["numbins"]);  
  float maxdata = Float(kws["max"]);  
  float mindata = Float(kws["min"]);  
  float binsize = (maxdata-mindata)/((float)numbins);
  Array outcounts(numbins,PyArray_INT);
  int* countPtr =(int*) outcounts.to_C();
  Array bincenters(numbins,PyArray_FLOAT);
  float* centerPtr = (float*) bincenters.to_C();
  Array binedges(numbins+1,PyArray_FLOAT);
  float* edgePtr = (float*) binedges.to_C();
  int  ibin;
  int lowcount=0;
  int highcount=0;
  int numpoints = data.dimension(1);
  for(int i = 0; i < numpoints; ++i){
    ibin = (int) ((*(dataPtr +i) - mindata) / binsize);
    if(ibin < 0){
      lowcount=lowcount + 1;
      continue;
    }
    if(ibin > (numbins - 1)){
      highcount = highcount + 1;
      continue;
    }
    *(countPtr + ibin) = *(countPtr + ibin) + 1;
  }
  for(int i = 0; i <= numbins; ++i){
    *(edgePtr + i) = mindata + (i*binsize);
  }
  for(int i = 0; i < numbins; ++i){
    *(centerPtr + i) = (*(edgePtr + i) + *(edgePtr + i + 1))/2.;
  }

  Dict retval;
  retval["edges"]=binedges;
  retval["centers"]=bincenters;
  retval["counts"]=outcounts;
  retval["lowcounts"]=Int(lowcount);
  retval["highcounts"]=Int(highcount);
  return retval;
}

_______hist.hh_______

#ifndef __hist__h
#define __hist__h

#include "Python.h"
#include "CXX_Objects.h"
#include "CXX_Extensions.h"

static std::string hist_h_ID=("$Id: hist.hh,v 1.1 2000/06/11 05:16:23 phil Exp $");

extern "C" void inithist();

class hist_module : public Py::ExtensionModule<hist_module>
{
public:
  hist_module();
  virtual ~hist_module() {}
private:
  Py::Object ex_hist (const Py::Tuple &args, const Py::Dict &kws);
};

#endif

_____demo  session______

>>> import  hist
>>> dir(hist)
['__doc__', '__file__', '__name__', 'hist', 'rcsid']
>>> hist.rcsid
'$Id: hist.cc,v 1.3 2000/06/11 16:30:36 phil Exp phil $'
>>> hist.hist.__doc__
'hist(data,max=max,min=min,numbins=numbins) -> dictionary'
>>> import Numeric
>>> import RandomArray
>>> myout=RandomArray.random((20000,))
>>> histout=hist.hist(myout.astype(Numeric.Float32),min=0.1,max= 1.,numbins=10)
>>> histout=hist.hist(myout,min=0.1,max= 1.,numbins=10)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: expecting float data
>>> histout=hist.hist(myout.astype(Numeric.Float32),min='oops',max= 1.,numbins=10)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: invalid literal for float(): oops
>>> histout=hist.hist(myout.astype(Numeric.Float32),minoops=0.1,max= 1.,numbins=10)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
KeyError: min
>>> histout=hist.hist(min=0.1,max= 1.,numbins=10)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IndexError: Unexpected SeqBase<T> length.



From tim.hochberg at ieee.org  Mon Jun 12 18:37:40 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Mon, 12 Jun 2000 22:37:40 GMT
Subject: NumPy arrays: how to remove extra axes
References: <slrn8kaov1.1ar.hzhu@rocket.knowledgetrack.com>
Message-ID: <m2og56k256.fsf@spiff.home.com>

hzhu at rocket.knowledgetrack.com (Huaiyu Zhu) writes:

> For arrays in Numeric, is there a general way to remove axes of length 1?
> For example I have an array of shape (1,3,1,4) I want to change it to (3,4).
> Thanks
> 
> 
> Huaiyu Zhu                               hzhu at knowledgetrack.com

Here's a function I just tossed together that should do what you
want. It's only barely tested, so I'd beat on it a bit before trusting
it.

-tim

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

from Numeric import *

def compactAxes(A):
    """Return an array with axes of length 1 removed"""
    index = []
    for length in shape(A):
        if length == 1:
            index.append(0)
        else:
            index.append(slice(None))
    return A[tuple(index)]

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



From mhammond at skippinet.com.au  Fri Jun  9 20:59:44 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 00:59:44 GMT
Subject: Linewrapping in Pythonwin?
References: <1Pb05.47081$WS3.490567@typhoon.we.rr.com>
Message-ID: <4qg05.712$cQ.9364@news-server.bigpond.net.au>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:1Pb05.47081$WS3.490567 at typhoon.we.rr.com...

>     I do have one big um.. sugguestion? (hides more from Mark) in
PythonWin.
> Or, rather, a question.Is there a way to have it linewrap in the
interactive
> window? I don't see an option -- maybe i'm just being blind, but when
i'm
> doing some debugging and looking at various attributes which are
possibly
> quite long, it'd be much easier to have it all wrapped so I don't have
to
> scroll anywhere to see anything.

Im a frayed knot.

Personally, I wouldnt use it - Id get more frustrated at losing the
horizontal space, as I tend to have a thin interactive window docked on
the bottom edge.

Look at the pprint module - it does a pretty good job.  If I did anything
like this, I would be more inclined to use pprint than use simple line
wrapping.

Mark.





From akuchlin at mems-exchange.org  Thu Jun  8 09:28:17 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Jun 2000 09:28:17 -0400
Subject: PyXML 0.5.5: fix for pyexpat.c
References: <m3k8g2fk8t.fsf@wilma.george.org>
Message-ID: <3d7lc0z4i6.fsf@amarok.cnri.reston.va.us>

Harry George <hgg9140 at seanet.com> writes:
> The command:
>   python setup.py build
> will work if you change the "New" to "NEW" on line 474

I've made a new release of the XML package that has a different fix
for this problem that should work with both 1.5.2 and 1.6.

http://www.python.org/sigs/xml-sig/files/PyXML-0.5.5.1.tar.gz

--amk


From gansevle at cs.utwente.nl  Thu Jun 29 10:31:38 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Thu, 29 Jun 2000 14:31:38 GMT
Subject: Mapping a network drive
References: <2B1262E83448D211AE4B00A0C9D61B03013011E2@msgeuro1.creo.be> <8iac3u$im8$1@nnrp1.deja.com> <8jflpo$a0t$1@nnrp1.deja.com>
Message-ID: <8jfmka$akk$1@nnrp1.deja.com>

> I just 'stumbeld' over the (not yet documented) win32wnet module in
> win32all-129
>
> With that module it is even simpeler
>
> Look at the 'new' wnet.py
>
[snip]
>
-----------------------------------------------------------------------
> ----          Linux/Windows-NT/IntraNetware Administrator
----
> -- Whenever it sounds simple, I've probably missed something! ... Me
--
>
-----------------------------------------------------------------------
Errr ... it probably sounded too simple ...

The following wnet.py *works*

# wnet.py
from win32wnet import WNetAddConnection2, WNetCancelConnection2, error

PROVIDER = 'Microsoft Windows Network'

def use (drive, host, user, passwd):
    remote = r'\\%s\%s' % (host, user)
    try:
        WNetAddConnection2 (1, drive, remote, PROVIDER, user, passwd)
        return 0
    except error:
        return 1

def unuse (drive):
    try:
        WNetCancelConnection2 (drive, 1, 1)
        return 0
    except error:
        return 1


--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From jkraska1 at san.rr.com  Thu Jun 15 02:46:41 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 15 Jun 2000 06:46:41 GMT
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie>
Message-ID: <39487C7C.A494198B@san.rr.com>

 
> As someone who uses and likes both C++ and Python, I think estimates
> like that are excessively simplistic.

Me, too. And yet I will claim that I've become *more* productive
in Python after a brief 4-6 weeks of use than I am in C++ after
15 years of use, and that's no exaggeration. If Python executed
as fast as other languages, I'd be tempted to *require* that our
staff use it. And please don't take this wrong; I'm no language
groupie -- I use C, C++, Java, Python, and Lisp -- all of them both
lately and frequently.

> I use C++ for commercial programming, because it gives me the power and
> control I need.  Productivity isn't the best when measured in time to
> get the first prototype running, but I've found nothing better when
> measured in time to develop, deploy, maintain and support something a
> customer will pay for.

Shrug. If it weren't for execution speed, your C++ environment would
be toast. Admit it. :)-



C/


From richard_chamberlain at ntlworld.com  Sat Jun 24 17:10:04 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 24 Jun 2000 22:10:04 +0100
Subject: tkinter, grid and resize
References: <39550101.AD8C7956@uniserve.com>
Message-ID: <gt955.953$F01.22132@news6-win.server.ntlworld.com>

Hi Bob,

Grid basically has an absence of fill, which is what pack uses to resize
widgets when their master resizes.

So you wouldn't use it to achieve what you want to do. The widgets fill
their allocated space according to the sticky options you pass but don't
resize when the master does. That is demonstrated in the following:

from Tkinter import *
root=Tk()
Text(root).grid(row=0,column=0)
Button(root,text='NS').grid(row=0,column=1,sticky=NS)
Button(root,text='N').grid(row=0,column=2,sticky=N)
Button(root,text='S').grid(row=0,column=3,sticky=S)
Button(root,text='EW').grid(row=1,column=0,sticky=EW)
Button(root,text='W').grid(row=2,column=0,sticky=W)
Button(root,text='E').grid(row=3,column=0,sticky=E)
root.mainloop()

You could rewrite your code in the following way:

from Tkinter import *
root=Tk()
frame=Frame(root)
frame2=Frame(root)
ys = Scrollbar(frame)
xs = Scrollbar(frame2)
dsp = Text(frame, yscrollcommand=ys.set, xscrollcommand=xs.set)
xs.config(orient='hor', command=dsp.xview)
ys.config(orient='vert', command=dsp.yview)
ys.pack(side=LEFT,expand=NO,fill=Y)
dsp.pack(side=LEFT,expand=YES,fill=BOTH)
frame.pack(side=TOP,expand=YES,fill=BOTH)
frame2.pack(side=TOP,fill=X)
Frame(frame2,width=16).pack(side=LEFT)
xs.pack(side=LEFT,expand=YES,fill=X)
root.mainloop()

achieves the same effect a little more verbosely. However I wrote this
pretty quickly so you should be able to rationalise it a little bit.

Richard

Bob van der Poel <bvdpoel at uniserve.com> wrote in message
news:39550101.AD8C7956 at uniserve.com...
>
> I'm having a problem getting widgets managed by grid to resize
> themselves when the screen size changes. This is easy to do using the
> pack manager and 'expand', however 'expand' doesn't exist in grid and
> the docs seem to indicate that the 'news' sticky option should do the
> same. The following snippet creates a text widget with scrollbars. If
> you resize the main window, the widgets do not resize....
>
> from Tkinter import *
>
> root=Tk()
>
> ys = Scrollbar(root)
> xs = Scrollbar(root)
>
> dsp = Text(root, yscrollcommand=ys.set, xscrollcommand=xs.set)
> xs.config(orient='hor', command=dsp.xview)
> ys.config(orient='vert', command=dsp.yview)
> xs.grid(row=1, column=1, sticky=E+W)
> ys.grid(row=0, column=0, sticky=N+S)
> dsp.grid(row=0, column=1, sticky=N+S+E+W)
>
> root.mainloop()
>
> What am I don't wrong? I've tried to insert columnspan stuff and played
> with the weight options--but I don't seem to be getting it.
>
> Thanks.
>
> --
>    __
>   /  )      /         Bob van der Poel
>  /--<  ____/__        bvdpoel at uniserve.com
> /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>




From ajwilson at ezy.net.au  Sat Jun  3 18:56:02 2000
From: ajwilson at ezy.net.au (A J Wilson)
Date: Sun, 4 Jun 2000 08:56:02 +1000
Subject: Combining two graphic images
References: <959996628.860207@centauri.ezy.net.au> <n3aijsoefu73lgv2h2h37750di9ckk7uh6@4ax.com>
Message-ID: <960072929.170200@centauri.ezy.net.au>

I downloaded PIL and tried to compile/install it.  However, when I try to
import it in python to test it I get the following error.

>>> import _imaging
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: ./_imaging.so: undefined symbol: jpeg_resync_to_restart
>>>

Does anyone know what I am doing wrong.
I am running Mandrake 7 on an intel box.

Also what other rpm packages do I need to have installed to make it work?

Thanks
Austin


"Anders Eriksson" <ame at swipnet.se> wrote in message
news:n3aijsoefu73lgv2h2h37750di9ckk7uh6 at 4ax.com...
> On Sat, 3 Jun 2000 11:45:12 +1000, "A J Wilson" <ajwilson at ezy.net.au>
> wrote:
>
> >Hi
> >
> >I would like to combine two BMP images so that one overlays the other.
Does
> >anyone know how this could be done using Python.
> >
>
> When it comes to imaging and working with images you can't beat PIL!
>
> PIL, Python Imaging Library, created by mr Fredrik Lundh is the best
> thing since sliced bread!
>
> http://www.pythonware.com/
>
> You may want to look up the function 'composite'
>
> // Anders
>




From andre at beta.telenordia.se  Mon Jun  5 13:16:23 2000
From: andre at beta.telenordia.se (André Dahlqvist)
Date: Mon, 05 Jun 2000 18:16:23 +0100
Subject: Scrolled canvas confusion
References: <8hg4a3$a09$1@zingo.tninet.se> <8hgd0r$1p8$1@nnrp1.deja.com>
Message-ID: <8hgjp8$5pt$1@cubacola.tninet.se>

> If you don't want to use Pmw (which is only Tkinter code anyway), 
> then look at PmwScrolledFrame.py and extract the bits you want and
> manage resources directly.

I definately don't have anything against Pmw, but I just don't want to 
add another requirement for using this program. The program I'm
working on is CML2, which is the new Linux kernel configuration tool.
People are pissed off enough as it is that Eric Raymond chose to use 
Python for this, so I don't want to add Pmw to the requirements:-)

If anyone wants to help out and add scrolling to the Tk interface of
CML2, or just help out in general, you can grab the source from:

http://www.tuxedo.org/~esr/kbuild
 
// Andr?



From dparsavand at aol.com  Wed Jun 21 15:48:02 2000
From: dparsavand at aol.com (Dparsavand)
Date: 21 Jun 2000 19:48:02 GMT
Subject: Is there no inversion of gmtime()?
References: <slrn8l11f0.16n.newsbunny@noether.freeserve.co.uk>
Message-ID: <20000621154802.07167.00000345@ng-fm1.aol.com>

J?rgen Hermann wrote:
> There are actually two such functions, time.mktime() and 
> rfc822.mktime_tz(). Now read the docs. :)

I did read the time module documentation before posting - the function
time.mktime is an inverse of time.localtime and thus at first, I did not think
of using it. So, in otherwords:

Set sec to an integer >= 0 
Import the time module
(mktime(localtime(sec))) correctly returns sec (as a float)
Is there a function f such that (f(gmtime(sec))) returns sec?

There is no function f explicitly listed in the time module (nor rfc822), but I
didn't see that it is easy to construct one with use of the timezone variable.
(mktime(gmtime(sec)) - timezone) returns sec correctly in all cases I've tried
so far.

Martin P Holland wrote:
> There was a thread about this in this newsgroup around 05/03/1999 and
> G. van Rossum posted a readable solution (which does what you'd expect)
> which I use in my ftpupdater app

I only briefly looked at this code, but I'll probably just use the timezone
offset. 

Thanks, Dara


From ivanlan at home.com  Wed Jun 14 11:58:24 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Wed, 14 Jun 2000 09:58:24 -0600
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB334@ALTNET> <8i890f$4fj$1@slb7.atl.mindspring.net>
Message-ID: <3947ABA0.4987D728@home.com>

Hi All--

Aahz Maruch wrote:
> 
> In article <65118AEEFF5AD3118E8300508B1248774CB334 at ALTNET>,
> Mike Steed  <MSteed at altiris.com> wrote:
> >
> >ctor
> 
> Okay, does anyone else think this is a misspelling of "Chtorr"?
> 

Yes.  When the hell is Gerrold going to get back to the series?

<topping-from-inside>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From rpm at wag.caltech.edu  Tue Jun  6 12:32:24 2000
From: rpm at wag.caltech.edu (Richard P. Muller)
Date: Tue, 06 Jun 2000 09:32:24 -0700
Subject: PyOpinion: Does Python Programming Marginalize You?
Message-ID: <393D2797.4E4EE7CC@wag.caltech.edu>

Does Programming in Python Marginalize You?

I love Python. Of the 10-15 programming languages I know, Python is
the only language which I actually look forward to using. I wake up in
the morning, ask myself, "Will I be using Python today?" and if the
answer is "yes" I know that it's going to be a good day.

And yet, I worry that using Python marginalizes one. Python is hard to
beat for writing applications that run on a single computer. But
that's yesterday's computing model. The future of computing, heck, the
present of computing, is in writing programs that run through a web
browser. And the fact of the matter is that 99% of web browsers
surfing the internet run Java and JavaScript, which gives these
languages a huge advantage over Python, an advantage that Python's
superior design may not be able to overcome. I have Ada-programming
friends who mumble about Ada's superiority over C++; that may be the
case, but does it really matter anymore?

I'm aware of CGI scripting with Python. It's great. But JavaScript
pages load faster than CGI pages, and are easier to program, to
boot. Plus, JavaScript can directly call Java Applet methods, enabling
it to do functions that are simply impossible with CGI scripts.

I'm also aware of the JPython module, which promises to enable
us to write applets in Python instead of Java. But who among us has
actually *used* JPython? Or, for that matter, who has come across a
random web page on the net that uses JPython applets? 

Do I recommend that students today learn Python? Certainly, because
it's easy, fun, it teaches one good OO habits, and it's incredibly
powerful. But I also recommend that students learn C/C++ for speed (I
no longer recommend Fortran), and Java and/or JavaScript for web
programming. Heck, I also recommend that they learn Perl, since so
many other people know Perl. That leaves a significant niche for
Python, but only a niche, and Python is too good a language to be a
mere niche player.

If we don't want to see Python wind up like Ada, what should we do?
I've been teaching a 6-lecture short-course on Python here at Caltech,
in an effort to introduce more people to the language. I'm thinking
about learning enough JPython to give an additional lecture on that
language. Hopefully, if enough people use JPython for writing Applets
we can help establish it as a reasonable alternative to Java. As for
JavaScript, I don't know of a solution. It would have been nice if
Netscape had implemented Python's syntax for LiveConnect rather than
Java, but it didn't happen that way. Think how wonderful it would have
been to write PythonScript functions in web pages rather than
JavaScript.

What else should we be doing to make Python *the* programming language
rather than *a* programming language?

Rick


From ovi at autornica.no  Mon Jun 26 04:27:16 2000
From: ovi at autornica.no (Olav Viken)
Date: Mon, 26 Jun 2000 10:27:16 +0200
Subject: Passing variables between scripts
Message-ID: <8j745m$qqj$1@oslo-nntp.eunet.no>

Can anyone tell me if there is a good way to pass a variable between two CGI
scripts?

Best regards Olav Viken






From rhicks at nospam.rma.edu  Fri Jun  2 12:04:12 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 02 Jun 2000 16:04:12 GMT
Subject: Python, Glade, & GTK
References: <8gb1gf$r0u$1@news.ihug.co.nz> <39295964.BD653227@bioreason.com>
Message-ID: <0WQZ4.29$vh.14640@news>

Glade ExampleWhy don't the GLADE maintainers just incorporate the ability to
save as Python code? It would be the thing that makes me swing from KDE to
GNOME.

Bob

Mitch Chapman <chapman at bioreason.com> wrote in message
news:39295964.BD653227 at bioreason.com...
Mafee wrote:
> Can anyone give me some clues of linking python code to a GUI interface
> generate by Glade or GTK or whatever please?

Here's an example.  Two files are attached.

GladeBase.py provides two base classes.  Class UI creates a widget
hierarchy from a Glade file.  Class Controller provides an
MVC-style controller for the UI.  The main() function of the module
shows how to use these base classes to build and control a GUI.

ui.glade is a simple user interface created using Glade.  If you
plop it into the same directory with GladeBase.py and then run
GladeBase.py, you should get an application window on your display.

The cool part about James Henstridge's libglade, on which this example
is based, is the way it wires up callback routines.  You pass it
a dictionary of callable objects, keyed by name.  It roots
through the Glade-based user interface hierarchy, and for every
connected signal declared in the interface, it looks for a callable
of the same name in the callback dictionary.  If it finds one,
it plugs it in.  It's cheap and easy.

In the example, Controller._connectToUI() assumes that you will
derive from class Controller and define all of the callback methods
in the derived class.

Hoping this helps...

--
Mitch Chapman
chapman at bioreason.com



#!/usr/bin/env python
"""This is an example of accessing Glade-built UIs from Python."""
import gnome.ui, gtk, libglade

class UI(libglade.GladeXML):
    """Base class for all UIs loaded from glade."""
    def __init__(self, filename, rootname):
        """Initialize a new instance."""
        libglade.GladeXML.__init__(self, filename=filename, root=rootname)
        self.root = self.get_widget(rootname)

    def show(self):
        """Make the UI visible."""
        self.root.show()

class Controller:
    """Base class for all controllers of glade-derived UIs."""
    def __init__(self, ui):
        """Initialize a new instance.  `ui' is the GladeXML UI to
control."""
        self.ui = ui
        self._connectToUI()

    def _connectToUI(self):
        """Wire up self to its UI.

        This method assumes that any method defined by self's class
        could be a Gtk+ callback.  It wires up any methods referenced by
        self's ui."""
        names = dir(self.__class__)
        d = {}
        for name in names:
            d[name] = getattr(self, name)

        self.ui.signal_autoconnect(d)

    def show(self):
        """Show the user interface."""
        self.ui.show()

def main():
    """Module mainline (for standalone execution)"""
    class SampleController(Controller):
        def on_pushBtn_clicked(self, *args):
            """Called when the 'push me' button is clicked."""
            print "Thanks for pushing me."
            gtk.mainquit()

        def on_win_delete_event(self, *args):
            """Called when the window is deleted."""
            gtk.mainquit()

    theUI = UI("ui.glade", "win")
    theController = SampleController(theUI)
    theController.show()
    gtk.mainloop()

if __name__ == "__main__":
    main()




CreateWorkspace create_workspace src pixmaps C True True GtkWindow win
delete_event on_win_delete_event Wed, 03 May 2000 00:16:13 GMT
GTK_WINDOW_TOPLEVEL GTK_WIN_POS_NONE False False True False GtkVBox vbox1
False 0 GtkLabel label1 Go on, push the button. GTK_JUSTIFY_CENTER False 0.5
0.5 0 0 0 False False GtkButton pushBtn True clicked on_pushBtn_clicked Mon,
22 May 2000 15:28:00 GMT Push Me 0 False False




From thomas at xs4all.net  Mon Jun 19 13:14:58 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 19 Jun 2000 19:14:58 +0200
Subject: Why is Python so slow ?- revisited.
In-Reply-To: <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu>; from bparsia@email.unc.edu on Mon, Jun 19, 2000 at 06:52:50AM -0400
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu>
Message-ID: <20000619191457.A26436@xs4all.nl>

On Mon, Jun 19, 2000 at 06:52:50AM -0400, Bijan Parsia wrote:
> William Dandreta <wjdandreta at worldnet.att.net> wrote:

> > The biggest improvement came (about a factor of 6) when I changed the
> > replace(x,y) function in string with
> > joinfields(splitfields(x,y),''). Considering that this is exactly what the
> > replace function in string does, I was quite surprised at the results.
> > Essentially I reduced 3 nested function calls to 2.

> And that can make an enormous difference, particularly with earlier
> versions. Remember that many "simple function calls" in fact involve
> rather expensive polymorphic lookup and dispatching. In a tight loop
> this can get very painful (Python doens't have, I believe even in the
> latest versions, a method/function cache, so every lookup will go
> through the whole rigamarole each time).

Indeed. Such a cache would have to be invalidated whenever the function
object or one of the objects used to lookup the function, was altered (or
rather, 'touched'.) The dynamic nature of Python makes this quite
complicated, and probably quite expensive.

What's more, currently 'bound methods' (methods which have a 'self' to which
they apply) are created on demand -- the unbound method exists on the class,
but it has to be 'bound' to an instance before it can be used. This is done
each time you access the method, though CPython is being smart and you have
to try hard to actually see that ;-)

>>> class Minister:
...     def walk(self):
...             print "The minister walks silly."
... 
>>> cleese=Minister()

>>> cleese.walk is cleese.walk
0
>>> x,y=(cleese.walk, cleese.walk)
>>> id(x), id(y)
(134879776, 134879248)

Adding methods to the instance dictionary works, but breaks pickle (and
cPickle, and of course all code that does something to the parent class or
the objects' __bases__) and seems to incur penalties in other areas. (At
least according to pybench ;) But if you want to speed up your inner loop
even more, move methods into your local namespace. (Up to a certain limit,
for sure, but if you do a lot of methods in an inner loop you shouldn't be
too worried about performance in the first case.)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From johnvert at my-deja.com  Mon Jun 19 16:06:01 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Mon, 19 Jun 2000 20:06:01 GMT
Subject: {Q} OOP Animal class.
Message-ID: <8ILUF0$LE$1@NNRP1.DEJA.COM>

Greetings,

I just the example and explanation of the classic Animal class in the
tutorial in the end of Programming Python by Orielly, and I have a few
questions.

This is the class (irrelevant parts snipped):

class Animal:
  def reply(self):  self.speak()
  def speak(self):  print 'spam'

class Mammal(Animal):
  def speak(self):  print 'huh?'

class Primate(Mammal):
  def speak(self):  print 'Hello world!'

class Hacker(Primate):
  pass

>>> data = Hacker()
>>> data.reply()
Hello world!

Programming Python says: ``Hacker has no methods of its own, but
inherits from Primate, which inherits from Mammal, which inherits from
Animal. When we call the reply method of the Hacker instance, it runs
Animal's reply, which in turn calls the Primate's speak method
(self.speak finds the first speak by looking at Hacker and above).
Finally, our message gets printed.'', p. 846

I want to make sure I understand the sequence of events right:

1. I call the method reply()
2. The interperter sees reply() does not exist in class Hacker
3. It goes one class up the hierarchy to class Primate
4. reply() doesn't exist in class Primate
5. It goes one class up the hierarchy to class Mammal
6. reply() doesn't exist in class Mammal either
7. It goes one class up the hierarchy to class Animal
8. reply() exists in class Animal, it calls it, and self.reply() gets
translated to `Primate.reply()' since Hacker is of type Primate
9. Primate.reply() is called
10. `Hello world!' is printed

Is that about right? The thing that I'm most unsure of is step #8..
hoping you could clarify.  I really want to have a solid understanding
of this because it's important to understand OOP.

Thanks,
--
John




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


From jhe at webde-ag.de  Fri Jun 23 15:33:58 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 23 Jun 2000 21:33:58 +0200
Subject: XML-code examples
References: <394f7bc2.365514882@news.online.no> <00062201025104.08285@quadra.teleo.net>
Message-ID: <8j0e0g$klc$1@pollux.ip-plus.net>

> [Thomas Weholt, on Wed, 14 Jun 2000]
> :: Is there more simple XML code out there? I mean the HowTos took about
> :: ten minutes, and it didn`t cover much.

For a real-world example, you can check out

http://center.oftheinter.net/~jh/xml/bookmarks/checkurls.py

and the other files in that directory. It parses a XBEL file, and checks the
status of the bookmarks in that file via the httplib module.

My very first non-trivial Python program, BTW. :)




From fiona at sitegnome.com  Tue Jun  6 08:02:22 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 6 Jun 2000 12:02:22 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 6th, 2000
Message-ID: <20000606120222.18409.qmail@synop.com>

Hi Guys,

Another installment of entries into http://python.faqts.com

cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Is there a version of Python that runs well on Windows 2000 Server?
http://www.faqts.com/knowledge-base/view.phtml/aid/3517
-------------------------------------------------------------
Fiona Czuczman
Grant Munsey

Python 1.5.2 and 1.6a2 both work as advertised on Win 2K pro and Win 2K 
server.


-------------------------------------------------------------
Where can I find examples of web programming with Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3515
-------------------------------------------------------------
Fiona Czuczman
Paolo Redaelli

www.zope.org a COMPLETE, Object oriented, enviroment with connection to
traditional SQL engines, with a Object oriented database and MANY other
things.


-------------------------------------------------------------
Does Zope have a version that runs well with Windows 2000 server?
http://www.faqts.com/knowledge-base/view.phtml/aid/3518
-------------------------------------------------------------
Fiona Czuczman
Wolfgang Strobl

I haven't had any difficulties running 2.1.6 on Windows 2000, so far. I
tried it on both Win2000 prof US and Win2000 Server Ger.


-------------------------------------------------------------
How can I strip unwanted characters out of my string? Octal 240 for instance.
http://www.faqts.com/knowledge-base/view.phtml/aid/3519
-------------------------------------------------------------
Fiona Czuczman
Ken Seehof

>>> import string 
>>> string.replace("s\240p\240a\240m","\240","") 
'spam'



But I want to actually strip them out of my string:

Can't exactly do that.  Strings are immutable.  If you don't know what 
I mean by immutable, find out (mutability is a key concept of python). 

(Hint: if strings were not immutable, they could not be safely used as 
keys in dictionaries.) 

So you'll have to settle for: 

>>> s = "s\240p\240a\240m" 
>>> s = string.replace(s, "\240", "") 
>>> print s 
spam






From hzhu at knowledgetrack.com  Wed Jun 14 20:04:03 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 15 Jun 2000 00:04:03 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com>

On 14 Jun 2000 19:11:08 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>
>How about just writing a function to verify methods against a
>reference class.  The reference class could either be an actual class
>that you are mimicing the interface for, or a dummy class that is just
>defined with the appropriate methods.
>
>For a crude quick example, perhaps something like:
>
>    def verify_class (Reference, Instance):
>
>	import types
>
>	print "Checking %s for conformity with interface %s" % \
>	      (Instance.__name__,Reference.__name__)
>
>	for curmethod in Reference.__dict__.keys():
>	    if type(Reference.__dict__[curmethod]) == types.FunctionType:
>		if (not Instance.__dict__.has_key(curmethod) or
>		    type(Instance.__dict__[curmethod]) != types.FunctionType):
>		    print "Method %s not supported" % curmethod

Cool!  Things always look easier in python. :)

Now only if somebody could enhance that to check inherited methods too (the
"more general case").  That is, if a method is not found, go recursively
into the superclasses.  The following example is not correct at the moment:

------------------------------------------------------------------
class MyInterface:
    def method1():       pass
    def method2():       pass
    def method3():       pass
    def method4():       pass

class MyBase:
    "Actual implementations"
    def method1():       pass
    def method2():       pass

class MyDerived(MyBase):
    "Actual implementations"
    def method1():       pass
    def method3():       pass


if __name__ == "__main__":
    verify_class(MyInterface,MyBase)
    verify_class(MyInterface,MyDerived)
    print MyDerived().method2

------------------------------------------------------------------
gives 

Checking MyBase for conformity with interface MyInterface
Method method3 not supported
Method method4 not supported
Checking MyDerived for conformity with interface MyInterface
Method method2 not supported
Method method4 not supported
<method MyBase.method2 of MyDerived instance at 80cb550>



-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From david_ullrich at my-deja.com  Tue Jun  6 13:09:48 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Tue, 06 Jun 2000 17:09:48 GMT
Subject: operator overloading
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com> <393C2386.307A160D@sightreader.com> <20000605235010.C469@xs4all.nl> <8hj2f9$1rn$1@nnrp1.deja.com> <oqwvk2df1q.fsf@titan.progiciels-bpi.ca>
Message-ID: <8hjb8c$96f$1@nnrp1.deja.com>

In article <oqwvk2df1q.fsf at titan.progiciels-bpi.ca>,
  =?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> wrote:
> david_ullrich at my-deja.com writes:
>
> > This reminds me of a question that I've never got around to asking:
Are
> > there implicit requirements on __cmp__, like the order it defines
has
> > to be transitive, __cmp(a,b)__ has to be the opposite of
__cmp(b,a)__, etc?
>
> If one wants tools like `sort()' to work properly, `__cmp__' should
establish
> a proper order relation, yes.  Otherwise, rather cryptic bugs might
pop out.

    Presumably. If so this should probably be mentioned in the docs.
(I recall once long ago trying to get a bug from a non-transitive
__cmp__, and failing. But I didn't try very hard...)

> However, to give a special meaning to `<=', say, one might surely play
> tricks and not use `__cmp__' as it was meant to be.  A butterfly is
surely
> free to travel nearby a fire :-).

    Heh-heh. I guess I'm getting old - things like using __cmp__
in unusual ways don't thrill me the way they used to.

DU

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


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


From dana at mmi.oz.net  Wed Jun 14 11:03:43 2000
From: dana at mmi.oz.net (Dana Booth)
Date: 14 Jun 2000 15:03:43 GMT
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com>
Message-ID: <8i86sf$e7c$0@216.39.141.36>

Grant Edwards <ge at nowhere.none> wrote:

GE: I've got a Tkinter-based GUI interface used to do
GE: administrative stuff for a device driver under Linux.  I know
GE: RedHat installs Python and Tkinter by default.  Is there a
GE: handy list somewhere of which Linux distributions can be
GE: expected to have Tkinter installed?

It's best to choose not to put on Python at install time, and then just
retrieve the latest version from the Internet. Uncomment the lines in the
Modules/Setup file before you build it pertaining to Tkinter, then you're
assured of having the latest version.

Of course, you'd need to make sure that you have TCL/TK installed... But
then, you should do that yourself, too.

This way makes it easier to keep track of when you want to upgrade. By doing
it yourself, you know exactly where the files went. When you upgrade after
an auto install, you don't know if the distribution's install put the files
in weird places, so that you'll have conflicting crap all over your drive.
For instance, I installed Mandrake 6.0 once, and it put a ton of KDE junk in
/usr/bin. What a stupid place, and what a clutter. With Python, if you
follow the configuration file defaults before you make, it'll always be nice
and cozy in /usr/local/lib/Pythonxx. Wanna upgrade? You can just move the
old directory out of the way, and then move your homemade modules directory
back once you've put a new version in.

-- 
-----
Dana Booth <dana at mmi.oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From jdgorman at att.net  Fri Jun  9 01:07:53 2000
From: jdgorman at att.net (John D. Gorman)
Date: Fri, 09 Jun 2000 05:07:53 GMT
Subject: Nympy for win
References: <Pine.LNX.4.21.0006061109170.30720-100000@chelin.chtf.stuba.sk> <fAdMjhAnqNP5Ew3P@jessikat.demon.co.uk> <m3aegy60g1.fsf@brant.geog.ubc.ca> <393FC2D2.F751741E@americasm01.nt.com>
Message-ID: <39407B16.122875DC@att.net>

Brad,
    You need to snag a copy of distutils and install it on your machine first.  Then use
distutils to install numpy.  There's a link to Distutils (
http://www.python.org/sigs/distutils-sig/ ) on the Numpy page (
http://numpy.sourceforge.net/ )

Cheers,
John Gorman

"Morris, Brad [WDLN2:2W40:EXCH]" wrote:

> Thanks, I've been wondering where Win NumPy could be found. Having
> looked through the files I couldn't see an obvious way to install it.
>
> Is there a README or web page that says where to put all the files from
> the zip archive into the python directories?
>
> Also, where is the python path information stored under windows to allow
> other directories to be searched by python (I'm a Mac user using a
> PC...)?
>
> Regards,
> BJM
>
> Phil Austin wrote:
> >
> > Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> >
> > > >I would like to use the Numpy package with my windoze/python1.5.2 system,
> > > >but it needs distutils and distutils need m$ visual c++ which I don't own.
> > > >Is here anybody who has it compiled and could send me the
> > > >binaries? Thanks.
> >
> > The binary file at http://download.sourceforge.net/numpy/python-numpy-15.2.zip
> > works for me with python 1.5.2.  Unfortunately, the new sourceforge site
> > hasn't made it onto the numerical topics webpage at:
> >
> > http://www.python.org/topics/scicomp/numpy_download.html
> >
> > and the LLNL archive no longer exists.  Is there someone besides webmaster at python.org
> > who should get this bug report?
> >
> > Regards, Phil



From wlfraed at ix.netcom.com  Fri Jun 16 23:08:13 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Fri, 16 Jun 2000 20:08:13 -0700
Subject: win32com and Excel
References: <3939D364.2EF22B13@zebrasoft.co.nz> <B_u_4.111323$E85.2192336@news1.rdc1.md.home.com> <8ie1n1$8d0$1@nnrp1.deja.com>
Message-ID: <1oolks80jo1p1vvkmsbhc8k1dl27111igq@4ax.com>

On Fri, 16 Jun 2000 20:12:34 GMT, v2ng at my-deja.com declaimed the
following in comp.lang.python:

> 
> My Python class is really small, a straight rip off of the PPOW
> O'Reilly book.
>
	Did you remember the part about how to obtain a clsid specific
to your machine? (Just a wild guess, since I've only glanced at the book
itself, and remember a warning about that).

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From info.jyv at videotron.ca  Sat Jun 10 11:03:48 2000
From: info.jyv at videotron.ca (Jean-Yves Vinet)
Date: Sat, 10 Jun 2000 11:03:48 -0400
Subject: windll
References: <3941ACC5.305B8830@videotron.ca> <3941B6B4.264FA7A1@san.rr.com> <l8j05.833$cQ.10319@news-server.bigpond.net.au> <3941D9BD.46BBEF0C@san.rr.com>
Message-ID: <394258D4.93633A93@videotron.ca>

I can't find the module "dl" on windows. Is the module
multi-plateform?


Courageous wrote:

> > > > Is there someting like to "windll" to use
> > > > dynamic librairy under Linux/Unix plateform.
> > >
> > > What does windll do? "Using" dynamic libraries
> > > under Unix is as easy as linking with them and
> > > then making certain that they are in your
> > > LD_LIBRARY_PATH.
> >
> > Using them from Python is that easy?  Really?
> >
> > windll tries to make using arbitrary DLLs, *from Python* that simple!
>
> Yeah, I had a discussion off line with someone else.
> For some reason I didn't consider that the person meant
> a python module. Anyway, what you're looking for on unix
> is "dl", as in:
>
> import dl
>
> C/



From morpheus at despair.rpglink.com  Thu Jun 22 14:00:11 2000
From: morpheus at despair.rpglink.com (Steve Lamb)
Date: Thu, 22 Jun 2000 18:00:11 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3950C0E2.7532D38@sightreader.com>
Message-ID: <slrn8l4kuo.6ah.morpheus@teleute.rpglink.com>

On Wed, 21 Jun 2000 06:19:30 -0700, Ken Seehof <kens at sightreader.com> wrote:
>if (x)
>    if (z)
>        do_something();
>else
>{
>     do_something_completely_different();
>}

    Just the be that guy that always points out proble with examples the
problem here isn't bad indention, the problem is not using braces explicitely.
Just because you /can/ shortcut the if statement in C doesn't mean that you
should, IMHO.  IF, to me, is always if{}, even if that if has a single
statement after it.  Kinda like how I always put parens around everything in
perl even though they aren't needed because you never know what is going to
grow from your code.

    With that said, I can see your point in the abstract.  ;)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From loewis at informatik.hu-berlin.de  Wed Jun 28 10:57:54 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Jun 2000 16:57:54 +0200
Subject: native Python text editor with Hebrew support?
References: <395A0B71.918A6673@nih.gov>
Message-ID: <p6qu2edyhsd.fsf@informatik.hu-berlin.de>

Jonathan Epstein <Jonathan_Epstein at nih.gov> writes:

> I realize that this is getting a bit specialized, but I am looking for a
> text editor written in Python which can handle some rudimentary facets
> of Hebrew; namely, writing right-to-left and the Hebrew fonts (e.g., ISO
> 8859).

As for displaying the text, I believe that Tkinter should work - at
least if you build it on a sufficiently recent version of Tk. I'm not
sure whether the Text widget has any right-to-left capabilities,
though.

Regards,
Martin



From wware at world.std.com  Mon Jun 19 17:04:09 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 19 Jun 2000 21:04:09 GMT
Subject: {Q} OOP Animal class.
References: <8iluf0$le$1@nnrp1.deja.com>
Message-ID: <FwF6Iy.G0I@world.std.com>

johnvert at my-deja.com wrote:
> 1. I call the method reply()
> 2. The interperter sees reply() does not exist in class Hacker
> 3. It goes one class up the hierarchy to class Primate
> 4. reply() doesn't exist in class Primate
> 5. It goes one class up the hierarchy to class Mammal
> 6. reply() doesn't exist in class Mammal either
> 7. It goes one class up the hierarchy to class Animal

Good so far. As you suspected, this is where things start to
get a bit dicey. My understanding (which perhaps others will correct
if necessary) is that each method search is a separate thing, and in
each case we are prepared to climb the inheritance tree. So when
reply() calls speak(), we begin the whole search process anew, starting
at Hacker. Hacker doesn't have a speak() method, but Primate does,
and that's why we use Primate.speak().

> I really want to have a solid understanding
> of this because it's important to understand OOP.

OOP is primarily a flexible way to let you make say things like this:

	A Blarndoogle is an object with this set of behaviors.
	A Flugestroid is almost exactly like a Blarndoogle, except
	that it splointers the way a Zargrove does, and it has its
	own unique way of flithering, different from either a
	Blarndoogle or a Zargrove.

The idea would be that the behaviors for Blarndoogles and Zargroves
would probably already be thoroughly debugged when you were ready to
define a Flugestroid, so the only really new thing you need to do is
define how it flithers. That is, all your effort is focussed entirely
on what's new about this situation, not on re-implementing and
re-debugging existing behaviors. At least that's the theory.

Some languages (Python generally not among them) stress a "privacy"
aspect of objects, that there should be information available only to
the object and not to anybody else, and the object can therefore define
how others get to view or modify that information. Python mostly
doesn't worry about object privacy much; you can create private
attributes by starting attribute names with two underscores, but you
don't have the elaborate machinations of public/protected/private you
get with C++ and Java.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From thantos at chancel.org  Wed Jun 14 08:19:13 2000
From: thantos at chancel.org (Alexander Williams)
Date: Wed, 14 Jun 2000 12:19:13 GMT
Subject: [Q:] hash table performance!
References: <8i7kro$i04$1@nnrp1.deja.com>
Message-ID: <slrn8keu20.ahe.thantos@brimstone.mecha>

On Wed, 14 Jun 2000 09:56:12 GMT, liwen_cao at my-deja.com
<liwen_cao at my-deja.com> wrote:

>P.S. Background of duplicate files check:
>My way of doing it is simple: walk the directories and files, do MD5
>coding for every file, use the MD5 code as the hash key, insert the
>file name into the hash table. When two files has the same key, check
>the contents by bytes.

Well, odds are that the slowness is coming from MD5, not the actual
dictionary implimentation.  Hashing in Python is what we in the
community call "damn fast" (this is a technical term, you are not
expected to understand it).

Were it me, and I was more interested in content than absolute
perfection, I'd use a less stringent content hash than MD5;as it
stands, you're reading files which are content-similar twice on every
run, once to gen the MD5, a second time, byte-for-byte, if it finds a
match.  I'm going to wager heavily the MD5 alone should be sufficient
in itself to determine similarity in any set of files you're likely to
have.  Are you, in fact, seeing several files with the same MD5 and
different content, or are you being paranoid above and beyond the call
of duty?

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From redaelli at inc.it  Sat Jun  3 18:13:23 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 00:13:23 +0200
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home> <39397FD4.67C7FD73@inc.it>
Message-ID: <39398303.6B224681@inc.it>

I'm a stupid beast!
I forgot asking HUMBLY to try my horrible code....
	Cheers
		Paolo

PS: the code that does not seems horrible come traight from PyGnome
examples.....


From exarkun at flashmail.com  Sat Jun 24 00:24:22 2000
From: exarkun at flashmail.com (Jp Calderone)
Date: Sat, 24 Jun 2000 00:24:22 -0400
Subject: realtime non-blocking keyboard input
Message-ID: <395437F6.73BFE6B3@flashmail.com>

 Any way *other* than using the curses library to check to see if a key
has been pressed, and if so, which one?  Or any way to prevent
curses.initscr()
from messing w/ normal prints?

 Thanks in advance...
  Jp

--
 12:23am up 50 min, 7 users, load average: 0.14, 0.13, 0.15


From breiter at usf.Uni-Osnabrueck.DE  Wed Jun 21 08:20:52 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 21 Jun 2000 12:20:52 GMT
Subject: Boa stalled? (GUI editor for wxPython)
Message-ID: <8iqbv4$4p$1@newsserver.rrzn.uni-hannover.de>

Is Boa development stalled?
It looks like nobody made changes in the last three month.

The program still needs developers. It is in competition to the nice
glade and libglade thing and not mature enough yet.
Why is nobody picking it up?

(Just to bring this to your attention.)

http://boa-constructor.sourceforge.net/

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From kain at cableadmin.com  Sun Jun  4 12:24:57 2000
From: kain at cableadmin.com (Scott)
Date: Sun, 4 Jun 2000 16:24:57 GMT
Subject: win32com and Excel
References: <3939D364.2EF22B13@zebrasoft.co.nz> <B_u_4.111323$E85.2192336@news1.rdc1.md.home.com>
Message-ID: <slrn8jl0rq.18e0.kain@cableadmin.com>

As a side note, I also recommend this book.  I really wouldn't even try
any python-com programming without reading it first.  If you can't afford
to buy it, it seems you can view it at http://www.netlibrary.com .  It's
a neat site that allows you to "check out" books for a period of time 
(usually 24 hours at a time) and read them online.  The only drawback is
they only allow so many "check outs" of each book so it may or may not
be available to read.

-Scott

PS.
Here is an exact URL:
http://www.netlibrary.com/SUMMARY.ASP?EV=1239265&ID=24666&advquery=

and chapter 9 is the main Excel chapter.

In article <B_u_4.111323$E85.2192336 at news1.rdc1.md.home.com>, 
Eric Hagemann wrote:
>Matthew,
>    Read Hammonds "Programming Python in Win32".  There is a very good
>headstart there with an example of a class that eases the interaction
>w/Excel via COM.  Next get "Writing Excel Macros" by O'Reilly.  It is all
>about how to program Excel via Visual Basic but translates very well into
>Python COM.
>
>Cheers
>Eric Hagemann
>
>"Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
>news:3939D364.2EF22B13 at zebrasoft.co.nz...
>> hi everyone,
>>
>> A couple o0f questions:-
>> 1)
>> can you access an excel (.xls) file via com without having excel loaded
>> on that machine?
>> 2)
>> once you've done:-
>>  blah = win32com.client.Dispatch("Excel.Application")
>> can anyone give details of
>> a) howto open a particular s/s in the current directory
>> and
>> b) howto access particular cells in that particular s/s
>> and
>> c) howto close that s/s
>>
>> I have to admit that I'm hoping some kind soul will help before I'm
>> consigned to grappling with what looks like grooling documentation!
>> thanks alot matthew.
>
>


From matt at nightrealms.com  Wed Jun 28 06:14:55 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Wed, 28 Jun 2000 02:14:55 -0800
Subject: Requiring a certain version of interpreter?
Message-ID: <Lqj65.4362$h8.209230@news-west.usenetserver.com>

Is there any builtin feature, or common module, that handles
requiring at least a certain version of the Python interpreter
to be used for a script?  Perl has 'require' for this purpose.

Thanks in advance.





From matt at null.net  Mon Jun 19 21:46:15 2000
From: matt at null.net (Matthew Schinckel)
Date: Mon, 19 Jun 2000 19:46:15 CST
Subject: Getting Password from User
References: <8ij9kr$qa5$1@bob.news.rcn.net> <B572AC79.517B%dgoodger@bigfoot.com>
Message-ID: <20000619194617.127655@210.9.22.17>

on 2000-06-18 15:49, Tom Santos (tomsantos at email.com) wrote:
> I'm trying to get a username and password from a user so that I can
> create an FTP connection. I'm using raw_input() to get the username
> but it's not acceptable to get their password. If I use raw_input() to
> get the password, the user's password is echoed to the screen. Is
> there any way to make it so that their password is not shown entirely,
> or characters like '*' are displayed instead of the characters in
> their password?

David Goodger <dgoodger at bigfoot.com> wrote:
> Use module getpass:
> 
> import getpass password = getpass.getpass()
> 
> If you still see the password echoed to the screen, then you're
> missing some functionality, probably termios. Check out the getpass.py
> code for the decision tree that is followed.

Hmm. I tried this under BeOS (python 1.5.2), termios is installed (as
termios.so, a shared lib), and password was echoed back.

Does this sound like a bug in my termios lib?

Matt.


From michael at stroeder.com  Wed Jun 14 03:51:02 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 14 Jun 2000 09:51:02 +0200
Subject: Web Session
References: <8i6lg3$htd$1@news.mty.itesm.mx>
Message-ID: <39473966.4EAA1C07@stroeder.com>

Dirk Xklbr wrote:
> 
> I need to make a cgi program that makes a session in web (like a email web
> session), where do I start from?, do I need cookies, how does it work?

Hmm, news:comp.infosystems.www.authoring.cgi might be a better start
for learning about different web session approaches: hidden fields,
random session-ids in hidden fields with session data stored on the
web server, session data stored in cookies, ...

Ciao, Michael.


From Wolfgang.Grafen at marconicomms.com  Fri Jun 16 03:15:39 2000
From: Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen)
Date: Fri, 16 Jun 2000 09:15:39 +0200
Subject: How does python invoke external C function?
References: <39498799.D28EDD46@esec.com.au>
Message-ID: <3949D41B.714D258E@marconicomms.com>

Sam Wun wrote:

> Hi,
>
> I would like Python call an external C function ( ie. char
> *getString()),
> Does anyone knows how to do that?
>
> Thanks
> Sam.

I would recommend SWIG for a first step. You don't have to learn a lot
about extension
modules with SWIG. Often it is easy like this (for sun solaris 2.5):

swig -python -module a a.c
gcc -O02 -DSWIG -c -fpic a.c a_wrap.c \
-I/user/freepool/sparc-sun-solaris2.5.1/lib/python/python1.52/include/python1.5/

ld -G a.o a_wrap.o -o a.so

Then in Python you can use it like it is.

Python 1.5.2 (#33, May 17 2000, 17:03:52)  [GCC 2.7.2] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import a
>>> dir(a)
['__doc__', '__file__', '__name__', 'hex2bin', 'hex2bin_char',
'hex2bin_expand']
>>>

SWIG has a very good documentation. You can build extensions for PERL
and TCL
the same way

http://www.swig.org

There is also another Corba-like Package called ILU. Probably SWIG is
more easy to use while
ILU might be more powerful.

ftp://ftp.parc.xerox.com/pub/ilu/ilu.html

Regards

Wolfgang



From pj at sgi.com  Mon Jun 19 15:49:07 2000
From: pj at sgi.com (Paul Jackson)
Date: 19 Jun 2000 19:49:07 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <uitv9qb72.fsf@ctwd0143.fitlinxx.com> <slrn8kobd5.1vs.hzhu@mars.localdomain> <04b401bfd960$7a9f95f0$6401a8c0@home>
Message-ID: <8iltfj$hm9u$1@fido.engr.sgi.com>

|> 'Tabs will be replaced with %d spaces\n Are you sure
|> Y/N:'%tabSize)

But a tab isn't 8 spaces, but rather the number of
spaces needed to get to the next multiple of 8 column,
which might be less than 8 spaces (s/8/tabSize/g, if
you must).

The expand/unexpand solution presented in another reply
look more promising, in my view.
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From emile at fenx.com  Thu Jun 15 11:04:49 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 15 Jun 2000 08:04:49 -0700
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <85em60nozb.fsf@one.net.au> <8i8o8b$83i$0@216.39.141.36> <crigkss2m7ick2imqdctubg5d4lqjpvtd1@4ax.com> <8iar91$gt6$0@216.39.141.36>
Message-ID: <133301bfd6db$15221400$1906a8c0@fc.fenx.com>

I've still got a CP/M based Otrona Attache
around here somewhere.  My first portable.


Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: Dana Booth <dana at mmi.oz.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, June 15, 2000 8:04 AM
Subject: Re: Does Python help with the no-Unix handicap?


> Dennis Lee Bieber <wlfraed at ix.netcom.com> wrote:
>
> >> TRS-DOS or NEW-DOS? :)
>
> DLB: The LS TRSDOS 6, or the Tandy original?
>
> Good old Tandy/Radio Shack original that came with my
TRS-80 III, complete
> with 48k ram. :)
>
> Ever try out CP/M on those old Osborne's or Kaypro's?
>
> --
> -----
> Dana Booth <dana at mmi.oz.net>
> Tacoma, Wa., USA
>
> key at pgpkeys.mit.edu:11371
> --
> http://www.python.org/mailman/listinfo/python-list
>




From johngrayson at home.com  Tue Jun  6 12:20:59 2000
From: johngrayson at home.com (John Grayson)
Date: Tue, 06 Jun 2000 16:20:59 GMT
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net> <8hj3g1$93l$1@gxsn.com>
Message-ID: <8hj8cl$6pv$1@nnrp1.deja.com>


> Of course if you don't know Tkinter then I recommend purchasing a book
- in
> Python and Tkinter programming by John Grayson there is a chapter on
graphs.
>

Don't tell my publisher I told you, but you can get the chapter
on graphs and charts in PDF format here:

http://www.manning.com/Grayson/chapt11.pdf

Of course, I hope you'll want to buy the book for the rest...  ;-)

    John


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


From MarkH at ActiveState.com  Fri Jun 23 19:24:44 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 23 Jun 2000 23:24:44 GMT
Subject: Python COM does not seem to like me.
References: <8j0ddv$fqb$1@nnrp1.deja.com>
Message-ID: <0lS45.2745$Sn2.11629@news-server.bigpond.net.au>

> First off, is there something I dont know about the import statement?
> Using it, it says that it cant find a module thats is the same
> directory as the .asp file.  I checked the file name and everything.

The import statement uses the PythonPath (ie, sys.path as shown from
inside Python).  The Python script doesnt really know the path it is being
executed in.

> The error is:
> TypeError: cant multiply sequence with non-int
>
> The statment is:
> CM1 = roll[1] * roll[2]
>
> I am pretty much positive roll[1] and roll[2] are ints.

Im certain they are not!  At least one of them is a "sequence" (ie, a list
or tuple), as the error message says.

>  As I said,
> this statement worked great in interpreter.

If it did, then the code that sets up 'roll' must be the problem ;-)

Mark.





From kpmurphy at my-deja.com  Wed Jun 28 12:26:29 2000
From: kpmurphy at my-deja.com (kpmurphy at my-deja.com)
Date: Wed, 28 Jun 2000 16:26:29 GMT
Subject: what is lambda?
Message-ID: <8jd8uu$gnh$1@nnrp1.deja.com>

I am a beginner at python and I would like to know what lambda does.
Could someone post an in-your-own-words definition and possibly an
example?  Thanks,

Keith


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


From harryt at gte.net  Fri Jun  2 02:10:15 2000
From: harryt at gte.net (Harry Tanovich)
Date: Fri, 02 Jun 2000 06:10:15 GMT
Subject: Copyright changes with move from CNRI to BeOpen?
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h7blm$j1b$1@slb2.atl.mindspring.net>
Message-ID: <39375113.5AF5BC0B@gte.net>

Will the copyright move to a more GPL style thing now?

Andrew Dalke wrote:

> t_e_sanders at my-deja.com wrote:
> >Do the terms of Python's distribution license require me to allow other
> >users to copy and distribute my software? I would guess not, but
> >inclusion of this notice in my package explicitly states so. There's
> >obviously something I'm missing,
>
> The key word is "use".
>
> The copyright only applies to the Python code.  You are free to add
> your own code to Python, which is under your own copyright.  The
> clause you quoted allows you to use the Python code with your own
> code, regardless of the license applied to your own code.
>
> The GPL, which you are probably thinking of, does not allow similarly
> unfettered use of their code.  Packages compiled with GPL code
> restrict their use to only GPL-compatible licenses.
>
> The inclusion of the statement in code you distribute only applies to
> Python code.  Someone could copy that portion of the source from your
> program and use it elsewhere.  However, since it's already available
> from python.org, there nothing new revealed.
>
> But remember, I-am-not-a-lawyer.
>
>                     Andrew
>                     dalke at acm.org



From andre at beta.telenordia.se  Thu Jun  1 08:38:15 2000
From: andre at beta.telenordia.se (André Dahlqvist)
Date: Thu, 01 Jun 2000 13:38:15 +0100
Subject: Marking hyperlinks in a Text widget
References: <8h04jo$lgh$1@zingo.tninet.se> <oq7lcc41zr.fsf@titan.progiciels-bpi.ca> <8h0kn5$6pv$1@cubacola.tninet.se> <slrn8j94er.ebv.cjc26@localhost.mindriot.net> <8h4d2e$ap4$1@cubacola.tninet.se> <slrn8jblag.jua.cjc26@localhost.mindriot.net>
Message-ID: <8h5hvi$djt$1@cubacola.tninet.se>

> That's actually a command-line argument of Unix versions of Netscape.  I
> don't know if it would work on Windows.  If you're using Unix, try
> typing the following at a shell prompt (with Netscape already running):
> 
> % netscape -remote 'openURL(www.python.org)'

Yeah, I'm on UNIX. I haven't really messed that much with the URL
opening yet, but this is how I have thought it should work:

common_browsers = ['mozilla', 'netscape', 'lynx', 'w3m']

for browser in common_browsers:
	if not os.system(browser + URL)
		# found browser
		break

> It actually returns the entire tuple we stored as its tag, not just the
> first tag name.  Since the url was already entered as the second element
> of this tuple, we just do 
> 
> textw.tag_names(...)[1]
> 
> to retrieve it.

But the thing is that I haven't used your URL inserting, because
it won't work in my application. I did however use the same code for
tagging the URLs.

So for me doing this will only return the name of the tag for me:

URL = textwidget.tag_names("@%d, %d" % (event.x, event.y)

URL will be 'url' after this call. The last braces [1] that you use doesn't
apply to my code since I haven't used that insert code you did, right?

Even if I try [0] it doesn't work.

> Nope; you just need to know the coordinate of one of the characters
> within the url.  The character closest to where the mouse was clicked is
> the most convenient one to use :)

Now this is what is driving me nuts, it should be so simple but I just
can't get it to work! If you want I can send you my code so you can
look at it.


> The callback function takes an argument (event), and you can get 
> the x and y positions where the mouse was clicked using event.x and
> event.y. So if you want to use tag_ranges, you could just look for the
> range which contains (event.x, event.y).

Aha, that I can do.

// Andr?





From digitig at cix.co.uk  Sun Jun  4 08:53:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 4 Jun 2000 13:53 +0100 (BST)
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <BXh_4.4616$Hz.35258@news-server.bigpond.net.au>
Message-ID: <memo.20000604135334.59489A@digitig.compulink.co.uk>

In article <BXh_4.4616$Hz.35258 at news-server.bigpond.net.au>, 
neilh at scintilla.org (Neil Hodgson) wrote:

> > >>> class Foo:
> > ...    pass
> > ...
> > >>> def bar(foo = None,
> > ...    # /lots/ of other optional parameters
> > ...    ):
> > ...    # some implementation
> > ...
> > >>> baz = Foo()
> > >>> bar(foo = baz)
> 
>    Wouldn't 'foo' be better used to describe its role rather than its 
> type.
> 
>    bar(output = baz)

Depends. For instance, I'm working on something at the moment that uses a 
flyweight character class, and being a flyweight the context is separated 
from the class itself. So, for instance, the method that renders needs to 
be passed a Tkinter.Canvas on which to render. It makes sense to me to 
define the method as:
	def render(canvas, ...)
If I call it "output", "context" or some such then it's not clear what to 
pass; calling it "canvas" makes it self-documenting. Using "output_canvas" 
or some such is more verbose and IMHO adds nothing at all to clarity -- if 
anything the verbosity detracts from the clarity. I /used/ to use the 
convention "a_canvas", but I found it to be nowhere near as clear & 
maintainable.


From Johannes.Nix at mail.uni-oldenburg.de  Fri Jun  2 15:41:12 2000
From: Johannes.Nix at mail.uni-oldenburg.de (Johannes Nix)
Date: 02 Jun 2000 21:41:12 +0200
Subject: mixed solution: unicase (unique allowed case) (was: Re: Case-sensitivity: )
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <20000522183552.A29224@mudge> <8gcoij$60n$1@news-int.gatech.edu>
Message-ID: <8h92ko$5e1@news.Informatik.Uni-Oldenburg.DE>

Andrew Henshaw <andrew.henshaw at gtri.gatech.edu> writes:

> require it.  In my opinion, that's the way programming should work.  I would
> like to read a program that followed case usage for readability; but, I
> don't want an interpreter to barf when case is ignored.  
> 

why don't turn it this way to get advantages of both alternatives:

- the language is case sensitive.

- an assignment to the same name with different case in the same
module is a Case Error exception.


Then, you would have to use matching case names (You can use the case
for readability or personal taste, or carrying extra information. You
can also do it all-lowercase or all-uppercase).

But you would _not_ allow any programmer to distinguish names only by
case (which in my eyes is not the best style, and most probably a
mistake which leads to a name error anyway). A small percentage of
code would have to be fixed up. (And this could be done
automatically).

A beginner couldn't ignore any more the mistake to refer to the same
name with different case, meaning the same variable. Rapid, clear and
consistent feedback is the best way to learn.

Err, if it becomes overly annoying, you could turn off the exception
with a switch to get only a less annoying warning. Semantics doesn't
change, but you will have to clean up your program afterwards if you
want to run it without switch. Also, you could automate this step
perfectly in a beginners development environment because you won't
have case clashes, e.g. you substitute all variations of a name with
the casing of the first occurrences.

As a plus, every beginner will be eager to show expertise to have
intelligently cased code!

Let's play around with the idea!

Johannes



From aiglauer at iname.com  Mon Jun  5 15:52:20 2000
From: aiglauer at iname.com (Alexis Iglauer)
Date: Mon, 5 Jun 2000 21:52:20 +0200
Subject: How to prevent a web client from timing out
References: <8hgfn8$3uq$1@nnrp1.deja.com> <393BCA8B.FC062262@rubic.com>
Message-ID: <8hh066$vor$10$1@news.t-online.com>

The job is different every time, so a cron solution wouldn't work.  The
problem is that when the client times out the server kills the python
process doing the number crunching, so it can't mail the result to the user
either (that is a good idea though, hadn't thought of that).

So how do I stop apache from killing the process?  How well does xmlrpc work
in the real world?

Thanx
Alexis

The problem I seem to have is that
"Jeff Bauer" <jbauer at rubic.com> wrote in message
news:393BCA8B.FC062262 at rubic.com...
> Alex,
>
> One solution would be to run your number-crunching
> in a separate process (e.g. xmlrpc server) that
> your cgi script accesses.
>
> Jeff Bauer
> Rubicon Research
>
> aiglauer wrote:
> > I have written a number-crunching app in python. which is being
> > accessed on our intranet.  All pages generated on-the-fly in python.
> > Some of the calculations can take a while though (calc time anything
> > from 5secs to 30 minutes) and I get the client timing out and the
> > server will then kill the calculation process.
> >
> > Can I rig it so that the calculation process will still complete and
> > write its output to a file so that I can use it later?  (Writing to a
> > file is not a prob - preventing the thing getting slapped is).
>




From mhammond at skippinet.com.au  Sun Jun 18 20:30:42 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 19 Jun 2000 00:30:42 GMT
Subject: Optimizing Object Instantiation (was Re: How Fast Does Python Run?)
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net> <39499C91.3A1586BE@sightreader.com> <8if21t$4u5$1@slb1.atl.mindspring.net> <20000617120003.S26436@xs4all.nl> <see-0F7889.18054217062000@news.dnai.com> <E0Y25.1541$qz4.13200@news-server.bigpond.net.au> <see-262267.15582418062000@news.dnai.com>
Message-ID: <SQd35.2175$qz4.17926@news-server.bigpond.net.au>

"Sam Penrose" <see at message.body> wrote in message
news:see-262267.15582418062000 at news.dnai.com...

> The code in question is based around nested __init__ calls:
> a main working class which is based on an abstract class
> (2 __inits__) and which calls a wrapper class (3) around the
> MySQLdb module (4).
>
> I'm starting to think for other reasons that I've been
> overcoding, but it would be nice to know whether the
> performance issue has any prospects of being addressed.

Not really!  The problem is simply that you execute a fair bit of Python
code, including a number of function calls (which are expensive) for each
object instantiation.

So, while we all try very hard to optimize Python, the fact remains that
you are running this code for each object, and it can never be faster than
the Python interpreter can chew through it.

Use the standard Python profiler - this should help you identify how you
can make the code that must execute as fast as possible - then you can
decide an appropriate strategy.

> OTOH, it's amazing how much faster $2,500 worth of current
> hardware is than the system we spent $2,500 on two years ago;
> perhaps the point will prove moot.

Too true - but the problem is Python is _too_ quick to develop - by the
time your app is finished, speeds havent moved too much.  OTOH, choosing
something like C++ often means the project takes far far longer, meaning
you do indeed see hardware speeds improve over the life of the project.
So ironically, it seems that you get the hardware speed increase for
"free" when using the languages that need it least :-)

Mark.




From ralph at inputplus.demon.co.uk  Sat Jun 10 07:43:46 2000
From: ralph at inputplus.demon.co.uk (Ralph Corderoy)
Date: 10 Jun 2000 12:43:46 +0100
Subject: RFC822 compliance in modern mail clients?
References: <86h05.51947$WS3.494817@typhoon.we.rr.com>
Message-ID: <8ht9li$74i$1@inputplus.demon.co.uk>

Hi,

> Do any clients pay attention to headers  such as Resent-etc? Anytime
> anyone's ever forwarded me something, its always seemed to come from
> them :)

Sure they do.  I've used MH, and now nmh, for years and if I can either
forward mail messages onto others with the `forw' command that lets me
insert my own message at the top or distribute on a message with the
`dist' command which just takes a list of email addresses for the
resent-to, resent-cc, resent-from headers;  I'm not allowed to add any
text.

Broken email clients, including Microsoft's, ignore the resent-*
headers so it looks to the recipient as if the original sender sent it
to them.  They then get confused because the email is addressed to me,
not them, so they send me a copy :-)

So, don't be a Microsoft;  follow the RFC.


Ralph.



From sdhyok at email.unc.edu  Fri Jun 16 14:18:57 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Fri, 16 Jun 2000 14:18:57 -0400
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <8ia4tv$e19$1@nnrp1.deja.com>
Message-ID: <8idr2h$fvi$2@news2.isis.unc.edu>

How can I get the DCI module?

Peter

"Fred Gansevles" <gansevle at cs.utwente.nl> wrote in message
news:8ia4tv$e19$1 at nnrp1.deja.com...
> In article <8i8ujt$hop$1 at news2.isis.unc.edu>,
>   "Daehyok Shin" <sdhyok at email.unc.edu> wrote:
> > How can I prevent illegal definition of a variable to an object?
> > For instance,
> >
> > class A:
> >     __init__(self):
> >             self.x = 1
> >
> >     __setattr__(self, name, value):
> >             if name == "x":
> >                     self.x = value
> >             else:
> >                 raise AttributeException
> >
> >     __getattr__(self, name):
> >             if name == "x":
> >                 return self.x
> >             else:
> >                 raise AttributeException
> >
> > >>> a = A()
> > >>> a.x = 10
> > >>> a.y = 20 # I want to raise an AttributeException.
> >
> > Peter
> >
> >
>
> Some time ago I posted my DCI module. This module also has a attribute
> module wich can, among other atribute-related stuff, do this for you.
>
> For instance,
> >>> import dci.attr
> >>> class A:
> ...     def __init__(self):
> ...         dci.attr.init(self) # initialize the attribute mechanism
> ...         dci.attr.declare(self, 'x')
> ...
> >>>
> >>> a = A()
> >>> a.x = 10
> >>> a.y = 20
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "dci/attr.py",
> line 52, in __setattr__
>     raise AttributeError, "Undeclared attribute: %s" % `name`
> AttributeError: Undeclared attribute: 'y'
>
> --
> -----------------------------------------------------------------------
> ----          Linux/Windows-NT/IntraNetware Administrator          ----
> -- Whenever it sounds simple, I've probably missed something! ... Me --
> -----------------------------------------------------------------------
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.






From neilh at scintilla.org  Fri Jun 16 23:29:17 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 17 Jun 2000 03:29:17 GMT
Subject: Ret status technique popen, popen2 (NT)
References: <3949a528.1161643@news.pipeline.com> <2Kq25.372$qz4.3272@news-server.bigpond.net.au>
Message-ID: <hgC25.714$qz4.6314@news-server.bigpond.net.au>

    Thanks to some comments from David Bolen, piper has been made more
robust. I've also fiddled with the code and added some comments.

http://www.scintilla.org/piper.zip

    Neil





From pinard at iro.umontreal.ca  Tue Jun  6 14:46:05 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 14:46:05 -0400
Subject: __getattr__ and __setattr__ troubles
In-Reply-To: "Adam Ruth"'s message of "Wed, 24 May 2000 08:15:33 -0600"
References: <8gehq1$kq6$1@news1.xs4all.nl> <8ggo6m$2v6h$1@news.aros.net>
Message-ID: <oq1z2ad4wi.fsf@titan.progiciels-bpi.ca>

"Adam Ruth" <aruth at intercation.com> writes:

> The __str__ and __coerce__, etc, will all be sent to __getattr__ because
> they are *not* defined in the class.  If they were, then __getattr__
> would never be called.

But would not this get fairly tedious?  For example:

---------------------------------------------------------------------->
class See: pass
print See()
----------------------------------------------------------------------<

actually yields:

---------------------------------------------------------------------->
pinard at titan:~ $ python see.py
<__main__.See instance at 81898b8>
----------------------------------------------------------------------<

which is convenient enough for me.  While a simple:

---------------------------------------------------------------------->
class See:

    def __getattr__(self, key):
        print '**', key
        return None

print See()
----------------------------------------------------------------------<

will yield:

---------------------------------------------------------------------->
pinard at titan:~ $ python see.py
 ** __str__
Traceback (innermost last):
  File "see.py", line 7, in ?
    print See()
TypeError: call of non-function (type None)
----------------------------------------------------------------------<

* By the way, and a bit unrelated, why the space before the two asterisks?

* The diagnostic would be rather cryptic, if the program was not so short.
  Usability studies could be driven to show that the diagnostic should rather
  be changed to something more informative :-).

* May I least get the default string representation, which was satisfying?

* So, as soon as I use `__getattr__', I suddenly get a lot of other things to
  define as well?  This would make `__getattr__' less attractive.

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





From ansible at typhoon.xnet.com  Wed Jun 21 10:21:04 2000
From: ansible at typhoon.xnet.com (James Graves)
Date: 21 Jun 2000 14:21:04 GMT
Subject: Transcendentals and Encryption (was Re: Nth digit of PI)
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <394a5ed7.83928743@10.1.1.28> <8idoj0$lea$1@nntp.itservices.ubc.ca> <394a680a.86283939@10.1.1.28>
Message-ID: <8iqj0f$kk5$1@flood.xnet.com>

Tim Dixon <tdixon.no at spam.fwi.com> wrote:
>On 16 Jun 2000 17:36:32 GMT, israel at math.ubc.ca (Robert Israel) wrote:
>
>>Tim Dixon <tdixon.no at spam.fwi.com> wrote:
>>>On Fri, 16 Jun 2000 16:33:59 GMT, Courageous <jkraska1 at san.rr.com>
>>>wrote:
>>
>>
>>>>> Suppose, however, that you could get digit 'n' of pi directly (or e,
>>>>> or any other inifinte, nonrepeating sequence).  All I have to do is
>>>>> communicate which digit of pi the message key starts at, and I can
>>>>> generate the rest of the key.
>>>>
>>>>This is a collosally bad idea, IMO.
>>
>>>Just for curiosity, why?
>>
>>Because the real key is the number of the digit that you start at, and
>>that's pitifully small by cryptographic standards.  The time to generate
>>the n'th digit is not much less than the time to generate all the digits
>>up to the n'th, given that you have lots of memory available.  So the
>>cryptanalyst with a computer somewhat bigger than yours could try all
>>the keys you could be using.  
>>
>
>Right, but in the hypothetical example where I can get *any* digit of
>pi, that could be any of an infinite number of starting points.

Using transcendentals for encryption is an interesting idea, one that's
been banging around in my head for a while.

I am under the impression that there are an infinite number of
transcendental numbers.  If there was some way to arbitrarily pick one
and the starting point, that might be harder to crack.  Then again, if
the black hats have your encryption algorithm (which is good to assume)
it wouldn't be any more secure than spending all the key bits on the
starting position inside of Pi.

This scheme still has the weaknesses of conventional stream
cyphers, in that it can be brute-forced.  A cypher can give you N years
of protection for your data (based on estimations of future computing
power).  But a one-time-pad is _forever_ if the pad was generated
correctly and used only once.

James Graves
--
____________________________________________________________________________
Free Speech Under Attack in the UK:	    http://www.liberty.org.uk/cacib/


From olczyk at interaccess.com  Tue Jun 13 00:12:19 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Tue, 13 Jun 2000 00:12:19 -0400
Subject: Testing for a file in python?
References: <Pine.GSO.4.10.10006122010090.15238-100000@sundial>
Message-ID: <642A954DD517D411B20C00508BCF23B001296888@mail.sauder.com>

On Mon, 12 Jun 2000 20:10:48 +0300 (IDT), Moshe Zadka
<moshez at math.huji.ac.il> wrote:

>> How would you go about testing if a certain file exists in python?
>> In most scripting languages you have the -f test:
>
>I don't know what you mean by "most".
For all those who emailed me ( especially the nasty ones ) suggesting
that I'm slamming Python in favor of Perl ( which I know well enough
to know, I don't like it )  by most I mean:

sh,csh,bash,tcsh,zsh ( and OK Perl )

ones that I seem to remember it being in but am not quite sure:

ksh, awk.

Ones that I'm sure don't do it
Python, sed ( and I'm not sure about sed, it's just not likely to be
there ).



From jbower at lims.com  Wed Jun 14 17:42:06 2000
From: jbower at lims.com (Jim Bower)
Date: Wed, 14 Jun 2000 21:42:06 GMT
Subject: Python and Windows Scripting Host
References: <9Ov15.1411$vM.56343@typhoon1.ba-dsg.net> <39y15.839$Zh1.63369@news.easynews.com> <8i8mbb$c3f$1@nnrp1.deja.com>
Message-ID: <O_S15.205$3y1.4771@typhoon1.ba-dsg.net>

Here is what I am trying to do:

import win32com

WScript.Echo("Hello")

Here is what is happening:

Traceback (innermost last):
  File "d:\Python\Pythonwin\pywin\framework\scriptutils.py", line 310, in
RunScript
    exec codeObject in __main__.__dict__
  File "C:\WINNT\Profiles\Administrator\Desktop\test.py", line 3, in ?
    WScript.Echo("Hello")
NameError: WScript

Thanks,
Jim Bower jbower at lims.com



"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:8i8mbb$c3f$1 at nnrp1.deja.com...
> In article <39y15.839$Zh1.63369 at news.easynews.com>,
>   "Roger Upole" <rupole at compaq.net> wrote:
> > The Wscript object is not referenced directly as a Python variable.
> > You create it via win32com.client.Dispatch:
>
> Im afraid that is not correct.  When Python is running under the
> Windows Scripting Host, the WScript variable is automatic.
>
>
> > "Jim Bower" <jbower at lims.com> wrote in message
> > news:9Ov15.1411$vM.56343 at typhoon1.ba-dsg.net...
> ...
>
> > > However, I am getting "NameError: wscript" anywhere the wscript
>
> Python is a case sensitive language.  You need to use "WScript"
>
> Mark.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From akuchlin at mems-exchange.org  Wed Jun  7 21:57:59 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 07 Jun 2000 21:57:59 -0400
Subject: Dutch.comp.python.org 
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain> <8hmmvu$ets$2@newshost.accu.uu.nl>
Message-ID: <3dr9a9hr2w.fsf_-_@newcnri.cnri.reston.va.us>

m.faassen at vet.uu.nl (Martijn Faassen) writes:
> Python's virtually perfect, but it does have a problem: those Dutch
> folks.

Still, at least they're not Austrian.  [1]

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
I have always had severe problems with Austrians. ... Musical, churchy,
uptight... nice legs [2]... hypocritical... authoritarian... always
insist their dustbins are very clean.
  -- Philip, in Peter Greenaway's _8 1/2 Women_ (1999)

[1] While we're drifting off topic... I'm going to be in Berkeley
    early next week.  Any Python or Linux-related goings-on around
    there Sunday or Monday?

[2] My dad is from a town near Salzburg whose name eludes me at the
    moment.  It *does* explain why I wear shorts a good deal, doesn't
    it?   


From nobody at nowhere.nohow  Sat Jun 17 12:56:29 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 17 Jun 2000 16:56:29 GMT
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com> <slrn8kn881.sdl.nobody@isis.visi.com> <394BA666.E5F3B19B@uniserve.com>
Message-ID: <15O25.1379$i06.308670@ptah.visi.com>

On Sat, 17 Jun 2000 09:25:10 -0700, Bob van der Poel <bvdpoel at uniserve.com> wrote:

>> Yes, I know.  The call to BitmapImage() works fine, the error occurs when I
>> try to use the bitmap in a button like:
>> 
>>    Button(root, bitmap='queen')
>> 
>> This is where the error occurs -- it doesn't matter if I use the name I
>> specified in the call to BitmapImage or let Tk generate one and retrieve it
>> with str().  What I can't figure out is how to use bobpic as the bitmap for
>> a Button.
>>
>
>More from my snippet:
>
>	l = Label(w, image=bobpic, borderwidth=4, relief='sunken')

Yes, using it as the widget's image works, but I'm trying to
use it as the bitmap for the widget.

-- 
Grant Edwards                   grante             Yow!  My CODE of ETHICS
                                  at               is vacationing at famed
                               visi.com            SCHROON LAKE in upstate
                                                   New York!!


From just at letterror.com  Sat Jun 10 05:40:04 2000
From: just at letterror.com (Just van Rossum)
Date: Sat, 10 Jun 2000 10:40:04 +0100
Subject: Stackless Python for the Mac?
In-Reply-To: <20000609213814.01734E2673@oratrix.oratrix.nl>
References: Message by Jeff Senn <senn@maya.com> ,	     09 Jun 2000
 15:10:52 -0400 , <zoouk6v7.fsf_-_@SNIPE.maya.com>
Message-ID: <l03102800b567b5427839@[193.78.237.152]>

At 11:38 PM +0200 09-06-2000, Jack Jansen wrote:
>There may be more complications, as the MacPython mainloop is a bit
>funny because of the frequent calls to WaitMextEvent to give cpu-time
>to other programs (MacOS is still in the basis non-preemptively
>scheduled).
>
>I've cc'd Just on this message, because he did a port of microthreads
>to macpython, and I seem to vaguely remember that this was related to
>stackless python. But I could be all wrong, i didn't really have the
>time to follow what happened. Just?

Erm, I don't think Stackless didn't involve any changes that affect what
you describe above. I have used the Stackless PythonCore for quite a while
now instead of the regular 1.5.2, (without using stackless' features that
much actually) and I haven't had a single problem...

Just





From adams_s at lab.eng.usyd.edu.au  Mon Jun 12 02:11:33 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Mon, 12 Jun 2000 16:11:33 +1000
Subject: Done (sortof) - (was Newbie: displaying binary numbers?)
References: <8i1q49$2nk3$1@news1.wire.net.au>
Message-ID: <8i1ujs$2to7$1@news1.wire.net.au>

there is probably a cleaner method in the modules already, but this seems to
work fine.

I converted any numbers to hex. then turned the hex into a string, for each
letter in the string, there is a corresponding 4 bit sequence.
####

hexvals={'0':'0000',  '1':'0001', '2':'0010',  '3':'0011',  '4':'0100',
'5':'0101',  '6':'0110',  '7':'0111',
'8':'1000',  '9':'1001',  'a':'1010',  'b':'1011',  'c':'1100',  'd':1101,
'e':'1110',  'f':'1111'}

def int2binary(num):
                     "int2binary( num, bits ): display 'num' as a binary
with 'bits' number of bits"
                     s=str( hex( num ) )
                     s=s[2:]
                     b=''
                     for i in s:
                                b=b+hexvals[i]
                     return b

#is this right, it seems to work.

now, whats the best way to go about converting back to a normal base 10
representation?
I'd thought of just quickly breaking the string of 1's and 0's into strings
4 chars long, then using another lookup table.




Steven Adams <adams_s at lab.eng.usyd.edu.au> wrote in message
news:8i1q49$2nk3$1 at news1.wire.net.au...
> Hi all,
>
> I'm trying to display integers as binary numbers, I know there are the
calls
> for making hexadecimal and octal strings, but couldn't find anything for
> binary?
>
> are there any modules, or operations not in the docs?
>
> all I need is to convert a number to a binary with n bits.
>
> thanks, Steven
>
>




From wtanksle at dolphin.openprojects.net  Sat Jun  3 14:42:27 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sat, 03 Jun 2000 18:42:27 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl>
Message-ID: <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net>

On 30 May 2000 10:59:24 GMT, Martijn Faassen wrote:
>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>>4. the point, that bothers me most about case insensitivity, is, that
>>>consistency in the spelling of identifiers will be lost. That means, because
>>>case doesn't count any more, why should a programmer enforce any kind of
>>>rules in the use of type case ? Why should one even consider using the same
>>>case for the same identifier in different places ? What would the right
>>>typing be anyway, if Python doesn't check it anymore ?

>> Your rhetorical questions answer themselves.  Why should a programmer
>> enforce these things?  Because it's stupid to NOT enforce them.  So
>> enforce them already!

>That's nice to say, but you don't agree with that yourself if you think
>Python's use of indentation is a good thing. It can help a lot if the
>language makes it impossible to be inconsistent. Unless you think the
>_only_ advantage of Python's indentation strategy is the lesser line 
>count.

This is a classical example of false dichotomy ;-).  I liked Python's
indentation not because it enforced consistency, but rather because it
makes code look good.  In fact, it could be argued that Python does _not_
enforce consistency; the programmer is free to choose any indentation for
each block, so long as each block can be parsed (which does require
internal consistency).

More recently I've come to realise that the reason Python indentation
looks good is that Python, like most modern languages, is a
tree-structured language, and so is best represented in two dimensions.
Most modern languages, however, attempt to represent themselves in one
dimension: an ASCII stream which ignores lines and whitespace.

So I don't really care about languages enforcing consistency; if I need
it, I'll enforce it myself.  Take a lesson from the people complaining
about Python's indentation: one of the leading false accusations is that
it's inflexible, that it won't allow people to choose thier own
indentation.

>Martijn

-- 
-William "Billy" Tanksley


From jcm at bigskytel.com  Sat Jun 10 06:25:48 2000
From: jcm at bigskytel.com (David Porter)
Date: Sat, 10 Jun 2000 04:25:48 -0600
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
In-Reply-To: <87em66owf6.fsf@globalreach.net>; from pgf@globalreach.net on Sat, Jun 10, 2000 at 01:57:33AM -0500
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net>
Message-ID: <20000610042548.A9887@bigskytel.com>

* Phil Fraering <pgf at globalreach.net>:
> Guido van Rossum <guido at python.org> writes:
> 
> > I am a very case-sensitive person myself: it bothers me when people
> > don't start their sentences with capital letters or when they refer to
> > themselves as "i" in lowercase -- or when people refer to Python as
> > "python" or "PYTHON".  (Come to think of it, it bothered me when you
> > wrote "german" instead of "German" :-).
> 
> Personally, I simply think that lowercase is more readable than uppercase,
> but I don't really know the reason why. 

maybe because there are more spacial cues to recognize in lower case than in
UPPER CASE, which looks MONOTONOUS and RUDE. lower case: each letter is well
defined and arranged. UPPER CASE: THE LETTERS ARE BLOCKY AND CRUDELY
DEFINED. PLUS THEY ARE BUNCHED TOGETHER.

  david.
  
-- 
Tac tic. -- e.e.cummings in a blender



From rbill at digisprings.com  Tue Jun 13 16:13:48 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Tue, 13 Jun 2000 14:13:48 -0600 (CST)
Subject: no SyntaxError?
Message-ID: <Pine.LNX.4.10.10006131359020.17071-100000@localhost.localdomain>

Just curious- I didn't find any references to this in a quick search of 
the faq and elsewhere...

This snippet generates "SyntaxError: invalid syntax"

def test():
    else: print "whoops"
test()

However, the following snippet generates "0 1 whoops" when it seems like
it should also generate "SyntaxError: invalid syntax" because there is no 
if before the else. 
 
def test1():
    for i in range(2):
        print i,
    else:
        print "whoops"

test1()

Why is it that I do not get a "SyntaxError" in the second example?

rb




From cut_me_out at hotmail.com  Fri Jun 16 21:13:00 2000
From: cut_me_out at hotmail.com (Alex)
Date: 16 Jun 2000 21:13:00 -0400
Subject: Are list mutations thread-safe?
Message-ID: <etd8zw5qfdv.fsf@w20-575-109.mit.edu>

Hi.  Can two threads get in each others' way if they simultaneously try
to mutate a list?  How could I make a reliable test to answer a question
like this?

Alex.


From cut_me_out at hotmail.com  Mon Jun 26 23:06:42 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Jun 2000 23:06:42 -0400
Subject: stopping threads
References: <3957C1DC.E7C77E2E@sec.noaa.gov> <8j8id2$v3h$1@nntp9.atl.mindspring.net> <etdg0q0hqvt.fsf@w20-575-109.mit.edu> <8j90s1$7vj$1@slb7.atl.mindspring.net>
Message-ID: <etdd7l3izzx.fsf@w20-575-109.mit.edu>

Oh, right.

Thanks, Aahz and Ken.

Alex.


From petersc at stud.ntnu.no  Sun Jun  4 07:12:16 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 13:12:16 +0200
Subject: ftplib storbinary method interception
References: <39399FBA.MD-1.4.4.angband@blueyonder.co.uk> <39399672.B35B7193@stud.ntnu.no>
Message-ID: <393A3990.1EB1F25B@stud.ntnu.no>

Oops, the counter might benefit from actually being
incremented ;-)

def mystorbinary(self, cmd, fp, blocksize):
  '''Store a file in binary mode and print stuff.'''
  self.voidcmd('TYPE I')
  conn = self.transfercmd(cmd)
  counter = 0
  while 1:
    print counter,"Bytes read yet."
    buf = fp.read(blocksize)
    if not buf: break
    conn.send(buf)
    counter = counter + blocksize
  conn.close()
  return self.voidresp()

excuse-or-execute-me-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From eviltofu at rocketmail.com  Sat Jun  3 13:58:47 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 03 Jun 2000 17:58:47 GMT
Subject: Erm... MacOS X?
References: <eviltofu-1AEA18.02454303062000@news.earthlink.net> <usenet-6DC18E.10333803062000@news.visi.com> <slrn8ji7hc.5b1.cjc26@localhost.mindriot.net>
Message-ID: <eviltofu-86F5A2.13584703062000@news.earthlink.net>

In article <slrn8ji7hc.5b1.cjc26 at localhost.mindriot.net>, 
cjc26 at nospam.cornell.edu (Cliff Crawford) wrote:

> * James Felix Black <usenet at inter-slice.com> menulis:
> | 
> | > Has anyone compiled Python for MacOS X?
> | 
> | Yes.  It builds right out of the box.  The only trap is if you build on 
> | an HFS+ as opposed to UFS volume; then, you have to be careful about 
> | case (HFS+ is case-insensitive but preserving.)
> 
> Specificially, if you use HFS+ and want it to "just work" when you run
> make, then you need to rename the Python directory to something else (I
> called it Python.d), then change Makefile.in, configure.in, and
> Python/Makefile.in to use the new directory name.  Then run autoconf,
> then configure, etc.

There is a man Page for Zlib and Gzip but I can't seem to find the 
location of the zlib.h file and the zlib library. Do you know where it 
is or do I have to download and compile it?


From dgoodger at bigfoot.com  Fri Jun 30 22:38:51 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Fri, 30 Jun 2000 22:38:51 -0400
Subject: ?Combination of a Dictionary and a List?
References: <395D33EE.75F1E36D@email.sps.mot.com>
Message-ID: <B582D1FB.5A1E%dgoodger@bigfoot.com>

on 2000-06-30 19:57, Norman Shelley
(Norman_Shelley-RRDN60 at email.sps.mot.com) wrote:

> Sometime  in the past I wrote a simple combination of a dictionary and a
> list, where __getitem__() would first look in a hidden dictionary and
> then if not found it would try looking in a hidden list.  I've seen this
> type of code out there somewhere also but I can not find my code or
> anyone elses. I guess I haven't figure our the "right" search criteria
> on www.python.org.  Any pointers?
> 

Many people have had such a need at some point (myself included; see my
signature below), but each with slightly different requirements. Do you want
a list that acts like a dictionary, or a dictionary that acts like a list,
and in what way?

Look for the c.l.p thread "indexed dictionaries?" of 2000-04-09, see
"seqdict" on Parnassus, and examine the "UserDict" and "UserList" modules in
the standard Python library.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From johnvert at my-deja.com  Fri Jun  9 20:11:14 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Sat, 10 Jun 2000 00:11:14 GMT
Subject: getting a URL
References: <8hop1b$c1b$1@nnrp1.deja.com> <BBF4FD9CA76E9DF9.BC74420340D17532.6B7363F9B8212DF1@lp.airnews.net>
Message-ID: <8hs12n$qd2$1@nnrp1.deja.com>

In article
<BBF4FD9CA76E9DF9.BC74420340D17532.6B7363F9B8212DF1 at lp.airnews.net>,
  Jeremy Hylton <jeremy at beopen.com> wrote:
> johnvert at my-deja.com writes:
>
> > But nothing is printed.  The problematic line seems to be:
> >
> > url = urllib.urlopen("http://www.server.org/foo.html")
> >
> > But the URL I gave it in the actual program is correct.  I tried a
few.
> > If I run this on the shell (it's meant to be a cgi) it raises
IOError
> > sayin 'network unreachable'.  What could be wrong?
>
> The network is unreachable <wink>.  The error message means that the
> operating system could not figure out how to deliver packets to the
> address you used.  Where you able to load the URL using your Web
> browser?  I would guess that your network connection had some problem,
> or that the end site was unreachable.

I gave it the URL of the server I was testing it on, which was
available.  And then I tried the URL they gave in the python.org in the
examples, and URLs of other sites which I know work like slashdot.org,
freshmeat.net, microsoft.com, netscape.com, etc.. same thing.  Is my
code at least correct?

Thanks,
--
John Vert
[insert witty computing joke here]


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


From thomas at xs4all.nl  Wed Jun 28 02:29:14 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 28 Jun 2000 06:29:14 GMT
Subject: Controlling an external program with Python
References: <3958C648.1D76@rd.francetelecom.fr> <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <8jc28s$8eb$0@216.39.151.169>
Message-ID: <slrn8lj6pa.l52.thomas@centurion.xs4all.nl>

>| In unix, AFAIK the only nonportable part is the *naming* of the pty
>| devices. That kind of decision is fairly trivial to make in
>| sh/awk/perl/python/tcl...
>
>The Berkeley pty, with the kind of naming scheme you allude to, is
>already supported in the "pty" module.  It's supposedly tested on
>Linux, and I'd expect it could work on FreeBSD/NetBSD/etc., and
>probably lots of others like Digital UNIX.

>But it's far from universal, and some of the alternatives do use
>special functions that would need to be supported in C.

>On the bright side, I am optimistic that the horrors of past
>pseudotty implementations are fading into oblivion and the
>survivors are fewer and cleaner.

I'd be interested in hearing about some of these pty-implementations. I know
about the SGI one, for which there is support in the sgi module, and I know
about openpty() (currently used by at least BSDI, FreeBSD and (glibc
2.1-based) Linux) -- in fact I submitted a patch to make the pty module use
'openpty()' and 'forkpty()' if it's available. And I know of the old
BSD/Linux way of 'just opening a pty' of course.

Are there any other 'mainstream' methods of opening a pty ? Could they be
implemented in terms of openpty() and/or forkpty() ? It would be pretty neat
if the 'emulation' could stay inside the posixmodule, without the need for X
different OS-specific modules and Y different function names for each method
of getting a pty ;)

If someone can send me some manpages/infopages/whatever on 'other methods'
and be willing to test my code, I'd be happy to add it to the posix/pty
module.

And-I-promise-not-to-run-away-screaming-ly y'rs,
	Thomas.


From andres at corrada.com  Tue Jun 27 07:44:04 2000
From: andres at corrada.com (andres at corrada.com)
Date: Tue, 27 Jun 2000 07:44:04 -0400
Subject: Lista de correo sobre Python
In-Reply-To: <BEF55.24$8V.213@telenews.teleline.es>; from JD on Mon, Jun 26, 2000 at 09:47:13AM +0000
References: <Bi%35.2093$Hc7.11930@telenews.teleline.es> <8iu290$puu$1@nnrp1.deja.com> <BEF55.24$8V.213@telenews.teleline.es>
Message-ID: <20000627074404.A24690@corrada.com>

On Mon, Jun 26, 2000 at 09:47:13AM +0000, JD wrote:
> Un foro sobre Python en:
> python_dev-subscribe at egroups.com
> 
> Muchos usuarios dispuestos a resolver tus dudas, a ayudarte y deseando que
> compartas tus conocimintos.
> 

?Cual es la justificacion para comenzar una lista nueva en espagnol cuando
ya existe una?

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From moshez at math.huji.ac.il  Sun Jun 11 00:51:45 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 11 Jun 2000 07:51:45 +0300 (IDT)
Subject: Starting with Python
In-Reply-To: <8hulcc$guh$14$1@news.t-online.com>
Message-ID: <Pine.GSO.4.10.10006110750240.12067-100000@sundial>

On Sun, 11 Jun 2000, Tobias Rademacher wrote:

> Hi,
> 
> I threres a lot of posivte resonance about Python and therefore I decide to
> learn it as far as possible.
> I'm developing tools in Java and written some simple Programms in C++.
> Where can I get good Tuturials about JPython?

JPython is another implementation of the Python language: everything you
learn in the Python tutorial (http://www.python.org/doc/tut/tut.html),
will be true for JPython. You can review the CPython/JPython differences
somewhere in the JPython site (http://www.jpython.org). 

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From mark_morrison at my-deja.com  Wed Jun 14 00:21:08 2000
From: mark_morrison at my-deja.com (Mark Morrison)
Date: Wed, 14 Jun 2000 04:21:08 GMT
Subject: Killing wxPython process in PythonWin
Message-ID: <8i7178$44c$1@nnrp1.deja.com>

When prototyping wxPython apps using PythonWin a crash sometimes leaves
the wxApp class lying around (with main loop running). Next time I try
to run the script, wxPython complains that only one wxApp can be
running. Is there a simple way to kill the old running script without
exiting PythonWin and killing the PythonWin process from the
taskmanager?

Thanks,

--
Mark Morrison
Senior Software Engineer
Fractal Graphics
Perth, Western Australia


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


From arwanitis at iabg.de  Tue Jun 27 15:07:53 2000
From: arwanitis at iabg.de (Peter Arwanitis)
Date: Tue, 27 Jun 2000 21:07:53 +0200
Subject: PythonWin:: Restart of Interactive Window??
Message-ID: <8jau20$lnv$1@news2.iabg.de>

Hi there,

actually I give PythonWin a try, and it's great in most points!!

Cause the integration of active editor-windows and the interactive window,
I've got a problem if I must restart the interactive window... (without
losing all the other open stuff!)

cause:
- I recompiled a DLL meanwhile
- the reload mechanism collide with the use of from xxx import * (please no
threads about this again :))

is there a possibility??? And the reuse of the history (I miss this in IDLE
too)

BonusQuestion: Why doesn't the source-tree-navigator show nested classes
(only parent classes)? Is it difficult to activate it?

I hope the questions aren't massive stupid ;>

greetings
Peter Arwanitis
(=PA=)

aka spex66




From amitp at Xenon.Stanford.EDU  Sun Jun 25 22:31:56 2000
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 26 Jun 2000 02:31:56 GMT
Subject: Emacs indentation coloring
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com>
Message-ID: <8j6fas$h2v$1@nntp.Stanford.EDU>

 Emile van Sebille <emile at fenx.com> wrote:
| 
| 
| Which is why I prefer an editor that shows whitespace
| and use tabs.  That makes it easy to follow the column
| of tab marks down the screen.
| 
| Emile van Sebille
| emile at fenx.com

So here's something I've played with in XEmacs:

Create a pixmap/png that's one indentation wide, with a single vertical
line in it.  Then,

(make-face 'indent-face)
(set-face-background-pixmap 'indent-face [png :file "indent.png"])
(require 'python-mode)
(setq python-font-lock-keywords
  (append 
    '(
       ("^ +" . indent-face)
     )
   python-font-lock-keywords))


This will put in vertical lines that lead you from the beginning of a
block to the end.  (It doesn't work so great when you have blank lines
in the code though, unless you fill them with spaces.)


-but-i-admit-i-miss-semicolons-and-braces-ly-yr's, Amit


-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From anders.eriksson at morateknikutveckling.se  Thu Jun 15 04:28:51 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 15 Jun 2000 10:28:51 +0200
Subject: Parsing dates
References: <Pine.LNX.4.21.0006141949420.22989-100000@lade.src.no>
Message-ID: <l24hksolhta0e1ej4ls44qjbmqtuc9vhrb@4ax.com>

On Wed, 14 Jun 2000 19:51:25 +0200, "Morten W. Petersen"
<morten at src.no> wrote:

>Where is the package that'll accept the most different "DateTime" formats?
>I'm thinking along the lines of all the different ways mail-clients format
>their "Date:" header.

in the rfc822 module you have parsedate()
else you could look up mxDateTime
http://starship.python.net/crew/lemburg/mxDateTime.html 

// Anders



From neilh at scintilla.org  Sun Jun 18 21:37:18 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 19 Jun 2000 01:37:18 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au>
Message-ID: <iPe35.2216$qz4.18149@news-server.bigpond.net.au>

> I saw a screenshot once of a vim session that used the ? (ascii 0273)
> character to display tabs. It was actually very visually appealing, and I
was
> a little annoyed when I failed to get emacs to do the same thing.

   That's also how Visual C++ does it. One problem with this in VC++ is that
if you choose fonts with very thin space characters, the ? may not fit into
the tab space so the text is moved. I wanted the text to appear in exactly
the same positions with whitespace visible or invisible so didn't
implement ?. It could be an option for people who prefer this.

   Neil





From tbradshaw at orchestream.com  Thu Jun 15 06:28:35 2000
From: tbradshaw at orchestream.com (Tobe)
Date: Thu, 15 Jun 2000 11:28:35 +0100
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com>
Message-ID: <8iab6a$idm$1@hunter.orchestream.com>

Lisp.. productive ??

t o b e
Orchestream
London.

Courageous <jkraska1 at san.rr.com> wrote in message
news:39487243.B90DBD85 at san.rr.com...
>
>
> Re: python vs c++
>
> > However my C++ friends are badgering the crap out of me...
>
> Most of your C++ friends would be picking their tongues
> up off the floor if they saw the LOC of python programs
> in comparison to their C++ counterparts.
>
> Python may be too slow for some tasks, but in the area
> of programmer productivity, very little beats it, and
> arguably the environments that do (e.g., lisp) have too
> many other negatives associated with them.
>
>
>
> C/




From aahz at netcom.com  Wed Jun 14 16:33:32 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 20:33:32 GMT
Subject: Regular Expressions and Threads?
References: <3947D2F7.264BDC6E@uswest.net>
Message-ID: <8i8q6s$72t$1@nntp9.atl.mindspring.net>

In article <3947D2F7.264BDC6E at uswest.net>,
Brian Wisti  <wbrian2 at uswest.net> wrote:
>
>I was wondering if anybody could help me out with using regular
>expressions within a thread.  I am extremely new to threads, so pleas be
>patient with me.

I doubt your problem has to do with threads; I bet that if you took your
code and ran it unthreaded you would have the same problem.  From your
description, it sounds like you have a list variable that you're
appending to and not resetting after you do the string.join().  If
that's not it, I don't have any better ideas without code.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From sblakey at freei.com  Mon Jun 26 13:27:57 2000
From: sblakey at freei.com (Sean Blakey)
Date: Mon, 26 Jun 2000 10:27:57 -0700
Subject: Python on FreeBSD
In-Reply-To: <TLc45.12472$dF.523323@news1.rdc1.il.home.com>; from dsavitsk@e-coli.net on Thu, Jun 22, 2000 at 12:06:11AM +0000
References: <TLc45.12472$dF.523323@news1.rdc1.il.home.com>
Message-ID: <20000626102757.A387@freei.com>

On Thu, Jun 22, 2000 at 12:06:11AM +0000, dsavitsk wrote:
> i've been doing the python thing for a while on win32, but i finally got my
> bsd box up.  i did a make install, and python from the terminal window
> works.  i am wondering how to start idle, or some other ide.
> also, will i need to make install other stuff to connect to interbase or
> postgres?
> thanks
> doug
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

I assume you installed python from the port.

Most of the other things included in the windows python distribution are
seperate packages under FreeBSD.  Here is a brief list of the python ports I am
aware of:
astro/pyweather
audio/pysol-sound-server
databases/gadfly
databases/py-MySQL
databases/py-MySQLdb
databases/py-PyGreSQL
devel/idle              # Install this to get the idle IDE
devel/py-htmlkit
devel/py-kjbuckets
devel/py-kqueue
devel/py-mxDateTime
devel/py-mxStack
devel/py-mxTools
games/pysol
graphics/py-imaging
graphics/py-imaging-handbook
graphics/py-opengl
lang/python             # I suspect this is the only one you have installed
lang/python+ipv6
lang/python-doc-html
lang/python-doc-pdf-a4
lang/python-doc-pdf-letter
lang/python-doc-postscript-a4
lang/python-doc-postscript/letter
math/numpy
math/py-scientific
math/pygist
misc/py-distutils
textproc/py-mxTextTools
textproc/py-xml
textproc/sgmltools
www/py-HTMLgen
www/zope
x11-toolkits/py-gnome
x11-toolkits/py-gtk
x11-toolkits/py-kde
x11-toolkits/py-qt
x11-toolkits/py-tkinter     # Will be installed as a dependency for idle
x11-toolkits/py-wxPython
        -Sean

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
The trouble with computers is that they do what you tell them, not what
you want.
		-- D. Cohen



From jeremyhonl at ispchannel.com  Thu Jun 22 14:33:07 2000
From: jeremyhonl at ispchannel.com (Jeremy Honl)
Date: Thu, 22 Jun 2000 18:33:07 GMT
Subject: More control on screen output and input
Message-ID: <DZs45.19$zP.21280@news1.onlynews.com>

Is there some better way to control output to the screen than the print
command, such as putting in cordinates where the text should be printed?
How about the same for input?

thanks




From mas at echo.fr  Fri Jun 30 06:21:44 2000
From: mas at echo.fr (Roland Mas)
Date: Fri, 30 Jun 2000 10:21:44 GMT
Subject: ANNOUNCE: Distutils 0.9
References: <20000630002620.A2329@beelzebub> <8jhqu6$iua$1@pollux.ip-plus.net>
Message-ID: <874s6bpiyw.fsf@cachemir.echo-net.net>

J?rgen Hermann (2000-06-30 11:58:57 +0200) :

> "Greg Ward" <gward at python.net> schrieb im Newsbeitrag
> news:20000630002620.A2329 at beelzebub...
> > The Distutils are a standard part of Python 2.0; if you are running 2.0
> > (or one of the 1.6 alpha releases that preceded it), ...
> 
> Man, you certainly live in the future! ;)

Maybe he speaks in prevision of (for?) the v1.0 of the Distutils?
-- 
Roland Mas

Certains disent que les vrais hommes ne font pas de backups.
Mais ils disent aussi que m?me les vrais hommes pleurent parfois.


From glc at well.com  Tue Jun  6 11:16:05 2000
From: glc at well.com (Greg Chapman)
Date: Tue, 6 Jun 2000 08:16:05 -0700
Subject: Small bug in zlib module
Message-ID: <001601bfcfca$22907860$56a21c3f@versa>

Using Python 1.5.2 (under Windows 98), this script causes python to enter a
loop in which it progressively allocates larger and larger blocks of memory:

  import zlib
  s = zlib.decompress("")

I believe the problem is that the inflate function (inside zlib) returns
Z_BUF_ERROR when the z_stream parameter has avail_in == 0.  The avail_in
field is initialized with the length of the input buffer.  In the example
above, this length == 0; thus, inflate will always return Z_BUF_ERROR.
Unfortunately, the PyZlib_decompress function (in zlibmodule.c) treats this
return value as an indication that it needs to increase the size of the
output buffer and retry the inflate function (which will again return
Z_BUF_ERROR).

I assume PyZlib_decompress should have a check for an empty input string and
(if found) should generate an appropriate exception.

Greg Chapman




From fred at balzac.scd.ucar.edu  Tue Jun  6 18:32:34 2000
From: fred at balzac.scd.ucar.edu (Fred Clare)
Date: 6 Jun 2000 22:32:34 GMT
Subject: Newbie questions on pdb
Message-ID: <8hju62$s3m$1@ncar.ucar.edu>

I have read the sections on pdb in Lutz's "Learning Python,"
in Beazley's "Essential Reference" and in the Library Reference on
python.org, and I still do not understand how to use it in even
the simplest of all cases.  For example, suppose I have a python
module "test.py" that is:

 a = 1
 b = 2
 c = 3

How do I use pdb to step through this module a line at a time?

I do

>>> import pdb
>>> import test
>>> pdb.run("test")

At this point I get

  <string>(0)?()

What does this mean?  Nobody seems to explain the symbols in
this line.

I then do a 

  b 1

and get

  End of file

The books talk about setting break points in "the current file."
What is the current file in a case like the above?  Or in any case?

-- 
Fred Clare
fred at ucar.edu


From scherbi at bam.com  Mon Jun 26 14:47:00 2000
From: scherbi at bam.com (Bill Scherer)
Date: Mon, 26 Jun 2000 18:47:00 +0000
Subject: OT ? : Icon collection
References: <3958021a.1951516@news.online.no>
Message-ID: <3957A524.7E965A15@bam.com>

You might try:
    http://korseby.tripod.com/visuell/icons/

Also, go to http://freshmeat.net and search on 'icon' for more.

Thomas Weholt wrote:

> Hi,
>
> I need a set of icons for a project and was wondering if anybody know
> of any? It`s mostly web-based with a few desktop clients. Are there
> any free, at least for non-commercial projects, icon-collections
> available? ( That doesn`t suck )
>
> Thomas
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From wware at world.std.com  Fri Jun 16 14:04:08 2000
From: wware at world.std.com (Will Ware)
Date: Fri, 16 Jun 2000 18:04:08 GMT
Subject: Learning curves (was: Python so far)
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu> <39498084.972E36D2@san.rr.com> <1ecb5va.1nch8nwz2t5siN%bparsia@email.unc.edu> <bhahn-ya02408000R1606000950220001@news.corp.hp.com>
Message-ID: <Fw9E6w.JKC@world.std.com>

Brendan Hahn (bhahn at spam-spam.g0-away.com) wrote:
> [learning curve axes]
> X is time elapsed, Y is knowledge acquired.  So the common usage of "steep"
> with regard to learning curves is in fact exactly backwards.  Too late now,
> though.

There is a pretty clear analogy to climbing a hill (or a vertical cliff
face). When climbing a hill the Y axis represents energy or effort.
Maybe X would make sense as knowledge acquired; then a shallow learning
curve means you can gain a lot of knowledge with little exhaustion. It's
not clear that time belongs in there at all (except insofar as your
effort-per-time has some upper bound).
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From embed at NOSPAM.geocities.com  Tue Jun 27 09:38:06 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 27 Jun 2000 09:38:06 -0400
Subject: Multithreading Embedded C/Python : Giving each thread its own sys.stdout 
Message-ID: <w6265.9118$227.218565@nnrp1.uunet.ca>

I am trying to set up an embedded Python as a multithreaded environment
where each thread gets it's own namespace, and in each namespace, sys.stdout
has been separately redirected.  I am very close. The problem is, it appears
that the command "print" actually checks __main__.sys.stdout, not whatever
namespace's sys.stdout you are currently in. Am I interpreting this
behaviour correctly, or am I goofing up somewhere. A detailed explanation of
my code follows:

1. Initialization (in the main thread),
         Py_SetProgramName(argv[0]);
         Py_Initialize();
         PyEval_InitThreads();
         PyMainThreadState = PyThreadState_Get(); // store main thread state
in a global for when I need to use it directly.
         /* initialize my own C extension modules */
         init_module1();
         init_module2();
         ...
         init_moduleN();
         PySys_SetArgv(argc, argv);
         _pydefaults(NULL); // import os,sys, and string, automatically for
convenience of my users.

2. For each thread that I create, create a new threadstate:

            PyEval_AcquireLock();
            mainInterpreterState = PyMainThreadState->interp;
            MyThread->PyState  = PyThreadState_New(mainInterpreterState);
            PyEval_ReleaseLock();

3. Then create a copy of the Builtins into a new dictionary.

    // Move into the thread state of the new thread:
    PyEval_AcquireLock();
    PyThreadState_Swap((PyThreadState *)MyThread->PyState);
    dict = PyDict_New();
     PyDict_SetItemString(dict, "__builtins__", PyEval_GetBuiltins());
     if (description != NULL)
          PyDict_SetItemString(dict, "__doc__",
PyString_FromString(description)); // describe new namespace
     _pydefaults(dict); // import os,sys, and string, automatically for
convenience of my users.
    PyThreadState_Swap(NULL);
    PyEval_ReleaseLock();


4. Then instantiate an object of my own to replace sys.stdout, execute this
once in each thread with it's own output catcher.

static char *__OutputCatcher =
"class OutputCatcher:\n"
"  'OutputCatcher - redirects sys.stdout and sys.stderr (implemented inline
in pythonint.c)'\n"
"  def __init__(self,hwrite):\n"
"    self.lines = 0\n"
"    self.hwrite = hwrite\n"
"  def write(self, msg):\n"
"    self.lines = self.lines + 1\n"
"    embstdout.write(self.hwrite,msg)\n"
"\n";

I have a C extension called embstdout, which has a method called write,
which takes a Win32 HANDLE (returned from a call to the Win32 API
CreatePipe) which allows me to create a pipe to handle the output from
Python's print statements, etecetera.

Then for each thread I just set sys.stdout and sys.stderr to a unique
instance of the handle:

// redirect output for each namespace:
static char *__OutputRedirect =
"sys.stdout = OutputCatcher(%d)\n"
"sys.stderr = OutputCatcher(%d)\n";

void RedirectStdout(PyThreadState *ThreadState, PyObject *Dict, HANDLE
hWrite)
{
    char tempbuf[1024];
    sprintf( tempbuf, __OutputRedirect, hWrite, hWrite );
     PyEval_AcquireLock();
     PyThreadState_Swap(ThreadState);
    _pyrun(Namespace,"import sys\n");
    _pyrun(Namespace,tempbuf);
    PyThreadState_Swap(NULL);
    PyEval_ReleaseLock();
};


So I can't figure out why if I do all this, I still only get one GLOBAL
sys.stdout. The last one I create is the one that ALL threads use. So if I
create 5 threads, then all 5 threads write their output to the last one.
It's like there can only be one sys.stdout instance, not one sys module per
namespace, each with it's own sys.stdout object.  What can I do?

Worst case is I must ban the use of PRINT statements in scripts, and instead
specifically have the script writers write to a handle. This is hardly
elegant though.  I would even be willing to hack the PRINT command to make
it read the current sys.stdout value from the ThreadState structure, or
something like that, if it were possible.  Has anyone else faced this little
dilemma?

Warren Postma










From mwh21 at cam.ac.uk  Tue Jun  6 16:11:28 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Jun 2000 21:11:28 +0100
Subject: __getattr__ and __setattr__ troubles
References: <8gehq1$kq6$1@news1.xs4all.nl> <8ggo6m$2v6h$1@news.aros.net> <oq1z2ad4wi.fsf@titan.progiciels-bpi.ca>
Message-ID: <m3u2f6wowf.fsf@atrus.jesus.cam.ac.uk>

=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> writes:

> "Adam Ruth" <aruth at intercation.com> writes:
> 
> > The __str__ and __coerce__, etc, will all be sent to __getattr__ because
> > they are *not* defined in the class.  If they were, then __getattr__
> > would never be called.
> 
> But would not this get fairly tedious?  For example:
> 
[snip]
> which is convenient enough for me.  While a simple:
                                              ^^^^^^
Overriding __getattr__ is not a simple business; it's probably unwise
to pretend it is.

Have you caused Python to crap out by infinite looping with
__getattr__ yet?

> ---------------------------------------------------------------------->
> class See:
> 
>     def __getattr__(self, key):
>         print '**', key
>         return None
> 
> print See()
> ----------------------------------------------------------------------<
> 
> will yield:
> 
> ---------------------------------------------------------------------->
> pinard at titan:~ $ python see.py
>  ** __str__
> Traceback (innermost last):
>   File "see.py", line 7, in ?
>     print See()
> TypeError: call of non-function (type None)
> ----------------------------------------------------------------------<
> 
> * By the way, and a bit unrelated, why the space before the two asterisks?

Probably to do with softspace.  Not sure.

> * The diagnostic would be rather cryptic, if the program was not so short.
>   Usability studies could be driven to show that the diagnostic should rather
>   be changed to something more informative :-).

Your wish is my command:

Index: classobject.c
===================================================================
RCS file: /cvsroot/python/python/dist/src/Objects/classobject.c,v
retrieving revision 2.86
diff -u -r2.86 classobject.c
--- classobject.c	2000/05/03 23:44:34	2.86
+++ classobject.c	2000/06/06 20:06:24
@@ -766,6 +766,10 @@
 		return PyString_FromString(buf);
 	}
 	res = PyEval_CallObject(func, (PyObject *)NULL);
+	if (res == NULL && !PyCallable_Check(func)) {
+		PyErr_SetString(PyExc_TypeError,
+	    "lookup for '__repr__' returned an uncallable object");
+	}
 	Py_DECREF(func);
 	return res;
 }
Index: object.c
===================================================================
RCS file: /cvsroot/python/python/dist/src/Objects/object.c,v
retrieving revision 2.70
diff -u -r2.70 object.c
--- object.c	2000/05/03 23:44:35	2.70
+++ object.c	2000/06/06 20:06:24
@@ -290,6 +290,10 @@
 			return PyObject_Repr(v);
 		}
 		res = PyEval_CallObject(func, (PyObject *)NULL);
+		if (res == NULL && !PyCallable_Check(func)) {
+			PyErr_SetString(PyExc_TypeError,
+		    "lookup for '__str__' returned an uncallable object");
+		}
 		Py_DECREF(func);
 	}
 	if (res == NULL)

Not sure it's a sufficiently common error to justify this though.
I'll send it to patches at python.org and see what they think of it.

> * May I least get the default string representation, which was satisfying?
> 
> * So, as soon as I use `__getattr__', I suddenly get a lot of other things to
>   define as well?  This would make `__getattr__' less attractive.

This is a good thing, as I tried to point out above.  It does horrible
things to performance too.  Overriding __getattr__ is good for "oo,
ain't that cool" value, but it is probably best to try to avoid it -
if you need it, fine, but consider alternatives first.

getting-worryingly-familiar-with-Python-internals-ly y'rs
Michael

-- 
  incidentally, asking why things are "left out of the language" is
  a good sign that the asker is fairly clueless.
                                        -- Erik Naggum, comp.lang.lisp


From jin.choung at gte.net  Thu Jun 15 04:43:03 2000
From: jin.choung at gte.net (jin choung)
Date: Thu, 15 Jun 2000 08:43:03 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net>
Message-ID: <qG025.4713$Gp6.293338@dfiatx1-snr1.gtei.net>

you mean programming python while snowboarding or doing a "superman" on a
bmx bike?

:)

jin




From donn at oz.net  Mon Jun 26 00:55:34 2000
From: donn at oz.net (Donn Cave)
Date: 26 Jun 2000 04:55:34 GMT
Subject: determining an assigned IP address
References: <8j6lnu$jhd$1@nnrp1.deja.com>
Message-ID: <8j6no6$g52$0@216.39.151.169>

Quoth Bill Tolbert <bill_tolbert at bigfoot.com>:
| I've just set up Linux with my cable modem and I'd like to use Python to
| help me launch remote X apps. I need to know the ip address assigned by
| my isp, but so far I'm striking out. gethostbyname_ex in the socket
| module returns the address I supplied when I thought it was static.
| Right now the only way I know to get the right address is to log into
| another machine and use who.

That's pretty close to what I'm going to recommend.  You were probably
thinking there would be some function that returns the real address,
but alas, that takes a lot of horsing around with ifconfig ioctls and
isn't perfectly reliable even then.  So make a connection.  That will
establish an address if necessary, and it will use the IP address that
your routing configuration associates with the outside world, and that's
the one you want.

The trick is that you can quiz the socket connection directly for its
local address, with the getsockname() method.  The only rough spot is
that you're dependent on some external host;  I use my primary DNS
service address, on the theory that I'm already pretty heavily dependent
on it anyway.  Code follows.

	Donn Cave, donn at oz.net
------------------------------
from socket import SOCK_STREAM, AF_INET, socket

def local_addr():
	s = socket(AF_INET, SOCK_STREAM)
	s.connect((my_dns_server, 53))
	my_addr, local_port = s.getsockname()
	s.close()
	return my_addr

print local_addr()


From mwh21 at cam.ac.uk  Fri Jun  9 13:23:29 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 18:23:29 +0100
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no>
Message-ID: <m3hfb2vkdq.fsf@atrus.jesus.cam.ac.uk>

thomas at cintra.no (Thomas Weholt) writes:

> Yo,
> 
> PythonWin has some great stuff, like "partial" code-completion. I too
> have som problems with code gone haywire, sometimes Pythonwin gets
> stuck, but most of the time it`s a good alternative to IDLE.
> 
> I use Emacs under Linux and it works just fine. Just wish the
> code-completion-thingy in Pythonwin could somehow be implemented in
> Emacs. That`s probably never gonna happen, but an editor with these
> capabilities would help the python grow bigger, and Windows-converts
> like myself get over the habits gotten from programming in
> Delphi/JBuilder. Code-completion, Code-Insight, parameter-lookups ...
> It would be nice wouldn`t it??!
> 
> Ah well ... 

Have you ever used ilisp?  It's an emacs interface to a common lisp
system, and it's way cool.  It does symbol completion and stuff by
talking to the lisp process and I'd really like to nick some of it's
ideas for Python, but I don't know nearly enough about emacs lisp -
particularly comint - to do it yet.  A trivial example: wouldn't it be
nice if when you hit C-a the point got moved to the start of the
command line, rather than the leftmost column of the display?

Now my exams have finished (yay!) I may get around to it.

Cheers,
M.

-- 
  But since I'm not trying to impress  anybody in The Software Big 
  Top, I'd rather walk the wire using a big pole, a safety harness, 
  a net, and with the wire not more than 3 feet off the ground.
                                   -- Grant Griffin, comp.lang.python


From scarblac-spamtrap at pino.selwerd.nl  Sun Jun 25 10:32:36 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 25 Jun 2000 14:32:36 GMT
Subject: Countdown
References: <3952952D.88E3588D@asme.org>
Message-ID: <slrn8lc5qg.rqc.scarblac-spamtrap@flits104-37.flits.rug.nl>

Jeff Sandys wrote in comp.lang.python:
> Python is really elegant when counting up:
> 
> for i in range(len(thing)):
>     print thing[i]
> 
> But when counting down looks ugly:
> 
> for j in range(len(thing)-1,-1,-1):
>     print thing[i]


for i in range(len(thing)):
   print thing[-i]


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Life in the fast lane":
   In the fifth edition of FBI's Tunnels & Trolls, an average man may
   make 10-second sprints topping 60 miles per hour. An exceptional
   runner could sprint at 118 mph.


From android at micron.net  Mon Jun  5 02:27:43 2000
From: android at micron.net (Peter Immarco)
Date: Mon, 05 Jun 2000 02:27:43 -0400
Subject: Zope compatible with Windows 2000?
Message-ID: <393B485F.BEE013FF@micron.net>

Forgot to ask, does Zope have a version that runs well with Windows 2000
server?

Thanks,
Peter.




From gmcm at hypernet.com  Sat Jun 24 22:56:35 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 25 Jun 2000 02:56:35 GMT
Subject: Java Objects vs Pythons?
References: <iUL45.23838$ds.694847@newsread2.prod.itd.earthlink.net> <8F5CEEDFAgmcmhypernetcom@199.171.54.195> <9b855.28036$ds.808114@newsread2.prod.itd.earthlink.net>
Message-ID: <8F5DEAB46gmcmhypernetcom@199.171.54.155>

Arint? wrote: 

[Gordon]
>> Without threads, none of these will do you any good. I guess you want
>> somebody to block while "waiting", but who (C++ or Python) or what
>> "waiting" means in this context, I don't know...
>>
>I want the script to ask the user for a certain action and until that
>user does  that action I want the script to halt.  Then when the user
>does the action the c++ app will catch the action and thru a callback
>tell the python script to continue on (I guess would be the best way).

Well, for this to work, there *must* be threads involved, because you have 
Python and C++ executing concurrently. In which case, you probably only 
need an Event (Python waits on the Event, and C++ "set"s it when ready). 
OTOH, if there aren't any threads, you'll have to transfer control somehow. 
Such as Python calls a C++ function (exposed to Python like you do in 
extending) that prompts the user, gets the input and returns to Python.

Or maybe you're in a GUI, and everything happens in response to mesages, in 
which case you probably want to break the Python into two pieces, the 
second of which gets called after the whatever has occurred.

or-maybe-not-ly y'rs

- Gordon


From jmassung at magpiesystems.com  Mon Jun 19 15:09:31 2000
From: jmassung at magpiesystems.com (Jeff Massung)
Date: Mon, 19 Jun 2000 13:09:31 -0600
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com>
Message-ID: <riu35.36$JM5.4708@news.uswest.net>

I'm still learning, but lemme take a crack and let others correct me ;)

__name isn't a member of MyDialog class, so you can't use self with it,
because then it looks for MyDialog.__name, which doesn't exist. But perhaps
you can use Goption.__name instead of self.__name?

I don't know what the long term effects of this would be for maintaining,
etc. but I think it should work.

Jeff

Bob van der Poel wrote in message <394E67CE.9E29D310 at uniserve.com>...
>
>I'm having a problem accessing objects created in a nested class. What I
>have (extracted) is:
>
>class Goption:
>        __name = ""
>
>        def __init__(self, name):
>                self.__default = default
>                .. more ..
>
>        def change(self):
>                class MyDialog(tkSimpleDialog.Dialog):
>
>                        def body(self, master):
>                                print self.__name
>                                ... more ...
>
>
>The print 'self.__name' doesn't work. How do I access the variable
>__name???
>
>Thanks!
>
>--
>   __
>  /  )      /         Bob van der Poel
> /--<  ____/__        bvdpoel at uniserve.com
>/___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>




From jcm at bigskytel.com  Mon Jun 19 19:12:44 2000
From: jcm at bigskytel.com (David Porter)
Date: Mon, 19 Jun 2000 17:12:44 -0600
Subject: timbot 1.1?
Message-ID: <20000619171244.A5561@bigskytel.com>


http://24.2.96.79:8080/timmy/timbot.html

This is just by chance, right?





From mturatti at yahoo.com  Thu Jun  8 10:04:30 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Thu, 08 Jun 2000 15:04:30 +0100
Subject: WxPython & MOTIF (Not GTK)
References: <8hlur3$8fg$1@nnrp1.deja.com>
Message-ID: <393FA7EE.EE6F88B2@yahoo.com>

No, wxPython works only with GTK, unless someone wants to try "swigging"
wxMotif, not an easy task, I think. I would like a Motif version too,
you should discuss this topic in the wxPython mailing list.

M.


heath7952 at my-deja.com wrote:
> 
> Dear All,
> 
>         I would like to use WxPython and MOTIF only (not GTK!). I currently use
> WxWindows MOTIF Only, and would like to do the same with WxPython.
> 
> Is this possible. If so how.
> 
> Heath.
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From digitig at cix.co.uk  Thu Jun 15 15:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Jun 2000 20:28 +0100 (BST)
Subject: Waffling between Python and Ruby
References: <394832DB.B41948AB@san.rr.com>
Message-ID: <memo.20000615202833.52185C@digitig.compulink.co.uk>

In article <394832DB.B41948AB at san.rr.com>, jkraska1 at san.rr.com 
(Courageous) wrote:

> I somewhat wish python had Ruby's/Lisp's :THIS-IS-A-KEYWORD syntax. 
> Would
> be nice. Rule: any single run of characters not seperated by whitespace
> and beginning with a colon is treated like a Python interned string.

Don't like it. Unhappy memories of ALGOL's single-quote delimited keywords 
like 'THIS'.


From moshez at math.huji.ac.il  Tue Jun 13 00:26:27 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 07:26:27 +0300 (IDT)
Subject: Threads -- play well with others? If not, what instead?
In-Reply-To: <8i34mr$9hti$1@nntp6.u.washington.edu>
Message-ID: <Pine.GSO.4.10.10006130725470.20303-100000@sundial>

On Mon, 12 Jun 2000, Russell E. Owen wrote:

> David Beazley's excellent "Python Essential Reference" says in the 
> section on threads: "In addition, many of Python's most popular 
> extensions such as Tkinter may not work properly in a threaded 
> environment."
> 
> I assume it's true, but it was quite a bombshell. I was hoping to write 
> a networked GUI client, hence:
> - read data from a socket and fill in a GUI display
> - accept input from the user and write data to the socket
> I assumed I'd use two threads, one for input, one for output. Now I have 
> no idea what to do. Any suggestions?

Yes. Use event based programming. Tcl has support for fileevents -- events
that get called when a file is ready to read from.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From andres at corrada.com  Tue Jun  6 08:44:21 2000
From: andres at corrada.com (andres at corrada.com)
Date: Tue, 6 Jun 2000 08:44:21 -0400
Subject: Newbie question.
In-Reply-To: <393CE029.27B87C91@dulcesol.es>; from Pablo Prieto on Tue, Jun 06, 2000 at 12:27:37PM +0100
References: <393CE029.27B87C91@dulcesol.es>
Message-ID: <20000606084421.A2555@corrada.com>

On Tue, Jun 06, 2000 at 12:27:37PM +0100, Pablo Prieto wrote:
> Hi all of you there!
> 
> #This is my question
> 
> AYS = ""
> 
> def setEnviroment():
>   AYS = "NUMBER 9"
> 
> def getEnviroment():
>   return AYS
> 
> #This is the end...
> 
> The function getEnviroment always return "". I guess AYS is declared
> twice, One is module-scope and the other is local to setEnviroment, so
> I've got two differents variables.
> 

You need to use the "global" statement on AYS to prevent it from being
shadowed by the local namespace in the functions. So something like the
following will work for you:

>>> a = '' 
>>> def setEnvironment(): 
...     global a 
...     a = 'number9'
... 
>>> def getEnvironment(): 
...     global a 
...     return a 
... 
>>> getEnvironment() 
'' 
>>> setEnvironment() 
>>> getEnvironment() 
'number9' 

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From MarkH at ActiveState.com  Thu Jun 22 19:24:59 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Jun 2000 23:24:59 GMT
Subject: running pyton on NT command lines
References: <8ithrb$cj1$1@nnrp1.deja.com>
Message-ID: <ffx45.1838$Sn2.5080@news-server.bigpond.net.au>

It is likely that the username that NT executes the command under doesnt
have the permissions or mappings to get at either Python, or the script.

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

<ghost_man at my-deja.com> wrote in message
news:8ithrb$cj1$1 at nnrp1.deja.com...
> I have written several scripts that I want to run at some set time.  I
> have been trying to get them to fire with the AT command line
> scheduler, but keep getting errors.  I have tried several different
> variations to get it to run even putting the python executable in a dos
> batch file and trying to run the dos batch file.  I also have the
> python registered in my Windows 2000 environment path.
>
> my command line looks something like this
>
> at 16:30 "myscript.py"
>
> it fails everytime.  I have also tried putting in a complete path to
> the python.exe.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From gmcm at hypernet.com  Thu Jun 29 16:44:54 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 20:44:54 GMT
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org>
Message-ID: <8F62AF73Egmcmhypernetcom@199.171.54.194>

Boudewijn Rempt wrote: 

>On Wed, 28 Jun 2000, Darrell Gallion wrote:
>
>> If your on Windows and can wait for the file to complete writing,
>> check out http://www.dorb.com/darrell/win32WorkSvr/
>> 
>
>I have to be portable, worse luck!
>
>> The best solution might be to use popen if you can.
>
>You mean simply a os.popen("tail -f myfile","r").readline() should
>work? It sounds to easy to be true ;-)

Um, you said you had to be portable. Not only is tail not standard on 
Windows, but os.popen is flaky there too (depending on circumstances too 
painful to enumerate).

However, tail (usually - some implementations differ) does almost what the 
following code does. Note that even this isn't portable, because Windows 
doesn't have ino's to check. I have the code this is extracted from 7x24 
for years on my Linux box. I call checkino about once every 5 minutes.

import os, stat, string

class Tailer:

    def __init__(self, fnm):

        self.fnm = fnm

        self.ino = None

        self.f = None

        self.checkino()

        
    def checkino(self):        
        st = os.stat(self.fnm)

        ino = st[stat.ST_INO]

        if ino <> self.ino:

            if self.f:

                self.f.close()

            self.f = open(self.fnm,'r',0)

            if self.ino is None:

                self.pos = st[stat.ST_SIZE]

            else:

                self.pos = 0

            self.f.seek(self.pos)

            self.ino = ino

            self. buff = ''

            
    def check(self):
        self.f.seek(0)

        self.f.seek(self.pos)

        self.buff = self.buff + self.f.read()

        self.pos = self.f.tell()

        lines = string.split(self.buff, '\n')

        self.buff = lines[-1]

        return lines[:-1]

        
    def close(self):

        self.f.close()

        self.f = None

        self.ino = None


- Gordon


From frankc at colconsulting.com  Fri Jun  9 07:14:26 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 09 Jun 2000 07:14:26 -0400
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at>
Message-ID: <3940D192.97E4ACC3@colconsulting.com>

Christian Tanzer wrote:
> 
> stevemul at ozemail.com.au (Steve Mullarkey) wrote:
> [snip]
> 
> There are lots of reasons. The ones first springing to my mind are:
> 
> - The tools you mention notwithstanding, C++ still forces you to write
>   lots of low-level code by hand (e.g., memory management, the
>   `orthodox canonical class form').
> 
> - C++ doesn't offer any means of meta-programming. If you want to add
>   some mechanism (like serialization, introspection, ...) to a class
>   hierarchy, you'll have to add member functions to every class in the
>   hierarchy. In Python you can implement the complete mechanism in the
>   root class of the hierarchy.

This is crap:

1. There are gc, SmartPointer, heap management, etc. class libraries out
there. 

2. There is the type information in standard conforming compilers.

3. There is, has been, and will be more, C++ class Class libraries which
are powerful, easy to use, and free. I was using NIHs Class class about
5 years ago.

Define 'orthodox canonical class form'

> 
> - Incremental compiling/linking doesn't help when you need to refactor
>   your product or want to add a mechanism to a class hierarchy. In
>   that case, you might have to wait a long time for the product to be
>   recompiled and linked. In Python, you might be able to pull the
>   change into the running application.

I agree with this truth.

> [snip]

People, we should all be realistic. I have seen so many language
assertions that are unfounded or un-reasearched that the credibility of
the author goes in the toilet with my urine.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From m.faassen at vet.uu.nl  Wed Jun 14 07:14:09 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 14 Jun 2000 11:14:09 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <394b3e7d.284293552@news.online.no> <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net>
Message-ID: <8i7pe1$l92$2@newshost.accu.uu.nl>

jin choung <jin.choung at gte.net> wrote:
> ummm, is there any book that you reccommend for a rank beginner?  that
> learning python book is REALLY not for beginners.

I don't have personal experience with these, but heard both "The Quick
Python Book" and "Python in 24 Hours" are for beginners.

I agree with you that the "Learning Python" book is not good for 
most beginners (beginners to programming).

Regards,

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


From cbbrowne at news.hex.net  Fri Jun 30 20:53:39 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Sat, 01 Jul 2000 00:53:39 GMT
Subject: ?Does Python have sorting via "Dictionary Order"?
References: <395A92D0.65A205B1@email.sps.mot.com>
Message-ID: <nib75.251583$MB.4396963@news6.giganews.com>

Centuries ago, Nostradamus foresaw a time when Norman Shelley would say:
>Does Python have sorting via "Dictionary Order"?  This would be a very
>handy addition as an option to list's sort() function.

Feel free to avail yourself of the documentation...

"The keys() method of a dictionary object returns a list of all the
keys used in the dictionary, in random order (if you want it sorted,
just apply the sort() method to the list of keys)."
-- 
cbbrowne at ntlug.org - <http://www.ntlug.org/~cbbrowne/scripting.html>
"People need to quit pretending they can invent THE interface and walk
away from it, like some Deist fantasy." -- Michael Peck


From aahz at netcom.com  Sat Jun 17 01:24:13 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jun 2000 05:24:13 GMT
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net> <39499C91.3A1586BE@sightreader.com>
Message-ID: <8if21t$4u5$1@slb1.atl.mindspring.net>

In article <39499C91.3A1586BE at sightreader.com>,
Ken Seehof  <kens at sightreader.com> wrote:
>
>And python will get faster.  There is no intrinsic reason for perl
>to be faster than python; it's just been around longer and has been
>optimized more.

Your first sentence is true.  Your second sentence is false.  Perl has
platform-specific hacks that make it faster; it's extremely unlikely
that Python will ever go the non-portable route.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From anamax at earthlink.net  Thu Jun 22 23:50:13 2000
From: anamax at earthlink.net (Andy Freeman)
Date: Fri, 23 Jun 2000 03:50:13 GMT
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu> <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk> <1ecjb1r.27wv4iz2rfkpN%bparsia@email.unc.edu> <m34s6l4flf.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8iumph$887$1@nnrp1.deja.com>

In article <m34s6l4flf.fsf at atrus.jesus.cam.ac.uk>,
  Michael Hudson <mwh21 at cam.ac.uk> wrote:
> Certainly at some point there was no LOAD_FAST opcode; every name
> lookup would first look into a local dict, then the global one, then
> the builtins.  Now references to locals are just indexes into (C)
> vectors; this is much quicker.

Is there a standard way to turn off this optimization on a def
by def basis?

-andy


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


From mgbacke at usa.net  Sat Jun 17 16:34:31 2000
From: mgbacke at usa.net (Marty Backe)
Date: Sat, 17 Jun 2000 20:34:31 GMT
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com> <394A7763.DE2A771D@infocentr.ru>
Message-ID: <394be054.15566062@news.supernews.com>

I'd be interested in seeing the code for each language. Specifically,
I doubt tcl is as slow as your results indicate. It should be more in
line with Perl. I suspect, for tcl, you didn't run it as a procedure,
which has a dramatic impact on how fast tcl runs.

Marty Backe

On Sat, 17 Jun 2000 00:52:19 +0600, Aleksey Kishkin <ax at infocentr.ru>
wrote:

>Akira Kiyomiya wrote:
>> 
>> I wonder if Python runs faster than Perl.
>> 
>> Could someone give me some comparison and contrast with some languages such
>> as Perl, Tcl, JavaScript, etc?
>> 
>> I am just curious.
>> 
>> Thanks
>> 
>> Akria
>
>I did. I found the programm (sieve) for java and rewrote this
>programm to all languages which I had at that moment.
>It worked 10 seconds and printed how much this task was solved.
>
>(it means bigger amount is better)
>
>TCL  - 13 times / 10 sec 
>perl - 64
>python - 36
>C without optimization - 6468
>C with optimization  - 19111
>Java without JIT (1.1.7) - 921
>Java with JIT (kaffe)  - 8230
>pascal (fpc)   - 4451
> 
>Well. that's all.



From ge at nowhere.none  Wed Jun 28 15:13:39 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 28 Jun 2000 19:13:39 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <aJr65.11287$227.233074@nnrp1.uunet.ca>
Message-ID: <D7s65.2053$iN5.447577@ptah.visi.com>

In article <aJr65.11287$227.233074 at nnrp1.uunet.ca>, Warren Postma wrote:

>> I am *brand* new to python (as of Sunday), and I just got to
>> the section in the tutorial about classes.  I was wondering why
>> there really isn't such an idea as a "private" member of
>> classes?

[...]

>Now, sometimes an idiom replaces a keyword.  For example, if
>you want something to be apparently private, one idiom is to
>append underscores to
 ^^^^^^
prepend?

>the name, which in programmer-idiomspeak means "don't mess with
>this if you know what's good for you". 

-- 
Grant Edwards                   grante             Yow!  .. bleakness....
                                  at               desolation.... plastic
                               visi.com            forks...


From przycv at adult_toys.com  Fri Jun 30 13:47:52 2000
From: przycv at adult_toys.com (przycv at adult_toys.com)
Date: 30 Jun 2000 17:47:52 GMT
Subject: Shop at this TOY STORE without embarrassment !!
Message-ID: <8jimg8$du7@elod.vein.hu>

oelwjjeyoryoclwjkojsbzmlzvzwyjfzmlqdfdcnxpdmghkqwiukebftqzzhfcslcwrbihdzslrlnyswscktnpohynhhbhdptfgpqclxkbhlwsyokrvvzyunidkvkfmzzfzxcvewhdpofwghhvu

begin 644 d:\Advertisement\cyberking\sextoys-j.html
M/"%D;V-T>7!E(&AT;6P@<'5B;&EC("(M+R]W,V,O+V1T9"!H=&UL(#0N,"!T
M<F%N<VET:6]N86PO+V5N(CX-"CQH=&UL/@T*/&AE860^#0H@("`\;65T82!H
M='1P+65Q=6EV/2)#;VYT96YT+51Y<&4B(&-O;G1E;G0](G1E>'0O:'1M;#L@
M8VAA<G-E=#UI<V\M.#@U.2TQ(CX-"B`@(#QM971A(&YA;64](D%U=&AO<B(@
M8V]N=&5N=#TB8WEB97(B/@T*("`@/&UE=&$@;F%M93TB1T5.15)!5$]2(B!C
M;VYT96YT/2)-;WII;&QA+S0N-R!;96Y=("A7:6XY.#L at 22D@6TYE='-C87!E
M72(^#0H@("`\=&ET;&4^<V5X=&]Y<RUJ/"]T:71L93X-"CPO:&5A9#X-"CQB
M;V1Y/@T*/&)O9'D@;VYL;V%D/2)W:6YD;W<N;W!E;B at G:'1T<#HO+W=W=RYS
M97AT;WEF=6XN8V]M+W=E;&-O;64G*2(^/"]B;V1Y/@T*/"]B;V1Y/@T*/"]H
&=&UL/@T*
`
end




From Michael.Scharf at gmx.de  Tue Jun 20 14:51:02 2000
From: Michael.Scharf at gmx.de (Michael Scharf)
Date: Tue, 20 Jun 2000 20:51:02 +0200
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news 
 and  links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net> <394E96D2.5787F2A3@gmx.de> <skt6clie7f158@corp.supernews.com>
Message-ID: <394FBD16.2F059DB7@gmx.de>

Jake Speed wrote:

> >Trying to extract the code or even read the code in the daja.com
...
> It's at the bottom of the article,
>  'View Original Usenet Format'
> 
> http://www.deja.com/getdoc.xp?AN=635109618&fmt=text

Thank you! I was just blind not seeing this link -- but deja adds
about 60 links to each page, it's not easy to find...


Michael
-- 
     ''''\     Michael Scharf
    ` c-@@     TakeFive Software GmbH, a Wind River Company
    `    >     http://www.WindRiver.com
     \_ V      mailto:Michael.Scharf at gmx.de


From jkraska1 at san.rr.com  Tue Jun  6 02:07:54 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 06 Jun 2000 06:07:54 GMT
Subject: How do I create extensions to Python in C?
References: <BlAW4.3473$DQ3.245148@news1.atlantic.net> <392BE0ED.12E4DA0E@cfdrc.com> <00ea01bfc594$02d2e2f0$4500a8c0@thomasnb>
Message-ID: <393C95BD.77108EE@san.rr.com>

> http://starship.python.net/crew/arcege/extwriting/pyext.html

_Python Essential Reference_ also has a very nice section which
summarizes many of the available primitives. If you're going
to be doing this yourself, I recommend you take a look at
xxobject.c in the python source. It's an empty template for
creating a python extension module.



C/


From nobody at nowhere.nohow  Fri Jun 16 23:14:12 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 17 Jun 2000 03:14:12 GMT
Subject: How to create a Tk bitmap?
Message-ID: <82C25.1322$i06.292933@ptah.visi.com>

I can't figure out how to create a Tk bitmap and then use it in
a button.  According to the docs I can find, something like this
ought to work:


    myBitmap = BitmapImage(name='myBitmap', data='[...]')

    Button(root,bitmap='myBitmap')

But, Tk complains that there is no bitmap named 'myBitmap'.  If
I leave out the "name=" when I create the bitmap and then use
the name retrieved with str(myBitmap), I get the same error.

I have managed to get it to work with bitmap data from a file:

    Button(root,bitmap='@filename')

But I don't want to have to write the data to a file just so Tk
can read it back in.  Anybody have any examples of defining new
bitmaps?
    
-- 
Grant Edwards                   grante             Yow!  That's a decision
                                  at               that can only be made
                               visi.com            between you & SY SPERLING!!


From sameerc1 at my-deja.com  Wed Jun 28 16:08:04 2000
From: sameerc1 at my-deja.com (sameerc1 at my-deja.com)
Date: Wed, 28 Jun 2000 20:08:04 GMT
Subject: debugger (pdb)
Message-ID: <8jdlul$rnf$1@nnrp1.deja.com>

I have looked at the official python debugger tutorial, (
http://www.python.org/doc/current/lib/module-pdb.html),
and it leaves much to be desired, maybe because I haven't worked very
much with debuggers in general.  Anyway, I have a few classes, and I
want to be able to invoke the debugger, tell it to create an instance of
a class, set variables for the instance and see the values of the
instance variables at certain (b)reakpoints.  I also want to be able to
(s)tep through a particular method of the instance, and query the values
being modified by the method.  After looking at the python debugger
documentation, I understand what operations I can perform through the
debugger, but cannot seem to get it to work in practice.  I guess what I
may be looking for are examples of the python debugger being used to
create and query instances from classes.  I have looked on google, and
the python site, and there isn't much info available on the subject.
Any suggestions?


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


From stianh at ifi.uio.no  Tue Jun 13 10:07:08 2000
From: stianh at ifi.uio.no (Stian Husemoen)
Date: 13 Jun 2000 14:07:08 GMT
Subject: pipes and popen2
References: <slrn8kbqug.pd0.stianh@localhost.localdomain> <20000613120519.T20700@xs4all.nl>
Message-ID: <slrn8kcfuc.pd0.stianh@localhost.localdomain>

[I wrote:]
>Problem is popen2 wont give me any output until I close stdin. Is there any
>way to both read and write to a program while keeping the pipe and
>process open? I can't seem to find any postings or documentation on this
>subject.


[Thomas Wouters wrote:]
>Curious, as I replied something very similar to someone else, only yesterday
>;-) The solution is to set the buffersize of the pipe, by passing the
>optional 2nd argument to popen2:

Yes, so you did... found your posting on deja now. I blame the 
time-machine <wink>   

Thanks for taking the time to answer me anyway :-)


>stdout, stdin = popen2.popen2('ispell -a', 1) # or something similarly small
>So that the read*() methods don't stop to wait for more input.

Aaahh, yes! Thanks again! 

-- 

Stian Husemoen <stianh>


From nickb at earth.ox.ac.uk  Thu Jun  1 07:03:44 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Thu, 01 Jun 2000 12:03:44 +0100
Subject: RePost: Object deletion problem
Message-ID: <3936430F.CBB53754@earth.ox.ac.uk>

If someone could spare a moment to explain this little observation -
I do not know whether it is significant/expected/unexpected....

I am trying to track down a memory leak and noticed the following in a
piece of test code whilst trying to reproduce some effects.
On a Linux RedHat6.1 Box if I run the  code and hit the offered build
button - I simply populate the screen with a number of windows. If I
then hit the destroy button, the windows are removed and the __del__
method of each child is called ok, but if I look at memory usage using
'Top' I see that no memory is released?
However - if I increase the dummy-data count  from  30000 to 40000 or
above, the memory IS released.

Is this ok or might it be giving me a problem?

Thanks for any help

Nick/Oxford

---------code-------------(probably mangled)
#!/usr/bin/python

import sys,Tkinter

class Child(Tkinter.Frame):
    def __init__(self, window):
        self.w = window
        Tkinter.Frame.__init__(self,self.w)
        self.data = [1.23456789]*30000     #<<<<----dummy data to eat
memory
        self.l = Tkinter.Label(self.w,text='hello')
        self.l.pack()

    def __del__(self):
        print 'see delete request ',self

class Parent:
        def depopulate(self):
                for each in self.childlist:
                        each.w.destroy()
                self.childlist = [0]*self.count


 def populate(self):
         for a in range(self.count):
                 new_window = Tkinter.Toplevel(self.W)
                 self.childlist[a] = Child(new_window)

 def __init__(self, W, n=10):
         self.count = n
         self.childlist = [0]*self.count
         self.W = W
         self.b1 =
Tkinter.Button(self.W,text='destroy',command=self.depopulate)
         self.b1.pack(side='right')
         self.b2 =
Tkinter.Button(self.W,text='build',command=self.populate)
         self.b2.pack(side='right')


W = Tkinter.Tk()

test = Parent(W,50)

W.mainloop()

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







From damonNOdaSPAM at unispace.com.invalid  Tue Jun 13 21:46:52 2000
From: damonNOdaSPAM at unispace.com.invalid (damonr)
Date: Tue, 13 Jun 2000 18:46:52 -0700
Subject: Python Sites
Message-ID: <1c99a8ea.c6f3e45c@usw-ex0103-023.remarq.com>

Heya,

I just added a site directory for Python related sites to my site
(http://www.devcritic.com) and was wondering if you guys had any
sites to add, either by using the "add a site" link on the menu
on the side or recommend them here and i will add them. The url
is: http://www.devcritic.com/sites/Python/ .

Thanks in advance

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From rhicks at nospam.rma.edu  Tue Jun  6 08:19:11 2000
From: rhicks at nospam.rma.edu (OPUS)
Date: Tue, 06 Jun 2000 12:19:11 GMT
Subject: Memory leak: assigning function object to instance variable
References: <8hh0pf$i7$1@phys-ma.sol.co.uk> <slrn8joise.ik.nascheme@cranky.arctrix.com>
Message-ID: <3%5%4.54$vh.18462@news>

Where is it??? Does it work with 1.5.2?

"Neil Schemenauer" <nascheme at enme.ucalgary.ca> wrote in message
news:slrn8joise.ik.nascheme at cranky.arctrix.com...
: John Parkey <jparkey at nospamrapidbs.com> wrote:
: >I have registered this bug, with code fragment below, but wondered if
: >anybody else was falling foul of memory leaks of this type.
:
: The latest version of my GC patch will prevent this leak.
: Hopefully it will be an optional feature in 1.6.
:
:     Neil
:
: --
: The internet: Learn what you know.  Share what you don't.




From tomsantos at email.com  Sun Jun 18 15:49:01 2000
From: tomsantos at email.com (Tom Santos)
Date: Sun, 18 Jun 2000 12:49:01 -0700
Subject: Getting Password from User
Message-ID: <8ij9kr$qa5$1@bob.news.rcn.net>

I'm trying to get a username and password from a user so that I can create
an FTP connection.  I'm using raw_input() to get the username but it's not
acceptable to get their password.  If I use raw_input() to get the password,
the user's password is echoed to the screen.  Is there any way to make it so
that their password is not shown entirely, or characters like '*' are
displayed instead of the characters in their password?

Thanks,

-Tom-

tomsantos at email.com





From kens at sightreader.com  Tue Jun 20 18:47:02 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Jun 2000 15:47:02 -0700
Subject: strings and integers?
References: <8ir75b$qfi$1@news2atm.raleigh.ibm.com>
Message-ID: <394FF466.51A5B77B@sightreader.com>

"Arint?" wrote:

> Is
> x = "txt %(somevar)"%vars()
> the only way to append text to a string?

No.  You can also use the + operator.

  >>> a = "abc"
  >>> b = "def"
  >>> c = a+b
  >>> print c
  abcdef

Also, your syntax for % is wrong.  Reread the chapter on %.

  >>> print "a = %s" % a
  a = abc

  >>> print "1+1 = %d, for sufficiently large values of 1." % 3
  1+1 = 3, for sufficiently large values of 1.

> Also, why do I get a len() of unsized object errors?  Do I need to rebuild
> the python dlls? Download 1.6?

No.  It is very unlikely that you have found a python bug.  Show me an
example.  When asking a question, always show an example.

The len() function can only be applied to things that have length (number of
items), such as strings, lists and tuples.  The len() function is /not/ the
same as the C function sizeof.

>>> len("hello world.")
12
>>> len([1,2,3,4,5,6])
6
>>> len(5)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: len() of unsized object

> Thanx

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000620/6a8b95d0/attachment.vcf>

From jwbnews at scandaroon.com  Wed Jun 28 00:14:15 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Tue, 27 Jun 2000 21:14:15 -0700
Subject: Guesstimate for 1.6 beta timeline?
References: <Cid65.2526$h8.109632@news-west.usenetserver.com> <39597711.960D7169@bellatlantic.net>
Message-ID: <jwbnews-A82046.21141527062000@news.olympus.net>

In article <39597711.960D7169 at bellatlantic.net>, 
sholden at BellAtlantic.net wrote:

> Would "real soon now" be any help? :-)

Actually, I'd prefer something before "real soon now" as that is so 
often used.  ;-)

But I'll gratefully accept whatever happens.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From m.faassen at vet.uu.nl  Wed Jun  7 19:48:14 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 23:48:14 GMT
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain>
Message-ID: <8hmmvu$ets$2@newshost.accu.uu.nl>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
> And now my own rant in Spanish:

> Python era un lenguaje casi perfecto, al menos hasta que empezaron
> a meterle tonterias sintacticas. Ahora estoy empezando a ver cual
> es el problema de python: los holandeses.

Python's virtually perfect, but it does have a problem: those Dutch
folks.

> Al parecer estan tan
> excitados con la idea de python que no pueden evitar soltar la
> parida en Holandes. La verdad es que los holandeses son muy imaginativos:
> - Las cafeterias de drogas

Ah, but they're just for drinking coffee..

> - El trafico de diamantes ( De Beers estara detras de Sierra Leona)

De Beers isn't a Dutch company; it's South African:

http://www.edata.co.za/debeers/

> - Tomates trangenicos. 
> ...

While they're chemical waterballoons, I didn't think they were genengineered
just yet -- that'd be the Evil Americans. :)

> Bueno, ya en serio, si realmente no se pueden aguantar, que 
> creen un Dutch.comp.python.org. 

So since I disproved all your evil allegations, there are no obstacles in
our way: Dutch should be the official language of this newsgroup. :)

Groeten,

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


From millja at my-deja.com  Thu Jun 15 18:05:08 2000
From: millja at my-deja.com (millja at my-deja.com)
Date: Thu, 15 Jun 2000 22:05:08 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com> <8i6u39$re3$1@nnrp2.deja.com> <8i8le9$b6u$1@nnrp1.deja.com> <8i95iq$o3a$1@nnrp1.deja.com>
Message-ID: <8ibju8$i13$1@nnrp1.deja.com>

Here is an example of what I mean.


<SCRIPT RUNAT = SERVER LANGUAGE=Python>
def Test4(number):
	return number + 3, number + 4
</SCRIPT>
Our numbers of the day are
<%
=Test4(10)
%>.


I get a Type Mismatch, An unhandled data type was encountered error.

thanks
jason


In article <8i95iq$o3a$1 at nnrp1.deja.com>,
  millja at wwc.edu wrote:
> Yup that works!  Thanks.  I now have another question.
>
> I need to (well, would like to) have my Python function returns a list
> (really just an array of integers).  VBScript dosnt seem to like this
> very much and throws up an invalid data type error.  I tryed messing
> around with VBSs array() function but thats not showing me any love.
> Anyone have ideas?
>
> thanks
> jason milliron
>
> btw, its freakn awesome to have mark hammond reply to your post :)
>
> In article <8i8le9$b6u$1 at nnrp1.deja.com>,
>   Mark Hammond <mhammond at skippinet.com.au> wrote:
> > In article <8i6u39$re3$1 at nnrp2.deja.com>,
> >   millja at my-deja.com wrote:
> >
> > > would like to integrate.  Is there some way to
> > > call this function from within the VBScript?
> > >  Or, it would work if I can get numbers,
> > > generated by the VBScript, into the Python
> > > function and the results of this function back to
> > > the VBScript.
> >
> > VB should be able to call any Python function defined on the page,
> > simply by calling it by name.
> >
> > Python should be able to call any VB function defined on the page
with
> > some magic syntax that Im not sure of :-)
> >
> > Under IE, you do it via "Window.VBFunc()".  I assume that under ASP,
> it
> > will be via the server or document object - most likely the
document -
> > eg "Document.VBFunc()"
> >
> > Feel free to post a sample once you get it working :-)
> >
> > Mark.
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From daniel.dittmar at sap.com  Wed Jun  7 04:42:53 2000
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Wed, 07 Jun 2000 10:42:53 +0200
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net>
Message-ID: <393E0B0D.F54CB7B6@sap.com>

Have a look at PIDDLE: http://piddle.sourceforge.net/

this allows you to draw on Tkinter, wxPython, GIF/PNG, PDF ...

-- 
Daniel Dittmar
SAP AG, Basisentwicklung Berlin
daniel.dittmar at sap.com


From petersc at stud.ntnu.no  Wed Jun 14 17:54:50 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 14 Jun 2000 23:54:50 +0200
Subject: Python and Video for Windows?
References: <39476A9C.DFE84D0@stud.ntnu.no> <8i8mlg$c84$1@nnrp1.deja.com>
Message-ID: <3947FF2A.21F2FAE2@stud.ntnu.no>

Mark Hammond wrote:
> 
> Pythonwin has a demo that uses the MS ActiveMovie (or whatever it is
> called) control.  This can play arbitary media clips from within
> Pythonwin.  By default, it tries to play the first .avi file it finds
> in your Windows directory.
> 
> Im not sure if that is what you were asking, tho!

Unfortunately it is not. My concern is capturing the image data
into Python. I don't care about saving or loading to a file,
I don't care about which OS I have to use (as long as I get
the USB interface to work) etc.

Basically I just want to have a VFW frame grabber wrapper for
Python. The problem is that I don't know anything about VFW
or Windows system programming.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From moshez at math.huji.ac.il  Sat Jun 10 18:09:08 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 11 Jun 2000 01:09:08 +0300 (IDT)
Subject: PATCH: Augmented assignment 
In-Reply-To: <8457258D741DD411BD3D0050DA62365907A202@huina.oceanic.com>
Message-ID: <Pine.GSO.4.10.10006110106360.10168-100000@sundial>

[Moshe Zadka] 
> Anyone doing *that* isn't  familiar with the pydiom:
> 
> s, i = nested_strucre[some_index], another_index
> s[i] = s[i] + 1

[Doug Stanfield]
> Wow, I like that.  Its great!!!
> 
> Could you pass on the link to that storehouse of great pydioms.
> I'd like to see more  ;-Q

Hmmmm....just read a lot of Python source code <wink>
Seriously, though, I learned a lot of Python from the standard library, 
and the Demo/ and Tools/ directories of the source distribution.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From bhahn at spam-spam.g0-away.com  Fri Jun 16 12:50:22 2000
From: bhahn at spam-spam.g0-away.com (Brendan Hahn)
Date: Fri, 16 Jun 2000 09:50:22 -0700
Subject: Learning curves (was: Python so far)
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu> <39498084.972E36D2@san.rr.com> <1ecb5va.1nch8nwz2t5siN%bparsia@email.unc.edu>
Message-ID: <bhahn-ya02408000R1606000950220001@news.corp.hp.com>

bparsia at email.unc.edu (Bijan Parsia) wrote:
>Er...the learning curve claims always confuse me since I never know what
>the axes measure.

X is time elapsed, Y is knowledge acquired.  So the common usage of "steep"
with regard to learning curves is in fact exactly backwards.  Too late now,
though.

bhahn at transoft.mmangle.net   <-- unmangle address to reply


From wjdandreta at worldnet.att.net  Mon Jun 19 21:51:54 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 20 Jun 2000 01:51:54 GMT
Subject: Python module for Dbase files?
Message-ID: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net>

Is there a Python module that has functions in it to manipulate Dbase tables
and indexes?

Bill




From thomas at xs4all.net  Sat Jun  3 18:08:36 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 4 Jun 2000 00:08:36 +0200
Subject: memory leak on solaris 2.5.1 / 2.6 ?
In-Reply-To: <3935429D.9C0C6491@marconicomms.com>; from Wolfgang.Grafen@marconicomms.com on Wed, May 31, 2000 at 06:49:33PM +0200
References: <3935126D.32AFBAA1@marconicomms.com> <3dln0qx271.fsf@amarok.cnri.reston.va.us> <jtya4qaitm.fsf@wazor.biostat.wisc.edu> <3935429D.9C0C6491@marconicomms.com>
Message-ID: <20000604000836.R469@xs4all.nl>

On Wed, May 31, 2000 at 06:49:33PM +0200, Wolfgang Grafen wrote:
> William Annis wrote:

> >         Right.  Solaris assumes that if you ask for a chunk of memory,
> > you are probably going to ask for it again later, even after a free().

> What is it for? After consuming the fast silicon main memory the
> process slows extremely down when swapping to disk although
> it could demand freed memory!

The memory that is malloc()ed and later free()ed (wether it's been used or
not, on most UNIX systems, or only if it's used, on systems that use GNU
libc (or at least I assume)) is marked as free, but stays part of the
process's memory segment. Later calls to malloc() that fit somewhere in the
unused parts of the memory segment (the 'heap') do not cause another memory
allocation.

The difference my seem strange, but comes from the way the UNIX kernel and
the C runtime (libc) share responsibilities. malloc() is a library call, and
does a lot of bookkeeping for the programmes' sake. The logic behind
malloc() is too complex for the average UNIX kernel; the kernel has a very
simplified take on processes: each has a stack, for automatic variables and
call frames, and a heap, for all other kinds of memory allocation. The heap
is just a block of memory, which you can grow using brk() or sbrk().

This is very nice for the kernel, because it has a lot less bookkeeping to
do, but not very nice for the programmer, who has a lot more bookkeeping to
do. So that's where the C runtime comes in, simplifying the programmers task
by allowing a simple malloc()/free() system. libc keeps a list of all
malloc()ed memory, increases the heap in size if it needs to, and does its
best to do that as little as possible.

And let's thank the gods for malloc()... Managing memory is hard enough
*with* malloc, let alone without it ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From fgbrown at ionet.net  Tue Jun 13 01:21:52 2000
From: fgbrown at ionet.net (F. G. Brown)
Date: Tue, 13 Jun 2000 00:21:52 -0500
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
Message-ID: <8i4g66$3f0$1@ionews.ionet.net>

Jim,
I'm as newbie as they come, but I would have to say that there is nothing
wrong with
your formula.  If you are using just the basic Python( 3 hacks >>> on a
black screen) and have saved the program where Python knows where it's at,
such as by the path or pythonpath statement (not sure how this would work in
NT (argh! newbie)) it should work.  I cut and pasted your lines and saved
them to a file called rad.py then at the Python command line did a "import
rad".  The question came up "Enter is the radius:" I said 22 and it said
"138.2 yada yada.
Sorry about the ramble, but hope I helped.  Everyone here has been helpful
to me, I had to jump at the opportunity.

have fun,
Fred

"jim roush" <jar at mminternet.com> wrote in message
news:q9abks07oq5q327ijvf6ejgf91kh5o5aci at 4ax.com...
> I'm just learning Python.  The following code does not work.  It's
> supposed to calculate the circumference after the user enters the
> radius.  The program displays the prompt, but when I enter the radius
> and the press enter, nothing happens.
>
> FYI, I'm using Python 1.5.2 on WinNT.
>
> pi = 3.141592654
> radius = input("Enter is the radius: ")
> circ = 2 * pi * radius
> print circ
>
>
> jar at mminternet.com
> James A Roush




From PClaerhout at CREO.BE  Fri Jun 30 14:24:00 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Fri, 30 Jun 2000 20:24:00 +0200
Subject: Web server executing Python scripts
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130125E@msgeuro1.creo.be>

Hello all,

I'm working on a web based system, which consist of a whole
bunch of Python scripts, which connect to a bunch of machines
on a network (either using ethernet or dailup networking) and
execute some to code to for example get things from the
windows registry of those machines.

What do you think would be the best idea?
- Use whatever webbrowser with CGI-scripting
- Use MS IIS with ASP (and Python)
- Use Zope to build the whole system?

It should also connect to a database running on the host.


Pieter



From root at rainerdeyke.com  Fri Jun  9 18:25:37 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 09 Jun 2000 22:25:37 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB333@ALTNET> <8hr8fj$fu$1@sunnews.cern.ch>
Message-ID: <B9e05.1145$vp3.81025@news1.rdc2.tx.home.com>

Wim Lavrijsen <wlav at atlas03.cern.ch> wrote in message
news:8hr8fj$fu$1 at sunnews.cern.ch...
> All that STL containers require is that your dtor and copy assignment
> don't throw. The compiler generated members fullfill that requirement.

This statement is both incorrect and misleading.  Incorrect because there
may be other requirements: for example, std::set (along with several others)
requires that operator< and operator== are defined for the object.
Misleading because for many classes the compiler generated functions cause
incorrect behavior.


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




From aahz at netcom.com  Tue Jun 27 17:36:07 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 21:36:07 GMT
Subject: Tutorials at IPC9
References: <39591877.535A96BC@sage.att.com>
Message-ID: <8jb6o7$us1$1@slb7.atl.mindspring.net>

In article <39591877.535A96BC at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>
>I'm starting to think about what kinds of tutorials we should have at
>the next Python conference.  The last conference had a number of great
>tutorials available (see http://www.python.org/workshops/2000-01/tut/
>for descriptions).  But I'm wondering what, if anything, we should
>change this year.  So I thought I'd try and get a sense of people's
>thinking on this.  If you were at IPC8, what did you like, or dislike,
>about the tutorials you attended?  If you skipped the tutorials, what
>might have tempted you to attend?  Would you prefer more tutorials
>covering different subjects?  Or fewer, covering things in more
>depth?

To what extent should people not at IPC8 involve themselves in this
discussion?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From petersc at stud.ntnu.no  Mon Jun 12 12:48:52 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Mon, 12 Jun 2000 18:48:52 +0200
Subject: Testing for a file in python?
References: <39450724.231994953@nntp.interaccess.com>
Message-ID: <39451474.7467C602@stud.ntnu.no>

"Thaddeus L. Olczyk" wrote:
> 
> stat doesn't work, it aborts if the file does not exist.

you could always catch the exception. But why not use
os.path?

>>> import os.path
>>> os.path.exists("python/")
1
>>> os.path.exists("spam/")
0
>>> os.path.exists("python/patches/submitted/array.pop-extend.patch")
1

If you only want a true for regular files have a look at isfile():

>>> os.path.isfile("python/")
0
>>> os.path.isfile("spam/")
0
>>> os.path.isfile("python/patches/submitted/array.pop-extend.patch")

for more info or other function look at:
http://python.org/doc/current/lib/module-os.path.html

good luck
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From wlav at hpatl26.cern.ch  Fri Jun  9 10:26:07 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 9 Jun 2000 14:26:07 GMT
Subject: Python Productivity over C++
References: <m130PPw-000wcEC@swing.co.at>
Message-ID: <8hqupv$pb0$1@sunnews.cern.ch>

tanzer at swing.co.at (Christian Tanzer) writes:
>"Frank V. Castellucci" <frankc at colconsulting.com> wrote:
>> 1. There are gc, SmartPointer, heap management, etc. class libraries out
>> there. =

>Guess what. Something like SmartPointer was one of the first classes I
>implemented. In C++, that doesn't relieve you of thinking about memory
>management, though.

As I understand it, Python objects are simply reference counted. So, a
pervasive use of smart pointers in C++ will give the exact same effect
as in Python.

>> 2. There is the type information in standard conforming compilers.

><rant>
>If you happen to use a standard conforming compiler. In 1997, there

If 1997 was the last time you touched C++, it explains a lot about your
arguments.

>were still several compilers that didn't even fully conform to the
>1990 edition of Stroustrup und Ellis's `The Annotated C++ Reference
>Manual'.

></rant>

>RTTI doesn't offer any mechanisms remotely approaching Python features
>like `__getattr__'/`__setattr__' though. Not to speak of `eval' and
>`exec'. =

If you wish to have such in C++, use a C++ interpreter, such as ROOT/CINT.

>> 3. There is, has been, and will be more, C++ class Class libraries which
>> are powerful, easy to use, and free. I was using NIHs Class class about
>> 5 years ago.

>So what? Your pains start as soon as you want to combine libraries
>from different sources which happen to use incompatible idioms (like
>overloading global operators, ...).

If programmers (library vendors in this case) mess up, you will always
have problems, in any language. I've seen several messages on this platform
about problems with modules redefining variables from other modules. Why
is that different than a library vendor who is so erratic to use global
operators? 

>> Define 'orthodox canonical class form'

>J.O. Coplien's `Advanced C++ Programming' will tell you all about it.

Also being to only book which advocated the canonical form. Compilers
do a reasonable job by providing the cctor, assignment copy and dtor.
(I worry about efficiency, so I write my own cctor and assignment copy
operators, see: http://www.lavrijsen.net/ctotw/ctotw21.html)

Best regards,
      Wim Lavrijsen


From denalione at my-deja.com  Thu Jun 29 10:28:22 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Thu, 29 Jun 2000 14:28:22 GMT
Subject: exiting threads
References: <8jfcf9$2rp$1@nnrp1.deja.com> <8jfiuj$lf1$1@slb1.atl.mindspring.net>
Message-ID: <8jfmdf$ah6$1@nnrp1.deja.com>

In article <8jfiuj$lf1$1 at slb1.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <8jfcf9$2rp$1 at nnrp1.deja.com>,
> Dale Burnett  <denalione at my-deja.com> wrote:
> >
> >I am using thread.start_new_thread(self.foo,(x,y))
> >
> >Do I have to explicitly exit the thread or does it automatically exit
> >when foo returns?
>
> I believe it does, but I strongly recommend that you subclass
> threading.Thread instead.
> --
>
why?  because of the greater functionality of threading or is there
something wrong with thread?


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


From darrell at dorb.com  Sun Jun  4 19:07:26 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 4 Jun 2000 16:07:26 -0700
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com> <39398EEE.C18EC987@tismer.com> <3939903D.79BD8D2D@san.rr.com> <393AA136.36C34915@tismer.com> <393AAC0F.FFB4B65@san.rr.com> <7sbljso0tvth02de6ql5at6100pns97f3h@4ax.com> <393AB2B8.1D1B3A87@san.rr.com>
Message-ID: <034301bfce79$a86306d0$6401a8c0@home>

// formatFlag is optional and defaults to 0
 int formatFlag=0;
 if (PyArg_ParseTuple(args,"s#|i",&bufStr, &bufSz, &formatFlag)==0)return
NULL;


--Darrell Gallion

From: "Courageous" <jkraska1 at san.rr.com>
> BTW, I was trying to figure out how to pass 'optional' args
> to external C functions and couldn't grok it. Anyone know how?
> (I wanted pqueue and dlist to default to 10 or something if no
> n was offered).





From aahz at netcom.com  Tue Jun  6 15:54:13 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Jun 2000 19:54:13 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net> <393C81D0.36732896@Lugoj.Com> <slrn8jql5v.6d4.wtanksle@dolphin.openprojects.net>
Message-ID: <8hjkt5$p4b$1@slb6.atl.mindspring.net>

In article <slrn8jql5v.6d4.wtanksle at dolphin.openprojects.net>,
William Tanksley <wtanksley at bigfoot.com> wrote:
>
>I'm not the one that's claiming that removing case-sensitivity will ruin
>Python.  My claim is very simple: first, it's a minor change for us to
>make, and second, it helps newbies.

I disagree with both of your assertions.  I don't think it's that minor
of a change, and I *do* think that case sensitivity helps newbies *read*
code.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From kens at sightreader.com  Thu Jun 15 23:58:19 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:58:19 -0700
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com> <20000615195014.A12612@fuf.sh.cvut.cz> <20000615200557.K26436@xs4all.nl>
Message-ID: <3949A5DB.5D4ECFCA@sightreader.com>

Actually C++ doesn't have true access control either.  Just insert before your
includes:

#define private public

. . . well, almost ;-)

Thomas Wouters wrote:

> On Thu, Jun 15, 2000 at 07:50:14PM +0200, Michal Vitecek wrote:
>
> [ access control ]
>
> > >Look at the big picture.  Do encapsulation, access control and type
> > >strictness really make C++ really less error prone than python?
>
> > >I think not.
>
> >  but what are the method names with '_' at the beginning in modules for?
> >  aren't they there as a kind of hack to implement some way of access
> >  control?
>
> No, they are a form of reducing namespace clutter and name-clashing.
>
> Variable names starting with a '_' do not get imported into the local
> namespace when you use 'from <module> import *'. They are accessible just
> fine, they just do not get bound to a *local* (to the 'from module import'
> statement) name.
>
> Class variables (including methods) starting with a double _ get 'mangled'
> to make them private -- so you can store 'class-related' data without the
> risk of a subclass accidentily shadowing it. The class-related data is still
> very accessible, it's perfectly valid to say 'self._BaseClass__private_data'
> or some such.
>
> --
> Thomas Wouters <thomas at xs4all.net>
>
> Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/c0e17f7c/attachment.vcf>

From erno at iki.fi  Sat Jun 24 14:42:32 2000
From: erno at iki.fi (Erno Kuusela)
Date: 24 Jun 2000 21:42:32 +0300
Subject: Classes: What is the standard trick for ...
References: <395235F9.94C0BBA1@erols.com>
Message-ID: <kuvgyyvs2v.fsf@lasipalatsi.fi>

>>>>> "Edward" == Edward C Jones <edcjones at erols.com> writes:

    Edward> I would like a function call to look like
    Edward> "function.f1()". I would like to spare the user the need
    Edward> to create an instance of the class. Can I do this? How?

you could make it a module (in a package, perhaps). or you could
create the instance yourself, for example like this..

class _functions:
        def f1(self):
                return 'spam'

functions = _functions()

now the users of your module can call yourmodule.functions.f2()
and get spam without needing to instantiate anything themselves.

  -- erno






From gmcm at hypernet.com  Mon Jun 26 19:47:53 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 26 Jun 2000 23:47:53 GMT
Subject: Questions about Installer/Builder/Standalone
References: <p9R55.17984$Yr4.303276@news1.rdc1.tn.home.com>
Message-ID: <8F5FCDEB5gmcmhypernetcom@199.171.54.194>

Mike Callahan wrote: 

>I am playing with Gordon McMillan's Installer 0.3f for Windows and have
>gotten it to work using freeze, simple, and standalone with non-Tkinter
>apps. Good job Gordon! I can make a Tkinter app using simple but I
>cannot get it to work using standalone. I know that simple does some
>magic to get the app to use the local Tk it installs and Gordon talks
>about it in the docs. But I cannot figure out what the magic is. 

Two pieces of "magic". Look in the gen_install.py from Simple. There's a 
snippet of code there that unpacks the bundled up tcl/tk:

installutils.installCArchive(os.path.join(idir,'TCLlib.pkg'), idir, 
'lib/tcl')

Basically, this is unpacking TCLlib.pkg into lib/tcl (relative to where 
you're installing). That happens once. Somehow you need to get 
InstallShield to do it for you.

Then, each time you run, there's some code that runs first (which you'll 
find in projects/BuildTK/runTK.py). All it does is set environment 
variables so that _tkinter will find the private tcl/tk stuff installed 
above. If you cut&paste from runTK to the top of your main script, it 
should run fine.

- Gordon


From mhammond at skippinet.com.au  Sat Jun 10 00:00:39 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 04:00:39 GMT
Subject: Does anyone has Mark Hammonds's IPC-8 presentation?
References: <R%h05.121727$R4.735202@news1.rdc1.nj.home.com>
Message-ID: <H3j05.832$cQ.10148@news-server.bigpond.net.au>

It should all be back at starship now...

Mark.

"S. Hoon Yoon" <cpu at bigfoot.com> wrote in message
news:R%h05.121727$R4.735202 at news1.rdc1.nj.home.com...
> Hi,
>
>   I am trying to implement and use Events in win32com. I found that the
> example is inside the new IPC-8 presentation by Mark (thanks); however,
the
> site (Starship) has not been recovered and I cannot get to the
explaination.
> If anyone has a copy, pls send it to me to this e-mail hyoon at home.com.
>   Thanks much in adv,
>
> Hoon,
>
>




From mfletch at tpresence.com  Wed Jun  7 03:17:15 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 7 Jun 2000 03:17:15 -0400 
Subject: Network byte ordering question ...
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3397@VRTMAIL>

Strange, I was under the impression that network byte ordering was '>', with
'<' being a little-endian (non-network) encoding.  You sure your headers are
really network ordered?

>>> import struct
>>> struct.pack( '>i', 1 )
'\000\000\000\001'
>>> # note the big-endian (network) encoding

Don't know if there are any _current_ Waterloovian Pythonistas, the two I
knew about (Paul and myself) having both graduated a while ago... and not
even an alumni newsletter to show for it in my case ;) .  Enjoy,
Mike

-----Original Message-----
From: istevens at calum.csclub.uwaterloo.ca
[mailto:istevens at calum.csclub.uwaterloo.ca]
Sent: Tuesday, June 06, 2000 8:08 PM
To: python-list at python.org
Subject: Re: Network byte ordering question ...


In article <CC59868A0A76D311B1F50090277B321C1F3394 at VRTMAIL>,
Mike Fletcher  <mfletch at tpresence.com> wrote:

>( val1, val2, val3, val4, val5 ) = unpack( ">16sBBBi", header )

Thanks!  Actually, the following works better:

    ( val1, val2, val3, val4, val5 ) = unpack( "<16sBBBI", header )

I tried this both on an Intel and a Sun machine and they both return
what I expect.  I also forgot to mention that the last value was
unsigned.

>Yay! another Waterloovian Pythonista!

Actually, ex-Waterloo ... graduated last year.


thanks again,
ian.
-- 
http://www.python.org/mailman/listinfo/python-list



From rcc at nospamthanks_jennic.com  Mon Jun 12 10:28:15 2000
From: rcc at nospamthanks_jennic.com (Robert Cragie)
Date: Mon, 12 Jun 2000 14:28:15 GMT
Subject: string.atoi() doesn't do negative base16 - why??
References: <8i2c9l$b0r$1@news1.wire.net.au>
Message-ID: <newscache$pnp1wf$x97$1@jenpc07.jennic.com>

Steven Adams <adams_s at lab.eng.usyd.edu.au> wrote in message
news:8i2c9l$b0r$1 at news1.wire.net.au...
| Hi,
|
| my quick hack for going back and forth between binary and base 10, 16 is
| backfiring a bit..
|
| I can form the values as a string, BUT, string.atoi() can't convert
negative
| base hexadecimal strings to integers.
|
| does that make sense?
| e.g
| >>> import string
| >>> hex(-20)
| '0xffffffec'
| >>> string.atoi('0xffffffec',16)
| Traceback (innermost last):
|   File "<pyshell#28>", line 1, in ?
|     string.atoi('0xffffffec',16)
| ValueError: atoi() literal too large: 0xffffffec

Try

string.atoi('0xffffffec',0)

Yes, I know it's inconsistent :-)

| why is this?

Don't know - I've had lots of fun and games with this.

Robert Cragie




From eviltofu at rocketmail.com  Thu Jun 15 22:54:11 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Fri, 16 Jun 2000 02:54:11 GMT
Subject: Extending Python with Obj-C on MacOSX(server)?
Message-ID: <eviltofu-D90BAE.22541415062000@news.earthlink.net>

Has anyone any examples of how I extend Python with Apple's Cocoa layer?


From n at n.com  Sat Jun 24 15:43:01 2000
From: n at n.com (Arinté)
Date: Sat, 24 Jun 2000 19:43:01 GMT
Subject: Java Objects vs Pythons?
References: <iUL45.23838$ds.694847@newsread2.prod.itd.earthlink.net> <8F5CEEDFAgmcmhypernetcom@199.171.54.195>
Message-ID: <9b855.28036$ds.808114@newsread2.prod.itd.earthlink.net>

> Without threads, none of these will do you any good. I guess you want
> somebody to block while "waiting", but who (C++ or Python) or what
> "waiting" means in this context, I don't know...
>
I want the script to ask the user for a certain action and until that user
does  that action I want the script to halt.  Then when the user does the
action the c++ app will catch the action and thru a callback tell the python
script to continue on (I guess would be the best way).

thanx




From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Tue Jun 13 08:44:24 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Tue, 13 Jun 2000 05:44:24 -0700
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
Message-ID: <2af14c7e.01880953@usw-ex0104-028.remarq.com>

Jim,

It works fine for me on Windows - except in IDLE. So either run
it from the command prompt or from PythonWin.

In PythonWin you'll get the result back to the interactive
window, which you may not be expecting.

Richard

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



From aahz at netcom.com  Sun Jun 18 00:49:33 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jun 2000 04:49:33 GMT
Subject: string.join question (was Re: urllib question)
References: <Pine.LNX.4.10.10006171424430.19831-100000@mail.fizzylab.com> <8igv24$5li$1@slb7.atl.mindspring.net> <394C4FF9.46B011DC@fizzylab.com>
Message-ID: <8ihkct$qao$1@nntp9.atl.mindspring.net>

In article <394C4FF9.46B011DC at fizzylab.com>,  <paul at fizzylab.com> wrote:
>
>u = urllib.urlopen(string.join(["http://cnn.com/","index.html"]))
>print u.read()
>u =
>urllib.urlopen(string.join(["http://some.host.com/","monstrousQueryString?withParameters"]))

Hmmmm....  string.join(["a","b"]) == "a b".  Perhaps you instead want
string.join(["a","b"], '')?  I have no idea why your first example
works.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From db3l at fitlinxx.com  Tue Jun 13 16:01:59 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Jun 2000 16:01:59 -0400
Subject: Client with select.select()
References: <#F2#YuI1$GA.106@net025s> <8i36rt$42mq$1@nntp6.u.washington.edu> <emt7cXW1$GA.180@net025s>
Message-ID: <uln09mjt4.fsf@ctwd0143.fitlinxx.com>

"A[r]TA" <arta at NOSPAMx-stream.nl> writes:

> But if you want to write a telnet-client. The server requests for data,
> isn't he?
> Then you can check with the write-set and send some data.
> Or am I missing something?

A telnet server never explicitly "requests" data from a telnet client
- it's always ready to receive data.  I suppose the initial options
negotiation could be considered a little more of a request since a
client does generate a response to the information being presented by
the server, but there are no such requests during normal data flow.
Once a telnet client has connected to a server, it is free to send
data at any point in time.

Now, it is possible that if a server is unresponsive, or slow, or if
there is network congestion, that the client may build up a queue of
stuff still waiting to be sent by the network layer on the client
machine.  So it can still be useful to use select() to decide if the
underlying network layer is ready to receive more data.

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


From shouldbe at message.com  Thu Jun 29 09:55:34 2000
From: shouldbe at message.com (Arinté)
Date: Thu, 29 Jun 2000 13:55:34 GMT
Subject: stderr and stdout
Message-ID: <qzI65.16987$_b3.366473@newsread1.prod.itd.earthlink.net>

I have python embedded in an app, so I need to redirect stdout and stderr
somewhere else:
    IOmod = PyImport_ImportModule("cStringIO");
    obj = PyObject_GetAttrString(IOmod, "StringIO");
    errorIO = PyObject_CallFunction(obj, NULL);
    stdIO = PyObject_CallFunction(obj,NULL);
    Py_DECREF(obj);

    /* redirect sys.stderr to a cStringIO instance */
    PySys_SetObject("stderr", errorIO);
    PySys_SetObject("stdout", stdIO);

Is there a better way to do this, especially so the output would be
synchronized?  Right now I print all the stdout then I print all the stderr
and that makes it hard to figure out which line in stdout came before such
and such a line in stderr.

Oh I print stdout and err like so:
   char *pix = PyString_AsString(temp);
   Py_XDECREF(temp);
   appcallback()->status(pix);
where temp is either errorIO or stdIO.





From samwun at yahoo.com  Sat Jun 10 23:44:27 2000
From: samwun at yahoo.com (Sam Wun)
Date: Sun, 11 Jun 2000 13:44:27 +1000
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21> <20000610203442.A497@better.net>
Message-ID: <39430B1B.33469112@yahoo.com>

doesn't compiled correctly, the _mysqlmodule.so is not found..
do you know how to fix this problem?

Thanks
sam.
William Park wrote:

> On Sun, Jun 11, 2000 at 12:18:44AM +0000, dana_booth wrote:
> > I've asked this before, but never did get a good response, so I hope that
> > one of you reading this might be able to help...
> >
> > The MySQLmodule1.4 does not compile properly on any platform. I've tried
> > removing the '&'s on the three lines which puke during compile. On
> > UNIX-like's Python won't make install when having done that, on Windows,
> > it'll compile, but Python chokes when trying to import.
> >
> > I've even tried using mSQL and PostgreSQL, on which the modules compile
> > fine, but I need to use MySQL, the only backend whos Python module seems to
> > be flawed in some way. :)
> >
> > Anyway do any of you know of, or have made yourself, a module for MySQL?
> >
> > --
> >
> > -----
> > Dana Booth <dana at oz.net>
> > Tacoma, Wa., USA
>
> There are 2 modules for MySQL:
>     - MySQLmodule-1.4 (old -- must remove 3 '&')
>     - MySQLdb-0.2.1 (new -- no problem)
> Both compiled and installed on Slackware Linux.
>
> --William



From aahz at netcom.com  Wed Jun 28 09:26:05 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 28 Jun 2000 13:26:05 GMT
Subject: long lines, long string
References: <j9tcj8.i0b.ln@iiab.soluma.de>
Message-ID: <8jcudd$gdk$1@slb7.atl.mindspring.net>

In article <j9tcj8.i0b.ln at iiab.soluma.de>,
Holger M. F??ler <fuessler at solutec.de> wrote:
>
>Is there a possibility to put long lines of python-source-code in multiple
>lines (for better clarity)?

myDict = {
  'foo': 1,
  'bar': 2,
  'baz': 3
  }

if myDict[xyzzy] == 'foo' \
  or myDict[xyzzy] == 'bar' \
  or myDict[xyzzy] == 'baz' :

>How can I accomplish the same with strings?

"""
This
is 
a 
long
string
"""
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From shapr at uab.edu  Sat Jun 17 21:35:34 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sun, 18 Jun 2000 01:35:34 GMT
Subject: dpkg-python??
References: <khlii8.j62.ln@127.0.0.1>
Message-ID: <394C27EE.1FEA81BD@uab.edu>

Fermin Martinez wrote:
> 
> well, does anybody knows where could I find documentation about this module???
> It doesn't seems to exist anything...

are you searching for the module? or the documentation?
If you're running a debian system, you just need to 'apt-get install
dpkg-python'

If you're searching for documentation, I don't know :(
Probably documentation on the Debian package system and the source code
for dpkg-python would do it.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From it at onestopent.com.au  Tue Jun 13 01:49:40 2000
From: it at onestopent.com.au (Jason)
Date: Tue, 13 Jun 2000 15:49:40 +1000
Subject: can python do this?
Message-ID: <3945c87a$0$505$7f31c96c@news01.syd.optusnet.com.au>

hi all
i am very new to python and was wondering if it was capable of doing to the
following and if so..how?
the following is an example of what i need to strip and turn into a tab
delimited file..
the entries are from a music database that is updated weekly..it is in lotus
notes so i export from there and it dumps this file out with the new titles
for the coming week (here the 19th June)....
i only need to get the rows with the double astericks at the start (i put
those in for this post, they are not there in the normal file) and then
strip the words at the start and then put a tab in between them so i can
bring it into excel for the sales team to look at .... i would need a first
row to desribe each column also maybe something like
Title tab Artist tab Cat no tab APN etc etc
the tracklisting is a little more complicated as it has multiple tracks
within the row..they are seperated by commas....
i think that it won't be that hard but i had no success in rebol so i
thought i might try python...

thanks in advance
jason

FORM:  Popular Recording
CMTYPESWITCH:  Popular
RELOCORIGINAL:  55864A2603AC94ACCA2568F200224388
CMCOUNTRY_ORIGIN:
CMGENRE_CODE:
**CPTITLE:  INFINITE POSSIBILITIES
**CPARTIST:  AMEL LARRIEUX
CPOARTISTS:
**CMCATNUMBER:  4948792
**CMAPN:  9399700067507
**CMARIAPRICECODE:  C9921
**CMARIAMEDIACODE:  21
**CMARIADISTRIBUTORCODE:  SMA
CMARIAPACKAGECODE:
**DMRELEASE:  19/06/2000
cMDateFormat:  19/06/2000 12:00:00 AM
DMDELETE:
cMDeleteStatus:  No
**CMLISTPTRACKS:  GET UP,I N I,SWEET MISERY,SEARCHIN' FOR MY SOUL,EVEN
IF,INFINITE POSSIBILITES,SHINE,DOWN,WEATHER,MAKE ME WHOLE,GET UP (THREAD HAD
FUN MAIN MI,GET UP (MV MIG MIX (RADIO))
CMLISTPARTISTS:  ,,,,,,,,,,,
RELOCMEDIA:  F73C80183A01A8D7CA2568F2002A5517
**CURMPRICE:  15.91
**CMTYPE:  Compact Disc
CMRECORDCOMPANY:  SONY MUSIC
CMPACKAGE:
CMARIADISTRIBUTORHOUSECODE:
cDistributorHouse:
CMLOCALE:  Y
$UpdatedBy:  CN=PPT/OU=AEROS/O=JUKEBOX
$Revisions:  02/06/2000 05:42:22 PM,02/06/2000 05:42:22 PM






From jaford at watford53.freeserve.co.uk  Tue Jun 20 13:21:56 2000
From: jaford at watford53.freeserve.co.uk (Jim Ford)
Date: Tue, 20 Jun 2000 18:21:56 +0100
Subject: Python and mSQL
Message-ID: <8io97i$d6u$1@newsg1.svr.pol.co.uk>

Hi

I use mSQL version 1.0.16 on Linux and want to write a script to access the
database. I'm currently fiddling around with Perl and am starting to get
somewhere, but what little I know about Python attracts me and I'm wondering
whether it would be a better choice.

What is the current state of a Python module to access mSQL, please?

Regards: Jim Ford




From webmaster at underlevel.net  Thu Jun  8 00:28:08 2000
From: webmaster at underlevel.net (Jordan Katz)
Date: 08 Jun 2000 00:28:08 -0400
Subject: Newbie Q: CGI
References: <8hn0c1$1a22$1@news1.wire.net.au>
Message-ID: <m3snuobxuv.fsf@underlevel.underlevel.net>

"Steven Adams" <adams_s at lab.eng.usyd.edu.au> writes:

> Hi,
> 
> How do you get the IP number from a request for a web page? gethostbyname()
> and gethostbyaddress() weren't mentioned anywhere in the cgi module docs
> (that I could see, did I miss it?)

The low-level socket module has what you need.  You can read all about
it at http://www.python.org/doc/current/lib/module-socket.html
-- 
Jordan Katz <webmaster at underlevel.net>


From m.faassen at vet.uu.nl  Thu Jun 15 09:48:12 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 13:48:12 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <394847C0.E237B43C@home.com>
Message-ID: <8iamqs$dro$2@newshost.accu.uu.nl>

Ivan Van Laningham <ivanlan at home.com> wrote:
>> Has anyone done any extreme programming in python? Care to share your
>> experiences?

> Not yet, but the company I just started with recently hired me to do
> Python programming (& other stuff), and we're all gearing up to do XP.

> I'd like to hear others' experiences, too, if there are any.  Otherwise,
> I'll let you know.

My main extreme programming experience has been unit testing. I don't do
that the complete XP way either; I tend to write the tests just after I write
the code, instead of just before. PyUnit is a nice unit testing framework.

Unit testing is not too bothersome in Python, and you can feel much more secure
about your code; you know better that it's doing what you want, and when
you change something, it's easy to check everything still works. That's
quite comfortable and helpful.

No experience with pair programming or the other XP ideas yet, though.

Regards,

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


From richard_chamberlain at ntlworld.com  Tue Jun 13 12:37:41 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 13 Jun 2000 17:37:41 +0100
Subject: Two newbie questions
References: <20000613165659.A7828@basajaun> <Mgt15.508$9K.10789@news6-win.server.ntlworld.com>
Message-ID: <Tit15.511$9K.7935@news6-win.server.ntlworld.com>

doh,

>>random.randint(0,i-1)

where i is maxint

Richard

richard_chamberlain <richard_chamberlain at ntlworld.com> wrote in message
news:Mgt15.508$9K.10789 at news6-win.server.ntlworld.com...
> or,
>
> import string
> st=string.join(l,'')
>
> will do it.
>
> if you do your random thing interactively and don't assign to i you'll
> notice you get an exception so you'll have to play around based on that.
>
> I'm sure I'll be corrected but I guess the range your trying to use
> (-maxint-1 to maxint) is not an int it's a long so it doesn't work.
>
> random.randint(0,i-1)
>
> does work though.
>
>
> Richard
>
> Arkaitz <ma9aba at bath.ac.uk> wrote in message
> news:20000613165659.A7828 at basajaun...
> > Hi all,
> >
> > I'm learning Python, as it seems everybody around here :-), and I found
> two
> > problems that I can't find the solution to:
> > 1. As the strings are inmutable, if I want to modify the order of the
> characters
> > in the string(e.g. sort them), what I do is:
> >
> > st = "asdfasfasdf"
> > l = []
> > for c in st:
> >     l.append(c)
> >
> > Then I process the list, and the problem is that I have no idea of how I
> can get
> > the string back, any help?
> > 2. The second question is about random numbers, how can I get a random
int
> between the minimum and maximum integers of the machine?
> > The next code doesn't worK
> > import sys, random
> >
> > i = random.randint(-sys.maxint-1, sys.maxint)
> >
> > Thanks in advance,
> >
> > Arkaitz.
> >
>
>




From stephen at cerebralmaelstrom.com  Thu Jun  8 20:50:13 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Fri, 09 Jun 2000 00:50:13 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <20000608155108.A8398@reepicheep.avalongroup.net>
Message-ID: <9bX%4.44746$WS3.478611@typhoon.we.rr.com>

Seriousely. Its free -- and its not the official IDE of Python, so you
really can't be quite so demanding unless you want to get off your ass and
fix any problems you find in it.

Sheesh.

--S

Timothy Grant <tjg at exceptionalminds.com> wrote in message
news:20000608155108.A8398 at reepicheep.avalongroup.net...
> On Thu, Jun 08, 2000 at 02:19:23PM -0700, Jeff wrote:
> >
> > If there are no answers to these questions, someone needs to work on
> > pythonwin's "development" capabilities.
>
> I vote for you to do it.
>
> --
> Stand Fast,
>     tjg.
>
> Timothy Grant                         tjg at exceptionalminds.com
> Chief Technology Officer              www.exceptionalminds.com
> Red Hat Certified Engineer                           MIG #1433
> Avalon Technology Group, Inc.
> >>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<
>




From jcm at bigskytel.com  Thu Jun  1 05:36:04 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 1 Jun 2000 03:36:04 -0600
Subject: ANN: bzip2 module
In-Reply-To: <20000601003920.A18615@dsl-only.net>; from eito@cs.pdx.edu on Thu, Jun 01, 2000 at 12:39:20AM -0700
References: <20000601003920.A18615@dsl-only.net>
Message-ID: <20000601033604.D1234@bigskytel.com>

* eito <eito at cs.pdx.edu>:
> 
> I would like to announce about a bzip2 module. This is my
> first attempt to write Python module. I really don't know
> how to package, announce, etc... I would appreciate if
> somebody can tell me what is the good way to do this kind
> of thing.

You probably already know more than me about this, but I recommend
submitting a reference to your module here:

http://www.vex.net/parnassus/submit/

It doesn't appear to have anything for bzip2 yet.


  david.

-- 
Impossibly motivated by midnight the flyspecked abdominous female
indubitably tellurian strolls emitting minute grins each an intaglio.
  -- e.e.cummings in a blender



From Dave at Thomases.com  Wed Jun 21 10:30:21 2000
From: Dave at Thomases.com (Dave Thomas)
Date: 21 Jun 2000 09:30:21 -0500
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <m2r99rrts2.fsf@zip.local.thomases.com>

"William Dandreta" <wjdandreta at worldnet.att.net> writes:

> fields = splitfields(line,',')
> 
> fields[Date][:5]  is the year
> 
> is more obvious and less cryptic than
> 
> fields[1][:5]

There have been many fine answers to this, but they all seem to me to
miss the point ;-)

The data you are reading is is not a list, it just happens to have
been stored as delimited strings externally. In reality, the data is
structured. So the OO solution would be to create a class to represent
this information and encapsulate the knowledge of the external
representation in that class. That way you don't have globals
containing magic numbers.

  timeRecord = TimeRecord(line)

  print timeRecord.date()

etc...


Regards


Dave

 ___________________________________________________________________________
| The Pragmatic Programmers, LLC     |  http://www.pragmaticprogrammer.com  |
| Read "The Pragmatic Programmer"    |  www.pragmaticprogrammer.com/ppbook/ |
 ---------------------------------------------------------------------------





From cut_me_out at hotmail.com  Wed Jun  7 21:48:20 2000
From: cut_me_out at hotmail.com (Alex)
Date: 07 Jun 2000 21:48:20 -0400
Subject: Can I explicitly request garbage collection from C?
Message-ID: <etdem692ba3.fsf@w20-575-109.mit.edu>

Hi.

I am having a little trouble keeping track of reference counts in C
extensions.  I thought it would be helpful if I could call garbage
collection from an arbitrary point in my program and check what was
destroyed.  Is that possible?

Alex.


From jkraska1 at san.rr.com  Thu Jun 15 02:00:15 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 15 Jun 2000 06:00:15 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <394847C0.E237B43C@home.com>
Message-ID: <3948719B.96B9367A@san.rr.com>

> Not yet, but the company I just started with recently hired me to do
> Python programming (& other stuff), and we're all gearing up to do XP.

Well, if I were joking, I'd say that python makes you
so productive that XP was overkill. :)-


C/


From bkc at Murkworks.com  Sat Jun 10 13:12:45 2000
From: bkc at Murkworks.com (Brad Clements)
Date: Sat, 10 Jun 2000 13:12:45 -0400
Subject: Porting CPython
References: <8hrbph$a4n$1@nnrp1.deja.com>
Message-ID: <gBu05.27$pa6.2183@news.clarkson.edu>

.com
<perl_monkey at my-deja.com> wrote in message
news:8hrbph$a4n$1 at nnrp1.deja.com...
> Hello,

> - SIGMAX in Modules/signalmodule.c

> checked /usr/include/signal.h, found that the highest numbered signal I
> could see was 28, so I #define SIGMAX 28 in signalmodule.c

Sounds okay to me.

>
> - Modules/pwdmodule.c
>
In my NetWare port of Python, I just eliminated this module.

> - Excluding Modules
>
> What is the build procedure for omitting certain built in C modules?
> (I"m talking about the stuff in Modules/)

Edit config.c to remove the module from the built-in list, then remove it
from your makefiles.






From rumjuggler at cryptarchy.org  Wed Jun  7 20:13:41 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Thu, 08 Jun 2000 00:13:41 GMT
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com> <m3em69w1a8.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <okotjsg0khkkgcpdt516a55eflade43bat@4ax.com>

On 07 Jun 2000 23:53:51 +0100, Michael Hudson <mwh21 at cam.ac.uk> wrote:

>You don't post the code for If, and I'm a bit confused, because in one
>place:
>
>> odds=odds, l=l,app=app: If(l.index(e)%2==0, app(evens,e), app(odds,e))
>
>you pass a value as the first parameter, and in another:
>
>> If(lambda c=c,fs=fs: c==len(fs)-1, Return(fs[-1]()), fs[c]))()
>
>you pass a thunk.  Which do you want?  I suspect this may be giving
>the error you're suffering, though I can't get it to work at all.

def If(cond, t, f):
	if callable(cond):
		cond = cond()
	if cond:
		return t()
	return f()


>Also:
>
>> a=a,e=e:a.append(e): Compose(apply(map, (lambda e, evens=evens,
>                               ^^^^^
>what's this apply for?

Good question.  Therein lie a lot of problems, I think.

-- 
Barnabas T. Rumjuggler

Ibi nullus timet mortem
Sed pro Baccho mittunt sortem
 -- Carmina Burana



From davec999 at my-deja.com  Thu Jun 29 12:20:10 2000
From: davec999 at my-deja.com (David Currie)
Date: Thu, 29 Jun 2000 16:20:10 GMT
Subject: Bug in PythonWin??
Message-ID: <8jfsv4$g53$1@nnrp1.deja.com>

Hi.  I've got this weird situation with PythonWin build 132.

I'm working with a spanish language keyboard under NT 4 (SP4) and every
time I hit one of the accent keys (the ones that expect a key to be
pressed afterwards to add an accent to it) PythonWin crashes in the
most horrible way (could not 'read' memory location bla, bla, bla.).
The version of NT I'm running is in Spanish.

Is this a bug or a misconfiguration (in PythonWin or NT itself)?  Does
anyone else out there manage to run this on a similar configuration
with no problems?

Thanks.

Dave


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


From pinard at iro.umontreal.ca  Wed Jun  7 09:14:39 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 07 Jun 2000 09:14:39 -0400
Subject: Spurious coercing? [was: Re: 'if foo' calls getattr?!?]
In-Reply-To: scarblac-spamtrap@pino.selwerd.nl's message of "7 Jun 2000 09:43:27 GMT"
References: <Pine.GSO.4.10.10006062045150.1650-100000@sundial>
	<oqsnuqbomf.fsf@titan.progiciels-bpi.ca>
	<slrn8js67e.npl.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <oqem69ljk0.fsf@titan.progiciels-bpi.ca>

scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) ?crit:
> Fran?ois Pinard wrote in comp.lang.python:

> > I saw many emails in which people were surprised by Python behaviour,
> > and I interpret the surprise as a sign that predictability suffers.
> > For example, A.__cmp__(B) will not be called for an equality comparison
> > when A "is" B.  Yet, given the meaning of __cmp__, this sounds absolutely
> > correct to me.

> > Also given the meaning of __coerce__, I would consider just equally
> > correct to avoid calling A.__coerce__(B) when type(A) "is" type(B).
> > The spontaneous predictable thing, here, is that __coerce__ should *not*
> > be used when it is evident it should not.

> But when both types are InstanceType, __coerce__ may still be needed,
> right?

I see your point.  `type()' has a rather generic value for instances,
and so, my way of putting it is not exact.  I really meant that if A and B
are instances from the very same class (which is expected to be a frequent
case), coercing should not be needed, and the __coerce__ call should then
be avoided, on the same principle that __cmp__ is avoided whenever possible.

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





From kain at cableadmin.com  Sun Jun 25 11:20:22 2000
From: kain at cableadmin.com (kain at cableadmin.com)
Date: Sun, 25 Jun 2000 15:20:22 GMT
Subject: [ANNOUNCE] PyX10 4.0.1 - Python Home Automation
Message-ID: <slrn8lc9da.ouj.kain@cableadmin.com>

PyX10 aims to bring X10 home automation to Python applications.

Release 4.0.1 is a wrapper for XAL 4.0.1 and currently implements
most of the functionality of the XAL library.

If you have X10 devices, please try this package and send me lots
of feedback.

Requirements:
	X10 Abstraction Library (XAL)
		http://xal.sourceforge.net

	PyX10
		http://pyx10.sourceforge.net

Thanks,
-Scott



From mhammond at skippinet.com.au  Mon Jun 12 16:13:14 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Mon, 12 Jun 2000 20:13:14 GMT
Subject: PythonWin and collections
References: <8hv12v$ol1$1@nnrp1.deja.com>
Message-ID: <8i3g8d$kev$1@nnrp1.deja.com>

In article <8hv12v$ol1$1 at nnrp1.deja.com>,
  barryhart at my-deja.com wrote:
> I'm trying to port the following fairly simple code to Python. It uses
> ADSI to get a list of members in an NT group.

I believe ADSI uses a custom enumeration technique.

You could try win32all build 132, where there is specific ADSI
support.  There is a demo there which should point you in the right
direction.  This does support ADSI enumerations correctly.

Mark.


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


From eric at estinc.com  Wed Jun 28 15:20:26 2000
From: eric at estinc.com (Eric Lee Green)
Date: Wed, 28 Jun 2000 19:20:26 GMT
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com>
Message-ID: <395A5033.DF103F0C@estinc.com>

robin.escalation at ACM.org wrote:
> Does anyone have any success stories (or gotchas) related to
> implementing a PostgreSQL database with Python that they'd like to
> share? I am hoping to use it as a back-end to a b2b web site.

Yes. Can't tell you exactly at the moment though (NDA). 

Basic plusses of PostGreSQL: transactions, cursors, stored procedures, joins
with subselects, all mostly work as you'd expect of any SQL database. The same
cannot be said to be true of most other embedded databases out there (e.g.
RAIMA, C-Tree, MySQL, ...). Read performance is good, and the DBMS engine
supports all the usual kinds of locking that you'd expect from an RDBMS. 

Minuses: The page length restriction on size of a record can be the pits, if
you're not careful about your design. If you need BLOBs, you're out of luck
(the Python module supposedly supports the PostGreSQL BLOB type, but it's not
reliable). Insert speed is horrible -- inserting data takes forever and a
half, though PostGreSQL 7.0 now has a bulk import facility that may make it
faster (I haven't had a chance to benchmark it yet). And, finally, the
documentation is cryptic, incomplete, and scattered all over the place... I
had to reference two different manuals in order to figure out how to create
(and destroy) a PL/SQL stored procedure, for example. Lack of explicit support
for replication is a pain too, though it's no worse than any other embedded
database (i.e., a 'modified' field and an explicit procedure can be used to do
it). 

The biggest minus for the back-end of a b2b web site, though, would be the
fact that it is bog-slow at opening a connection. PostGreSQL forks a new
process for each connection, and the new process must do all sorts of things
to establish authentication etc... PHP3's PostGres support gets around this by
creating "persistent" connections that hang around with the Apache process,
but I have no idea if Python's Apache module support is robust enough to
handle this situation. Not to mention that Python's Apache module support is
almost entirely undocumented... it is quite unclear, for example, how to run
your former Python CGI scripts via the PyApache module. 

My basic conclusion is that PostGreSQL makes a fine alternative to current
embedded databases. It has better functionality than the xBase types, and
unless you need massive bulk inserts, the performance is excellent.  On the
other hand, I personally would not use Python with PostGreSQL and Apache for
implementing a b2b web site. Python with ZOPE/Zserver and PostGreSQL, however,
might be reasonable (I have not investigated that yet). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From paul at fizzylab.com  Sun Jun 18 00:28:42 2000
From: paul at fizzylab.com (paul at fizzylab.com)
Date: Sat, 17 Jun 2000 21:28:42 -0700
Subject: string.join question (was Re: urllib question)
References: <Pine.LNX.4.10.10006171424430.19831-100000@mail.fizzylab.com> <8igv24$5li$1@slb7.atl.mindspring.net>
Message-ID: <394C4FF9.46B011DC@fizzylab.com>

Nothing. That's the problem. The item I want to append to the first never gets
evaluated/appended. I get the first item -- the host name -- and I make a valid
request for the root of the webserver, not what I want. It's as if I requested
'http://www.site.com/' rather than 'http://www.site.com/thispage.html'  which is
not what I want.

See the code I submitted initially . . . . the first example works, the second
does not.

#!/usr/bin/python
import string, urllib

u = urllib.urlopen(string.join(["http://cnn.com/","index.html"]))
print u.read()
u =
urllib.urlopen(string.join(["http://some.host.com/","monstrousQueryString?withParameters"]))

print u.read()

I added an additional loop to ask for a flat html page (no query string), still
no joy. I made a telnet connection to port 80 and requested the page, no problem.

If I omit the string.join and request the page, it works fine.
u = urllib.urlopen("http://some.host.com/monstrousQueryString?withParameters")
print u.read()

This suggests something in the join is not working. And the webserver log just
tells me there was no second item to be joined/requested.

Don't make me do this in perl.

Aahz Maruch wrote:

> In article <Pine.LNX.4.10.10006171424430.19831-100000 at mail.fizzylab.com>,
> paul beard  <paul at fizzylab.com> wrote:
> >
> >The length isn't the issue. I'm not able to do the string.join for some
> >reason. I need to feed many URLs from a common host. I have the URLs (from
> >an access log) and want to just append the URL to the host.
>
> What happens when you do a string.join()?
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>    http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "The only problem with Microsoft is they just have no taste." --Steve Jobs
> (From _Triumph of the Nerds_ PBS special)



From morten at src.no  Tue Jun 20 08:07:29 2000
From: morten at src.no (Morten W. Petersen)
Date: Tue, 20 Jun 2000 14:07:29 +0200
Subject: How to get the filename for mail-attachments
Message-ID: <Pine.LNX.4.21.0006201405040.20804-100000@lade.src.no>

How do I do that? There's a lot of ways to inclue the filename, either by
adding it in content-disposition, or content-type, or whatever; are there
any programs or pacakges that can handle this? Or somewhere I can get a
list of possible combinations?

-Morten

-------------------------------------------------------------
How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From petersc at stud.ntnu.no  Wed Jun 14 07:29:24 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 14 Jun 2000 13:29:24 +0200
Subject: tkinter/windows/starting html
Message-ID: <39476C94.DA0AA2AD@stud.ntnu.no>

I am writing a start menu for a "multimedia" CD which contains
a lot of .htm and .jpg files.

To execute the index.htm in the main directory I use.

cmd = ["start","index.htm"]
os.execpv(cmd[0],cmd)

This runs fine except when I use it together with tkinter.
I assign self.view_content to a button.

def view_content(self):
    cmd = ["start","index.htm"]
    os.execpv(cmd[0],cmd)

This works fine except that after starting my browser
the menu does not work anymore. Even when I close the
browser the menu is dead and cannot even be killed.

I am using win98 and the pythonware 1.5.2 distribution.

Can I use some other construct? What am I doing wrong?

rebooting-the-computer-very-confused-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From cpu at bigfoot.com  Fri Jun  9 22:48:17 2000
From: cpu at bigfoot.com (S. Hoon Yoon)
Date: Sat, 10 Jun 2000 02:48:17 GMT
Subject: Does anyone has Mark Hammonds's IPC-8 presentation?
Message-ID: <R%h05.121727$R4.735202@news1.rdc1.nj.home.com>

Hi,

  I am trying to implement and use Events in win32com. I found that the
example is inside the new IPC-8 presentation by Mark (thanks); however, the
site (Starship) has not been recovered and I cannot get to the explaination.
If anyone has a copy, pls send it to me to this e-mail hyoon at home.com.
  Thanks much in adv,

Hoon,




From wjdandreta at worldnet.att.net  Wed Jun 21 12:16:33 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 16:16:33 GMT
Subject: an help for every function
References: <395078C0.5105D26E@tcomhp43.epfl.ch>
Message-ID: <BT545.13336$C44.784772@bgtnsc05-news.ops.worldnet.att.net>

Hi cedric,

There is a Windows help file called Pythlp.chm that I use. You can download
from the python webside. It has an index and you can seach for keywords.

Bill
Cedric Lefeuvre wrote in message <395078C0.5105D26E at tcomhp43.epfl.ch>...
>i am a beginner in python, and i would be happy if anyone can answer me
>if there is an help or a man function like in unix-like OS that we can
>use to describe any function of python.
>




From mwh21 at cam.ac.uk  Fri Jun 30 20:37:42 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 01 Jul 2000 01:37:42 +0100
Subject: backslash in regular expression problem
References: <394A55BA.69269793@wolfson.co.uk> <02de01bfd826$cf05bd80$6401a8c0@home> <8jivfr$qc2$1@nnrp1.deja.com>
Message-ID: <m3g0pu7kix.fsf@atrus.jesus.cam.ac.uk>

Keith Murphy <kpmurphy at my-deja.com> writes:

> what if instead of having '\\stuff' you have a str = '\\stuff'  ?
> 
> i know that you can print r'\\stuff' but you can't print rstr...
> 
> how do you use the raw with a variable?

You don't.  You do know about re.escape?  Asking that question
suggests you might not...

Cheers,
M.

-- 
  I saw `cout' being shifted "Hello world" times to the left and
  stopped right there.                                -- Steve Gonedes


From ahujache at cse.msu.edu  Tue Jun 13 15:44:57 2000
From: ahujache at cse.msu.edu (Chetan Ahuja)
Date: 13 Jun 2000 19:44:57 GMT
Subject: python or java for my graphical app?
Message-ID: <8i62vp$2fa3$1@msunews.cl.msu.edu>

 Hi,
 (Short version)
   I had created a graphical app, including some animation font
   mangling (rotation of characters etc)  in Java2 last year. Now I am
   thinking of porting it to python. Should I go for it?

(Some Details)
    The application is an education tool for kids etc. which uses
    stuff like dancing, moving words on the screen. It needed to be
    cross platform (at least Windows and Mac and maybe linux if
    possible) so I used Java2. I know Java2 was not available for many
    different platforms at that point but I knew it would be
    eventually. Now I'm thinking of porting it to python for the
    following reasons: (I don't know any python at the moment but I
    don't mind spending a few days learning what seems to be a useful
    tool going by the press )

   1) Many JVM's seem to interact very badly with the OS due to the
      varying threading models used. This screws up the timings of the
      animation ( how fast the word moves across the screen etc.)
      a program that worked more or less OK on windows ran horribly on
      the Linux JDK. 
   2) The general slowness and heavyness of Java. Anytime you start a
      JVM, my PII 400 system with 128 Meg of RAM feels like an old
      386. What about people, schools etc with slightly older
      systems. Again, animation timing will suffer.

  3) I like developing on linux and Java support, specially for
      graphics and fonts etc is nothing short of awful on linux.

      I know python is OK on count (3). I'd like to hear experienced
      python programmer's opinions on points 1 and 2.

       Also, one of my concerns is how much of python runtime I have
       to ship with my app? Can I remove what is not needed by my
       application or the whole thing has to be shipped.

       Another thing is, is there any choice of graphic toolkits at
       all or Tkinter is the only game in town? I have nothing against
       TKinter but I'd like to ship as small a package as I possibly
       can... So would like to avoid shipping tcl/tk with my app.
       (why not just use tcl to build the app otherwise?)

       What is the general suitability of python for what I want to do?

       I'll  appreciate any answers, tips, hints etc. 

      Thanks
      Chetan
       




From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 28 06:54:12 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 28 Jun 2000 03:54:12 -0700
Subject: Help with launching acrobat
References: <hib65.8$dm4.455036@news0.optus.net.au>
Message-ID: <160a1a5c.bcd2ac53@usw-ex0104-032.remarq.com>

Hi Douglas,

I'm ignoring both your questions, and showing a different way:

import win32api
win32api.ShellExecute(0, 'open','c:\\yourPdf.pdf', None, None, 0)

You'll need to install the win32 extensions obviously.

The parameters are (parent window handle or 0 for no
parent,either 'open' or 'print',the file path,parameters if an
exe,initial directory,show the application (0 if a document)).

There is a current thread on COM and acrobat but I think that's
probably not necessary if all you want to do is open acrobat.

Richard


Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From pete at visionart.com  Wed Jun 14 14:54:43 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 14 Jun 2000 11:54:43 -0700
Subject: can this be optimized
Message-ID: <3947D4F3.1D315713@visionart.com>

i have this routine which is running pretty quick
on my machine. i may be getting overly greedy, but
if i could shave off more time i'd be pleased
(btw, this is with numeric python)
(btw2, WIDTH and SAMPLES are defined elsewhere as global)


def pumpmeup_func(img):
    rndpos = RandomArray.uniform(0, WIDTH, SAMPLES).astype(Int)
    rndcol = RandomArray.uniform(100, 255, SAMPLES).astype(Int)
    for x, color in map(None, rndpos, rndcol):
	img[x,-3:-1] = color



my thoughts for optimization are...
- a quicker way to create the random arrays?
  this is already by far the quickest way i've found.
  i'm not truly worried about nonuniform distribution
  of my random arrays, and i'm not sure if this
  uniform() function is taking the time to bother
  with that.

- replacing the for loop with a map/reduce thingy
  and a lamda. i've never done that before and couldn't
  get anything i tried to parse.

- somehow creating a single random array, and avoiding
  the need to map then together? (2 dimensional?)
  i question this would gain me anything
  (the map is still quicker than looping through
   range(SAMPLES) and indexing each array)

- i was wondering about stackless as well. it seems with
  a chunk of code like this it wouldn't make a noticeable
  difference, but maybe if the loop was processed with
  a lambda thingy it could offer benefits?

- creating a list/array of the slices and then use numpy
  to assign the random colors to that?  (like to see this!)


anyways, i've been twiddling with this for awhile and have
taken it from .045 seconds per call, down to around .013
WOOHOO! i'm excited as it is, but want to see if it can go
further

thanks all you gurus!


From m.faassen at vet.uu.nl  Tue Jun  6 14:21:01 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 6 Jun 2000 18:21:01 GMT
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <8hjfed$n0c$1@newshost.accu.uu.nl>

Richard P. Muller <rpm at wag.caltech.edu> wrote:
> Does Programming in Python Marginalize You?

[snip]

> And yet, I worry that using Python marginalizes one. Python is hard to
> beat for writing applications that run on a single computer. But
> that's yesterday's computing model. The future of computing, heck, the
> present of computing, is in writing programs that run through a web
> browser.

It is? I mean, given what you say next, you seem to think as if the
future/present of computing is writing programs that run *inside*
browsers. That was the hype of 3 years ago, when Java was new. It turned
out not to be the case. At least not yet, and not wholly.

What is actually happening is that people write server applications,
with HTML interfaces (which may have a bit of JavaScript or Java glue;
the second much rarer than the former). Web applications, yes, they're
important. Web browser applets are not very important as of the moment, 
though their support role in supporting web server applications may
grow (but only in beefing up the user interface).
 
> And the fact of the matter is that 99% of web browsers
> surfing the internet run Java and JavaScript, which gives these
> languages a huge advantage over Python,

Which doesn't matter a lot for most web applications, as they run on
the server, where you can run anything.

> an advantage that Python's
> superior design may not be able to overcome.

Fortunately, Python already overcame this hurdle at least in part, 
as there is JPython, which runs on top of Java and can be used to
produce web browser applets (http://www.jpython.org -- ah, I see
later on you know about this).  The second hurdle is currently being taken;
Mozilla is gaining Python support (though initially this will be used more 
to extend Mozilla than for web interfaces). I also think Internet Explorer
can have some Python support with the Windows Scripting Host; I'm not sure
about this though.

But all this is beside the main point, which is that web applications
tend to happen more on the server than on the client.

> I have Ada-programming
> friends who mumble about Ada's superiority over C++; that may be the
> case, but does it really matter anymore?

Fortunately, Python's been rapidly gaining in popularity for quite
a while now, at an accelerating pace.

> I'm aware of CGI scripting with Python. It's great. But JavaScript
> pages load faster than CGI pages, and are easier to program, to
> boot.

That's, excuse me, a grave misunderstanding of the way the web works,
and comparing apples with oranges. CGI pages (or other server generated
pages, for instance those generated by Zope) are not inherently slower
than JavaScript pages, and in fact, a JavaScript page can be generated
on the server-side. It's just not a valid comparison at all, as you
can't do the same things with a JavaScript page as you can do with a
server system.

> Plus, JavaScript can directly call Java Applet methods, enabling
> it to do functions that are simply impossible with CGI scripts.

Can it? I wasn't aware of this. Then again, I seldom encounter Java
Applets. You may be navigating a different web? I generally encounter
lots of HTML pages with forms on them, lots of server-generated pages,
with a little bit of JavaScript glue here and there.

> I'm also aware of the JPython module, which promises to enable
> us to write applets in Python instead of Java. But who among us has
> actually *used* JPython? Or, for that matter, who has come across a
> random web page on the net that uses JPython applets? 

How often do you come across a web page that uses Java applets? And now,
how often do you come across a web page that uses Java applets to do
something *useful* (instead of display gimmickery)?

But you know about JPython. All right.

> Do I recommend that students today learn Python? Certainly, because
> it's easy, fun, it teaches one good OO habits, and it's incredibly
> powerful. But I also recommend that students learn C/C++ for speed (I
> no longer recommend Fortran), and Java and/or JavaScript for web
> programming. Heck, I also recommend that they learn Perl, since so
> many other people know Perl. That leaves a significant niche for
> Python, but only a niche, and Python is too good a language to be a
> mere niche player.

Is there any priority in such recommendations? I mean, there's a law of
diminishing return here. C is useful to know because it's everywhere,
fast and low level. C++ is useful to know because of speed and fairly
high level abstractions, but I don't think it's for everyone. Python
I *do* think is for everyone. :) But if you know Python, it doesn't make
that much sense to learn Perl afterwards, unless you have to or are 
dissatisfied with Python. This because Perl and Python fill approximately
the same niche.

Java may also be useful because it's everywhere, though I haven't found
a use for it yet. JavaScript is not really very learnable on the present
web; it's mostly a bunch of hacks that aren't cross-browser compatible.
There's hope that this will improve with newer browsers. But it's
not really an application programming language.

But there should be a list of priorities here. Python may be a good
first language, for instance. Will probably bring you up to speed
with C++ or Java faster than starting with C++ or Java itself will.

> If we don't want to see Python wind up like Ada, what should we do?

Just wait. Really, look at the amount of postings on this newsgroup
over time for one datapoint. Look at the explosion of Zope use for
another. Look at the amounts of books on Python in your local bookstore
(hint: a year and a half ago there were 2 if you were lucky). You don't
have to do a thing, just use Python, and tell others.

> I've been teaching a 6-lecture short-course on Python here at Caltech,
> in an effort to introduce more people to the language.

Ah, so you're doing that already. Great!

> I'm thinking
> about learning enough JPython to give an additional lecture on that
> language. Hopefully, if enough people use JPython for writing Applets
> we can help establish it as a reasonable alternative to Java.

But as far as I'm aware, Java's main use these days is for
application programming, and server programming. We already have
reasonable alternatives to lots of Java technology; often enough they're
even *more* reasonable alternatives. :)

> As for JavaScript, I don't know of a solution. It would have been nice if
> Netscape had implemented Python's syntax for LiveConnect rather than
> Java, but it didn't happen that way. Think how wonderful it would have
> been to write PythonScript functions in web pages rather than
> JavaScript.

Lots is happening on the Mozilla front. Have you seen the latest
ActiveState announcements? Python will definitely be usable to extend
Mozilla with new components. PythonScript will take a while longer,
though. But perhaps one day it'll get there. I still think you're
overestimating the importance of JavaScript and Java Applets on the
web, though. Could you list a typical application? I do web programming
for a living and I can't really think of any common ones.

Regards,

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


From johngrayson at home.com  Wed Jun 21 12:51:00 2000
From: johngrayson at home.com (John Grayson)
Date: Wed, 21 Jun 2000 16:51:00 GMT
Subject: Widget resizing
References: <394F8774.138B@cs.bham.ac.uk> <ABO35.4876$fw6.96380@news6-win.server.ntlworld.com>
Message-ID: <8iqrpd$e7o$1@nnrp1.deja.com>

In article <ABO35.4876$fw6.96380 at news6-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:

>
> If you look at www.manning.com/grayson you can download two chapters
of
> Python and Tkinter programming one of which is screen layout. (of
course
> you'll be much better buying a copy ;)
>
> Richard
>
> Elliot Smith <e.smith at cs.bham.ac.uk> wrote in message


You'll also find over 20,000 lines of example code at
the same site...  :-)

    John


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


From Jonathan_Epstein at nih.gov  Fri Jun 16 16:50:21 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Fri, 16 Jun 2000 16:50:21 -0400
Subject: Module index information in machine readable form?
Message-ID: <394A930D.77CA9CB9@nih.gov>


Is there any way to obtain the module information (module names and the
names of their methods) in a form which is suitable for machine
processing?  I am interested in processing this data into a form which
can be digested by ELSE:
  http://www.lirmm.fr/doc/python/lib/modindex.html
  http://members.xoom.com/pmilliken/

Thanks,

- Jonathan


From thomas at xs4all.net  Thu Jun  8 15:53:18 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 8 Jun 2000 21:53:18 +0200
Subject: Successful Solaris 2.6 Compile???
In-Reply-To: <393F9E65.80E9C383@usa.alcatel.com>; from Sara.Fowler@usa.alcatel.com on Thu, Jun 08, 2000 at 08:23:49AM -0500
References: <393F9E65.80E9C383@usa.alcatel.com>
Message-ID: <20000608215318.A20700@xs4all.nl>

On Thu, Jun 08, 2000 at 08:23:49AM -0500, Sara Fowler wrote:

> Has anyone successfully compiled the Python Interpreter on a Solaris
> 2.6?  If so, do you know where DIR is declared?

DIR is a standard define in one of the standard UNIX header files, usually
dirent.h (either directly or indirectly.) If you are missing dirent.h, you
have a very broken system. Are you able to compile anything else ? Did you
install the proper Solaris 'packages' ? (From what I recall, long ago, about
SunOS installs and compilers, and gcc and the like, you have to be careful
about what you install where.)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From mhammond at skippinet.com.au  Tue Jun  6 04:44:59 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 06 Jun 2000 08:44:59 GMT
Subject: set Registry User Security
References: <8hghbt$5bc$1@nnrp1.deja.com>
Message-ID: <fS2%4.6572$Hz.48327@news-server.bigpond.net.au>

The win32api module supports all the security parameters to the win32api,
and the pywintypes and win32security modules contain all the support for
manipulating the SECURITY_ATTRIBUTES and other win32 security related
objects.

Mark.

<tiddlerdeja at my-deja.com> wrote in message
news:8hghbt$5bc$1 at nnrp1.deja.com...
> In regedt32 you can assign user privilleges for a Key.
>
> Can you do this from within Python, perhaps using the win32api?
>
>
> Any help appreciated.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From hungjunglu at hotmail.com  Thu Jun 15 19:57:02 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Thu, 15 Jun 2000 16:57:02 PDT
Subject: Python Productivity over C++ (Encapsulation)
Message-ID: <20000615235703.24571.qmail@hotmail.com>

--- In python-list at egroups.com, m.faassen at v... (Martijn Faassen) wrote:
>Encapsulation is necessary if you want abstraction.

Hmm... I guess we are talking about different encapsulations. I was talking 
about the "private" keyword alone (hence data hiding.) Sorry if I have not 
been more clear. Instead of "encapsulation" I should have said "data hiding" 
or simply the "private" keyword.

I agree that data members are best accessed through accessor methods, and 
that's the way I (and most people) program. But I really can't live with the 
"private" keyword. If I have been criticizing "encapsulation", it's because 
I was referring to the "private" keyword, not the accessor methods.

There is no need for the "private" keyword, if a program works through 
accessor methods. If one doesn't want a particular accessor method to be 
used by others, one can simply use a naming convention (like an initial 
underscore _).

>That doesn't mean a Python programmer should forget about encapsulation.

Not encapsulation... "abstraction" would be a more correct word, I guess. 
Encapsulation conveys the image of data hiding and data ownership.

regards,

Hung Jung

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From olczyk at interaccess.com  Tue Jun 13 00:12:19 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Tue, 13 Jun 2000 04:12:19 GMT
Subject: Testing for a file in python?
References: <Pine.GSO.4.10.10006122010090.15238-100000@sundial>
Message-ID: <394db29a.275888218@nntp.interaccess.com>

On Mon, 12 Jun 2000 20:10:48 +0300 (IDT), Moshe Zadka
<moshez at math.huji.ac.il> wrote:

>> How would you go about testing if a certain file exists in python?
>> In most scripting languages you have the -f test:
>
>I don't know what you mean by "most".
For all those who emailed me ( especially the nasty ones ) suggesting
that I'm slamming Python in favor of Perl ( which I know well enough
to know, I don't like it )  by most I mean:

sh,csh,bash,tcsh,zsh ( and OK Perl )

ones that I seem to remember it being in but am not quite sure:

ksh, awk.

Ones that I'm sure don't do it
Python, sed ( and I'm not sure about sed, it's just not likely to be
there ).



From dbroadwell at mindspring.com  Tue Jun  6 00:37:32 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Tue, 06 Jun 2000 00:37:32 -0400
Subject: How to find out height of a Canvas
References: <memo.20000603235615.14417A@digitig.compulink.co.uk> <8hdth5$8i1@cs.vu.nl>
Message-ID: <393C800C.4237A8AB@mindspring.com>

> --
> There is a theory which states that if ever anyone discovers exactly what the
> Universe is for and why it is here, it will instantly disappear and be
> replaced by something even more bizarre and inexplicable. There is another
> theory which states that this has already happened.
>         -- Douglas Adams, 'The Restaurant at the End of the Universe'

And then there is annother theory that says that this has already
happened, Multiple times.



 --
	Got CODE?
 --


From nbridges at interlog.com  Mon Jun  5 13:30:58 2000
From: nbridges at interlog.com (Neal Bridges)
Date: Mon, 05 Jun 2000 17:30:58 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com>
Message-ID: <8hgo4i$aqn$1@nnrp1.deja.com>

In article <3939FFB4.A7824367 at mindspring.com>,
  David Broadwell <dbroadwell at mindspring.com> wrote:
> > Not that this exactly belongs in a python ng but I'm told the
> > only usable onboard programming environment is quartus, an ansi
> > forth system by Neal Bridges.
>
> Actually forth is the bios level language of the chip, before palm os
> loads. Just like a sparc, only smaller. Nope it dosen't really belong
> here, but is an option for the palm people. Still given the option of
> python of forth, i'll take python. (i used to know forth)

That's not correct.  There's no Forth in the PalmOS at all.

Quartus Forth is an on-board development environment and compiler that
generates stand-alone applications that require no run-time module.

--
Neal Bridges
<http://www.quartus.net/> Quartus Handheld Software!


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


From fsteele at mindspring.com  Thu Jun 15 13:53:37 2000
From: fsteele at mindspring.com (Frank Steele)
Date: Thu, 15 Jun 2000 13:53:37 -0400
Subject: Python in home automation
Message-ID: <fsteele-E5789E.13533715062000@news.inet-systems.net>

I'm beginning to think about extending the home network to include some 
simple home automation/X10 functions. I've started looking around at 
products and software, and even went so far as to pick up the latest 
copy of <A HREF="http://www.tpj.com">The P*&l Journal</A>, which has a 
lengthy story on <A HREF="http://www.misterhouse.net">MisterHouse</A>. 

MisterHouse is a cross-platform system, all in P*&l, that allows voice 
and web control of X10 components, and uses text-to-speech to 
communicate alerts to the home occupants.

Here's the problem/question -- if it's my system, I don't want it in 
P*&l. I would prefer to stick to the Python world. Unfortunately, in a 
quick sweep through the Vaults of Parnassus, The Daily-URL, 
Python.FAQts, and the O'Reilly Python dev-center, I don't see any 
modules, extensions, or projects on this.

So, is anyone aware of Python modules or extensions to do X10 home 
automation? I'm interested hearing about them on _any_ platform, but I 
intend to implement on Linux and MacOS....

Thanks,
  Frank
  fsteele at mindspring.com


From vivienne_van_velzen at my-deja.com  Mon Jun  5 10:16:17 2000
From: vivienne_van_velzen at my-deja.com (vivienne_van_velzen at my-deja.com)
Date: Mon, 05 Jun 2000 14:16:17 GMT
Subject: Regular expressions trouble
Message-ID: <8hgcn2$1lh$1@nnrp1.deja.com>

I'm fairly new to the use of regular expressions, and I've come upon a
problem I can't seem to fix. I'm trying to extract a keyword from a
HTML-page and color it red. This works fine, except that it also
extracts the keywords within tags.
Code example:

newline = re.sub('([^<>]*?)((?:<[^<>]*?>[^<>]*?)*?)('+query+')','\g<
1>\g<2><b><font color=\"red\">\g<3></font></b>',newline)

Query is the keyword and newline is a line from the HTML page.
The expression works in Perl with the use of \G. I couldn't find a
replacement in Python's re-module.
I hope someone can help me out.
TIA,

Vivienne


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


From herzog at online.de  Thu Jun 22 06:47:46 2000
From: herzog at online.de (Bernhard Herzog)
Date: 22 Jun 2000 12:47:46 +0200
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <LOz35.14014$Xx5.685990@bgtnsc06-news.ops.worldnet.att.net> <slrn8l2nce.q2.hzhu@rocket.knowledgetrack.com>
Message-ID: <m37lbikn59.fsf@greebo.nodomain.de>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> On Tue, 20 Jun 2000 01:30:19 GMT, William Dandreta
> <wjdandreta at worldnet.att.net> wrote: 
> 
> >
> >My error was that I did the following:
> >
> >from strop import joinfields, splitfields
> >from string import replace
> >
> >I erroneously assumed that replace would use the joinfields, splitfields
> >from strop, but it really used the joinfields, splitfields from string which
> >are very slow.
> 
> 
> If you really want to replace an internal method in an imported module,
> maybe you can try this (not tested):
> 
> import string, strop
> string.joinfields = strop.joinfields
> string.splitfields = strop.splitfields
> from string import replace

Messing around with strop like that shouldn't be necessary. string.py
already imports all of strop at the end:

# Try importing optional built-in module "strop" -- if it exists,
# it redefines some string operations that are 100-1000 times faster.
# It also defines values for whitespace, lowercase and uppercase
# that match <ctype.h>'s definitions.

try:
	from strop import *
	letters = lowercase + uppercase
except ImportError:
	pass # Use the original, slow versions


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


From qfcqor at speedyinternet.com  Mon Jun 26 21:46:55 2000
From: qfcqor at speedyinternet.com (qfcqor at speedyinternet.com)
Date: Tue, 27 Jun 2000 01:46:55 GMT
Subject: -  Learn how to SPEED UP your INTERNET connection by 70-100% for FREE!
Message-ID: <jIT55.93851$m5.59849@typhoon.kc.rr.com>

Hey,
Sign up at Netsetter (for free!), and it will install some free software that will speed up you Internet connection by 100%!  It helped me a lot with my connection!  I now download a lot faster than before, and get lower pings in games, which eliminates lag!
Read more about it on their site if you doubt it!
Get it here:
http://speedyinternet.cjb.net



From pablo.prieto at dulcesol.es  Tue Jun  6 07:27:37 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Tue, 06 Jun 2000 12:27:37 +0100
Subject: Newbie question.
Message-ID: <393CE029.27B87C91@dulcesol.es>

Hi all of you there!

#This is my question

AYS = ""

def setEnviroment():
  AYS = "NUMBER 9"

def getEnviroment():
  return AYS

#This is the end...

The function getEnviroment always return "". I guess AYS is declared
twice, One is module-scope and the other is local to setEnviroment, so
I've got two differents variables.

How can I make the variable AYS global to setEnviroment, so I'd be able
to return "NUMBER 9" in getEnviroment?. (Don't say return "NUMBER9".
I'll get annoyed :))

Well, Have a nice summer all of you (or Winter if you are in the other
side of the floor).

Pablo.



From moshez at math.huji.ac.il  Tue Jun  6 02:38:58 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 6 Jun 2000 09:38:58 +0300 (IDT)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
In-Reply-To: <393C81D0.36732896@Lugoj.Com>
Message-ID: <Pine.GSO.4.10.10006060938200.4711-100000@sundial>

On Mon, 5 Jun 2000, James Logajan wrote:

> If God had meant for languages to be case insensitive, He would not have
> invented both upper AND lower case letters. If it is good enough for Him, it
> is good enough for me!

GOD CREATED UPPERCASE LETTERS. everything else is the work of man.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From z3penguin at penguinpowered.com  Sat Jun 10 21:04:05 2000
From: z3penguin at penguinpowered.com (Z 3 Penguin)
Date: Sat, 10 Jun 2000 21:04:05 -0400
Subject: Wrapper for pcap?
Message-ID: <8huolp$olv$1@bob.news.rcn.net>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Does anyone have/Has anyone seen a Python wrapper for libpcap?


- -----
Z 3 Penguin
z3p at penguinpowered.com
http://www.z3p.f2s.com/

  ()
  (o_
  //\
<-\_/

This is a virus that works on the honor system.  Send this .sig to
everyone in your address book and randomly delete some of your files.

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.3 for non-commercial use <http://www.pgp.com>

iQA/AwUBOULlglBXOEphiE7DEQJbSACgoQP50ZVO0KTkkwXZc6FxLANK8Y8AoKsb
ZNzptLlQ9UZjRdTSifZz61jh
=ocCV
-----END PGP SIGNATURE-----




From dyeung at ust.hk  Thu Jun 22 03:54:31 2000
From: dyeung at ust.hk (David YEUNG)
Date: Thu, 22 Jun 2000 15:54:31 +0800
Subject: Python interface to Oracle
Message-ID: <3951C636.2E985CE@ust.hk>

Does anyone know if any python module has been developed for
interfacing with Oracle8i database on Linux?

Thanks

david


From tom-main at REMOVEME.home.com  Sat Jun 17 09:41:11 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Sat, 17 Jun 2000 13:41:11 GMT
Subject: Can't find win32all-132
References: <Tvr25.13782$7o1.355124@news2.rdc1.on.home.com> <8ifann$pte$1@phys-ma.sol.co.uk>
Message-ID: <XdL25.17082$7o1.433010@news2.rdc1.on.home.com>

Thanks.

Tom.

"John Parkey" <jparkey at nospamrapidbs.com> wrote in message
news:8ifann$pte$1 at phys-ma.sol.co.uk...
> Tom,
>
> go to
>
> http://starship.python.net/crew/mhammond/downloads/win32all-132.exe
>
> jp
>
> "Tom" <tom-main at REMOVEME.home.com> wrote in message
> news:Tvr25.13782$7o1.355124 at news2.rdc1.on.home.com...
> > A week or two ago Mark Hammond announced that win32all-132 (based on
> 1.5.2)
> > was now available to all.  But I can't find it.  Anywhere.
> >
> > Anyone else?
> >
> > Tom.
> >
> >
>
>




From cjensen at bioeng.ucsd.edu  Wed Jun 28 18:26:36 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Wed, 28 Jun 2000 15:26:36 -0700
Subject: pdb causes seg. fault
Message-ID: <395A7B9C.129C1EE4@bioeng.ucsd.edu>

If I run my script from the debugger, it segment faults.  If I run it
from the interpreter, it works fine.  Any ideas?  The code is too big to
post, and I can't make an example.  Thanks.

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


From thomas at xs4all.nl  Sat Jun 24 12:39:38 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 24 Jun 2000 16:39:38 GMT
Subject: PyType_Type?
References: <Ebo45.13372$Uw3.865124@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <slrn8l9p1v.elg.thomas@centurion.xs4all.nl>

On Thu, 22 Jun 2000 13:06:44 GMT, William Dandreta
<wjdandreta at worldnet.att.net> wrote:
>I am trying to compile the C source code for Python and the linker gives me
>an error message:

>undefined symbol PyType_Type in module bltinmodule.c.

>The problem is I cannot find PyType_Type in bltinmodule or any other module.

The type of a type object is not defined in a module, but in an Object file.
'Objects/typeobject.c' to be exact. 

>Does anyone know what might cause this?

Probably Fawlty compiler/linker flags. What architecture/OS ? What error
message ? Did you pass configure alternative flags ? Do you get any other
error messages ? What version of Python ? Or perhaps you're seeing the
effects of a broken archive or a disk-full during unpacking or such.

Thomas.


From m.faassen at vet.uu.nl  Wed Jun  7 20:03:01 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Jun 2000 00:03:01 GMT
Subject: What is Python3000??
References: <393e426f.77267284@news.online.no> <8hlkdf$nun$1@slb7.atl.mindspring.net>
Message-ID: <8hmnrl$ets$5@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <393e426f.77267284 at news.online.no>,
> Thomas Weholt <thomas at cintra.no> wrote:
>>
>>I want to know more about Python3000. How is it going to be different
>>from Python as we know it? More features, new syntax ?? 

> Nobody -- repeat, NOBODY -- has the answers to these questions.  Py3K is
> a codename for a project THAT HAS NOT YET STARTED.  Py3K is predicated
> to a certain extent -- to what extent, NOBODY KNOWS -- on the Fred
> Brooks maxim, "Build one to throw away."

And it'll be case-insensitive! Hope not. Let the arguments begin! And it'll
have optional type declarations. Hope so. It may even have use '->' in
its syntax to support this. Hope for better syntax. Oh, and it'll use more
intuitive constructs instead of 'and' and 'or'. :) (not!)

<ob flame dousers>
Haskell.
Old TTYs.
Evolution theory.
Cosmic Dutch conspiracy.
</ob flame dousers>

Regards,

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


From Johannes.Nix at mail.uni-oldenburg.de  Fri Jun  2 14:58:26 2000
From: Johannes.Nix at mail.uni-oldenburg.de (Johannes Nix)
Date: 02 Jun 2000 20:58:26 +0200
Subject: planning ALSA PCM audio interface
Message-ID: <8h904j$5e1@news.Informatik.Uni-Oldenburg.DE>

Hi,

I am considering to make a simple interface to the ALSA (Advanced
Linux Sound Architecture) driver library, to allow audio I/O for some
sound cards which don't support the OSS interface. (Actually many
cards are already supported by ALSA, which offers better capabilities
for duplex I/O, for example; however, few applications exist, partly
because the ALSA API ist still alpha and changes frequently). As I
need to use ALSA, I am probably forced to keep up with the API for the
next time, and insulate it pretty nicely from my number crunching.

I am especially interested in doing I/O of digital sample data
(/dev/dsp in OSS notation, PCM interface in ALSA notion), also for
mmap'ed I/O (that means using asynchronous direct access to a DMA
buffer which is mapped to part of Python's virtual memory).
Contrarily I am not interested at all in dealing with MIDI data or
that like. Oh, it will be also 64-bit clean from the beginning.

Does there exist a useful interface for similar tasks ? Perhaps
somebody already is doing this kind of work ?


What would you propose for a simple audio API? (which includes sound
cards for different channels, or voices)?

The module should do the following:

- Open a Soundcard/channel object

- possibly, query hardware preferences from the environment

- get a kind of configuration object

- configure and open the audio device (or throw an exception if the
hardware cannot satisfy the desired setting)

- provide a kind of read() / write() method for data exchange

- alternatively, provide buffer space and sychronization methods for
asynchronous I/O

- watch for buffer underruns/overruns and throw exceptions if they
should crop up

- close the audio interface

What it won't do is data type conversion.

If you want to have a look at the API:

http://www.alsa-project.org/alsa-doc/alsa-lib/

As you see, the API is very clean, but not kind of small. It supports
various soundcards, independent playback and capture, soundcard with
different audio devices at once and multichannel soundcards with
digital I/O.

I think the simplest approach would be to make a hierarchy of objects:
soundcard, device, channel, voice. But perhaps it is not smart to
reproduce the whole C API...

In the SGI audio interface, a lot is done with specific member
functions which set / retreive parameters. I think it could become
perhaps a bit easier with dictionaries and functions with keyword
arguments.


(I am also generelly interested to get contact to people which do
audio programming with Python/NumPy. Here everyone is using MATLAB all
over the place...)

P.S.

Does there exist a "tostring" or "tofile" implementation for the NumPy
array class ? It would be quite useful for me !

Johannes

jnix at medi.physik.uni-oldenburg.de

(our newsserver is cronically unreliable, so I would ask you kindly to
send me a carbon copy. However I am very interested in a thoughtful
discussion!)





From Jyrki.Alakuijala at mail.dosetek.varian.com  Thu Jun 22 03:31:40 2000
From: Jyrki.Alakuijala at mail.dosetek.varian.com (Jyrki Alakuijala)
Date: Thu, 22 Jun 2000 10:31:40 +0300
Subject: Problem in _tkinter.c with extending both tk and python with C++
Message-ID: <3951C0DC.8D2F15C0@iki.fi>

I have an old python/tkinter application that used to work fine
with Python 1.4 and 1.5. Roughly one year ago it got
broken. The application is mostly C++ and Python with some
"embedded" tcl. Now, with 1.5.2 the program crashes
similarly in both Windows NT and Linux environments.

The problem occurs with the following pattern:
First, I load a (self-made) tk widget as a dynamic library.

Second, I load the (self-made) python extensions as dynamic
libraries.

Third, I make an operation in the user interface that
starts as a Tk-event, going to python. Python updates
my widget, which calls C++ code, which calls
(Tcl_VarEval) which tries to visit Python code.
This is when the program crashes.

The error message it gives before termination is:
Fatal Python error: PyEval_RestoreThread: NULL tstate

I believe this has something to do with the thread
safety of tkinter (ENTER_PYTHON and
LEAVE_PYTHON). Perhaps I should take these
macros into my code for visiting Tcl code from C++.
However, I cannot get access the Tcl-lock declared in
_tkinter.c, because it is static.

What should I do to get the code working with
newer releases of Python? Should I give up with
Tk and move to one of the newer UI libraries?
Should I write my own _tkinter.c?
Have I messed it up somewhere else?

--Jyrki




From egbert at bork.demon.nl  Mon Jun 19 08:02:52 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Mon, 19 Jun 2000 14:02:52 +0200
Subject: Three Tkinter questions
Message-ID: <20000619140251.A6860@bork>

Three questions only:

When and why should i use Intvar() etc ?
When do you need mainloop() more than once ?
When and how do i use the textvariable option ?
Examples are very welcome.
egbert
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From jjlucsy at concentric.net  Tue Jun 20 13:10:47 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Tue, 20 Jun 2000 13:10:47 -0400
Subject: Roman Number Representation
References: <S9Q1YMA7B5T5Ew1U@jessikat.demon.co.uk>
Message-ID: <TyN35.7946$Q7.118988@news-east.usenetserver.com>

I remember saving this from the discussion.

--
Joel Lucsy (jjlucsy at concentric.net)


"Robin Becker" <robin at reportlab.com> wrote in message
news:S9Q1YMA7B5T5Ew1U at jessikat.demon.co.uk...
> I could have sworn I had seen a discussion of Roman Numbers somewhere on
> clp, but can't find it. Anyone got any clues?
> --
> Robin Becker


begin 666 roman.py
M#0II;7!O<G0@<F4-"@T*:78@/2!R92YC;VUP:6QE*"))5B(I#0II>" ](')E
M+F-O;7!I;&4H(DE8(BD-"GAL(#T@<F4N8V]M<&EL92 at B6$PB*0T*>&,@/2!R
M92YC;VUP:6QE*")80R(I#0IC9" ](')E+F-O;7!I;&4H(D-$(BD-"F-M(#T@
M<F4N8V]M<&EL92 at B0TTB*0T*:2 @/2!R92YC;VUP:6QE*"))(B I#0IV(" ]
M(')E+F-O;7!I;&4H(E8B("D-"G@@(#T@<F4N8V]M<&EL92 at B6"(@*0T*;" @
M/2!R92YC;VUP:6QE*"),(B I#0IC(" ](')E+F-O;7!I;&4H(D,B("D-"F0@
M(#T@<F4N8V]M<&EL92 at B1"(@*0T*;2 @/2!R92YC;VUP:6QE*")-(B I#0H-
M"F1E9B!2;VUA;DYU;65R86QS5&]$96-I;6%L*')O;6%N7W-T<FEN9RDZ#0H@
M(" @:6UP;W)T(')E+"!S=')I;F<-"B @("!S='(@/2!S=')I;F<N=7!P97(H
M<F]M86Y?<W1R:6YG*0T*(" @(&5X<" ]("(B#0H@(" @:68@:78N<V5A<F-H
M*'-T<BD@(3T at 3F]N93H@('-T<B ](&EV+G-U8B at B*S0B+" @("!S='(I#0H@
M(" @:68@:7 at N<V5A<F-H*'-T<BD@(3T at 3F]N93H@('-T<B ](&EX+G-U8B at B
M*SDB+" @("!S='(I#0H@(" @:68@>&PN<V5A<F-H*'-T<BD@(3T at 3F]N93H@
M('-T<B ]('AL+G-U8B at B*S0P(BP@("!S='(I#0H@(" @:68@>&,N<V5A<F-H
M*'-T<BD@(3T at 3F]N93H@('-T<B ]('AC+G-U8B at B*SDP(BP@("!S='(I#0H@
M(" @:68 at 8V0N<V5A<F-H*'-T<BD@(3T at 3F]N93H@('-T<B ](&-D+G-U8B at B
M*S0P,"(L("!S='(I#0H@(" @:68 at 8VTN<V5A<F-H*'-T<BD@(3T at 3F]N93H@
M('-T<B ](&-M+G-U8B at B*SDP,"(L("!S='(I#0H@(" @:68@(&DN<V5A<F-H
M*'-T<BD@(3T at 3F]N93H@('-T<B ]("!I+G-U8B at B*S$B+" @("!S='(I#0H@
M(" @:68@('8N<V5A<F-H*'-T<BD@(3T at 3F]N93H@('-T<B ]("!V+G-U8B at B
M*S4B+" @("!S='(I#0H@(" @:68@('@N<V5A<F-H*'-T<BD@(3T at 3F]N93H@
M('-T<B ]("!X+G-U8B at B*S$P(BP@("!S='(I#0H@(" @:68@(&PN<V5A<F-H
M*'-T<BD@(3T at 3F]N93H@('-T<B ]("!L+G-U8B at B*S4P(BP@("!S='(I#0H@
M(" @:68@(&,N<V5A<F-H*'-T<BD@(3T at 3F]N93H@('-T<B ]("!C+G-U8B at B
M*S$P,"(L("!S='(I#0H@(" @:68@(&0N<V5A<F-H*'-T<BD@(3T at 3F]N93H@
M('-T<B ]("!D+G-U8B at B*S4P,"(L("!S='(I#0H@(" @:68@(&TN<V5A<F-H
M*'-T<BD@(3T at 3F]N93H@('-T<B ]("!M+G-U8B at B*S$P,# B+"!S='(I#0H@
M(" @97AE8R B;G5M(#T@(B K('-T<@T*(" @(')E='5R;B!N=6T-"B @(" -
M"@T*9&5F($1E8VEM86Q4;U)O;6%N3G5M97)A;',H8F%S93$P7VEN=&5G97(I
M. at T*(" @("<G)U1R86YS;&%T960 at 9G)O;2!A('!U8FQI8R!D;VUA:6X at 0R!R
M;W5T:6YE(&)Y($II;2!786QS:"!I;B!T:&4-"B @("!3;FEP<&5T<R!C;VQL
M96-T:6]N+ at T*(" @("<G)PT*(" @(')O;6%N(#T@(B(-"B @("!N+"!B87-E
M,3!?:6YT96=E<B ](&1I=FUO9"AB87-E,3!?:6YT96=E<BP@,3 P,"D-"B @
M("!R;VUA;B ](")-(BIN#0H@(" @:68 at 8F%S93$P7VEN=&5G97(@/CT at .3 P
M. at T*(" @(" @("!R;VUA;B ](')O;6%N("L@(D--(@T*(" @(" @("!B87-E
M,3!?:6YT96=E<B ](&)A<V4Q,%]I;G1E9V5R("T at .3 P#0H@(" @=VAI;&4@
M8F%S93$P7VEN=&5G97(@/CT at -3 P. at T*(" @(" @("!R;VUA;B ](')O;6%N
M("L@(D0B#0H@(" @(" @(&)A<V4Q,%]I;G1E9V5R(#T at 8F%S93$P7VEN=&5G
M97(@+2 U,# -"B @("!I9B!B87-E,3!?:6YT96=E<B ^/2 T,# Z#0H@(" @
M(" @(')O;6%N(#T@<F]M86X@*R B0T0B#0H@(" @(" @(&)A<V4Q,%]I;G1E
M9V5R(#T at 8F%S93$P7VEN=&5G97(@+2 T,# -"B @("!W:&EL92!B87-E,3!?
M:6YT96=E<B ^/2 Q,# Z#0H@(" @(" @(')O;6%N(#T@<F]M86X@*R B0R(-
M"B @(" @(" @8F%S93$P7VEN=&5G97(@/2!B87-E,3!?:6YT96=E<B M(#$P
M, T*(" @(&EF(&)A<V4Q,%]I;G1E9V5R(#X](#DP. at T*(" @(" @("!R;VUA
M;B ](')O;6%N("L@(EA#(@T*(" @(" @("!B87-E,3!?:6YT96=E<B ](&)A
M<V4Q,%]I;G1E9V5R("T at .3 -"B @("!W:&EL92!B87-E,3!?:6YT96=E<B ^
M/2 U,#H-"B @(" @(" @<F]M86X@/2!R;VUA;B K("),(@T*(" @(" @("!B
M87-E,3!?:6YT96=E<B ](&)A<V4Q,%]I;G1E9V5R("T at -3 -"B @("!I9B!B
M87-E,3!?:6YT96=E<B ^/2 T,#H-"B @(" @(" @<F]M86X@/2!R;VUA;B K
M(")83"(-"B @(" @(" @8F%S93$P7VEN=&5G97(@/2!B87-E,3!?:6YT96=E
M<B M(#0P#0H@(" @=VAI;&4 at 8F%S93$P7VEN=&5G97(@/CT@,3 Z#0H@(" @
M(" @(')O;6%N(#T@<F]M86X@*R B6"(-"B @(" @(" @8F%S93$P7VEN=&5G
M97(@/2!B87-E,3!?:6YT96=E<B M(#$P#0H@(" @:68 at 8F%S93$P7VEN=&5G
M97(@/CT at .3H-"B @(" @(" @<F]M86X@/2!R;VUA;B K("))6"(-"B @(" @
M(" @8F%S93$P7VEN=&5G97(@/2!B87-E,3!?:6YT96=E<B M(#D-"B @("!W
M:&EL92!B87-E,3!?:6YT96=E<B ^/2 U. at T*(" @(" @("!R;VUA;B ](')O
M;6%N("L@(E8B#0H@(" @(" @(&)A<V4Q,%]I;G1E9V5R(#T at 8F%S93$P7VEN
M=&5G97(@+2 U#0H@(" @:68 at 8F%S93$P7VEN=&5G97(@/CT at -#H-"B @(" @
M(" @<F]M86X@/2!R;VUA;B K("))5B(-"B @(" @(" @8F%S93$P7VEN=&5G
M97(@/2!B87-E,3!?:6YT96=E<B M(#0-"B @("!W:&EL92!B87-E,3!?:6YT
M96=E<B ^(# Z#0H@(" @(" @(')O;6%N(#T@<F]M86X@*R B22(-"B @(" @
M(" @8F%S93$P7VEN=&5G97(@/2!B87-E,3!?:6YT96=E<B M(#$-"B @("!R
M971U<FX@<F]M86X-"@T*:68 at 7U]N86UE7U\@/3T@(E]?;6%I;E]?(CH-"B @
M(" G)R=792=L;"!T97-T('1H92!C;VYV97)S:6]N(')O=71I;F5S(&)Y(&-O
M;G9E<G1I;F<@9G)O;2!A(&1E8VEM86P-"B @("!I;G1E9V5R(&X@=&\@82!2
M;VUA;B!N=6UE<F%L(&%N9"!T:&5N(&)A8VL at 86=A:6XN("!)9B!T:&4@;W!E
M<F%T:6]N<PT*(" @(&%R92!N;W0@=&AE(&ED96YT:71Y('1R86YS9F]R;6%T
M:6]N+"!I="=S(&%N(&5R<F]R+ at T*(" @("<G)PT*(" @(&QA<F=E<W1?;G5M
M8F5R(#T at -3 P, T*(" @(&9O<B!N=6T@:6X@>')A;F=E*#$L;&%R9V5S=%]N
M=6UB97(K,2DZ#0H@(" @(" @('-T<B ]($1E8VEM86Q4;U)O;6%N3G5M97)A
M;',H:6YT*&YU;2DI#0H@(" @(" @(&YU;6)E<B ](%)O;6%N3G5M97)A;'-4
M;T1E8VEM86PH<W1R*0T*(" @(" @("!I9B!N=6UB97(@(3T@;G5M. at T*(" @
M(" @(" @(" @<')I;G0@(E)O=71I;F5S(&9A:6QE9"!F;W(B+"!N=6T-"B @
M(" @(" @(" @(')A:7-E(")497-T(&9A:6QU<F4B#0H@(" @<')I;G0@(E1E
-<W0@<&%S<V5D+B(-"@``
`
end



From mhammond at skippinet.com.au  Wed Jun 14 15:22:30 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:22:30 GMT
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com> <dCL15.4673$cQ.38267@news-server.bigpond.net.au>
Message-ID: <8i8m11$bni$1@nnrp1.deja.com>

In article <dCL15.4673$cQ.38267 at news-server.bigpond.net.au>,
  "Neil Hodgson" <neilh at scintilla.org> wrote:

> # Module 'commands'
> #
> # Various tools for executing commands and looking at their output and
> status.
> #
> # NB This only works (and is only relevant) for UNIX.
>
>    It'd be nice if commands was fixed to work on Windows.

It appears the only real Unixism (apart from hard-coding "ls -l" etc)
is the use of os.popen().

It appears very unlikely that os.popen() will be made portable in the
1.6 timeframe.

An option that _could_ make it would be to hack os.py, and attempt to
use win32pipe.popen for os.popen.  There is a precedent -
os.path.abspath() attempts to use win32api.GetFullPathName() if
available.

As other people start making additional/competing Python distributions
for Windows, we will (hopefully) find many of these win32 extensions as
standard.  If we dont, then someone who _does_ bundle them is likely to
have a more successful distribution, IMO.

Any thoughts on that?

Mark.


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


From sabren at manifestation.com  Tue Jun 13 22:53:17 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 13 Jun 2000 22:53:17 -0400 (EDT)
Subject: Just-In-Time importing
Message-ID: <Pine.BSF.4.10.10006132236080.23561-100000@chi.pair.com>

Hey all,

   I think I brought this up a few months ago, but I just now got
around to deciding on an implementation. I have a multi-file module
called "zikebase" that has various classes defined, one per file. For
example, the file, zikebase/Node.py contains a class called Node. I
wanted all of these classes to be available with a simple "import
zikebase", but I *didn't* want the overhead of importing a bunch of
files I wasn't going to need. Here's what I did:

# BEGIN ####################

dbc = my_database_connection

__objs=(
    'Content',
    'Picture',
    'User',
    'UserAuth',
    'ObjectEditor',
    'Node'
    )

for item in __objs:
    _ = "def %s(*args, **kwargs):"           "\n" \
        "    import zikebase, %s"            "\n" \
        "    zikebase.%s = %s.%s"            "\n" \
        "    return apply(%s.%s, args, kwargs)\n"
    _ = _ % ((item,) * 7)
    exec(_)


# END #####################


The nice thing about this technique is that you can 
configure or replace the classes on the fly:

>>> class Vertex():
>>>   pass
>>>
>>> import zikebase
>>> zikebase.dbc = some_other_database
>>> zikebase.Node = Vertex

Now, if zikebase.Content were usually a subclass of zikebase.Node (it
isn't), creating a zikebase.Content() would produce something that was
a subclass of Vertex instead. This can come in really handy if you
want to add something to every class in an object hierarchy and don't
want to (or can't) actually change the original top-level code. (It
sure beats subclassing every item in the tree, or writing a
wrapper-factory)


Cheers,

- Michal
------------------------------------------------------------------------
Zike Interactive    http://www.zike.net/    http://zike.sourceforge.net/
------------------------------------------------------------------------





From alex at magenta.com  Mon Jun 12 05:06:12 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 12 Jun 2000 11:06:12 +0200
Subject: How to tell if I can open a file or not
References: <Pine.GSO.4.10.10006081936380.15963-100000@sundial> <8hokva$2rqa$1@nntp6.u.washington.edu>
Message-ID: <8i29au01om8@news2.newsguy.com>

Donn Cave <donn at u.washington.edu> wrote in message
news:8hokva$2rqa$1 at nntp6.u.washington.edu...
    [snip]
> After you check, there's a period of time before you actually open
> the file.  It's a brief window, but it's real, and many things can
> happen in that time if the system is busy, including changes to the
> file's permission bits or even whether it exists.  Might as well
> just open the file, and if it works then you have a green light.

*EXACTLY*!!!  In the immortal words of Commodore Hopper:

    It's easier to get forgiveness than permission

I think she had another context in mind, but it works great for
the present one too: don't check if you're allowed to do X Y Z
(because things might change between your check and your doing
it), but rather, attempt to do the action ("just do it" comes to
mind:-) *AND*, of course, be ready to handle the exception ("get
forgiveness":-) if it turns out the action-attempt fails.


Alex





From hat at se-46.wpa.wtb.tue.nl  Mon Jun 26 12:12:12 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 26 Jun 2000 16:12:12 GMT
Subject: boolean true and false values.
References: <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl> <memo.20000622231736.45677A@digitig.compulink.co.uk> <slrn8l69kp.geg.hat@se-46.wpa.wtb.tue.nl> <slrn8l6dff.k96.quinn@photo.ugcs.caltech.edu>
Message-ID: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl>

On 23 Jun 2000 10:00:29 GMT, Quinn Dunkan <quinn at photo.ugcs.caltech.edu> wrote:
>
>In C, variable names have types.  So if you say
>int i = 4;
>the symbol "i" has the type int, the number 4 doesn't.  In my (non CS-trained)
>eyes, C doesn't really have any types, since you can change them 'en passant'
>with a cast... or it sorta does, but there's weird holes and tricks.  I guess
>that's why they call it weakly typed.  Anyway, I find the situation
>non-intuitive.
>
>In python, *values* have types.  So 4 is an integer, '4' is a string, and i is
>a symbol, which is just a name for whatever it refers to.  It doesn't have any
>type, it's just a name, the actual data has the type.  Like in a filesystem:
>the file *name* doesn't have a type, it's the contents (text, binary, jpeg,
>etc.).

I got a personal email which said the same thing using a rather nice metaphor
of pointing fingers to objects. I know that quite some things are said about
this stuff in the Python tutorial, but they didn't hit home, until this email.
Tnx for pointing it out to me.

[ for the other readers, consider real-life objects like coffee-mugs or
  telephones as the data or value, and your finger as the variable, pointing to
  the value. The values all have a type (e.g. a telephone is a thing of type
  'telephone'), the finger has no concept of 'point only to things of type X'
]

>So python never dynamically adapts types, except for integer -> float
>promotion things.  It's just that names don't have types at all, values do.

I understand that now.
However, I am not quite sure that that is what one wants.

>
>>Then the lack of booleans.
>>In my work I noticed that having clean and orthogonal concepts helps starting
>>programmers. In our language design the language tries to force users into
>>thinking cleanly and orderly, otherwise they end up with a messy program which
>>the machine doesn't like.
>
>Yeah, well I do think it's easier to grasp "functions that return true are
>true and functions that return false are false, and everything else is an
>error" than "well, this this this and that are false, everything else is
>true".  But having implicit booleans is just plain easier to type.  And if I

Easier to type is imho no argument to justify introducing chances for making
errors. Unless the differences is very large, any time spend trying to find the
error is always larger than time spend to type the extra characters.

>>I realize that introducing new variables on the fly is extremely important for
>>ease of hacking^H^H^H^H^H^H^Hprogramming, but letting it mess up the type
>>concept seems an unwanted side effect to me.
>
>Sounds like you're making a case that all newbies should be taught haskell or
>scheme.  After all, invariants should be preserved.  The best way to do that
>is to make them not vary.

I think I am (is that bad ? :-) ).
I really like functional programming, and i am quite convinced that it is a
very good language to start really clean thinking. However, unless one
understands mathematics, functional programming is not easy to grasp. It gets
even more difficult if users are already contaminated with some other
non-functional programming language (e.g. try explaining to a C programmer that
you can specify complex behaviour without variables).
For this reason, Python is a better alternative imho (of course, our own
language is even better, but nobody on this list knows about the language, so
there is little point in telling that :-) ).

There is imho an important reason why the Python way of typing may be harmfull
for newbies.
Basically, you'd want them to program cleanly, which means that they should use
invariants even if they are not aware of it.

These invariants are always about variables, basically because they are valid
in a general case rather than a specific one. (e.g. one would write 'total
contains the sum of all elements dealt with upto now', rather than '4 is total
of the elements of the list starting with the elements 1,2,1, or the list
starting with 4, or the list starting with 1,3, or ..., 5 is the total of ' etc
etc)

In other words, variables are in reality quite tightly connected to their
values because a meaning is given to the variable.
For example, pointing the variable to a value of another type does not fit in
the idea of invariants. For example, by executing 'total="kjerg"', the
invariant is not longer true, since "kjerg" is not the sum of elements (it is
not even a number).

Thus, to a good programmer, the dynamic nature of variables is not needed. To
bad programmers, it is a tool which can be used quite abusively.
I once met some-one who was programming DB3, another dynamic language aimed at
data bases.
He did things like

count = 234
<use count as integer value>
count = str(count) # assuming that str(234) delivers "234"
<use count as string value>
count = int(count) # assuming that int("234") delivers 234
<use count as integer value>

etc

The first thing I did when I saw it was deciding on a single representation of
count, and inserting the appropiate conversions everywhere necessary.

This person was self-thaught so blaming him is difficult, but a less dynamic
language would have prevented him from doing this in the first place.

I tried explaining why the above was a bad idea, but even the difference
between 234 and "234" is already non-trivial to explain, let alone the above
stuff.


As a wild shot, what would happen if one would limit the use of variables to
values of the same type as the first occurence ?
(thus after x=3, i can only assign integers to x).



Albert
---
Look ma, windows without Windows !!


From jkraska1 at san.rr.com  Fri Jun 23 22:11:36 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 24 Jun 2000 02:11:36 GMT
Subject: Python COM does not seem to like me.
References: <8j0ddv$fqb$1@nnrp1.deja.com>
Message-ID: <39541965.45DCE978@san.rr.com>

> First off, is there something I dont know about the import statement?

The python module must either be in your PATH or your PYTHONPATH
for python to be able to import the module, IIRC.

> TypeError: cant multiply sequence with non-int 
> The statment is:
> CM1 = roll[1] * roll[2] 
> I am pretty much positive roll[1] and roll[2] are ints.

You'd know for sure if on the line before you wrote:

print `roll[1]` + " " + `roll[2]`

I suspect you're going to be surprised by the results.



C/


From anders.eriksson at morateknikutveckling.se  Tue Jun 20 04:38:56 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 20 Jun 2000 10:38:56 +0200
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com>

On Tue, 20 Jun 2000 01:51:54 GMT, "William Dandreta"
<wjdandreta at worldnet.att.net> wrote:

>Is there a Python module that has functions in it to manipulate Dbase tables
>and indexes?
>

Hello!

You don't say which version of dBASE and not on which platform you
are, so let's assume that you are using Windows 9x/NT/2000.

The only real alternative today is to use ODBC!

The win32-all comes with an ODBC modules that works OK, if you want to
spice it up a bit get mxODBC (seach on Vaults of Parnassus)

// Anders




From hungjunglu at hotmail.com  Fri Jun 16 16:21:03 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Fri, 16 Jun 2000 13:21:03 PDT
Subject: Reloading modules dynamically
Message-ID: <20000616202103.23388.qmail@hotmail.com>

This is probably not the most efficient hack. But it works. Make the 
following 3 Python files, and then run M1.py. After than, modify the 
messages in M2.py and/or M3.py, and you should see the changes in real-time. 
I don't know how to write a load_module() function without the extra scope 
argument... I attempted to simplify things, but did not succeed... scope 
management was kind of tricky. The hack on __builtins__ is usually not 
recommended, but since you need load_module() all over places, I guess it's 
OK. Let me know if there is any problem.

regards,

Hung Jung

---- M1.py ----------------------------------
import sys
import time

def load_module(module_name, scope):
    if scope.has_key(module_name):
        print '... reloading ' + module_name
        reload(scope[module_name])
    else:
        print '... importing ' + module_name
        exec 'import %s' % module_name in scope

__builtins__.__dict__['load_module'] = load_module

while 1:
    load_module('M2', globals())
    M2.print_me()
    time.sleep(5.)

---- M2.py ----------------------------------
load_module('M3', globals())

def print_me():
  print 'M2 loaded nice'
  M3.print_me()

---- M3.py ----------------------------------
def print_me():
  print 'M3 loaded cool'

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From Sttscitrans at tesco.net  Wed Jun 14 12:44:36 2000
From: Sttscitrans at tesco.net (Iain Davidson)
Date: Wed, 14 Jun 2000 17:44:36 +0100
Subject: Wondering about Domingo's Rational Mean book"
References: <pn5djsceirmrdhdd2ncht9c26hv0q5gkdq@4ax.com> <ralyh3rlx4xa@forum.mathforum.com> <8hp1g8$47o$1@barcode.tesco.net> <8i475e$48n$1@nnrp1.deja.com> <8i5vsn$llr$1@epos.tesco.net> <8i7vo0$p8v$1@nnrp1.deja.com>
Message-ID: <8i8cdd$ndh$1@barcode.tesco.net>

<paraguana at my-deja.com> wrote in message news:8i7vo0$p8v$1 at nnrp1.deja.com...
> In article <8i5vsn$llr$71 at epos.tesco.net>,
>   "Iain Davidson" <Sttscitrans at tesco.net> wrote:
> >
> > > "Iain Davidson" <Sttscitrans at tesco.net> replied:

You haven't explained yet how your method finds best
approximations to sqrt(n) without first  knowing
a best approximation.
How would your method find all best approximations
to sqrt(61) ?


> Now, considering that this not a one-way discussion I?m sure you
> will be so kind and take some time to formulate here those CFs method
> you mentioned for finding a sequence of best approximations for the
> cube root of any number, I?m sure it will be amusing and very useful
> not only for me but for the whole audience of this newsgroup.

No I was not taking about best approximations to the cube root of a
number - you can do that with CFs.

What I was talking about was simultaneous approximation, say, of the
ratio cubrt(25): cubrt(5):1. As rational means are the basis of all
rational approximation theory,  you should be able to come up
with an effective method. After all, this is the claim you are
making.




From mfletch at tpresence.com  Tue Jun 13 17:17:06 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 13 Jun 2000 17:17:06 -0400
Subject: help formatting number
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33B9@VRTMAIL>

http://x53.deja.com/[ST_rn=ps]/getdoc.xp?AN=630397831

And the related thread may help.

Enjoy,
Mike

-----Original Message-----
From: Mario [mailto:mtbros at tin.it]
Sent: Tuesday, June 13, 2000 4:01 PM
To: python-list at python.org
Subject: help formatting number


Hi..
this is a simple question but I can't find a simple answer... so help me.

I have an integer variable and I want it to become a string variable
formatted so that there is a dot between each three digits of the original
number. An example:
var =3456789
It must become:
"3.456.789"
Ideas ?
 Thanks
        Mario





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



From kmself at ix.netcom.com  Mon Jun 26 23:30:57 2000
From: kmself at ix.netcom.com (kmself at ix.netcom.com)
Date: Mon, 26 Jun 2000 20:30:57 -0700
Subject: profiler -- numeric overflow?
Message-ID: <h579j8.grg.ln@kmself.nntp.ix.netcom.com>

I'm new to Python.

I'm maintaining, extending, and optimizing in inherited weblog analyzer.
It runs in about an hour on a 420,000 line logfile, which strikes me as
longer than necessary, so I ran the code through the Python profiler.
Interesting results, but apparently a number of overflow errors, output
copied inline below.  I'm assuming an internal representation of 1/1000
second, which will overflow for large values -- I'm assuming 2^20 is the
maximum 

I'm not fully up on all my Python functions and internals, The intensive
calls appear to be:
    - <string> ( 1 call, -2605.767 cumtime).
    - __call__ (line 76, 4,135,458 calls, -4029.837 tottime)
    - _regularize (166 seconds -- internal function


>From this, does anyone have suggestions on how to modify the profiler so
that it doesn't overflow?  I've located what appears to be the source
under /usr/lib/python1.5/profile.py (Debian GNU/Linux).  Would it be
possible to change the time values to longs, and/or would this
negatively impact performance.

...yes, I've also considered running a smaller sample of the log, but I
suspect that runtime may grow nonlinearly, large input samples are good.


13166270 function calls (12889814 primitive calls) in -2605.747 CPU seconds

   Ordered by: standard name

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.000    0.000 -2605.767 -2605.767 <string>:1(?)
        1    0.000    0.000    0.000    0.000 getopt.py:12(?)
        1    0.000    0.000    0.000    0.000 getopt.py:20(getopt)
        2    0.000    0.000    0.000    0.000 getopt.py:64(do_longs)
        2    0.000    0.000    0.000    0.000 getopt.py:85(long_has_args)
   362448   50.870    0.000  166.910    0.000 hitcount3.py:131(_regularize)
        5    0.000    0.000    0.000    0.000 hitcount3.py:141(__init__)
   190108    9.360    0.000    9.360    0.000 hitcount3.py:153(increment_dict)
        3    0.020    0.007    0.190    0.063 hitcount3.py:158(__repr__)
      321    0.050    0.000    0.050    0.000 hitcount3.py:169(<lambda>)
        1    0.000    0.000    0.000    0.000 hitcount3.py:177(__init__)
   384665   71.950    0.000  145.580    0.000 hitcount3.py:180(process_line)
        1    0.000    0.000    0.000    0.000 hitcount3.py:192(__init__)
   384665   69.810    0.000  143.600    0.000 hitcount3.py:195(process_line)
        1    0.000    0.000    0.000    0.000 hitcount3.py:205(__init__)
   384665   68.060    0.000  371.640    0.001 hitcount3.py:209(process_line)
        1    0.000    0.000    0.000    0.000 hitcount3.py:216(__repr__)
        1    0.000    0.000    0.000    0.000 hitcount3.py:221(__init__)
   384665   33.220    0.000  158.360    0.000 hitcount3.py:224(process_line)
        1    0.000    0.000    0.000    0.000 hitcount3.py:232(__init__)
   384665   33.650    0.000   54.130    0.000 hitcount3.py:235(process_line)
        1    0.000    0.000    0.010    0.010 hitcount3.py:241(__repr__)
      134    0.000    0.000    0.000    0.000 hitcount3.py:249(<lambda>)
       33    0.010    0.000    0.010    0.000 hitcount3.py:253(<lambda>)
        3    0.080    0.027    0.120    0.040 hitcount3.py:258(sort_by_value)
     1677    0.040    0.000    0.040    0.000 hitcount3.py:262(<lambda>)
        1    0.000    0.000    0.000    0.000 hitcount3.py:282(__init__)
   405299   35.170    0.000   65.460    0.000 hitcount3.py:284(__call__)
        1    0.000    0.000    0.000    0.000 hitcount3.py:289(__init__)
   424910   18.380    0.000   18.380    0.000 hitcount3.py:293(__call__)
        2    0.000    0.000    0.000    0.000 hitcount3.py:299(datestring_to_time)
        1  309.710  309.710 -2605.767 -2605.767 hitcount3.py:311(main)
   424910   71.730    0.000  265.550    0.001 hitcount3.py:62(__init__)
4135458/3859002 -4029.837   -0.001 -3649.137   -0.001 hitcount3.py:76(__call__)
  1172023  193.480    0.000  395.050    0.000 hitcount3.py:89(_compute_data)
        1    0.020    0.020 -2605.747 -2605.747 profile:0(main())
        0    0.000             0.000          profile:0(profiler)
  1192657  181.410    0.000  220.470    0.000 re.py:112(match)
  1057390   52.320    0.000   52.320    0.000 re.py:290(__init__)
   344808   41.110    0.000   41.110    0.000 re.py:335(group)
   424910   78.570    0.000   78.570    0.000 re.py:360(groupdict)
        1    0.000    0.000    0.000    0.000 re.py:76(compile)
        1    0.000    0.000    0.000    0.000 re.py:89(__init__)
  1105824  105.070    0.000  118.330    0.000 re.py:95(search)
        2    0.000    0.000    0.000    0.000 string.py:213(index)



-- 
Karsten M. Self <kmself at ix.netcom.com>         http://www.netcom.com/~kmself
  Evangelist, Opensales, Inc.                       http://www.opensales.org
   What part of "Gestalt" don't you understand?      Debian GNU/Linux rocks!
     http://gestalt-system.sourceforge.net/      K5: http://www.kuro5hin.org
GPG fingerprint: F932 8B25 5FDD 2528 D595  DC61 3847 889F 55F2 B9B0


From wjdandreta at worldnet.att.net  Fri Jun  9 01:22:59 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 09 Jun 2000 05:22:59 GMT
Subject: Why is Python a memory hog?
Message-ID: <Ta%%4.1757$iy.138198@bgtnsc06-news.ops.worldnet.att.net>

I downloaded Python 1.2 to run under DOS without a memory extender.  The
interpreter seems to take up almost all of the 640K. When I try to run all
but the tinyest programs I get an out of memory error.

Is there anything I can do to improve this situation?

Bill Dandreta




From avle at rogn.ii.uib.no  Sat Jun 17 09:39:54 2000
From: avle at rogn.ii.uib.no (Helge Avlesen)
Date: Sat, 17 Jun 2000 09:39:54 -0400
Subject: Python vs. Matlab
Message-ID: <642A954DD517D411B20C00508BCF23B001296E71@mail.sauder.com>

I have become almost addicted to Matlab after using it extensively for
some years, however, the licencing feels very strict after getting
used to free and very good quality software like Emacs, Linux, etc.

My main use of Matlab is reading in large binary files into
arrays(output from simulations), do some processing with various
matrix routines, then extract sections of the 3D arrays for plotting.

Spotted Python some months ago, and liked it!  Can Python, Numpy and
the many plotting extensions be used as a replacement for Matlab, is
Python slower/faster etc.?  

It would be nice to hear some experiences!

-- 
Helge Avlesen



From dgomezm at etheron.net  Wed Jun 14 19:59:02 2000
From: dgomezm at etheron.net (dgomezm at etheron.net)
Date: Wed, 14 Jun 2000 23:59:02 GMT
Subject: Wondering about Domingo's Rational Mean book"
References: <pn5djsceirmrdhdd2ncht9c26hv0q5gkdq@4ax.com> <ralyh3rlx4xa@forum.mathforum.com> <8hp1g8$47o$1@barcode.tesco.net> <8i475e$48n$1@nnrp1.deja.com> <8i5vsn$llr$1@epos.tesco.net> <8i7vo0$p8v$1@nnrp1.deja.com> <8i8cdd$ndh$1@barcode.tesco.net>
Message-ID: <8i967u$oh3$1@nnrp1.deja.com>




You insist to rise differences between Cfs and the Rational
Process, however, there is no difference, at all. You were not aware of
that,
however by this time you certainly  know that Continued fractions are
just a particular case of the rational process ruled by the rational
mean.

I clearly stated that point, moreover, at the beginning of this thread
Kirby Urner?s remark was very much to the point, he got the point and
cleverly told you that this was not a problem of CFs vs. Rational
process.
Continued fractions are ruled by the rational mean as well as
Bernoulli?s, Newton?s and Halley?s, among many other new methods, all
of them are just a rational process ruled by the rational mean (as
stated in my web page and book). That?s why the rational process is a
new fundamental and general concept.

If you need to see some best approximations then go to my web page,
enjoy some of them and do some homework by yourself, there you have
many rational processes i.e.: Bernoulli?s, Newton?s, Halley?s and many
others, I hope you enjoy them and please any question on best
approximations ask to Bernoulli, Newton and Halley, the problem is
that  they were unfortunately  unaware of the __trivial and
obvious__    fact  that their methods can be __trivially and
obviously__   found  by means of simple ARITHMETIC: The Rational
process).
No Geometry, no decimals, no derivatives, no Cartesian system.
JUST the most simple ARITHMETIC.
WORST!, there are no precedents on this matter, at all


Of course, you are free to  continue your self-talk on CFs vs. Rational
Process. You insist to ignore all kind of questions and remarks, and it
is clear that even by this time you haven?t read my web page, so I
shall reply _your_ messages by repeating my previous one.


Again:
If you have some doubts, something to say,  about the rational mean and
the rational process, then all your arguments are straightaway directed
to the CFs, Bernoulli?s, Newton and Halley?s method, because all of
them are ruled by the rational mean (They are just rational processes)
as has been trivially and obviously stated in my book and summarized in
my home page:
www.etheron.net/usuarios/dgomez/default.htm

That?s why it is just striking to realize that western culture needed
more than two or three thousand years to find Bernoulli?s, Newton?s and
Halley?s methods by means of previously inventing the infinitesimal
calculus and the cartesian system, also that all  their most
outstanding mathematicians never realized that such methods could have
been  easily developed by using simple arithmetic.

Just striking, indeed, specially for a humble purchaser
of "rigurous" math books containing so much rigorous talk about best
approximations and roots solving.

Finnally, considering that this must be a two-way discussion I?m sure
you will be so kind and take some time to formulate here those CFs
method you mentioned for finding a "fundamental and general" sequence
of best approximations for some cube roots, I?m sure it will be amusing
and very useful not only for me but for the whole audience of this
newsgroup.
 Also I would like to get any response from you about the other issues
I rised on any  precedents on developing
Bernoulli?s, Newton?s, and specially Halley?s method
by means of very simple arithmetic, if I?m wrong on this then
YOU SHOULD BRING OUT, at least, two or three thousands years,
Oooops...sorry,
I meant to say "two of three thousands references" (books, articles,
etc) on such a very simple arithmetical matter.


Domingo





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


From aahz at netcom.com  Thu Jun 22 10:16:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jun 2000 14:16:18 GMT
Subject: Am I the only one to receive the messages several times????
References: <7BD10B680501D411B9DF009027E06F32011AD2@EXCHANGE>
Message-ID: <8it73i$of7$1@slb2.atl.mindspring.net>

In article <7BD10B680501D411B9DF009027E06F32011AD2 at EXCHANGE>,
Max M?ller Rasmussen  <maxm at normik.dk> wrote:
>
>This starting to be a drag.

There was one spew a couple of days ago; I wrote to python-list-admin
and haven't seen the problem since.  Note that I'm on the newsgroup, not
the mailing list, so there may be a different problem involved.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From nobody at nowhere.nohow  Sat Jun 10 10:32:35 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 10 Jun 2000 14:32:35 GMT
Subject: Tkintr: how to seize focus in a transient?
References: <voR%4.1057$zm3.238149@ptah.visi.com>
Message-ID: <7ks05.1254$zm3.340136@ptah.visi.com>

>In a transient dialog window, how do I disable focus for the
>rest of the app?  I only want to prevent focus for the rest of
>the app, not the whole X11 system.

I found it. In case anybody else is wondering, you call the
grab_set() method of the transient window.

[I always hate when I find on deja.com that somebody else has
asked a question which I am researching, and never posts any
results.]

-- 
Grant Edwards                   grante             Yow!  Now, I think it would
                                  at               be GOOD to buy FIVE or SIX
                               visi.com            STUDEBAKERS and CRUISE for
                                                   ARTIFICIAL FLAVORING!!


From andy.elvey at paradise.net.nz  Sun Jun  4 03:39:02 2000
From: andy.elvey at paradise.net.nz (Andy Elvey)
Date: Sun, 04 Jun 2000 19:39:02 +1200
Subject: Query about using "DataHandlers" module
Message-ID: <393A0796.4E5254D0@paradise.net.nz>

Hi all.  I have a query about the use of the DataHandlers module
(available at "Vaults of Parnassus").

 I have attached a simple  example which uses this module (this example
works fine).  But .... when I amend the SQL code to say  'select * from
test where team == "Yankees" '  ,  the code still retrieves all teams -
the where statement is ignored.

  My question is - "does anyone have an example (or two!) of some Python
code similar to my example , but which has a select statement that has a
"where" clause in it?

 ( and -  are there any examples which use the DataHandlers module and
use * 2 tables*   with a join ... ? )

 Many thanks in advance for any help offered .... :-)

 ( and by the way .. this DataHandlers module is *great*  ..... :-)


-------------- next part --------------
#!/usr/bin/env python

# Illustrate the parsing a data file and creating SQL insert statements
# The table definition comes from an XML file

from DataHandlers.SQLTable import SQLTable
from DataHandlers.dbdefXMLParser import dbdefXMLParser
from DataHandlers.FileConnection import Connection

# create a parser
xml = dbdefXMLParser()
# return a dictionary of table definitions
tables = xml.parseFile( 'test.xml' )
# get the baseball table
tbl = tables['baseball']
# the data file is delimited by commas
tbl.setReader( 'Delimited' )
tbl.setDelimiter( ',' )
# open a file connection
fc = Connection()
fc.connect( 'teams.txt' )
# we need to do this fake query so we can get the cursor
# it actually means something when we are doing an SQL thing
cursor = fc.query('')
if not cursor:
    print 'Unable to execute query' 
    sys.exit(1)
# now walk thru the lines of data
while cursor.next():
    tbl.parse( cursor.getData())
    # print out the insert statements
    print tbl.getInsert()
    
print "Query data returned..."
cursor = fc.query( 'select * from test' )
while cursor.next():
    print cursor.getData( )

cursor.close()
    
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.xml
Type: text/xml
Size: 316 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000604/2ced7da0/attachment.xml>
-------------- next part --------------
Orioles,Baltimore,Camden Yards,A,E
Red Sox,Boston,Fenway Park,A,E
Angels,Anaheim,Anaheim Stadium,A,W
White Sox,Chicago,Comiskey Park,A,W
Indians,Cleveland,Jacobs Field,A,E
Tigers,Detroit,Tiger Stadium,A,E
Royals,Kansas City,Royals Stadium,A,W
Brewers,Milwaukee,County Stadium,A,E
Twins,Minnesota,Hubert H. Humphrey Stadium,A,W
Yankees,New York,Yankee Stadium,A,E
Athletics,Oakland,Oakland Coliseum,A,W
Mariners,Seattle,Kingdome,A,W
Rangers,Texas,Arlington Stadium,A,W
Blue Jays,Toronto,Toronto Stadium,A,E
Braves,Atlanta,Fulton County Stadium,N,W
Cubs,Chicago,Wrigley Field,N,E
Reds,Cincinnati,Three Rivers Stadium,N,W
Astros,Houston,Astrodome,N,W
Dodgers,Los Angeles,Dodger Stadium,N,W
Expos,Montreal,Exposition Stadium,N,E
Mets,New York,Shea Stadium,N,E
Phillies,Philadelphia,Veterans Stadium,N,E
Pirates,Pittsburgh,Pirate Stadium,N,E
Cardinals,St. Louis,Busch Stadium,N,E
Padres,San Diego,Jack Murphy Stadium,N,W
Giants,San Francisco,Candlestick Park,N,W

From richard_chamberlain at ntlworld.com  Wed Jun  7 03:47:13 2000
From: richard_chamberlain at ntlworld.com (RS Solutions)
Date: Wed, 7 Jun 2000 08:47:13 +0100
Subject: Real basic question about Entry box's data
References: <8hkqde$9cv$1@leopard.it.wsu.edu>
Message-ID: <8hkuld$fd$1@gxsn.com>

from Tkinter import *
root=Tk()
entry=Entry(root)
var=StringVar()
entry2=Entry(root,textvariable=var)
entry.pack(side=TOP)
entry2.pack(side=TOP)

def calculate():
    var.set(entry.get())

button=Button(root,text="Calculate",command=calculate)
button.pack(side=RIGHT)
root.mainloop()

Getting and setting an Entry widget is straightforward. In the above example
I've used entry.get() to get the contents of a widget and I've used the
textvariable option to set the value.

var=StringVar()

This creates an instance of StringVar that we can assign to our entry.

entry2=Entry(root,textvariable=var)

So every change to var is now reflected in the widget.

var.set(entry.get())

I'm simply changing var to the contents of entry. Obviously here you'll need
to do some calculations and pass that as a string.

In the Grayson book look at the appendices at the back to check out the
methods and options of Entry.

Richard



kent snyder <__kent__ at excite.com> wrote in message
news:8hkqde$9cv$1 at leopard.it.wsu.edu...
> I'm trying to capture an Entry box's value and then print it out.
Actually
> I'd like to do some math with it and return the answer to another Entry
box.
> I have the Grayson book and I can't find any SIMPLE, DIRECT guidance on
> this.  Can anyone offer direction or a link to some sample code that's not
> all cluttered up?
>
> I'd like to NOT use PMW if at all possible.
>
> Thanks a LOT in advance,
> Kent
>
>




From gmcm at hypernet.com  Mon Jun  5 17:03:26 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 05 Jun 2000 21:03:26 GMT
Subject: win32Service.OpenService not working
References: <8hggdu$4fa$1@nnrp1.deja.com>
Message-ID: <8F4AAD3B8gmcmhypernetcom@199.171.54.155>

tiddlerdeja at my-deja.com wrote in <8hggdu$4fa$1 at nnrp1.deja.com>:

>Can you help?
>
>I'm trying to get a handle on a service but can't. If I use the
>following code, it works:
>h = win32service.OpenService(hscm, "UPS",
>win32service.SERVICE_ALL_ACCESS)
>
>However, if I try to get a service with a space in the name, I get an
>error:
>>>> h = win32service.OpenService(hscm, "Windows installer",
>win32service.SERVICE_ALL_ACCESS)

Are you sure that's not the "descriptive name" instead of the "service 
name"? The services applet shows you descriptive names. Off hand, I'm not 
sure how to find the real name.

- Gordon


From jhe at webde-ag.de  Thu Jun 29 12:38:53 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Thu, 29 Jun 2000 18:38:53 +0200
Subject: traceback string
References: <8jfnn5$1fpo$1@news2atm.raleigh.ibm.com>
Message-ID: <8jfua8$f8g$1@pollux.ip-plus.net>

"Arint?" <shouldbe at message.com> schrieb im Newsbeitrag
news:8jfnn5$1fpo$1 at news2atm.raleigh.ibm.com...
> How do you print the traceback string from an exception
> try:
>    whatever()
> except:
            import traceback, string
            msg = string.join(traceback.format_exception(
                    sys.exc_info()[0], sys.exc_info()[1],
sys.exc_info()[2]), "")
            print msg





From jam at quark.emich.edu  Wed Jun 21 01:22:48 2000
From: jam at quark.emich.edu (Jeff)
Date: Wed, 21 Jun 2000 01:22:48 -0400
Subject: EMail Monitoring in Linux
In-Reply-To: <8iph0p$ekl$1@nnrp1.deja.com>; from buttnerw@my-deja.com on Wed, Jun 21, 2000 at 04:40:58AM +0000
References: <8iph0p$ekl$1@nnrp1.deja.com>
Message-ID: <20000621012248.D18093@quark.emich.edu>

On Wed, Jun 21, 2000 at 04:40:58AM +0000, buttnerw at my-deja.com wrote:
> Good Day
> 
> I need to know the easiest way to do the following.  Whenever the
> mailing program (sendmail) on a mail server (linux) receives a mail for
> a user, a "program" needs to be notified about the incoming mail and a
> UDP message has to be send with the user's information and the amount
> of mails in his inbox.  All I need is the "program" which will listen
> to incoming mails, and send the UDP packet.
> 
> Kind Regards
> 

greetings,

I haven't tried this personally, but couldn't you just hold a path to the
user's mailbox file, and check it's modification time? if it differs from
the last sample, you could then open it up and scan it to get the count of
messages. you have to consider locking issues, so I'd probably still check
out an implementation of 'biff' or something similar first just to be sure,
but without thinking about it further, it *seems* like it would work. once
you have *that* part figured out, you can build the network stuff on top.

you can check 'http://freshmeat.net/' (or, for example,
'http://freshmeat.net/search.php3?query=biff') for existing biff-ish
programs.

hope that helps..

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 



From manni.heumann at gmx.de  Wed Jun 14 14:29:11 2000
From: manni.heumann at gmx.de (Manni Heumann)
Date: Wed, 14 Jun 2000 18:29:11 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <394b3e7d.284293552@news.online.no> <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net> <8i8ebi$5et$1@nnrp1.deja.com>
Message-ID: <3947cef9$1_1@news.uni-bielefeld.de>

spenrose at my-deja.com wrote:
[SNIP]
>FWIW, I agree with those who think Python an ideal choice for a first
>language--it worked for me! It's just the orientation of the two
>O'Reilly books I'm referring to here.
>
[SNIP]

If anybody cares: there was a discussion on slashdot recently about 
teaching kids how to program, and especially what language they should 
use. Ok, so I'm not a kid and I already know C and C++, but this 
discussion made me take a closer look at python.

http://slashdot.org/askslashdot/00/05/21/1919205_F.shtml


--

Manni



From thomas at xs4all.net  Sat Jun 10 15:45:59 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Jun 2000 21:45:59 +0200
Subject: Type problem on Windows
In-Reply-To: <eviltofu-9D1688.14430510062000@news.earthlink.net>; from eviltofu@rocketmail.com on Sat, Jun 10, 2000 at 06:43:11PM +0000
References: <eviltofu-B4ED2B.12472010062000@news.earthlink.net> <02d101bfd318$8091ae10$6401a8c0@home> <eviltofu-9D1688.14430510062000@news.earthlink.net>
Message-ID: <20000610214559.K20700@xs4all.nl>

On Sat, Jun 10, 2000 at 06:43:11PM +0000, Jerome Chan wrote:
> In article <02d101bfd318$8091ae10$6401a8c0 at home>, "Darrell Gallion" 
> <darrell at dorb.com> wrote:

> > sockets are platform dependent as expected :)

> Eeek. So maybe I should wrap the socket object in my own class and then 
> test for that class.

No. The best thing to do is (usually) to assume that the object passed in is
indeed a socket, and use it as such. Document that the function takes a
socket *or a socket-like object* as argument, possibly stating which methods
of sockets it is going to use, and let users of the function decide for
themselves wether they *really* want to pass you a socket, or a
makebelieve/wrapper/debug class.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From sugaking at gis.net  Wed Jun 21 19:55:41 2000
From: sugaking at gis.net (SugarKing)
Date: Wed, 21 Jun 2000 18:55:41 -0500
Subject: Any other good tutorials?
Message-ID: <sl2i51h0e7f149@corp.supernews.com>

     Are there any other good tutorials besides Guido van Rossum, I don't
think he's very explainatory. I like full explainations:)





From scarblac-spamtrap at pino.selwerd.nl  Thu Jun 22 06:29:44 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 22 Jun 2000 10:29:44 GMT
Subject: url directory listing
References: <39518289.D4175618@bioeng.ucsd.edu>
Message-ID: <slrn8l3qfm.nol.scarblac-spamtrap@flits104-37.flits.rug.nl>

Curtis Jensen wrote in comp.lang.python:
> I little while ago, there was a post about websucker.  I'm writing a
> similar script. What is the socket command or url command to get a
> directory listing?  Especialy if there is an index.html file in the
> directory.  Thanks.
 
This is not possible with HTTP.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From petersc at stud.ntnu.no  Sun Jun  4 07:09:54 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 13:09:54 +0200
Subject: Found the problem! Was Re: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home>
			<39397FD4.67C7FD73@inc.it> <867lc64a47.fsf@g.local> <393A2B6D.4B33A5C8@inc.it>
Message-ID: <393A3902.F67CD1F8@stud.ntnu.no>

Paolo Redaelli wrote:
> 
> I found the problem. It's related to lists; it seems to me that python
> does not allow smoothly things like:
> numbers = [n,m] =  map(int, numbers)

Well, that seems to work for me. At least with numbers as a
list of two floats.

strange-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thomas at cintra.no  Fri Jun 16 03:51:26 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 16 Jun 2000 07:51:26 GMT
Subject: PalmPilot computing on the Windows platform
Message-ID: <3949dc45.1104177@news.online.no>

Has anybody tried using Python to program the 3COM Palm
handheld-device on Windows? 

Pyrite seem to work only with Pilot-Link.

Thomas


From sdm7g at virginia.edu  Sat Jun 24 15:00:34 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Sat, 24 Jun 2000 15:00:34 -0400 (EDT)
Subject: Darwin (Not Python, yet...)
Message-ID: <Pine.A32.3.90.1000624142526.18059A-100000@elvis.med.Virginia.EDU>

OK -- Question first and then the report: 

  What's the state of Python on Darwin? 
  Any one do a build yet with X11, Tkinter, ( or WxWindows or Qt or ... )?
  Any tips? 
  
  Darwin and OSX integration: Is this close enough to make one version 
    (with DLLs to support all of the options, for example, if you're 
     running Darwin, you may have X11, but you don't have Carbon, etc.)


  Python isn't included in Apple's Scripting CVS directory, and I don't 
   see any links in the Darwin pages at Stepwise or elsewhere. 

Report:

I just loaded Darwin 1.0.2 binary on my 8500 Mac at work, connected
to apple's CVS server, checked in and built XFree86 (overnight!) 
and got it running, all with only a few hitches:

 1. Although the installer seems to indicate that it will install
    (and erase) on a partition, it actually needs a free disk with
    one partition -- whenever I tried it on a one-gig partition, 
    the installation hung up on me. 

 2. Read the FAQs about starting DNS. 
    ( Other than DNS, the networking came up live -- it must have 
      used BOOTP or DHCP to get an IP address ? ) 

 3. There is a binary distribution of lynx available (insert URL here...)
    Get it so that you can read the FAQs and mailing list archives once
    you've installed it and don't have you Mac's browser available. 

 4. You probably want to get registered on Apple's publicsource web
    BEFORE you install, so that you can log into the CVS server -- 
    you'll need to keep that password. 

 5. The product name for checking in X is "XFree86" 
    ( I had to use the CVS Web browser interface to find this -- one
      of those many things lynx came in handy for. ) 

  setenv CVSROOT :pserver:USER at anoncvs.publicsource.apple.com:/cvs/Darwin
  cvs login 
    <asks password...>
  cvs -z3 checkout XFree86

    <cd to directory>

  make World
    <on an 8500/120, go home an get some sleep!>
  make install  

  <add /usr/X11R6/bin to path>

  startx 



Reading thru the mailing list archives, I see that it's going to be 
impossible (right now) to boot Darwin on my 6100 at home. I'm really
thinking about going out and getting an IBook real soon so  I can 
play with this. 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

"There's a new programming language called C+++. It's 50% better than C++, 
         but every time you compile, your modem hangs up."





From neilh at scintilla.org  Sat Jun 17 06:16:13 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 17 Jun 2000 10:16:13 GMT
Subject: Ret status technique popen, popen2 (NT)
References: <3949a528.1161643@news.pipeline.com> <2Kq25.372$qz4.3272@news-server.bigpond.net.au> <hgC25.714$qz4.6314@news-server.bigpond.net.au> <394b2fed.3375663@news.pipeline.com>
Message-ID: <NdI25.1027$qz4.8474@news-server.bigpond.net.au>

> Thanks, I'm on W95a for the weekend so have given it a try on that
> with mixed results. The zop test script works well but I haven't
> manged to get output from anything else ("dir") as the pipe appears to
> close immediately. e.g.

   The dir command is part of the command processor rather than an external
executable so has to be run like this on NT:
p=piper.PipedCommand("cmd /c dir")
    or on 95
p=piper.PipedCommand("c:\\command.com /c dir")
   although the current command processor should be found from the
environment variable COMSPEC instead of hard coding it.

   You could run all commands with the additional layer of the command
processor if you want. I didn't in piper as there is a bit of overhead to
doing this.

> >>> p=piper.PipedCommand("c:\\utils\where lha.exe")
    Try adding the command processor in here. Is "where" a batch file?

    Neil





From aahz at netcom.com  Wed Jun 28 10:10:12 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 28 Jun 2000 14:10:12 GMT
Subject: Ptyhon Thread Safety
References: <8jcu92$7op$1@nnrp1.deja.com>
Message-ID: <8jd104$ubt$1@slb0.atl.mindspring.net>

In article <8jcu92$7op$1 at nnrp1.deja.com>,
Dale Burnett  <dburnett at mindspring.com> wrote:
>
>What is the status of thread safety in Python?  I saw some information
>stating that Python wasnt thread safe but not which version.

This is a somewhat complicated question.  The basic answer is that
Python is thread-safe if you use it correctly, particularly in the
context of pure Python; Python is *not* thread-safe if you have multiple
C/C++ threads making calls into a single interpreter session unless you
are very, very, very careful.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From erno at iki.fi  Sat Jun 24 14:21:18 2000
From: erno at iki.fi (Erno Kuusela)
Date: 24 Jun 2000 21:21:18 +0300
Subject: realtime non-blocking keyboard input
References: <395437F6.73BFE6B3@flashmail.com>
Message-ID: <ku7lbfvt29.fsf@lasipalatsi.fi>

>>>>> "Jp" == Jp Calderone <exarkun at flashmail.com> writes:

    Jp>  Any way *other* than using the curses library to check to see
    Jp> if a key has been pressed, and if so, which one?  Or any way
    Jp> to prevent curses.initscr() from messing w/ normal prints?

yep, you can do it with the termios module (or with system() & stty) if
you dare.

here is some code for one-key-at-a-time input that i had laying
around (might be borrowed from somewhere):

import termios, TERMIOS, sys, os

def getkey():
        fd = sys.stdin.fileno()
        old = termios.tcgetattr(fd)
        new = termios.tcgetattr(fd)
        new[3] = new[3] & ~TERMIOS.ICANON & ~TERMIOS.ECHO
        new[6][TERMIOS.VMIN] = 1
        new[6][TERMIOS.VTIME] = 0
        termios.tcsetattr(fd, TERMIOS.TCSANOW, new)
        c = None
        try:
                c = os.read(fd, 1)
        finally:
                termios.tcsetattr(fd, TERMIOS.TCSAFLUSH, old)
        return c

if __name__ == '__main__':
        print 'type something'
        s = ''
        while 1:
                c = getkey()
                if c == '\n':
                        break
                print 'got', c
                s = s + c

        print s

however this doesn't do it in a non-blocking way. you should
probably check for data with the select module prior to the read()
call. set the timeout to 0 and it returns immediately iirc.
(it's also possible to set stdin to non-blocking mode, or use the
FIONREAD ioctl, or ...)

  -- erno


From michaelahoward at my-deja.com  Tue Jun 13 10:29:48 2000
From: michaelahoward at my-deja.com (Michael A. Howard)
Date: Tue, 13 Jun 2000 10:29:48 -0400
Subject: A simple question
Message-ID: <642A954DD517D411B20C00508BCF23B0012968DA@mail.sauder.com>

I am developing an application using Python/Tkinter that needs to look
similar to a VB application (a language I am far more familiar with) as
it will extend some functionality & we want the user interface to be as
close as possible to the existing system.

What I need to do is have a tabbed switchboard type menu screen with
buttons for each program that the user will interact with. This has
already been developed and went far faster than doing the same thing
with VB.
>From the switchboard the user will select the program they want to run
and it should open a new window for data entry/edit.

None of the code examples I look at show me how to do this.

Am I missing something really simple or is this task beyond the
capabilities of Tkinter or Python?



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



From edcjones at erols.com  Tue Jun 13 22:32:09 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Tue, 13 Jun 2000 22:32:09 -0400
Subject: Standalone http server, cgi software?
Message-ID: <3946EEA9.DFCBAB4D@erols.com>

I am a fairly experienced Python programmer but when it comes to
http servers, etc, I am a newbie. I would like to be able to fill
out an html form, press the submit button, and execute a Python
program with data from the form. My newbie questions are: What
URL should I use in the http form? Do I need to change any
preferences in Netscape or configuration files on my Linux
system? Are the Python httpserver and cgi modules the way to go?
Does anyone have a simple, working, piece of software that does
what I want that has detailed instructions? (I would use the
software as a model.)

The following keywords found several articles on Dejanews: local
standalong cgi server python forms. Unfortunately, no detailed
code was given.

Thanks,
  Ed Jones




From mhammond at skippinet.com.au  Tue Jun  6 22:01:07 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 07 Jun 2000 02:01:07 GMT
Subject: Perl and NT/2000 resources please
References: <zXf%4.522$t56.6648@news.uswest.net>
Message-ID: <D1i%4.7137$Hz.53100@news-server.bigpond.net.au>

In an unabashed plug for Andy and my book, we do cover this stuff in some
detail.  We cover services, processes, and user/share etc administration,
just to name a few topics.

http://www.ora.com/catalog/pythonwin32/


Mark.

"Eric Lengvenis" <elengvenis at uswest.net> wrote in message
news:zXf%4.522$t56.6648 at news.uswest.net...
> I currently use Perl to administer NT. I mostly do simple things like
change
> settings on every user on the network in one fell swoop. Or to allow
call
> center to unlock accounts without explicitly giving the Help Desk Tech
the
> rights to do this. Or to explicitly check for and stop processes before
a
> scheduled reboot of the NT servers (And you all know they need regular
> rebooting to stay healthy).
>





From kern at caltech.edu  Fri Jun  2 16:20:41 2000
From: kern at caltech.edu (Robert Kern)
Date: 2 Jun 2000 20:20:41 GMT
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com>
    <8h8jg3$d66$1@slb6.atl.mindspring.net>
Message-ID: <8h94up$euo@gap.cco.caltech.edu>

In article <8h8jg3$d66$1 at slb6.atl.mindspring.net>,
	aahz at netcom.com (Aahz Maruch) writes:
> In article <8h87u1$sga$1 at news2atm.raleigh.ibm.com>,
> Arint? <shouldbe at message.com> wrote:
>>
>>I am trying to write this escape seq to a device from python
>>"H\x1b\x23\x12art\x1b\x23\x01\n"
>>The \x1b\x23\x12 is supposed to be a command tell the device to do something
>>special with "art", but python is making it look like this \x1b\x23\x12a
>>"rt".  Am I missing something on how to work with escape sequences?  Is
>>there another way to write hex values in a string.  I rather us \x00 because
>>that is what are testers are familiar with.
> 
> I'd like to see some of the actual code.  Firing up the Python
> interpreter produces correct resulte for me.

It doesn't produce the desired result for me.

Python 1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian GNU/Linux)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> s="H\x1b\x23\x12art\x1b\x23\x01\n"
>>> s
'H\033#*rt\033#\001\012'
>>> print s
Hrt

>>> '\x12'
'\022'
>>> '\x12a'
'*'
>>> chr(0x12a % 256)
'*'

The 'a' disappears into the escape sequence, which is not the behaviour that
was desired.

>From http://www.python.org/doc/current/ref/strings.html :

"""\
In strict compatibility with Standard C, up to three octal digits are 
accepted, but an unlimited number of hex digits is taken to be part of the 
hex escape (and then the lower 8 bits of the resulting hex number are used 
in 8-bit implementations). 
"""

A possible solution is the following

>>> s = "H\x1b\x23\x12%s\x1b\x23\x01\n" % 'art'
>>> s
'H\033#\022art\033#\001\012'

It's kludgy (depending on the use; if 'art' isn't the only thing you can put
in that string, then it's actually more flexible this way), but it should
work.

> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)

-- 
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From mikael at isy.liu.se  Thu Jun  8 10:26:43 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 08 Jun 2000 16:26:43 +0200 (MET DST)
Subject: How to tell if I can open a file or not
In-Reply-To: <8ho4to0m8n@enews2.newsguy.com>
Message-ID: <XFMail.000608162643.mikael@isy.liu.se>

On 08-Jun-00 Skip Hollowell wrote:
 >  I have a indexer that I am writing that goest through a user-specified
 >  directory and all subdirectories and indexes all the .htm* file contained
 >  therein.  This works great until I get to a file that someone has chmod-ed
 >  so that I can't read it.  Then I bomb after the open().
 >  
 >  Is there a way I can check the attrbutes of the file before I open it? 
 >  This
 >  is easy enough in c or perl, but have yet to find an equivalent in Python.

Well, the easiest thing you can do is to wrap your offending code inside 
a try-clause, like so:

    try:
        offending_code
    except:
        deal_with_breakdown

If you really want to check the settings of the file, I guess you should 
read up on module os, and especially os.stat. Also, take a look at module
stat.

Good luck!

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    08-Jun-00
Time:    16:09:03

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


From jamarijr at my-deja.com  Fri Jun  2 10:47:24 2000
From: jamarijr at my-deja.com (Arinté)
Date: Fri, 02 Jun 2000 14:47:24 GMT
Subject: Escapse sequences???
Message-ID: <8h8hdn$nsq$1@nnrp1.deja.com>

I am trying to write this escape seq to a device from python
"H\x1b\x23\x12art\x1b\x23\x01\n"
The \x1b\x23\x12 is supposed to be a command to tell the device to do
something special with "art", but python is making it look like this
\x1b\x23\x12a "rt".  Am I missing something on how to work with escape
sequences?  Is there another way to write hex values in a string.  I
rather use \x00 because that is what are testers are familiar with.

I guess I need 8bit sized escapes and not 16bit.

Thanks


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


From vigenereNOviSPAM at leland.stanford.edu.invalid  Sun Jun  4 07:47:53 2000
From: vigenereNOviSPAM at leland.stanford.edu.invalid (cheesecake)
Date: Sun, 04 Jun 2000 04:47:53 -0700
Subject: file.seek misbehaving?
References: <203da3d8.c492c8fd@usw-ex0103-019.remarq.com> <Lyc_4.33464$WS3.397965@typhoon.we.rr.com>
Message-ID: <067d003a.25f504d0@usw-ex0104-028.remarq.com>

Thanks for the info. I hadn't thought about using the binary
mode. It's not a problem with eoln characters. When I said the
strings were identical, I meant including the eoln.

I can still use readline() for binary files, so for what I'm
doing, I won't need to change anything in my code except to open
in "rb" mode. I'll try this out tomorrow at work and post the
results.

TD

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



From wjdandreta at worldnet.att.net  Fri Jun  9 01:19:19 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 09 Jun 2000 05:19:19 GMT
Subject: Why is Python so slow?
Message-ID: <r7%%4.1754$iy.137191@bgtnsc06-news.ops.worldnet.att.net>

I am just learning Python and I wrote a little program to modify a text file
with 120,000 lines. It took 1hr and 30 mins. I wrote a C program that
created the text file from a .dbf and it only took 2 mins.

Both programs ran on the same machine (486-133mhz) under Concurrent DOS (a
multi-user mutitasking OS).

The Python version was 1.2.

Why does it take Python 50 times longer than C?

Bill Dandreta




From fiona at sitegnome.com  Sat Jun 10 02:41:54 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sat, 10 Jun 2000 16:41:54 +1000
Subject: Why is Python so slow?
References: <r7%%4.1754$iy.137191@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <3941E332.C94B5DB8@sitegnome.com>

William Dandreta wrote:
> 
> I am just learning Python and I wrote a little program to modify a text file
> with 120,000 lines. It took 1hr and 30 mins. I wrote a C program that
> created the text file from a .dbf and it only took 2 mins.
> 
> Both programs ran on the same machine (486-133mhz) under Concurrent DOS (a
> multi-user mutitasking OS).
> 
> The Python version was 1.2.

I'm thinking this is a typo... in an earlier mail William was having
trouble compiling Python 1.5.2 ... so, I don't think the version for the
test was the problem.....

-F.


From zessin at my-deja.com  Wed Jun 28 11:53:24 2000
From: zessin at my-deja.com (Uwe Zessin)
Date: Wed, 28 Jun 2000 15:53:24 GMT
Subject: float literal too large
References: <ulsntz2kra.fsf@runner.princeton.idaccr.org> <8jat5b$2t4@stochastic.eng.umd.edu>
Message-ID: <8jd71a$f49$1@nnrp1.deja.com>

In article <8jat5b$2t4 at stochastic.eng.umd.edu>,
  moron at Glue.umd.edu (Warren B. Focke) wrote:
>
[...]
> Python floats are stored in platform doubles.  On IEEE-754 compliant
> hardware (pretty much anything other than a Cray (or VAX, but I don't
> think they make those anymore (unless alphas have a VAX math mode)))

Correct, VAX hardware does not support IEEE floating point formats.
The last VAX chips have been produced a long time ago and Digital/
Compaq have lived from a big warehouse. Last time I checked there
is only a very limited supply left and orderability ends soon. Note
that it's format is older than IEEE. (So much for the 'proprietary'
argument ;-)
Operating system support for VAX (OpenVMS) is still available, and
active. New things, of course, go into Alpha - if possible they appear
on VAX, too.

Yes, Alpha chips do support (some) VAX floating point formats in
addition to 2 (if I recall correctly) IEEE fp formats. The port of
Python for OpenVMS, however, uses the VAX formats for now.
I have tried to compile in IEEE mode and it seemed to work, but I
have not done any more tests.

[ remaining explanations cut ]

--
Uwe Zessin


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


From arnold at dstc.monash.edu.au  Tue Jun 20 12:31:13 2000
From: arnold at dstc.monash.edu.au (David Arnold)
Date: Wed, 21 Jun 2000 02:31:13 +1000
Subject: Why whitespace denotation of blocks doesn't work. 
In-Reply-To: Your message of "Tue, 20 Jun 2000 08:33:21 MST."
             <ted.drain-2006000833210001@mjordan.jpl.nasa.gov> 
Message-ID: <200006201631.CAA29255@xevious.dstc.monash.edu.au>

-->"Ted" == Ted Drain <ted.drain at jpl.nasa.gov> writes:

  Ted> I'll write some code, and as I start debugging/prototyping/etc
  Ted> etc, I find that I have to enclose some existing code in a
  Ted> control block (if,etc).  With C/C++, I just put the code in
  Ted> brace, tell xemacs to re-indent the block, and everything's
  Ted> good.  With Python, I have to hand edit everything because the
  Ted> editor can't figure out how to do the indentation after I've
  Ted> inserted the control block.

if you're using xemacs, i presume you're using python-mode.  in
python-mode, C-c > and C-c < indent and dedent the current region.



d



From punck at PenguinMints.cx  Wed Jun 28 07:08:36 2000
From: punck at PenguinMints.cx (Chris Armstrong)
Date: Wed, 28 Jun 2000 06:08:36 -0500
Subject: Requiring a certain version of interpreter?
References: <Lqj65.4362$h8.209230@news-west.usenetserver.com>
Message-ID: <s2k65.1080$MZ3.2911@newsfeed.slurp.net>

In article <Lqj65.4362$h8.209230 at news-west.usenetserver.com>, Matthew
Cline <matt at nightrealms.com> wrote:
> Is there any builtin feature, or common module, that handles requiring
> at least a certain version of the Python interpreter to be used for a
> script?  Perl has 'require' for this purpose.
> 
> Thanks in advance.

>>> import sys 
>>> print sys.version
1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian GNU/Linux)]

There may a way to just get the version number only, but if nothing else this 
will work.
-- 
There is a 90% chance that this message was written when the author's been
up longer than he should have. Please disregard any senseless drivel.
--
Chris Armstrong http://penguinmints.cx



From aahz at netcom.com  Wed Jun 14 11:47:04 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 15:47:04 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com>
Message-ID: <8i89do$gn7$1@slb7.atl.mindspring.net>

In article <3947A59F.5016D9F0 at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>Aahz Maruch wrote:
>>
>> Like Grant, I've been using Unix about fifteen years, and I find Perl
>> bizarre and difficult.
>
>after reading several responses like this, i wonder if perl is
>particularly irritating to old timers. twenty years ago, this kind of
>obscure but powerful notation (as embodies in awk and such), was really
>cool. but after a while the appeal kind of wears thin, particularly
>when you find something that gives you the same power but without the
>obscurity.

I think that might be on the right track, but I think it'd be more
accurate to say that many of us used those tools because they were the
best available then; had something like Python been around, we would
have used that instead.

OTOH, I do still delight in writing the occasional really grotty regex
in vi....
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From hamish_lawson at yahoo.co.uk  Wed Jun 21 18:18:22 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Wed, 21 Jun 2000 15:18:22 -0700
Subject: Any Python project similar to Jserv and Cocoon?
References: <6D8A17398E28D3119F860090274DD7DBF13C74@pces.cadlab.it>
Message-ID: <12e3a089.c1a07541@usw-ex0105-040.remarq.com>

Alessandro Bottoni wrote:

> I wonder if anybody is thinking to use Python for building a
> set of XML server tools, more or less equivalent to Jserv and
> Cocoon, for the Apache web server

Zope (Z Object Publishing Environment) is a component-based web
application server and content-management system written in
Python. Though Zope is not an Apache technology, it can be
integrated with Apache and other web servers. Zope addresses
many of the same issues as Java servlets and EJBs, and uses a
template system similar to Java Server Pages. Recently there has
been work on publishing XML documents through Zope; see
http://www.zope.org/Products/XMLDocument.

Hamish Lawson


Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From ullrich at math.okstate.edu  Sat Jun 10 11:03:40 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 10 Jun 2000 15:03:40 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no> <8hrhqa$afm$1@news1.xs4all.nl> <G2g05.688$cQ.9254@news-server.bigpond.net.au>
Message-ID: <01bfd2ee$caf27540$29cf8ad1@daves-dell>


Mark Hammond <mhammond at skippinet.com.au> wrote in article
<G2g05.688$cQ.9254 at news-server.bigpond.net.au>...
> "Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
> news:8hrhqa$afm$1 at news1.xs4all.nl...
> 
[...] 
> > The lookups aren't that great - they only find what you've already used
> > in that file. I'd like loopup to look in every file that's included for
> > completion. But I'm not going to complain - it's good for my memory not
> > to have it ;-).
> 
> This is amazingly hard to do correctly - and the hard bits are not
related
> to Pythonwin!
> 
[...]
> 
> So if anyone can come up with generic code that can do this, all current
> and future IDEs for Python could benefit.

        My conjecture is it's literally/provably impossible to do it
perfectly.
Not that I'm going to try to prove that, and not to say that an imperfect
version could not still be very useful.

> Mark.
> 
> 
> 


From mwh21 at cam.ac.uk  Sun Jun  4 12:29:27 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 04 Jun 2000 17:29:27 +0100
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <m3itvpxvdk.fsf@atrus.jesus.cam.ac.uk>

"Bablos" <angband at blueyonder.co.uk> writes:

> Hi There...
> 
> Now that I've got my FTP client doing some basic stuff, I now have a
> progress report which looks something like the following;
> 
> uploaded X bytes
> uploaded Y bytes
> uploaded Z bytes
> 
> Is there a way that I can get Python to print this all on the same
> line, so that it looks as if the line is being updated constantly, as
> opposed to having a ridiculously long list of 'uploaded...' lines?

Maybe 

print "uploaded %s bytes"%(bytecount,),

?  If you wat to replace the line you just wrote with a new line,
you're into escape code territory.  This sort of works on an xterm:

>>> for i in range(10):
...  print "uploaded %s bytes\033[A"%i 
...  time.sleep(1)
... 

but I think you probably want to look at curses to do this sensibly.

Cheers,
M.

-- 
  [Perl] combines all the  worst aspects of C and Lisp:  a billion
  different sublanguages in one monolithic executable. It combines 
  the power of C with the readability of PostScript.
                                                     -- Jamie Zawinski


From phil at geog.ubc.ca  Thu Jun 15 17:09:10 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 15 Jun 2000 14:09:10 -0700
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu> <150620000713119061%pecora@anvil.nrl.navy.mil> <etdu2evf3dv.fsf@w20-575-109.mit.edu>
Message-ID: <m31z1y1wjt.fsf@brant.geog.ubc.ca>

Alex <cut_me_out at hotmail.com> writes:

> Well, it wasn't a good idea anyway, since the information it would have
> gotten is already in the doc's.  But I was thinking writing a script
> that would write a C program that would create the python objects needed
> by each API function, and record how their refcounts change after they
> are passed to those functions.
> 
> Alex.

A similar idea which helped me was to insert print statements in the
CXX functions that call XINCREF (5 functions), INCREF (1 function),
XDECREF (1 function) and DECREF (1 function) and watch the output
for sample extension modules.  It did manage to clear up some of the
C-language cases I wasn't sure about.

Phil



From aaron.ginn at motorola.com  Mon Jun 26 14:10:37 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 26 Jun 2000 11:10:37 -0700
Subject: Tkinter: problems running as a script.
Message-ID: <sng0q05n4y.fsf@motorola.com>

I'm unable to run a simple "Hello, world" python script, but I
can run the script if I paste the lines one at a time into the
python interpreter.  Here's the script called Tkinter.py:

#! /home/ginn/python_tk/bin/python
from Tkinter import *

root = Tk()

w = Label(root, text="Hello, world!")
w.pack()

root.mainloop()

When I run the script, I get the following output:

coronado <213> ~/bin/Tkinter.py
Traceback (innermost last):
  File "/home/ginn/bin/Tkinter.py", line 2, in ?
    from Tkinter import *
  File "/home/ginn/bin/Tkinter.py", line 4, in ?
    root = Tk()
NameError: Tk

However, when I run the script from the interpreter, it works.  Here's
the output from that:

coronado <216> /home/ginn/python_tk/bin/python
Python 1.5.2 (#14, Jun 26 2000, 10:29:29)  [GCC 2.95 19990728 (release)] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from Tkinter import *
>>> root = Tk()
>>> w = Label(root, text="Hello, world!")
>>> w.pack()
>>> root.mainloop()

The widget pops up just as expected.

What am I doing wrong here?

Thanks,
Aaron Ginn


From olczyk at interaccess.com  Sat Jun 17 19:36:51 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 17 Jun 2000 23:36:51 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com>
Message-ID: <39520785.690843718@nntp.interaccess.com>

On Sat, 17 Jun 2000 08:12:03 -0500, "Bob Hays, Computer Geek"
<rhays at interaccess.com> wrote:

>On Sat, 17 Jun 2000 07:49:43 GMT, olczyk at interaccess.com (Thaddeus L.
>Olczyk) wrote:
>
>The following are just my opinions to take or leave (probably the
>latter) as you wish.
>
>>1) The first is rather stylistic and may not appeal to most, but I
>>have often found it useful. When debugging code I often put in
>>code to help me figure out whats happening. I also avoid indenting
>>such code so that it easy to remember where it is later when I want to
>>take it out. An example ( assuming I use begin and end to denote
>>blocks ):
>
>I've gotten in the habit of putting the comment DEBUG on such lines in
>all programming languages; depending upon indention is very dangerous
>for debug lines since someone else may reindent the line properly and
>then it gets left in (since they didn't know it was a debug line).
>
I never check in code with debug lines in it, so I never have to worry

about others changing it. That's the whole point, it makes it easier
to remove.

>>2) You can easily insert a tab introducing a bug.  Consider the
>>following code ( similar to a bug I recently fixed ):
>{Example skipped}
>>Initially the line in error was indented correctly, but someplace
>>along the line the extra tab was introduced. A lot of effort then had
>>to go in to finding this bug. With begin...end or braces this is less
>>likely to happen.
>
>Yep, this happens.  Its a tradeoff - speed vs. occasional extra work.
What increases in speed do you get by having to type braces?
>
>In addition, I had an example, in C++, where someone had commented out
>a block of code, but then had an embedded inline comment inside that
>block, and, to top it all off, the less-than-desired commented out
>block would still compile - the guy worked on that bug for a day
>before coming to me.  I loaded it into emacs and the color coding told
>him immediately what had happened; he was very ashamed and angry, but
>I told him sh*t happens:-).
>
But that's the point. His code was illegal, it wouldn't compile. Easy
to catch with emacs, and the compiler will catch it too. The little
typo I show is easy to make and perfectly legal code. Its hard to
accidentally bump against a keyboard and make a similar nasty typo.
EVen something that looks like it might work like }{ would have to be
put in just the right spot.

>>3) Finally. Python is a language where it is almost easy to write
>>extensible programs. To illistrate what I mean consider this example
>>of a python program I recently wrote:
>
>Here's were we will agree to disagree, I think.  I think its way
>faster and easier to extend Python stuff than C++ (and Java) due to
>the lack of compile/link cycle and due to inheritance inside of
>Python.  Yep, you have to be careful how you indent, but if you have a
>good editor it will cut/copy/paste as-is without reindenting.  I see
>more problems due to bad cut/copy/paste reuse in all programming
>languages than I do from mis-indenting so far.
>
I think you misunderstood what I said. I think that Python is not now
easily extensible do to the  problems with whitespaces. If that were
fixed then Python would be easily extensible. 

The point is not that I can copy and paste with out reindenting.
The point is that sometimes I have to. let me show a generic example

if a:
   for b in B:
      if c(b):
          block statement.

Assume that you are editing another part and you realise that you
are about to essentially retype the block statement. So you do

def fun():
  block statement

if a:
   for b in B:
      if c(b):
           fun()
you have to reindent the block statement. This is not copy and paste
reuse. It is cut and paste refactoring, and you have to reindent.

>>Anyway I hope this causes TPTB to think about how blocks should be
>>handled in Pyhton 3000.
>
>My personal moral - no programming language will save you from
>yourself - use good tools and do your best.
>
But there are some features in a language that make errors more
likely. Bocking by indentation is one of them. I shouldn't have
to be saved from this error in the first place.


From aiglauer at iname.com  Mon Jun  5 11:07:33 2000
From: aiglauer at iname.com (aiglauer)
Date: Mon, 05 Jun 2000 15:07:33 GMT
Subject: How to prevent a web client from timing out
Message-ID: <8hgfn8$3uq$1@nnrp1.deja.com>

Not a purely python question, but.....

I have written a number-crunching app in python. which is being
accessed on our intranet.  All pages generated on-the-fly in python.
Some of the calculations can take a while though (calc time anything
from 5secs to 30 minutes) and I get the client timing out and the
server will then kill the calculation process.

Can I rig it so that the calculation process will still complete and
write its output to a file so that I can use it later?  (Writing to a
file is not a prob - preventing the thing getting slapped is).

Thanks
Alexis


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


From sblakey at freei.net  Tue Jun  6 15:13:05 2000
From: sblakey at freei.net (Sean Blakey)
Date: Tue, 6 Jun 2000 12:13:05 -0700
Subject: strange list behavior
In-Reply-To: <393D4811.9989DE96@magellan.umontreal.ca>; from seefelds@magellan.umontreal.ca on Tue, Jun 06, 2000 at 06:50:16PM +0000
References: <393D4811.9989DE96@magellan.umontreal.ca>
Message-ID: <20000606121305.E19776@freei.com>

On Tue, Jun 06, 2000 at 06:50:16PM +0000, Stefan Seefeld wrote:
> I'v run into a strange problem with lists. I'm not sure
> at all where the problem is so I'll provide some context:
> 
[snip]
> class Visitor:
>     def visitDeclarator(self, node):
>         name = self.scope()
>         name.append(node.identifier())
>         self.__result_declarator = AST.Declarator(name)
>     def visitModule(self, node):
>         name = self.scope()
>         name.append(node.identifier())
>         module = AST.Module(name)
> #...
> 
> what I *though* was happening is that I created temporary variables 'name'
> in the respective scopes of all the visitSomething methods and then hand over
> the objects to the newly created Node types (Module, Declarator, etc.)
> However, what I *observe* is that even though a Module is initialized with
> name : ['foo', 'bar']
> it may end up being
> name : ['foo', 'baz']
> i.e. it seems the list is not properly released but instead modified from the outside. 
> printing out 'id(name)' reveals indeed that the various temporary 'name' objects 
> in the Visitor class have all the same id, i.e. they are the same object. Given
> that the AST nodes are created with the list copied by reference, not by value,
> I should not wonder that the value changes. Is this a fault of GC ?
> 
> Could anybody explain what's going on here and how I get the behavior I want ?
> Thanks a lot !
> 
> Stefan
> 
> _______________________________________________________              
>               
> Stefan Seefeld
> Departement de Physique
> Universite de Montreal
> email: seefelds at magellan.umontreal.ca
> 
> _______________________________________________________
> 
>       ...ich hab' noch einen Koffer in Berlin...
> -- 
> http://www.python.org/mailman/listinfo/python-list

It looks like what is happening is that your self.scope() method is
returning a reference to an existing list instead of a new list.  You do
not provide any code for your scope() method so I cannot verify this,
but I would guess that you are doing something like this:
    def scope(self, default_scope=[]):
	# manipulate default_scope here
	return default_scope

When you use a mutable object as the default argument of a function or
method in python, that object is only initialized once.  Each call to
self.scope() will return a reference to the same list.
    -Sean
-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
It is easier to change the specification to fit the program than vice versa.



From thomas at xs4all.net  Sun Jun 11 08:41:45 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 11 Jun 2000 14:41:45 +0200
Subject: time.clock() always returns 0?
In-Reply-To: <GEL05.4911$HD6.127120@iad-read.news.verio.net>; from dougfort@downright.com on Sun, Jun 11, 2000 at 08:31:34AM -0400
References: <GEL05.4911$HD6.127120@iad-read.news.verio.net>
Message-ID: <20000611144145.M20700@xs4all.nl>

On Sun, Jun 11, 2000 at 08:31:34AM -0400, Doug Fort wrote:

> On my system (RedHat Linux 6.2 Python 1.5.2 as shipped with RedHat)
> time.clock() always returns 0.  time.time() returns a good value.  Is this
> known behavior?  Am I doing something wrong?

What do you think time.clock() does ? The docstring says this:

clock() -> floating point number

Return the CPU time or real time since the start of the process or since the
first call to clock().  This has as much precision as the system records.

Notice how it says 'or since the first call to clock()'. Notice also how it
says it has as much precision as the system records -- if your Linux system
for some reason isn't keeping proper records, you wont get proper results.
If you need better results, os.times() may be what you need:

times() -> (utime, stime, cutime, cstime, elapsed_time)
Return a tuple of floating point numbers indicating process times.

Then again, if you need what time.time() returns, you do not need
time.clock() :-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas at cintra.no  Tue Jun 20 10:03:11 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 20 Jun 2000 14:03:11 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <8inq3u$p0i$1@newshost.accu.uu.nl> <39586f3d.366407626@news.online.no>
Message-ID: <395978c6.368848746@news.online.no>

On Tue, 20 Jun 2000 13:34:02 GMT, thomas at cintra.no (Thomas Weholt)
wrote:

>>> Linux with Windows  2000/ NT 4.0.
>>
>>Why switch from Linux to Windows NT/2000, though? ;)
>
>Well, first off it`s my OS of choice, but mostly it *was* because it
>costs money. I`ve realized later that the cost of an os is scraps in a
>huge project so that`s not a very important issue anymore. But I would
>still go for Linux. It has remote-administration out-of-the-box with
>telnet/ssh, ready-and-running web-server ( Apache ) , dns, ftp - well,
>do I need to go on. AND PYTHON by default.
>
>Ok, a little off topic, but Linux is probably THE development platform
>at the moment, and have been for some time. If you look at
>distributions like Mandrake Linux 7.1 you`ll see that developers get
>alot of free and great stuff. When I had installed for about a week
>ago I had MySQL, PostgreSQL, PHP, Apache, Python, Perl, C/C++ and so
>on up and running in less than 2 hours. It at the cost of a blank cd.
>And that`s including documentation. 
>
>Oh, the big-ol` Raving-Linux-Fanatic took over, sorry.
>
>Thanks for your input anyway.
>
>Thomas

Eh ... uhm ... what did I just say ?? Let me clarify; I`m not a
schizo. I meant use Linux instead of Windows. Ok?

I better get some sleep tonight. Insomnia is eating at my brain.

Thomas


From aahz at netcom.com  Tue Jun  6 11:04:14 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Jun 2000 15:04:14 GMT
Subject: author of htmllib/formatter?
References: <m2og5ec2q9.fsf@netmeg.net> <Pine.GSO.4.10.10006061736170.26389-100000@sundial> <14653.3965.113400.816770@gargle.gargle.HOWL>
Message-ID: <8hj3te$gb9$1@slb7.atl.mindspring.net>

In article <14653.3965.113400.816770 at gargle.gargle.HOWL>,
Les Schaffer  <godzilla at netmeg.net> wrote:
>
>i did already, no one answered....  ;-)

I hope I'll have some time to look at it, but I'm real busy right now.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From clive at pisquaredoversix.force9.co.uk  Sat Jun 17 04:08:31 2000
From: clive at pisquaredoversix.force9.co.uk (Clive Tooth)
Date: Sat, 17 Jun 2000 09:08:31 +0100
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <8idie3$l4i$1@mail.pl.unisys.com> <394AC273.24A9@esatclear.ie>
Message-ID: <394B31FF.5D56487D@pisquaredoversix.force9.co.uk>

Russell Wallace wrote:

> Clive Tooth wrote:
> >
> > Russell Wallace wrote in message <394889C3.1ED5 at esatclear.ie>...
> > >Correct me if I'm wrong, but I thought memory was the limiting factor?
> > >i.e. that a good algorithm on typical hardware can fill all the
> > >available memory with digits of pi in a few days or so?
> >
> > You are wrong. The discussion was about methods of calculating digit N of pi
> > (in some base) without computing digits 1 thru N-1 . This can be done in
> > about O(log(N)) bits of memory.
> 
> Yes, that's what I mean.  The O(log N) memory consumption would, it
> seems to me, make it possible to reach far higher values of N than if
> you had to calculate digits 1 thru N-1, even if the amount of
> computation required isn't much different.

A calculation of the Nth digit of pi, where the limit was the amount of
_memory_ available, say 100MB on some typical PC, would mean an
extremely large N and a truly stupendous amount of time. Greater than
the age of the universe, etc.

-- 
 
Clive Tooth
http://www.pisquaredoversix.force9.co.uk/
End of document


From michael.drumheller at boeing.com  Wed Jun 28 00:45:53 2000
From: michael.drumheller at boeing.com (Michael Drumheller)
Date: Wed, 28 Jun 2000 04:45:53 GMT
Subject: Python with braces.
References: <395683ae.299281703@nntp.interaccess.com> <etdsnu1byth.fsf@w20-575-101.mit.edu>
Message-ID: <39598301.8BBC4621@boeing.com>

I'd agree, it seems like a terrible way to write code, at least for
humans. But maybe not for machines.  If you're doing a lot of automatic
code generation, as in some genetic algorithms I've seen that generate
code to solve some problem or another, then I can see how it might be
more natural to write a prorgam that generates properly brace'd code
than properly whitespace'd code.  (The example I'm thinking of was in
Lisp, incidentally.)
MD

Alex wrote:

> like a terrible way to code, for me, even though it's a cool idea.  But
> you can find it here:
> http://www.python.org/pipermail/python-list/1999-June/066547.html

-- 
Michael Drumheller--Boeing Mathematics & Computing Technology
PO Box 3707 M/S 7L-40 Seattle 98124-2207 (425)865-3520 Fax:2965
michael.drumheller at boeing.com


From brunns at beer.com  Tue Jun 20 07:24:53 2000
From: brunns at beer.com (Simon B.)
Date: Tue, 20 Jun 2000 11:24:53 GMT
Subject: request: web site copy utility
References: <grg3jsg899nhuhl5sua80ookjop9cuk8j0@4ax.com> <8gsr3m$ufr$1@nntp9.atl.mindspring.net> <393207A4.C04E6BDB@earthlink.net>
Message-ID: <8inka0$20i$1@nnrp1.deja.com>

In article <393207A4.C04E6BDB at earthlink.net>,
  Simon <tomega at earthlink.net> wrote:
> websucker.py, while an awesome utility, doesn't actually suck the
entire
> site.  For instance, it will pull the image files on a page if those
> images are statically SRC-ed.  If there is an onmouseover directive,
for
> instance, which changes the image source to another file, it will not
> grab the other file, even if it lies in the same directory as the
first
> image.

I have problems with some sites (see <http://publib.boulder.ibm.com/cgi-
bin/bookmgr/BOOKS/QBKAQV00/CCONTENTS> for example) where the URLs are
fully qualified, rather than relative. At least I *think* that that's
what the problem is. Oh, and java applets aren't sucked, either.

I've had a quick look into the websucker source, but I haven't spotted
anything yet. I'll keep looking, but can anyone help me out here?

--
Simon B


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


From chris.strombergerNOchSPAM at wnco.com.invalid  Mon Jun 19 15:38:29 2000
From: chris.strombergerNOchSPAM at wnco.com.invalid (cstromberger)
Date: Mon, 19 Jun 2000 12:38:29 -0700
Subject: Python vs Perl
References: <8ikai5$v3d$1@nnrp1.deja.com> <394ebe95.255648282@news.online.no> <slrn8krpbf.gc.andyl@azaal.dircon.co.uk>
Message-ID: <058a78be.c653bdf6@usw-ex0104-026.remarq.com>

andyl at azaal.dircon.co.uk (Andy Leighton) wrote:
>On Mon, 19 Jun 2000 06:41:39 GMT, Thomas Weholt
<thomas at cintra.no> wrote:
>>This question has been asked several times before so you might
find a
>>better answer than mine searchin deja.com, but :
>>
>>Perl vs. Python :
>>
>>* Perl is harder to read than Python. This makes debugging and
re-use
>>of other peoples code harder.
>
>
>Of course if you only know Perl, I would think Perl would be
>the easiest to read.


Dubious.

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From db3l at fitlinxx.com  Fri Jun  9 01:34:30 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 09 Jun 2000 01:34:30 -0400
Subject: Newbie: Simple question about string slices
References: <8hpsl6$mve@newshost.fujitsu.com.au>
Message-ID: <ud7lrquxl.fsf@ctwd0143.fitlinxx.com>

"Ray Smith" <Ray.Smith at fujitsu.com.au> writes:

> Hi there,
> 
> I'm completely new to Python and was reading the standard tutorial for
> Python and came across string slices., The tutorial says:

(...)

> "Here's a useful invariant of slice operations: s[:i] + s[i:]equals s."

(...)

> My question is:
> If given a slice s[a:b]
> can I assume index "b" means upto but not including the index "b"?

Precisely - that's required or else the invariant wouldn't be true.
It's also generally a very useful condition since it fits the base 0
addressing nicely, and helps simplify various computations of pieces
of strings and lists that otherwise would be adjusting computed
offsets by 1 a lot.

For one example, if a slice was inclusive then s[0:n] (or s[:n]) where
n was len(s) would have an upper bound one too large.

> I guess another way to ask my question is:
> Why is s[a:a] always '' (empty) ?

Because that particular slice doesn't cover a full element.  It's sort
of like a piece of the string just before the 'a'th element.  For
mutable objects (like lists) you can actually assign to a slice like
this to insert elements into the list at the point just before element
'a'.

> Hope someone out there in Python land can make sense of my question!

The tutorial you're reading does a reasonably good job of explaining
slices and providing examples, I think.  You might also peek at the
page just after the statement about the invariant that you mention
above also has a small ASCII diagram that provides one sample manner
of a way to think of slices (pointing between elements) that may also
help.

And as you've done, experimenting in the interactive interpreter is
one of the easiest ways to test out some of the language features.

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


From hinsen at cnrs-orleans.fr  Wed Jun  7 11:18:53 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 17:18:53 +0200
Subject: OT: nederlands (Re: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <PCp%4.109780$Qp2.291904@afrodite.telenet-ops.be>
Message-ID: <m3pupth63m.fsf@chinon.cnrs-orleans.fr>

ben at co.and.co writes:

> Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
>  
> > Er zijn dus merdere rationale verklaringen, dus geen behoefte om
>               ^^^^^^^
>   Toch nog een beetje een franstalige tongval :-).

Merde!  ;-)
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From shouldbe at message.com  Thu Jun 29 11:54:56 2000
From: shouldbe at message.com (Arinté)
Date: Thu, 29 Jun 2000 10:54:56 -0500
Subject: traceback string
Message-ID: <8jfnn5$1fpo$1@news2atm.raleigh.ibm.com>

How do you print the traceback string from an exception
try:
   whatever()
except:
   print sys.exc_traceback

gives me <traceback object at 351900>
thanx





From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Thu Jun 29 04:00:46 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Thu, 29 Jun 2000 01:00:46 -0700
Subject: [Tkinter] Centering a window in the screen ?
References: <962215194.1421627556@news.libertysurf.fr>
Message-ID: <00dbb77f.cec66729@usw-ex0104-032.remarq.com>

Hi Jerome,

Your method works fine if you root.update() before entering the
center method. This method forces Tk to complete the geometry
calculations. It also processes all pending events on the event
queue so use it sparingly to avoid hogging system resources.

The behaviour your getting with winfo_width returning 1 is
simply that the negotiations between the window manager and
widgets hasn't completed so window size is actually still 1x1.

>> Is there a "center-window" in the Tkinter module?

Not that I know of.

>> Which functions am I to call to get the size of a window / a
widget?
winfo_width() and winfo_height are fine. There is also a
winfo_reqwidth() and winfo_reqheight(). These both return the
dimensions the widget requested - which of course may not be
exactly the same as what it finished with.

Some widgets, Photoclass and Bitmap spring to mind, have .width
() .height() functions which you can call directly.

>> What are all those winfo_* functions? Are we supposed to use
them or should we use some more user-oriented functions (such as
geometry(), configure(), etc.)?

Nobody has ever said not to use them ;-)


- By the way, what are those tuples returned by widget.configure
()?

configure() returns a dictionary of options for that widget.

>> Also, is there a Tkinter FAQ? I went on www.python.org, but
there
isn't much information on Tkinter... Where can I find some (and
if
possible in a downloadable form)?

http://www.python.org/topics/tkinter/doc.html is about as good
as it gets online.

You can download the pythonware docs by using websucker.py in
your tools directory where python is installed.

I'm considering starting a Tkinter web site (including a more
comprehensive FAQ) if anyone shows any interest.

Richard


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

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From MarkH at ActiveState.com  Wed Jun 21 11:00:26 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 21 Jun 2000 15:00:26 GMT
Subject: win32all-132 / Word Automation
References: <394e2b8d.515788975@news.mch.sni.de> <lAy35.3131$qz4.24329@news-server.bigpond.net.au> <394f269c.580059221@news.mch.sni.de> <ZVI35.3597$qz4.27900@news-server.bigpond.net.au> <395068d0.58350693@news.mch.sni.de>
Message-ID: <eM445.4664$qz4.34929@news-server.bigpond.net.au>

This turns out to be a bug in Pythoncom.dll.  A fix has been checked into
CVS, and will be in the next build.  It shows up whenever there is a
"byref" param that is not also marked as "out" - the C++ and the Python
code didnt agree on the size of the result.

A workaround is to clone the method that fails from the makepy generated
file (or simply modify that file in place), and add "[0]" to the end of
the InvokeTypes() line - ie, it is returning a tuple when it shouldnt!!

Ive mailed the specific work-around for this problem to Nikolai.

Mark.




From Marc.Poinot at onera.fr  Thu Jun 22 03:20:28 2000
From: Marc.Poinot at onera.fr (Marc Poinot)
Date: Thu, 22 Jun 2000 09:20:28 +0200
Subject: French Python documentation
References: <oqwvjj58f7.fsf@titan.progiciels-bpi.ca>
Message-ID: <3951BE3C.D1401773@onera.fr>

Fran?ois Pinard wrote:
> 
> Hello, my friends.  Robert, Cc:ed, is asking me what exists, in French,
> to document Python.  He would like to get as much as possible for the
> unilingual members of his team.  Please reply to the list, or to me at
> least, so I can save the collective knowledge and wisdom on this point! :-)
> 
> --
> Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard
> 
> --
> http://www.python.org/mailman/listinfo/python-list

Il y en a. Commence par la...

http://perso.club-internet.fr/olberger/python

Marcvs [alias La traduction joual reste a faire ;) ]



From scott at lund.com  Thu Jun 29 22:12:53 2000
From: scott at lund.com (Scott Chapman)
Date: Thu, 29 Jun 2000 19:12:53 -0700
Subject: Book for sale - Programming Python w/CD $30.00 + Shipping
Message-ID: <cf0olsg5im5akgaqjt237pjkg22m581dus@4ax.com>

Email scott at lund.com if interested. Payment via PayPal preferred.



From thiele at muc.das-werk.de  Wed Jun 28 15:52:26 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 28 Jun 2000 21:52:26 +0200
Subject: "+" does not match?
Message-ID: <395A577A.C83AE15D@muc.das-werk.de>

>>> import regex 

>>> s1 = "c++test/"
>>> s2 = "c++test/Derivetest.c++"
>>> regex.match(s1,s2)
-1

but if I substitude the "+" I'll get the correct behaviour.

>>> s1 = "c11test/"
>>> s2 = "c11test/Derivetest.c++"
>>> regex.match(s1,s2)
8

Why? Can anybody explain me that?


From Bgvinyard at aol.com  Mon Jun 19 12:56:55 2000
From: Bgvinyard at aol.com (Bgvinyard at aol.com)
Date: Mon, 19 Jun 2000 12:56:55 EDT
Subject: [Litestep] <.OT> New filemanager needs people with files
Message-ID: <61.4a531bd.267faad7@aol.com>

In a message dated 6/19/2000 2:41:54 AM Eastern Daylight Time, 
etoffi at softhome.net writes:

> Tell you the
>  truth I'm not good at naming things. (If I got a dog, I would name it 
"dog". 
> 
I named my dog "cat" heh

Message



From mwh21 at cam.ac.uk  Fri Jun  9 08:37:35 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 13:37:35 +0100
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <m3snunuj1s.fsf@atrus.jesus.cam.ac.uk>

stevemul at ozemail.com.au (Steve Mullarkey) writes:

> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated. I am genuinely interested
> to ask "C++" programmers :-
> 
> 1. What productivity increase do you achieve ?
> 
> 2. How long did you use Python before you achieved increased
> productivity ?

One of my habits is answering questions on comp.lang.python with
quotes from the Python Quotation file at

http://starship.python.net/crew/amk/quotations/python-quotes.html

The one for this occasion is:

  I never got beyond starting the data-structures in C++, I never got
  beyond seeing how it would work in Scheme. I finished it in one
  Python-filled afternoon, and discovered the idea sucked big time. I
  was glad I did it in Python, because it only cost me one afternoon
  to discover the idea sucks.             --  Moshe Zadka, 13 May 2000

Cheers,
M.

-- 
3. Syntactic sugar causes cancer of the semicolon.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From swdevpatrick at my-deja.com  Mon Jun 26 23:32:44 2000
From: swdevpatrick at my-deja.com (swdev)
Date: Tue, 27 Jun 2000 03:32:44 GMT
Subject: PicklingError on DateTime objects
Message-ID: <8j978s$flb$1@nnrp1.deja.com>

Hey All,

just taking a wild shot here, but... someone may have seen this.

I'm trying to write a pickle containing an object instance.  Within this
object instance is an attr containing a DateTime object instance.

When I call pickle.dump(obj, somefile), i get different outcomes
depending on platform; on Linux it works famously, no pause, just works.
On Solaris, it throws the following exception:

>>> pickle.dump(u, of)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/pickle.py", line 818, in dump
    Pickler(file, bin).dump(object)
  File "/usr/local/lib/python1.5/pickle.py", line 97, in dump
    self.save(object)
  File "/usr/local/lib/python1.5/pickle.py", line 198, in save
    f(self, object)
  File "/usr/local/lib/python1.5/pickle.py", line 412, in save_inst
    save(stuff)
  File "/usr/local/lib/python1.5/pickle.py", line 198, in save
    f(self, object)
  File "/usr/local/lib/python1.5/pickle.py", line 361, in save_dict
    save(value)
  File "/usr/local/lib/python1.5/pickle.py", line 159, in save
    raise PicklingError, \
pickle.PicklingError: can't pickle 'DateTime' objects

Any ideas?  Thanks for any help at all!
--p


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


From davidw at linuxcare.com  Wed Jun 28 05:43:17 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 28 Jun 2000 11:43:17 +0200
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com> <39593888.E7091570@estinc.com>
Message-ID: <877lbaw37u.fsf@eugene.prosa.it>

Eric Lee Green <eric at estinc.com> writes:

> Greg Gallagher wrote:

> > I'm a little disheartened here... I know how to do this in Perl,
> > but I'm lost in Python and searching for 'complex data structures'
> > is landing me nowhere.  Basically, I want a dictionary whosevalues
> > are lists of dictionaries.  Simple.  I want to basically loop
> > through some data and so something like this:

> You may want to simply this by using classes. For example, here's a class
> "IP_Address":

> class IP_Address:
>     def __init__(self,ip,hostname,mac_address):
>          self.ip=ip
>          self.hostname=hostname
>          self.mac_address=mac_address

> Then:
>       new_forward=IP_Address(ip_address,hostname,mac_address)

> Then:

>       masq_table['forward'].append(new_forward)

> then later:

>    for ip in masq_table['forward']:
>        ...do something here...

I think this is the best solution, in the long run.  I used to use
Perl a lot, and did all the list of list of hashes of lists, and so
on.  But using something like a struct will improve the long-term
readability, maintainability and extensibility of the code, IMO.  

-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From jilanik at tin.it  Sun Jun  4 06:10:22 2000
From: jilanik at tin.it (Jilani Khaldi)
Date: Sun, 04 Jun 2000 12:10:22 +0200
Subject: Windows application with python - Tcl/Tk. How?
Message-ID: <393A2B0E.6879595F@tin.it>

Hi All,
could somebody send me a little example, or point me where can find it,
about how to use python with Tcl/Tk to write GUI windows applications?

Thank you.

Jilani
--
<jk>
___________________________________________________
http://space.tin.it/scuola/jkhaldi
FreeBSD ~ The Power to Serve







From jae at ilk.de  Sat Jun  3 15:54:26 2000
From: jae at ilk.de (Juergen A. Erhard)
Date: Sat, 03 Jun 2000 21:54:26 +0200
Subject: String.join revisited (URGENT for 1.6)
In-Reply-To: <8hb67j$71h$1@slb6.atl.mindspring.net> (aahz@netcom.com)
References: <28052000.5@sanctum.jae.ddns.org> <39346FAC.F2797EA5@cosc.canterbury.ac.nz> <8h320n$6fo$1@newshost.accu.uu.nl> <03062000.1@sanctum.jae.ddns.org> <8hb67j$71h$1@slb6.atl.mindspring.net>
Message-ID: <03062000.2@sanctum.jae.ddns.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>>>>> "Aahz" == Aahz Maruch <aahz at netcom.com> writes:

    Aahz> [cc'd to Guido for his delectation upon vacation return]
    Aahz> In article <03062000.1 at sanctum.jae.ddns.org>,
    Aahz> Juergen A. Erhard <jae at ilk.de> wrote:
    >> 
    >> It would seem (at least to me) to be the nicest (and cleanest) design
    >> to be able to do:
    >> 
    >> "some string".split().join()

    Aahz> That's what some people advocate, but that's semantically
    Aahz> equivalent to allowing [].join().  As has been pointed out
    Aahz> several times, adding a join() method to sequence types is a
    Aahz> Bad Idea.

I don't see why it should be a `Bad Idea'.  Or rather, I don't buy the
arguments put forward for calling it bad.

For one, in my original mail at the start of this thread I showed how
this could be done without `putting string knowledge into sequences'
(disregarding performance issues for now).

The other objection was that [].join would be 'lists and tuples are
*heterogeneous* containers. you can put *any* kind of Python object in
them, not just strings' (as quoted from a mail by the effbot).

Which is a non-argument... the exact same thing can be said against
string.join(list, sep).  So why should list.join(sep) not simply do
the same thing for non-strings that string.join(list, sep) does and
raise an TypeError exception?[1]

    Aahz> So far, nobody has pointed out *ANY* disadvantages of
    Aahz> retaining the current string.join() mechanism, with one
    Aahz> exception: it destroys the "orthogonality" of moving all
    Aahz> string functions to string methods.  Given that various
    Aahz> people have demonstrated that join() has disadvantages as a
    Aahz> method for both string or sequence types, I think
    Aahz> maintaining the 1.5.2 status quo is probably the best
    Aahz> option.

I repeat: none of the arguments against list.join really has any legs.
IMNSHO, of course.  When one day (in the hopefully not-so-distant
future) classes and types are merged, there (also hopefully) will be a
Sequence class that would be the ideal place for a .join method.

I'll shut up now ;-)

Bye, J

[1] The exact error is: "TypeError: first argument must be sequence of
strings".

- -- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
          My WebHome: http://members.tripod.com/Juergen_Erhard
             Internet Movie Database (http://www.imdb.com)
          Codito, ergo sum - I code, therefore I am -- Raster
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: Use Mailcrypt and GnuPG <http://www.gnupg.org/>

iEYEARECAAYFAjk5Ym8ACgkQN0B+CS56qs1VewCgmxv/Qfq0HhQx9xPvyXs6XUD4
VyoAn3tUly3MqE/exoigU6gHXzFJ5pnz
=3ezs
-----END PGP SIGNATURE-----



From gansevle at cs.utwente.nl  Thu Jun 29 10:17:47 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Thu, 29 Jun 2000 14:17:47 GMT
Subject: Mapping a network drive
References: <2B1262E83448D211AE4B00A0C9D61B03013011E2@msgeuro1.creo.be> <8iac3u$im8$1@nnrp1.deja.com>
Message-ID: <8jflpo$a0t$1@nnrp1.deja.com>

In article <8iac3u$im8$1 at nnrp1.deja.com>,
  Fred Gansevles <gansevle at cs.utwente.nl> wrote:
> In article
<2B1262E83448D211AE4B00A0C9D61B03013011E2 at msgeuro1.creo.be>,
>   Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> > Hello to all,
> >
> > anyone an idea how I can map a drive using pythonwin, and see
> > if it was successfull or not? I'm running python 1.5.2 on WinNT 4.
> >
> > Pieter
> >
> >
>
> This code I wrote using the 'dynwin' package.
>
I just 'stumbeld' over the (not yet documented) win32wnet module in
win32all-129

With that module it is even simpeler

Look at the 'new' wnet.py

# 'new' wnet.py -- uses win32wnet form win32all/build 129

from win32wnet import WNetAddConnection2, WNetCancelConnection2

PROVIDER = 'Microsoft Windows Network'
# If PROVIDER is an empty string, all the NetworkProviders are probed
# in the order defined in the ControlPanel->Network->Properties

def use (drive, host, user, passwd):
    remote = r'\\%s\%s' % (host, user)
    return WNetAddConnection2 (1, drive, remote, PROVIDER, user, passwd)

def unuse (drive):
    return WNetCancelConnection2 (drive, 1,
1)

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From dougfort at downright.com  Sun Jun 11 08:31:34 2000
From: dougfort at downright.com (Doug Fort)
Date: Sun, 11 Jun 2000 08:31:34 -0400
Subject: time.clock() always returns 0?
Message-ID: <GEL05.4911$HD6.127120@iad-read.news.verio.net>

On my system (RedHat Linux 6.2 Python 1.5.2 as shipped with RedHat)
time.clock() always returns 0.  time.time() returns a good value.  Is this
known behavior?  Am I doing something wrong?

--
Doug Fort
Meat Manager
Downright Software
http://www.dougfort.com/~dougfort





From scherbi at bam.com  Tue Jun  6 13:49:07 2000
From: scherbi at bam.com (Bill Scherer)
Date: Tue, 06 Jun 2000 17:49:07 +0000
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <001101bfcfde$0a6a6f20$48a2143f@tim>
Message-ID: <393D3993.84B507C9@bam.com>

> [Michael Hudson]
> > ...
> > Tangentially, does anybody know of any implementations of JavaScript
> > (not necessarily connected to a browser) that let you type interactive
> > commands into a listener and genereally muck about with the language?

There's the NJS Javascript Interpreter:
     http://www.bbassett.net/njs/

-Bill Scherer




From phil at geog.ubc.ca  Sat Jun 10 20:26:02 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 10 Jun 2000 17:26:02 -0700
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no>
Message-ID: <m33dml6p34.fsf@brant.geog.ubc.ca>

Peter Schneider-Kamp <petersc at stud.ntnu.no> writes:

> I have to create a tuple (of two integer values) in a
> C extension. At the moment I use the following (rather
> ugly in my eyes) construction:
> 

[snip]

> Is there "A Better Way(TM)"?
> 

If you're able to use the CXX extension, the sequence is:

Tuple t(2);
t[0]=Int(a);
t[1]=Int(b);

If the conversion fails, Int will throw an exception that
can be caught on the python side.

Regards, Phil


From wbrian2 at uswest.net  Wed Jun 14 14:46:15 2000
From: wbrian2 at uswest.net (Brian Wisti)
Date: Wed, 14 Jun 2000 11:46:15 -0700
Subject: Regular Expressions and Threads?
Message-ID: <3947D2F7.264BDC6E@uswest.net>

Hi all,

I was wondering if anybody could help me out with using regular
expressions within a thread.  I am extremely new to threads, so pleas be
patient with me.


My current project involves grabbing a large number of HTML files from
different servers around the Web (sort of a meta-search thing). These
files are then parsed into a single file. Since speed of response was an
issue, I thought it would be best to use a separate
threading.Thread-derived object to grab and parse each page.  The code
to parse the HTML is kept in separate files which get compiled and
exec()'d inside the thread. (I plan on adjusting that so the code gets
compiled at startup & exec()'d in the thread once I iron this oddness
out)

On the first run, it works great.  Everything is grabbed, everything is
parsed, and it comes out looking all nice and pretty for the user.

Subsequent runs are another story, though.  Everything is still grabbed
and parsed and made pretty, but the results from the previous request
are sent as well.  After three separate requests, you are looking at a
very long page indeed!  It ends up looking sort of like this:

--request 1--
<results from request 1>

--request 2--
<results from request 1>
<results from request 2>

--request 3--
<results from request 1>
<results from request 2>
<results from request 3>

I've been beating my head against the monitor for a few days on this,
and I _think_ I've narrowed it down to the regular expression objects
which are compiled within the exec'd code (in its own thread,
remember).  If I just dump the raw HTML rather than parsing it, there's
no repetition.

I thought I'd be clever and del() the regular expression object at the
end of the code to be exec()'d, but that has unpredictable results. 
Sometimes it works, sometimes it doesn't.

No luck finding anything helpful via searching python.org, or any of the
search engines I frequent.  Mostly what I've found is 're is
automatically threadsafe.'

So ... umm... what am I probably missing about threads and / or the
Python re module?  Anything would be helpful right now, even which Fine
Manual I should read.

Thanks,
Brian Wisti
 
----------------------------------------------------------------------
| "All you need is ignorance and confidence; then success is sure."  |
|                                                    - Mark Twain    |
----------------------------------------------------------------------
| Brian Wisti   |  wbrian2 at uswest.net  | http://www.COOLNAMEHERE.com |
----------------------------------------------------------------------



From thantos at chancel.org  Thu Jun 29 07:01:50 2000
From: thantos at chancel.org (Alexander Williams)
Date: Thu, 29 Jun 2000 11:01:50 GMT
Subject: How to concatenate external files ..?
References: <slm1odapjev11@corp.supernews.com> <8jf6lf$v2j$1@nnrp1.deja.com>
Message-ID: <slrn8lmb4t.lpi.thantos@brimstone.mecha>

On Thu, 29 Jun 2000 09:59:15 GMT, richard_chamberlain at my-deja.com
<richard_chamberlain at my-deja.com> wrote:

>               file=open(x,'r')

Teach me to post while still mostly asleep here at the office.  I
forgot entirely about opening the iterated filename for reading.
Bugger.

To try and make up for it, how about this wee gem?

> # "fls" is the globbed list of filenames
> # "outfile" is the file object already opened to write to
> flsHandles = map(lambda f: open(f, 'r').read, fls)
> for fH in flsHandles:
>     outfile.write(fH())

Note that you never explicitly give each opened file its own name;
instead, you create a list of all the read methods of all the file
objects created dynamically from the list of filenames.  (That refered
to by "open(FILENAME, 'r').open" is just a function/method reference,
it can still be invoked.)

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From neilh at scintilla.org  Tue Jun 20 19:29:48 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 20 Jun 2000 23:29:48 GMT
Subject: Output from a python script to a calling program.
References: <3951f692.948648187@nntp.interaccess.com>
Message-ID: <M7T35.3891$qz4.30140@news-server.bigpond.net.au>

> When I call a python script from another program, I notice several
> things.
>
> 1) It sort of bunches the output. Specifically it sits and waits for a
> large block of output, then prints, stops printing, sits and waits
> then prints the next batch.
> 2) Sometimes it seems that parts of the output are printed in
> different order them I would expect.

   1) The -u option turns off buffering. However, performance may be lower
without buffering.

   2) You may be seeing standard output being buffered and standard error
not being buffered. So output to standard error appears before some earlier
output to standard output.

   Neil





From aahz at netcom.com  Thu Jun 15 08:59:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 12:59:56 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <8iagu3$ici$1@newshost.accu.uu.nl>
Message-ID: <8iak0c$aau$1@slb3.atl.mindspring.net>

In article <8iagu3$ici$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Garry Hodgson <garry at sage.att.com> wrote:
>> Aahz Maruch wrote:
>>>
>>> Like Grant, I've been using Unix about fifteen years, and I find Perl
>>> bizarre and difficult.
>>
>> after reading several responses like this, i wonder if perl
>> is particularly irritating to old timers.
>
>I'm definitely not an old timer and it's frustrating to me too.
>(Gordon McMillan makes comments like "Even Martijn knows it, and he's
>young" :)

If you recall the age-of-FORTRAN-use discussion, I'm on the youngish
side for comp.lang.python, myself, though I expect I'm slowly creeping
up on the average.  ;-)

Say, whatever happened to Gerrit, anyway?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From akuchlin at mems-exchange.org  Wed Jun 14 11:30:05 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 14 Jun 2000 11:30:05 -0400
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i34pv$2ruu$1@nntp6.u.washington.edu> <8i5p73$6eo$1@nnrp1.deja.com>
Message-ID: <3dr9a046wy.fsf@amarok.cnri.reston.va.us>

david_ullrich at my-deja.com writes:
>     One of the very few things that bugs me about the Python
> docs... let's say one of the _incredibly_ few things that
> bugs me about the Python docs is where the explanation for
> how something works refers to a UNIX manpage.

The problem is that the semantics of some operations can be quite
complicated, and sometimes they'll vary across different Unix
implementations.  It would be a big job to document all this in
Python, so it's easiest to just point people to the man pages for
their platform.  Perhaps the best compromise would be to give a brief
explanation of the most important points, and refer to the platform's
docs for the fine points of semantics.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
On a scale of one to ten I'd give it a tim.
  -- William Tanksley, 13 Sep 1999

 


From robinporter72 at yahoo.ca  Mon Jun 12 08:32:44 2000
From: robinporter72 at yahoo.ca (Robin Porter)
Date: Mon, 12 Jun 2000 08:32:44 -0400 (EDT)
Subject: NumPy
Message-ID: <20000612123244.5898.qmail@web1503.mail.yahoo.com>

Does anyone know where I can get NumPy?


Mr. Robin Porter
St. Lawrence College
Computer Engineering

Database Coordinator - Raiser's Edge

Alumni Development

 



---------------------------------
Do You Yahoo!?
Get your free @yahoo.ca address at Yahoo! Mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000612/d1804433/attachment.html>

From jcoruna at umd.es  Tue Jun  6 06:45:49 2000
From: jcoruna at umd.es (Juan Carlos =?ISO-8859-1?Q?Coru=F1a?=)
Date: Tue, 06 Jun 2000 10:45:49 GMT
Subject: MimeWriter and Excel file
References: <20000605.17343286@jcoruna2.bio.umd.es> <8766rnj3eq.fsf@parus.parus.no>
Message-ID: <20000606.10454973@jcoruna2.bio.umd.es>


>>>>>>>>>>>>>>>>>> Mensaje original <<<<<<<<<<<<<<<<<<

El 05/06/00, 22:09:33, Jon K Hellan <hellan at acm.org> escribi? sobre el 
tema Re: MimeWriter and Excel file:


> juan Carlos Coru?a <jcoruna at umd.es> writes:

> > Hello all,
> > I have the following problem:
> > I send an email with python using the MimeWriter module and an excel
> > file attached. When I send the to myself inside our intranet the
> > attchmente arrives well, but if I send the email to anyone outside our
> > intranet and open the attachment with MS Excel I see the base64
> > encoded characters. Any ideas.

> I would be very surprised if this turned out to be a Python
> problem. The MIME apparently doesn't arrive intact at the recipient.

> As to why, I can only guess. It could be an anti-virus measure at the
> receiving site.

> Regards

> Jon K?re

It must be a Python problem, because sending it with the mutt program 
it arrives correctly.





From rjroy at takingcontrol.com  Tue Jun  6 22:04:03 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 07 Jun 2000 02:04:03 GMT
Subject: Seek in urllib.urlopen()?
References: <393ca88a.228109774@news.online.no>
Message-ID: <393dabe7.80124968@news1.on.sympatico.ca>

On Tue, 06 Jun 2000 07:32:22 GMT, thomas at cintra.no (Thomas Weholt)
wrote:

>Hi,
>
>I was wondering if you could use some method similar to seek in the
>urllib-module. If I want to read just the 256 bytes in the middle of a
>file, not read the entire crap in front of those bytes, can I do that,
>or do I have to read all the data in front too?
>
>Thomas

>From the documentation for urllib.urlopen(

This supports the following methods: read(), readline(), readlines(),
fileno(), close(), info() and geturl(). 

so no seek()...
nothing preventing you from doing 
x = urlopen('http://www.python.org')
# read into the file and throw away result
x.read(1024)
stuffIWant = x.read(256)

Bob



From thantos at chancel.org  Thu Jun 29 05:47:22 2000
From: thantos at chancel.org (Alexander Williams)
Date: Thu, 29 Jun 2000 09:47:22 GMT
Subject: How to concatenate external files ..?
References: <slm1odapjev11@corp.supernews.com>
Message-ID: <slrn8lm6p8.lk2.thantos@brimstone.mecha>

On Thu, 29 Jun 2000 04:20:49 -0400, Randolph MacKenzie <rmack at eznet.net> wrote:
>import sys, string, glob, os
>
>#First I create a list:
>fls = glob.glob('C:\\wxpython\\wxPython-2.1.16\\docs\\wx\\wx*.htm')
>
>#Then I guess I should open an output file for writing
>outfile = open('temp.txt,'r')
>
>Now what  .. how do I iterate over the list and specify that each file is to
>be read and appended to "outfile" ?

Well, at this point, you're in good shape.  Except outfile needs to be
opened with 'w', not 'r'.  Writing, not reading.

> for f in fls:
>     outfile.write(f.read())
>     print 'File %s written.' % f

There you go.  Bingo bango bongo, and its done.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From PClaerhout at CREO.BE  Thu Jun 22 06:31:14 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 12:31:14 +0200
Subject: List of all running processes under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301212@msgeuro1.creo.be>

Hello,

anyone knows a fuction in Python to get a list off all currently running
processes under NT, so that we can get something similar like the
task manager?

Pieter



From gmcm at hypernet.com  Fri Jun  2 17:18:06 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 02 Jun 2000 21:18:06 GMT
Subject: Question on embedding. www.mcmillian-inc.com
References: <8h8id5$ooa$1@nnrp1.deja.com>
Message-ID: <3938248e$0$25427@wodc7nh6.news.uu.net>

Arint? <jamarijr at my-deja.com> wrote:
>
>I just looked at that web page and saw a sample of embedding.  The
>source is located here http://www.mcmillan-inc.com/embed2.html.  
I am
>not sure if I am reading this right, but it seems the c program runs
>the script by calling "doit" in the *.py file.  How would you call a
>*py file that didn't have any functions just the main thing.

When a file is first  imported, the top level stuff runs. So all you 
need is the PyImport_Import. Unless you want it to happen again 
in the same run.

- Gordon



From jvickroy at sec.noaa.gov  Mon Jun 26 16:49:32 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 26 Jun 2000 14:49:32 -0600
Subject: stopping threads
Message-ID: <3957C1DC.E7C77E2E@sec.noaa.gov>

Hello,

This may not be a Python question per se, but ...

Why is there no stop_execution() function/method in either the thread or
the threading modules?  It seems that if I want to stop an executing
(sub)process, I must launch it using os.fork() which is platform
specific.



From david_ullrich at my-deja.com  Wed Jun  7 12:01:40 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 07 Jun 2000 16:01:40 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news>
Message-ID: <8hlrkn$5t2$1@nnrp1.deja.com>

In article <Wbr%4.69$vh.19893 at news>,
  "OPUS" <rhicks at nospam.rma.edu> wrote:
>
> "Conrad Schneiker" <schneiker at jump.net> wrote in message
> news:8hl866$bt5$1 at news.jump.net...
>
> :     Ruby is a relatively new, very high level, fully OO language
that
> :     integrated many good ideas from Perl, Python, Smalltalk, Eiffel,
> :     ADA, Clu, and Lisp. (Ruby is more fully OO than Python.)
>
> I thought everything in Python was an object and that Python was
designed
> from the ground up to be object oriented...how can you be more "fully"
OO?

    I don't know anything about Ruby, but a language could certainly
be more fully OO than Python. For example in Python although everything
is an object not every type is a class (you can't subclass list, for
example).

> And stop trolling this newgroup for Ruby converts! If someone wants to
learn
> it they will.

    Not if they never hear about it. I thought he went out of his
way to be polite, to make it clear that he wasn't saying Python
bad Ruby good. Sounds to me like something that might be of
interest to a lot of Python people.
    Maybe it doesn't matter to people who are somehow instantaneously
informed of everything, but the rest of us appreciate a little bit
of marginally on-topic information. (I first heard about Delphi when
I was lurking in a VB group trying to figure out whether it was more
awful or less awful than the BASIC I was using at the time...)


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


From m.faassen at vet.uu.nl  Wed Jun 21 10:16:06 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 21 Jun 2000 14:16:06 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <8iqin6$muc$1@newshost.accu.uu.nl>

William Dandreta <wjdandreta at worldnet.att.net> wrote:
> What I am doing is reading a comma delimited text file like this

> line = AB12345,20000621,3.56,...

> fields = splitfields(line,',')

> recognizing that

> fields[Date][:5]  is the year

> is more obvious and less cryptic than

> fields[1][:5]

In many cases I'd use tuple/list unpacking for this:

date, address, name, whatever = string.split(line, ",")

If you don't know how many fields you're going to get, this can be
more tricky. But imagine you know the first 4:

date, address, name, whatever = string.split(line, ",")[:4]

Of course changing a field is harder here. Generally I use the idiom of
constructing new lists (or dictionaries or objects) in that case, 
though. After all, if you're really going to manipulate your fields
it makes sense to wrap them up in objects in any case.

> Trying to put this into a dictionary seems a contrivance to me that doesn't
> make things clearer and only adds overhead.

Pythoneers don't care a lot about overhead. :) But it's true that a dictionary
in this case may not be a lot of help.

Regards,

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


From Gareth.McCaughan at pobox.com  Sat Jun  3 19:29:12 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: 04 Jun 2000 00:29:12 +0100
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home>
	<39397FD4.67C7FD73@inc.it>
Message-ID: <867lc64a47.fsf@g.local>

Paolo Redaelli wrote:

> Oh, well. It seems I'm missing an obvious thing.
> I post here my program and the data that caused the problem

I tried running this. It didn't work, but not for reasons
connected with float(). First of all, it complained (after,
so far as I can tell, successfully parsing two floats --
so actually it looks like it's already got past the point
at which it failed for you) at "Node(node,xx,yy)" because
there were no variables called xx,yy. I replaced them with x and y.

Then it complained at an assignment to load_case.nodes[node]
with an out-of-range error. I made it set loadcase.nodes.size
and loadcase.rods.size after it had got loadcase.nodes_n and
loadcase.rods_n.

Then it complained at an assignment to load_case.rods[rod_n].
I fixed the preceding line so that it assigned the result
of map(int, [rod_n, etc]) to the obvious thing instead of
throwing it away.

At that point it made it through the creation of the Telaio
object without complaining.

                       *  *  *

The output you quoted doesn't look complete. It has a
traceback but doesn't say what kind of exception was
raised. Could you post the whole message?

I wonder whether your file might contain some funny
characters (though I'd have thought they'd show up if
so). Try adding, after that diagnostic "print", another
line that says

    print map(ord, x)

and see if there are any values in the list it prints
outside the range 32..126. Or change the print line
to
  print "x(%s)=%s y(%s)=%s" % (type(x),`x`,type(y),`y`)
for more readable output that should give the same
information: look for unexpected escapes in the output.

                          *

Does the following program produce a similar error for you?

for line in open("LCASE.001").readlines():
  for item in string.split(line):
    print "About to try parsing:", item
    x = float(item)
    print "Result is:", x

                          *

Is the LCASE.001 file you posted the only LCASE* file
in the relevant directory, or might it actually have been
during the parsing of a *different* file that it fell over?

                       *  *  *

Something else I happened to spot: you have "lenght"
instead of "length" in a few places. Shouldn't make
any difference to the running of the program, though.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From fstajano at uk.research.att.com  Fri Jun 30 10:11:57 2000
From: fstajano at uk.research.att.com (Frank Stajano)
Date: Fri, 30 Jun 2000 14:11:57 GMT
Subject: "+" does not match?
References: <395A577A.C83AE15D@muc.das-werk.de>
Message-ID: <395ca9cf.12064290@localhost>

On Wed, 28 Jun 2000 21:52:26 +0200, Thomas Thiele
<thiele at muc.das-werk.de> wrote:

>>>> import regex 
>
>>>> s1 = "c++test/"
>>>> s2 = "c++test/Derivetest.c++"
>>>> regex.match(s1,s2)
>-1
>
>but if I substitude the "+" I'll get the correct behaviour.
>
>>>> s1 = "c11test/"
>>>> s2 = "c11test/Derivetest.c++"
>>>> regex.match(s1,s2)
>8
>
>Why? Can anybody explain me that?

Because "+" is a metacharacter in regexps, meaning "1 or more
occurrences of the previous regexp", and as such would need to be
escaped in the pattern.

However I suspect that in your case you might be better off using
string.find() rather than regexp.match(), since from the example you
give you don't seem to actually need regular expressions.


From claird at starbase.neosoft.com  Tue Jun 20 23:22:46 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 20 Jun 2000 22:22:46 -0500
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net>
Message-ID: <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net>

In article <8ip8rf$1ig$1 at slb7.atl.mindspring.net>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <QoU35.15352$Xx5.796205 at bgtnsc06-news.ops.worldnet.att.net>,
>William Dandreta <wjdandreta at worldnet.att.net> wrote:
>>
>>Is there a Python type like the C enum type?
>
>I generally prefer using a dict to simulate it.
			.
			.
			.
I'll say that more forcefully than Aahz does here:  while
enums are often a marker of good taste in C coding, Python
generally achieves the same results with a considerably
cleaner reformulation in terms of dicts.  If you feel a
need for enums while writing Python, you're probably missing
out on a better (and not-so-reminiscent-of-C) coding.

I've become rather dogmatic on this point.  I need the
practice of working to falsify it.  If you have a specific
algorithm you can share that you see as ripe for a Python
enum, please tell us here; maybe we can suggest an
alternative.
-- 

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


From irmen at -nospam-bigfoot.com  Tue Jun  6 15:03:41 2000
From: irmen at -nospam-bigfoot.com (Irmen de Jong)
Date: Tue, 6 Jun 2000 21:03:41 +0200
Subject: [ANNOUNCE] Pyro 1.2 - distributed objects
Message-ID: <8hjhsg$p1$1@news1.xs4all.nl>

PYRO - Python Remote Objects - version 1.2

by Irmen de Jong - irmen at bigfoot.com


WHAT IS PYRO?

Pyro is an acronym for PYthon Remote Objects. It is a basic Distributed
Object Technology system written entirely in Python, and for use in Python
only. With this, it closely resembles Java's Remote Method Invocation (RMI).
It has less similarity to CORBA - which is a system- and language
independent Distributed Object Technology and has much more to offer than
Pyro or RMI.

The most important changes since the previous version (1.1) are:


- Direct remote object attribute access with normal Python syntax.
- Servers can create new Pyro objects and return proxies back to the
  clients. They can then access these new objects.
  (So, it's no longer necessary to have all Pyro objects ready on
  the server when it starts, you can add or remove them as you wish).
- It's now much easier to start multiple servers on a single machine.
- Some small bug fixes and documentation updates.


Some interesting features of Pyro are:

- Naming Service which keeps record of the location of objects. The location
  of the NS can be discovered using a broadcast mechanism, amongst others.

- Support for Python types only restricted by those that the standard
  'pickle' module handles (all builtin types, and user created classes).
  Objects as arguments are also supported when the code for those user
  defined classes is also available on the server side.

- Support for all Python argument types (variable argument lists and keyword
  arguments, *args and **keywords).

- Exceptions that occur in the remote object will actually be raised too on
  the client, just as if the error occured locally. Pyro's exception
  transportation mechanism makes this possible.

- Dynamic Proxies.

- Full source code in Python.



WHERE CAN I GET IT ?

>From my Pyro homepage: http://www.bigfoot.com/~irmen/pyro.html

It comes with a Python-style license, but is otherwise free for
commercial and non-commercial use.


REFERENCE:

<P><A HREF="http://www.bigfoot.com/~irmen/pyro.html">
Pyro 1.2</A> - Distributed Object system for Python (4-Jun-2000)

--
Irmen de Jong -- irmen @
                bigfoot.com






From gmcm at hypernet.com  Mon Jun 12 21:33:20 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 13 Jun 2000 01:33:20 GMT
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu>
Message-ID: <8F51D9DB3gmcmhypernetcom@199.171.54.154>

Michael Vanier: 

>I wrote a little python script using the thread module the other day.  I
>wanted to be able to set a flag in the parent thread that would be read
>in the child thread.  So I set a flag variable (an integer in this case)
>in the parent, created the child thread, and then checked the value of
>the flag every time through a loop in the child thread to see if it had
>changed.  It turns out that the child thread apparently made a copy of
>the flag, because when it changed in the parent it didn't change in the
>child. 

>To fix this, I changed the flag to be a list with one integer element. 
>Now, when I change the list element in the parent, the child sees it
>too.  


Let me guess. You did something like make "flag" a module level global, and 
then did "from ... import *" in the other.

If you had referenced "module.flag", all would have worked. But that 
stinks, too. Your thread should have a flag member, and the parent thread 
should set it (through a setter, if you really want to be virtuous <wink>). 
That way the flag becomes part of the thread's API, instead of a bass-
ackwards dependency.


> I gather that what this means is that there is a fundamental
>distinction between atomic types like integers and reference types like
>lists or dictionaries.  However, since python treats everything as a
>reference this seems inconsistent to me.  So what I want to know is:
>what is the cause of this behavior, and what is the justification for
>it?  Or could it be a bug? 

Aahz is correct (as usual). It has to do with assignment and immutability. 
To see assignment to any immutable (not in your global or local namespace), 
you have to see the immutable through some larger context. Say, a module 
namespace, or (as you did) through a container object.

- Gordon


From gmcm at hypernet.com  Sun Jun 11 16:37:02 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 11 Jun 2000 20:37:02 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net>
Message-ID: <8F50ADB75gmcmhypernetcom@199.171.54.155>

Cameron Laird wrote: 

>On the other hand, I'm still searching for the best
>way to speak about the "well you Python people only
>find those errors at run-time" objection.  I find
>it a challenge to communicate effectively to
>newcomers how much better it is to have a language
>that encourages unit testing as Python does. 

Sometimes you can shock them a bit by claiming they're using the compiler 
as a crutch. When the tools are that different, you need to change working 
styles; something the static-typing dependent don't seem to consider.

>Python's developmental productivity is so much
>greater, in my experience, as also to defy easy
>explanation for those who haven't already used it.

Interesting, though, that this thread (at least as much of it as I've 
tracked) talks about the libs and the syntactic simplicity. Good stuff, no 
question; but not order of magnitude stuff.

One point worth reiterating is that where the app (in a C++ design) calls 
for lots of related classes, you will probably only have a handful in 
Python. Of course, this means discarding classical "OOP" thinking, and 
desinging more in the "component" style. But this can yield an order of 
magnitude all by itself.

The real killer is when you are working on something you don't have a model 
for. Long before you've completed your UML for your best guess of how to do 
it in C++, you can be on your 4th or 5th functional prototype in Python. In 
other words, almost done.

In fact (as I'm sure you've heard me say before), even if I have to deliver 
in C / C++ or Java, I write it first in Python. The few times in the last 3 
years that I've skipped that step, I've regretted it.

- Gordon


From fcastel at ibm.net  Fri Jun 23 11:07:54 2000
From: fcastel at ibm.net (Frank V. Castellucci)
Date: Fri, 23 Jun 2000 11:07:54 -0400
Subject: Dictionary class in C++
References: <8ivrup$nqe$1@news.rchland.ibm.com>
Message-ID: <39537D4A.C62F3798@ibm.net>


Larry Whitley wrote:

> I am writing a python program to process some traces in preparation for
> running the traces through a model written in C++.  Python's dictionary
> class seems to give me what I'm looking for on the python side of the
> equation and I am trying to translate the dictionary to C++ constructs.
>
> Looking through the IBM Open Class library (I haven't upgraded my compiler
> to one that can do STL yet) I find a number of "collection classes".  The
> first level of organization is into Flat Collection Classes or Trees.  Of
> the two Flat seems to be right.  The next level of organization is a table
> too complicated to reproduce here.  The net is that it looks to me like the
> Dictionary is: Unordered and Keyed.   Given that the corresponding
> collection class is one of four posibilities depending on the element and
> key characteritics:
>
> Element equality and element uniqueness = Map
> Element equality and no element unizueness = Relationship.
> No Element equality and element uniqueness = Key Set.
> No Element equality and no element uniqueness = Key Bag.
>
> Can anyone help me sort the above out - which is the python dictionary?
>
> Larry

Its a Map.




From thantos at chancel.org  Tue Jun 27 13:33:00 2000
From: thantos at chancel.org (Alexander Williams)
Date: Tue, 27 Jun 2000 17:33:00 GMT
Subject: Algoritm that'll find usable account names
References: <Pine.LNX.4.21.0006271644130.14053-100000@lade.src.no>
Message-ID: <slrn8lhpab.g2r.thantos@brimstone.mecha>

On Tue, 27 Jun 2000 16:51:46 +0200, Morten W. Petersen <morten at src.no> wrote:
>Does anybody have any snippets of code that'll morph names ?

Would it not be simpler to decide on the algorithm you want to use and
then simply impliment it?  There are likely limits in username
selection for your system that will exclude characters or combinations
thereof, or make them less than choice, or there is a preferential
hierarchy of tries you'd like.

I see people, all too often, forgetting that the first step in
implimenting something is planning it out.  Python  makes implimenting
easy.  Use that.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From loewis at informatik.hu-berlin.de  Tue Jun 27 04:23:51 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Jun 2000 10:23:51 +0200
Subject: French Python documentation
References: <oqwvjj58f7.fsf@titan.progiciels-bpi.ca>
Message-ID: <p6qituvzg4o.fsf@informatik.hu-berlin.de>

=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> writes:

> Hello, my friends.  Robert, Cc:ed, is asking me what exists, in French,
> to document Python.  He would like to get as much as possible for the
> unilingual members of his team.  Please reply to the list, or to me at
> least, so I can save the collective knowledge and wisdom on this point! :-)

A list of such documents is kept at

http://www.python.org/doc/NonEnglish.html#french

If anybody knows of any other resources, please have them added to that list.

Regards,
Martin


From andres at corrada.com  Wed Jun  7 22:28:25 2000
From: andres at corrada.com (andres at corrada.com)
Date: Wed, 7 Jun 2000 22:28:25 -0400
Subject: toString()
In-Reply-To: <8hmspg$r7n$1@bob.news.rcn.net>; from David Allen on Wed, Jun 07, 2000 at 08:26:42PM -0500
References: <8hmspg$r7n$1@bob.news.rcn.net>
Message-ID: <20000607222825.A17061@corrada.com>

On Wed, Jun 07, 2000 at 08:26:42PM -0500, David Allen wrote:
> 
> Is there such thing as a toString() for objects like java has that
> will be automatically called when an object is used "as a string"?
> 

Yes, in Python it's called __str__.

> I.e. when you create an object, __init__ gets called automatically,
> so if I say:
> 
> x = Spam()
> print x
> 
> can I put a method in Spam such that it will print out something
> useful based off of the class fields rather than the ugly data 
> definition?
> 
> Thanks,

In your class definition for Spam include a __str__ method, something like
the following:

class Spam:
  .
  .
  .
  def __str__( self ):
     return self.someAttribute
     
------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From neurocrat at one.net.au  Wed Jun 14 13:37:12 2000
From: neurocrat at one.net.au (Neurocrat)
Date: 15 Jun 2000 03:37:12 +1000
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com>
Message-ID: <85em60nozb.fsf@one.net.au>

Garry Hodgson <garry at sage.att.com> writes:

> Aahz Maruch wrote:
> 
> > Like Grant, I've been using Unix about fifteen years, and I find Perl
> > bizarre and difficult.
> 
> after reading several responses like this, i wonder if perl
> is particularly irritating to old timers.  twenty years ago,
> this kind of obscure but powerful notation (as embodies in awk
> and such), was really cool.  but after a while the appeal kind
> of wears thin, particularly when you find something that gives
> you the same power but without the obscurity.
> 
> i find i'm more enamored of elegance these days.

Me too.
Now if only I could find it in an OS ...


From mhammond at skippinet.com.au  Tue Jun  6 04:55:46 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 06 Jun 2000 08:55:46 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com>
Message-ID: <m03%4.6579$Hz.48365@news-server.bigpond.net.au>

"Tom" <tom-main at REMOVEME.home.com> wrote in message
news:w3Q_4.160138$55.3354931 at news2.rdc1.on.home.com...

> Studio 7, but what does this all mean for the future of the PythonWin?

Dunno.  Not much will change.  As usual, no one will be paying me to work
on it! :-)

> Will PythonWin (the IDE and the MFC wrapper library) continue to be
actively
> and publicly developed?

Dunno.  I like it :-)  If Active State end up developing an IDE that I
would personally prefer to use over Pythonwin, then I doubt I will spend
too much time on Pythonwin as an IDE.  I seriously doubt that will be MSVC
based to qualify.  Certainly, the extra resources AS can dedicate should
mean the Mozilla based IDE ends up qualifying, but Pythonwin exists today
and is quite slick on Windows (as opposed to the cross-platform focus of
AS) meaning I doubt it will qualify in the short term.

The good side of this is that is doesnt "compete" directly with any AS
products.  One IDE is focussed on existing MSVC (and maybe new) owners,
while the other will be a great cross-platform solution.  Pythonwin doesnt
try and be either of these!  So while it is unlikely that AS will fund
further development of Pythonwin, they also will certainly not discourage
me from continuing to tinker.

Re MFC: Microsoft dont seem too interested in MFC any more.  Ive certainly
heard nothing about it, and everything new I see has alternatives that
compete directly with it.

Unfortunately, my personal opinion is that MFC itself will die pretty
soon, kept alive only by the weight of all the old code written for it.
So with that in mind, I dont believe that MFC will undergo enough
significant enhancements that keeping up will be a problem!

Mark [and obviously not speaking for ActiveState]




From M.Vitecek at sh.cvut.cz  Wed Jun 14 04:24:57 2000
From: M.Vitecek at sh.cvut.cz (Michal Vitecek)
Date: Wed, 14 Jun 2000 10:24:57 +0200
Subject: SV: Python Productivity over C++
In-Reply-To: <20000613225554.27650.qmail@hotmail.com>; from hungjunglu@hotmail.com on Tue, Jun 13, 2000 at 03:55:54PM -0700
References: <20000613225554.27650.qmail@hotmail.com>
Message-ID: <20000614102457.A1870@fuf.sh.cvut.cz>

Hung Jung Lu <hungjunglu at hotmail.com> wrote:
>--- In python-list at egroups.com, Neurocrat <neurocrat at o...> wrote:
>>for (int i = 0; i < 10; ++i);
>>         cout << "Hello" << endl;
>>
>>This, of course, prints "Hello" only once, not 10 times as we
>
>When I started to use Python a few years ago, the thing that told me that 
>Guido had some brain was exactly little tiny features like that.
>
>Try to talk to some hardcore Perl or C++ programmers, and most likely they 
>will balk back at you. "What? Python does not even support increment 
>operator like i++? What kind of a retarded language is that?"
>
>I have had hard time convincing to fellow C++ programmers that an assignment 
>inside an "if" statement is a bad thing, to no avail. E.g:
>
>    if ( a = getNumber() ) {...}
>
>Sure, totally valid statement in C++, but why would anyone introduce such a 
>stupid feature so you and your compiler get confused all the time with
>
>    if ( a == getNumber() ) {...}

 i have no idea why you call this a stupid feature - IMO this all makes
 the program shorter and as such more readable. this is really something i
 hate about python, missing ++ and -- operators plus having to do the
 assignment stupidly before if comment - more & longer source.

>C++ programmer get bitten all the time with the == && || operators, and 
>refuse to acknowledge that there is a problem.

   :) maybe C++ programmers like you? i personally never had any problem
 with ==, && and ||, what's more - in the source those 'glyphs' help to
 increase the readability - there's less of real text. can you describe
 your problems with those operators? oh, and don't forget C/C++ has been
 here for years - it's like with VI - you hate it until you
 get to know it in deep.
   and one more thing - most programming languages are similar to C/C++ in
 syntax - and most programmers are/were coming from C/C++ world. making
 differences like and, or instead of && ||, or no assignments in if
 statements means much less people actually like python and refuse to use
 it. it has some really great features, other are really stupid and
 uncalled-for.

>
>Guido was smart enough to get rid of the confusing operators. You can't do
>
>   if a = b:
>
>and you have to use explicitly the "or" "and" keywords instead of the 
>cryptic "&&" "||". And the superflous operators "+=" "++" are eliminated. 
>(Why? If you are trying to save nanosecond execution time, or 3 bytes in the 
>code by using Python, you are in the wrong business.)
>
>And Perl people would issue you a stern warning about the indentation in 
>Python, they'd say: "be very careful with the unusual indentation of Python, 
>you miss one space and all hell breaks loose." Yeah right. I've been 
>programming in Python for 3 years, now. It might have happened 2 or 3 times. 
>Compare with Perl where every single program I wrote was a nightmare to 
>debug, even if they are just a few lines. And guess what? No matter whether 
>it is Perl or C++, I always do proper indentation and match up all the 
>statements and parentheses neatly by hand, so do all the dozens programmers 
>that I have worked with. So the parenthesis and the semicolons sit there for 
>... I guess, to be pretty? :) For avoiding a bug that might happen once a 
>year, Perl and C++ prefer that you introduce bugs that happen daily or even 
>hourly. Great logics.
>
>After I saw what Guido did with the == && || ++ operators, I knew some brain 
>was put into Python. And since then, I have never looked back.
>
>Hung Jung
>
>
>________________________________________________________________________
>Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
>
>
>-- 
>http://www.python.org/mailman/listinfo/python-list

-- 
			fuf


------------------------------ na IRC -------------------------------------
 BillGates [bgates at www.microsoft.com] has joined #LINUX
 ...
 mode/#linux [+b BillGates!*@*] by DoDad
 BillGates was kicked off #linux by DoDad (banned: We see enough of Bill
          Gates already.)
 




From hzhu at knowledgetrack.com  Wed Jun 28 19:30:33 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 28 Jun 2000 23:30:33 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <vndwvj9bgzh.fsf@camelot-new.ccs.neu.edu> <slrn8ll00c.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8ll2qj.25h.hzhu@rocket.knowledgetrack.com>

On Wed, 28 Jun 2000 22:42:33 GMT, Huaiyu Zhu <hzhu at knowledgetrack.com> wrote:

>To avoid any further unnecessary digressions, I state my actual problem
>here.  I'm developing the MatPy pacakge which works quite well on Linux and
>Solaris, but some Windows user's keep complaining about name clashes.  It
>turns out that I have, in the file MatPy/gnuplot.py, this line
>
>from Gnuplot import Gnuplot
>
>Since on Windows Gnuplot.py and gnuplot.py is indistinguishable, this line
>tries to import itself!  
>

Let me describe this more precisely.  Within $PYTHONPATH/MatPy/gnuplot.py,
the statement

import Gnuplot

will import $PYTHONPATH/Gnuplot/__init__.py on Unix, which is intended, but
$PYTHONPATH/MatPy/gnuplot.py on Windows, which causes havoc.


Now the simplest solution for this particular problem is just to rename
MatPy/gnuplot.py into something else.  I am objecting to this on principle.
Because this essentially requires every module in every package to be
uniquely named (and case insensitive) from the beginning, as one is never
sure that in some future implementation one might not want to import a
module from a different package with an identical name.

I'm open to any solutions.  Be there any doubt as to what the problem is,
you can test it on the MatPy package listed below as well.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From MarkH at ActiveState.com  Sun Jun 25 02:37:22 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 25 Jun 2000 06:37:22 GMT
Subject: Serial port IO Python module?
References: <39554BAE.6C00C44E@es.co.nz> <39554DD6.C9B45E88@zebrasoft.co.nz>
Message-ID: <CMh55.3859$Sn2.24571@news-server.bigpond.net.au>

"Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
news:39554DD6.C9B45E88 at zebrasoft.co.nz...

> OK, OK, so I should have checked the ng before posting. the sio-151 is a
> pretty good fit for what's needed. Out of interest though, anything else
> out there?

Another alternative is to use a COM object that provides serial port
access - Pythonwin contains a demo of such an object.

Also, I only recently posted that Active State sponsored the completion of
the win32file module (completion WRT serial IO, anyway:), but they havent
yet been released as binaries.    So quite soon, or right-now if you work
from CVS sources, you will have complete access to the "low-level" win32
API for serial IO.

Mark.





From mfletch at tpresence.com  Mon Jun 19 19:04:24 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 19 Jun 2000 19:04:24 -0400
Subject: Binary Input: read how?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33D3@VRTMAIL>

You're currently trying to & a string, only numbers respond to &'ing in
Python.  ord(b[0]) gives you the integer value for the character.

>>> for x in range(30):
... 	if ord(b[0]) & (2**x):
... 		print 1,
... 	else:
... 		print 0,
... 
0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

See also various recent postings about getting the "bits" of a number
(though you'll still need to do an ord on the character).

HTH,
Mike

-----Original Message-----
From: Michael Morrison [mailto:borlak at home.com]
Sent: Monday, June 19, 2000 6:49 PM
To: python-list at python.org
Subject: Binary Input: read how?


Here's the basic problem: I can read a binary file using open(...,'rb') but
when I assign it to a object (b = file.read()) it stores it as an escape
character or \300 - whatever range.

I want to test this byte of binary data by comparing it's bits.  b[0] & 1 ..
b[0] & 2
for x in range(30):
    if b[0] & 2**x:
        print 1,
    else
        print 0,

This doesn't work.  Tried all sorts of things.... Can't get it to work :(

What am I missing here?

thanks


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



From tanzer at swing.co.at  Sat Jun 10 14:24:43 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sat, 10 Jun 2000 20:24:43 +0200
Subject: Python Productivity over C++ 
In-Reply-To: Your message of "Fri, 09 Jun 2000 11:34:10 EDT."
             <39410E72.3DDE39B7@colconsulting.com> 
Message-ID: <m130pw3-000wcDC@swing.co.at>

"Frank V. Castellucci" <frankc at colconsulting.com> wrote :

> > > 2. There is the type information in standard conforming compilers.
> > 
> > Nowadays, there is.
> 
> Agreed, but the language had the ability for you to write your own, or
> use someone else's library which provided the same.

Sure it provides it. But you have to implement it. And that costs
time.

The decisive difference is:

- in C++ you have to add RTTI code to every class in the class
  hierarchy -- it is simply impossible to factor the behavior into the
  root of the hierarchy. 

- In Python, you can implement the entire mechanism needed in the root
  of the hierarchy.

Of course, nobody forces you to do that, and you might end up with
exactly the same design as in C++.

> Every language has a painful aspect to itself. Some take more time to
> correct than others.

Sure. But you can still compare the total amount of pain and the time
it costs you. And here, Python comes out way ahead.

If the problem mandates it, you might still have to use C++. But you'll
need more development time. And it is less fun to use C++.

> I try not to pick a language for implementation until I am happy that
> problem space is adequately defined, a reasonable degree of requirement
> specification have been put forth, a sound analysis and design has been
> done, and a language is available which satisfies the solution space. As
> a contract developer this time alloted to perform these tasks may span
> from one weekend by myself to many months with teams of people.
> Sometimes the people who hire me decide the language before these steps.
> I do the best I can.

Great. I hope we all do (though I met too many people who don't).

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From aa8vb at yahoo.com  Tue Jun  6 07:24:30 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 06 Jun 2000 07:24:30 -0400
Subject: rld failure on IRIX
In-Reply-To: <393839a7_4@corp.newsfeeds.com>; from soper@delanet.com on Fri,
 Jun 02, 2000 at 06:43:08PM -0400
References: <393839a7_4@corp.newsfeeds.com>
Message-ID: <20000606072430.A111487@vislab.epa.gov>

Paul Soper:
 |After building Numerical Python 15.3 and MMTK 2.0 on IRIX 6.46 without any
 |errors during the 'build' or 'install' steps, both packages generate error
 |messages like the following;
 |
 |ImportError: 135737:python: rld: Fatal Error: Cannot Successfully map soname
 |'/usr/local/lib/python1.5/site-packages/MMTK/irix646/MMTK_universe.so' under
 |any of
 |the filenames
 |/usr/local/lib/python1.5/site-packages/MMTK/irix646/MMTK_universe.so
 |
 |The file referred to does exist and has all of its permissions set
 |correctly.  Why is the dynamic linker upset?

I'm going to guess mismatched binary architectures:

> file /usr/local/lib/python1.5/site-packages/MMTK/irix646/MMTK_universe.so
> file /usr/local/lib/python1.5/site-packages/Numeric/multiarray.so
> file `which python`

These should all agree.  For example, the following indicates a problem:

MMTK_universe.so: ELF 32-bit MSB mips-2 dynamic lib MIPS - version 1
multiarray.so:    ELF N32 MSB mips-3 dynamic lib MIPS - version 1
python:           ELF N32 MSB mips-3 dynamic executable (not stripped) MIPS - version 1
                      ^^^^^^

Use $SGI_ABI or $COMPILER_DEFAULTS_PATH to set your architecture flag
preference [see abi(5) and pe_environ(5) for more details, or feel free to
drop me a line if you have questions].

-- 
Randall Hopper
aa8vb at yahoo.com



From te at first.gmd.de  Wed Jun  7 08:47:40 2000
From: te at first.gmd.de (Thilo Ernst)
Date: Wed, 07 Jun 2000 14:47:40 +0200
Subject: Python under VMware/NT?
References: <r3h%4.23$cT1.480@news> <14653.56846.80327.424290@anthem.python.org>
Message-ID: <393E446C.665CA10A@first.gmd.de>

"Barry A. Warsaw" wrote:

> I very briefly ran Linux under NT-native VMware and of course ran
> Python inside the Linux vm.  Worked fine.  I had the general
> impression that networking was much slower for the entire OS but spent
> no time benchmarking or otherwise investigating.  I do plan to play
> with NT running inside Linux-native VMware pretty soon, but I don't
> expect it to be all that different than running in native NT.
>

I could almost repeat Barry's words, just with the OSes reversed. I tried
some Python stuff on a virtual NT4.0 under Linux-native VMware recently,
and noticed no severe performance degradations. My Python programs
weren't network-intensive, though.

The general network performance (i.e. with non-Python apps) is, ahem,
acceptable, with some (reproducible) exceptions in the "absolutely lousy"
range. E.g. saving a PowerPoint document to a Samba share on the Linux
"mother" machine takes AGES as soon as the presentation has more
than around 10 slides. This is getting off-topic, but ... hints anybody?
(Perhaps I should simply upgrade to VMware 2.0; still running 1.1)

Nevertheless, by and large I just love having this NT window around in a
corner  of my KDE desktop. Especially, the setup was great for Python-based
interoperability experiments I did some months ago, e.g. remote-controlling
MS-Office apps from Linux via DOPY and the win32 extensions.

Best regards, Thilo




From stevena at permanent.cc  Fri Jun  9 10:20:17 2000
From: stevena at permanent.cc (Steven D. Arnold)
Date: Fri, 09 Jun 2000 10:20:17 -0400
Subject: = vs. == in Py3k
In-Reply-To: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <4.3.1.2.20000609101313.01add2f8@phear.dementian.com>

At 11:18 PM 6/8/2000 -0700, Lawrence Kesteloot wrote:

>But this isn't worth the trouble it gives new programmers.  The
>syntax could be changed to make = and == equivalent tokens
>and have the parser figure out which is meant from the context,
>just as it's done in Basic.  The documentation would only ever
>mention =, but the language would support == for backward
>compatibility.

Or we could have `==' always be a comparison operator, but if `=' was used 
in `==' context, it would be interpreted as `=='.  One possible negative is 
that this could lead to inconsistent coding and sloppiness, but it seems 
like a virtue of a good scripting language to be as understanding as 
possible about this sort of error.  (I do think at a minimum a user should 
at least be able to set a flag to get compiler warnings, a la perl `-w', 
and when this flag is set, an `=' in an if clause should raise a warning.)


steve




From m.faassen at vet.uu.nl  Fri Jun  2 10:01:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Jun 2000 14:01:37 GMT
Subject: Status of List
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com>
Message-ID: <8h8eo1$7c0$3@newshost.accu.uu.nl>

Emile van Sebille <emile at fenx.com> wrote:
> Can anyone confirm the problem? Or should I look on my
> end.  Or is it just that quiet, and what I saw in usenet
> only the result of propogation delay.

With others, I think it's more than that. I've received some messages in
email that were cc to the list, that haven't appeared in this newsgroup
yet (and it's been a while).

Regards,

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


From mhammond at skippinet.com.au  Tue Jun 13 14:47:22 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 13 Jun 2000 18:47:22 GMT
Subject: WIN32API GetLastError()
References: <8i3460$ppm@news.or.intel.com>
Message-ID: <8i5vjl$bh9$1@nnrp1.deja.com>

In article <8i3460$ppm at news.or.intel.com>,
  "Jeff" <no_spam at umich.edu> wrote:

> ther than 0.  I am suspecting that this means that the GetLastError
> it is trying to "get" is an error from the pythonwin process
> and not from an API call I make (example writing to a file when
> only read access was given, or from an extension).  So my question

win32api.GetLastError() simply wraps the API function.  The error is
for the last function called, if it supports GetLastError().

Note that many functions DO NOT support GetLastError() - they return
the error code in a result, so it is pointless.  GetLastError() is
mainly used for functions that return bools, or (say) handles, where
only failure is indicated, and not the specific failure.

> is, what can GetLastError be used for,
> and is it at all possible to get the windows error code from a bad
> win32file.CreateFile??

I believe not - I dont think CreateFile uses GetLastError.  Note that
if CreateFile fails, an exception will be raised with the error code -
so GetLastError() is not necessary anyway - simply get it from the
exception.

Mark.


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


From donn at oz.net  Wed Jun 21 01:49:18 2000
From: donn at oz.net (Donn Cave)
Date: 21 Jun 2000 05:49:18 GMT
Subject: Python sockets/select
References: <8ip6kt$1lc$1@bob.news.rcn.net>
Message-ID: <8ipl0u$9l3$0@216.39.151.169>

Quoth "David Allen" <s2mdalle at titan.vcu.edu>:
...
| When I select() sockets for reading, writing, and error conditions,
| there are sockets (usually connected to clients who have already
| quit or died unexpectedly) that don't show up in the error queue,
| but continually flag select() as being readable.  When I read them,
| of course, '' comes back.
|
| For some reason, sometimes '' comes in for valid clients, so I 
| can't just disconnect a client because I get that from them.  Is
| there a way to detect which clients are alive and which aren't?
| What causes a client to send '' when I didn't tell it to, according
| to select()?  (The clients are also written in python)

When you get 0 bytes back from a read, that's EOF (End Of File.)
As you apparently surmised.  That is indeed a readable event, at
least the first time.  I'd say behavior after the first 0 byte
read is probably undefined, but anyway it's not going to do anyone
any good.

You should never get a 0 byte read before the socket has been
closed on the other end, so it sounds like there's something
wrong (in your server, not in your clients.)

	Donn Cave, donn at oz.net


From jwbnews at scandaroon.com  Fri Jun 23 18:00:39 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 23 Jun 2000 15:00:39 -0700
Subject: writelines() or write()
References: <3953C7B9.E32E8599@seas.upenn.edu>
Message-ID: <jwbnews-7C68B0.15003923062000@news.olympus.net>

In article <3953C7B9.E32E8599 at seas.upenn.edu>, dwhite2 at seas.upenn.edu 
wrote:

> strings = ['one \r', 'two \r', 'three \r}
> outfile.writelines(strings)
> ---
> outfile looks like this:
> 'three '
> ---
> What I want to have happen of course is:
> 'one '
> 'two '
> 'three '

Use of \n (newline, meaning whatever that is on the machine you're 
working on) rather than \r would be a good first step.  Clearly you 
didn't cut and paste, otherwise the code as shown wouldn't have done 
anything.

The file I get on my BSDi system is (thanks, hexdump)
6e6f 2065 740d 6f77 0d20 6874 6572 2065 000d
just as you directed.

Note that the \r is faithfully reproduced as 0x0d.  With some commands, 
nothing will be seen because the file has no complete lines or because 
the trailing CR wipes out the 'three' (cat on BSDi)...with others, the 
[virtual] Carrier will Return with each \r, and the next line overwrites 
the prior line...that's likely what you're seeing.  And the old less on 
BSDi will show

one ^Mtwo ^Mthree 

(CR is control-M).

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From thomas at xs4all.net  Sun Jun 18 14:45:55 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 18 Jun 2000 20:45:55 +0200
Subject: Rude question about BeOpen.com
In-Reply-To: <Pine.HPX.4.10.10006181239180.21364-100000@merle.acns.nwu.edu>; from nickgeo@merle.acns.nwu.edu on Sun, Jun 18, 2000 at 12:45:02PM -0500
References: <Pine.HPX.4.10.10006141744570.3588-100000@merle.acns.nwu.edu> <20000615010758.D26436@xs4all.nl> <Pine.HPX.4.10.10006181239180.21364-100000@merle.acns.nwu.edu>
Message-ID: <20000618204554.W26436@xs4all.nl>

On Sun, Jun 18, 2000 at 12:45:02PM -0500, Nicholas Geovanis wrote:
> > On Wed, Jun 14, 2000 at 05:47:15PM -0500, Nicholas Geovanis wrote:
> > > How will the move of the Python crew to a commercial venture affect the
> > > Computer Programming For Everybody (CP4E) project?

> > Q. 7. What is happening to CP4E (Computer Programming for Everybody)?

> > A. We love this project and are looking for industry sponsors to help us
> >    expand this multi-disciplinary work to include educators and usability
> >    experts. Contact sponsor at beopen.com if you are interested.

> OK, but that doesn't answer my question.
> It seems then that the attempt to obtain US government funding for CP4E
> has been abandoned, and therefore that there is no upcoming funding
> decision to be made (possibly could have elicited an answer of "no" from
> them). True or false? 

I'm not sure on the status, but I dont think it the attempts to find funding
been 'abandoned' -- they might've been fruitless, but I doubt Guido and the
other people involved in CP4E would reject government funding. Indeed, I
read the FAQ entry as 'any funding would be more than welcome.' I cant give
you definitive answers on these issues, though, you might have to wait for
Guido to come back (or ask a more CP4E-targeted platform, like the EDU-SIG)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From MarkH at ActiveState.com  Fri Jun 23 19:33:14 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 23 Jun 2000 23:33:14 GMT
Subject: CurrentUser for remote machine under NT
References: <2B1262E83448D211AE4B00A0C9D61B0301301217@msgeuro1.creo.be> <8j0gl2$i9k$1@nnrp1.deja.com>
Message-ID: <_sS45.2748$Sn2.11688@news-server.bigpond.net.au>

<kellyk at my-deja.com> wrote in message news:8j0gl2$i9k$1 at nnrp1.deja.com...

> You can try this. It does not always work.

For workstation statistics, you can try this - it _should_ always work (on
NT, anyway!):-)

>>> import win32net, win32api
>>> import pprint
>>> pprint.pprint(win32net.NetWkstaGetInfo(win32api.GetComputerName(),
502))
{'buf_files_deny_write': 0,
 'buf_named_pipes': 1,
 'buf_read_only_files': 0,
...

To get info on the logged-in user, try:
>>> pprint.pprint(win32net.NetUserGetInfo(None, win32api.GetUserName(),
3))
{'acct_expires': -1,
 'auth_flags': 0,
 'bad_pw_count': 0,
 'code_page': 0,
 'comment': u'',
...


Mark.





From rwallace at esatclear.ie  Fri Jun 16 20:14:45 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Sat, 17 Jun 2000 01:14:45 +0100
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <1ecb5k1.1s3tce6xg2hfxN%bparsia@email.unc.edu>
Message-ID: <394AC2F5.23C3@esatclear.ie>

Bijan Parsia wrote:
> Actually, the "simplistic" part is not taking into account things like
> problem domain, tools, and, most importantly, programmer "type" (skill,
> experience, etc.). Strangely enough, I would bet that *temperment*
> counts for oodles

I completely agree.

> (Er...ok,
> Smalltalk/Squeak makes me way more productive

Good example: I tried Smalltalk, never got anywhere at all with it.  But
I know a good many people swear by it.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From kbdavidson at my-deja.com  Mon Jun 12 12:11:23 2000
From: kbdavidson at my-deja.com (kbdavidson at my-deja.com)
Date: Mon, 12 Jun 2000 16:11:23 GMT
Subject: OpenDX Python Wrapper -- any interest?
References: <20000515112219.C1304991@vislab.epa.gov>
Message-ID: <8i323b$c4s$1@nnrp2.deja.com>

Add my vote for wanting to see this...


In article
<20000515112219.C1304991 at vislab.epa.gov>,
  Randall Hopper <aa8vb at yahoo.com> wrote:
>      A while back I wrote a Python wrapper for
OpenDX, complete with
> callbacks to Python and exceptions thrown
across the DXL layer.  Before I
> devote any time toward preping, packaging, and
posting it, I thought I'd
> first see if any others are interested in this
tool (e.g. to build Python
> apps which use DX for graphics/visualization
under-the-hood).
>
>      For those that don't know, OpenDX
(formerly IBM Data Explorer) is an
> industrial-strength scientific/data
visualization package, now open-source:
>
>      http://www.research.ibm.com/dx/
>      http://www.opendx.org/
>
> --
> Randall Hopper
> aa8vb at yahoo.com
>
>



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


From mhammond at skippinet.com.au  Wed Jun 14 15:31:29 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:31:29 GMT
Subject: Problem assigning timevalue to Excel - COM
References: <8i7pp7$1vl$1@newstoo.ericsson.se>
Message-ID: <8i8mih$c6s$1@nnrp1.deja.com>

In article <8i7pp7$1vl$1 at newstoo.ericsson.se>,
  "Bror Johansson" <bror.johansson at esavionics.se> wrote:
> I haven't found out what type and/or format to use when assigning a
value to
> a cell in MS Excel having a predefined time format like this:
> 1997-04-03  12:15:00

The formatting of a cell is completely seperate from setting its
value.  So you need 2 steps - first you set the cell to the "Time"
object, then you ask Excel to format the time.

Generally, you can convert to and from the Python time module's
concepts of time, and the PythonCOM time object - eg, when converting
the tine object to an int, the int is compatible with the time module,
and can be used with any functions that take a tick count.

Ditto the other way - generally you convert the time module object to
the standard time tuple used, and pass this to the Time constructor (as
found in the pywintypes and pythoncom modules).

For a working example, check out win32com\test\testMSOffice.py - there
we set a time value, read it back, and compare them.

Mark.


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


From donn at u.washington.edu  Mon Jun 12 12:57:35 2000
From: donn at u.washington.edu (Donn Cave)
Date: 12 Jun 2000 16:57:35 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <8i34pv$2ruu$1@nntp6.u.washington.edu>

Quoth mkx at excite.com:
| In many Perl vs. "other scripting language" discussions, I have often
| seen Perlers state that those who dislike Perl's syntax and structure
| are often non-Unix types, and thus are not used to common Unix
| concepts. And, since Perl was born as a Unix admin tool, us "Windows
| types" will likely be lost.

Could be.  I'll go along with everyone else so far, UNIX competence
is no guarantee you'll get along with Perl, but it does have a lot
of sed, awk and shell heritage.

Python doesn't, but there are places where platform neutrality isn't
really feasible, and there it seems to me UNIX often prevails.  The
main thing I notice is the documentation, e.g., invoking the python
interpreter in the tutorial, but I really only use it on UNIX and
similar (BeOS), you may notice more functional issues.

	Donn Cave, donn at u.washington.edu


From cut_me_out at hotmail.com  Sat Jun  3 01:15:30 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Jun 2000 01:15:30 -0400
Subject: How to free a port that has been used by a python server?
Message-ID: <etdg0qvtkel.fsf@w20-575-109.mit.edu>

Hi.  I have some code like this:

######################################################################
class Server (SocketServer.ThreadingMixIn,
	      BaseHTTPServer.HTTPServer):
    pass
    
if __name__ == '__main__':

    port = 1729 # Called from interactive session, debug mode.

    proxy =  Server (('127.0.0.1', port), Handler)
    proxy.serve_forever ()
######################################################################
    
The problem is, when I run this more than once in interactive mode, I
get the error:

socket.error: (98, 'Address already in use')

Is there some way to free up port 1729 when proxy gets destroyed?  

I'm running RedHat Linux, but I have also seen this problem with
Solaris. 

Alex.


From david_ullrich at my-deja.com  Thu Jun  8 13:24:29 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 08 Jun 2000 17:24:29 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news> <8hlrkn$5t2$1@nnrp1.deja.com> <1ebvwau.af3wl21301xdcN%bparsia@email.unc.edu>
Message-ID: <8hokro$8gi$1@nnrp1.deja.com>

In article <1ebvwau.af3wl21301xdcN%bparsia at email.unc.edu>,
  bparsia at email.unc.edu (Bijan Parsia) wrote:
> <david_ullrich at my-deja.com> wrote:
> [snip]
> >
> >     I don't know anything about Ruby, but a language could certainly
> > be more fully OO than Python. For example in Python although
everything
> > is an object not every type is a class (you can't subclass list, for
> > example).
> [snip]
>
> Of course, this presumes that the more "class" you have the more fully
> "OOPY" you are. *Really* ools have *class*. Yuk yuk.
[...]

    I'm not qualified to argue about what's more OOP than what -
with any luck I'd decline to do so even if I were so qualified.
Was just commenting on the preceding 'I thought everything in
Python was an object and that Python was designed from the
ground up to be object oriented...how can you be more "fully"
OO?' - the facts that Python was [etc] do not imply that it's
impossible to be more fully OO than Python.

    Whatever that means.

DU

>
> --
> Bijan Parsia
> http://monkeyfist.com/
> ...among many things.
>


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


From mwh21 at cam.ac.uk  Sat Jun  3 05:26:38 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Jun 2000 10:26:38 +0100
Subject: overriding import
References: <IGVZ4.2412$r7.584924@news1.rdc2.tx.home.com>
Message-ID: <m37lc7yv1t.fsf@atrus.jesus.cam.ac.uk>

"Rainer Deyke" <root at rainerdeyke.com> writes:

> I am writing a C++ program with embedded python.  I want to redirect
> import commands to read the file from a different source, possibly
> not a true file at all but an entry in a zip file.  How would I do
> this?

I don't know if being embedded offers other ways of doing this, but in
Python code (which should work here), you'd probably want to use
Greg Stein's imputil.py:

   http://www.lyra.org/greg/python/

If you want to override *all* import unconditionally, you can assign
to __builtin__.__import__, but that requires a certain amount of
Python internals knowledge.

HTH,
Michael

-- 
  please realize that the Common  Lisp community is more than 40 
  years old.  collectively, the community has already been where 
  every clueless newbie  will be going for the next three years.  
  so relax, please.                     -- Erik Naggum, comp.lang.lisp


From cut_me_out at hotmail.com  Tue Jun  6 09:50:53 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 09:50:53 -0400
Subject: C extension using Numeric arrays -- do I really need all this error
References: <etdpupvdccz.fsf@w20-575-109.mit.edu> <393CE2F0.74B32153@sightreader.com>
Message-ID: <etdbt1e6hqa.fsf@w20-575-109.mit.edu>

> In general it is a good idea to be very prissy about error checking.
> In particular, it should be impossible to crash an extension module,
> or corrupt memory, no matter how hard you try from python.

Yes, I fully agree, but I'm really asking if there is some way to make
the Numeric extension handle some of the prissiness for me.

Spoiled-by-python,-perhaps-'ly yrs
Alex.


From fiona at sitegnome.com  Sun Jun  4 08:21:13 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 4 Jun 2000 12:21:13 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 4th, 2000
Message-ID: <20000604122113.27178.qmail@synop.com>

Hi Guys,

Sorry I haven't sent the summaries over the last couple of days.  I'm 
sure you'll all be happy to hear I had a great weekend :)

Cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
How can I setup Python on MacOS X?
Has anyone compiled Python for MacOS X?
http://www.faqts.com/knowledge-base/view.phtml/aid/3464
-------------------------------------------------------------
Fiona Czuczman
Joseph Block, James Felix Black, cliff crawford

You can download it from softrak.stepwise.com

Just search for Python and MacOS X

It builds right out of the box.  The only trap is if you build on  an 
HFS+ as opposed to UFS volume; then, you have to be careful about case 
(HFS+ is case-insensitive but preserving.)

Specifically, if you use HFS+ and want it to "just work" when you run
make, then you need to rename the Python directory to something else (I
called it Python.d), then change Makefile.in, configure.in, and
Python/Makefile.in to use the new directory name.  Then run autoconf,
then configure, etc.


-------------------------------------------------------------
How can I free a port that has been used by a python server?
http://www.faqts.com/knowledge-base/view.phtml/aid/3465
-------------------------------------------------------------
Fiona Czuczman
Johannes Stezenbach

Problem:

######################################################################
class Server (SocketServer.ThreadingMixIn,
              BaseHTTPServer.HTTPServer):
    pass
    
if __name__ == '__main__':

    port = 1729 # Called from interactive session, debug mode.

    proxy =  Server (('127.0.0.1', port), Handler)
    proxy.serve_forever ()
######################################################################
    
The problem is, when I run this more than once in interactive mode, I
get the error:

socket.error: (98, 'Address already in use')

Is there some way to free up port 1729 when proxy gets destroyed?

Solution:

Normally you should close the socket properly before exiting. You could 
try to catch KeyboardInterrupt or whatever signal you use to break out 
of serve_forever(), but it's problematic with multithreaded servers.

Workaround:
-----
class Server (SocketServer.ThreadingMixIn,
              BaseHTTPServer.HTTPServer):
    def server_bind(self):
        import socket
        self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 
1)
        BaseHTTPServer.HTTPServer.server_bind(self)
-----

Note that this doesn't free up the port, instead it allows more than  
one process to bind to it, under certain circumstances (bind() still
fails if there's already a socket actively listening on that port).
See socket(7) (Linux) for details.


-------------------------------------------------------------
Is it possible for Python to call commands, e.g. "find, perl scripts, nmap, and others"?
http://www.faqts.com/knowledge-base/view.phtml/aid/3469
-------------------------------------------------------------
Fiona Czuczman
Sean Blakey

There are many ways to call other programs from python.

First, and simplest, is os.system.  This simply fires off a sub-shell to
run the specified program, than returnsthe exit code of that program.
For example:
>>>import os
>>>os.system('wget ftp://ftp.python.org/pub/python/src/py152.tgz')
will download the python source distrobution.

If you want the output of a command (or want to write to a command on
stdin) you can use os.popen, which returns a python filehandle.  For
example, to get a listing of allfiles ending in .mp3 on your system, you
can do:
>>import os
>>import string
>>fd = os.popen('locate *.mp3')
>>filenames = string.split(fd.read, '\n')
You can also use popen to write to a process:
>>>import os
>>>fd = os.popen('/usr/sbin/sendmail nobody at localhost', 'w')
>>>fd.write('''Subject: test

this is a test.
''')
>>>fd.close()
Since you are working on a linux box, you probably also have access to
the popen2 module.  This allows you to use coprocesses (giving you
access to both stdin and stdout of a process).

The popen2 module is documented at
http://www.python.org/doc/current/lib/module-popen2.html

Note that os.popen works unreliably on Windows boxes, and the popen2
module is unix-specific.


-------------------------------------------------------------
I am writing a C++ program with embedded python.  How can I redirect import commands to read the file from a different source, possibly not a true file at all but an entry in a zip file?
http://www.faqts.com/knowledge-base/view.phtml/aid/3467
-------------------------------------------------------------
Fiona Czuczman
Michael Hudson

I don't know if being embedded offers other ways of doing this, but in
Python code (which should work here), you'd probably want to use Greg 
Stein's imputil.py:

   http://www.lyra.org/greg/python/

If you want to override *all* import unconditionally, you can assign
to __builtin__.__import__, but that requires a certain amount of Python 
internals knowledge.


-------------------------------------------------------------
How can I combine two BMP images so that one overlays the other?
http://www.faqts.com/knowledge-base/view.phtml/aid/3466
-------------------------------------------------------------
Fiona Czuczman
Robert Roy, Anders Eriksson

I think PIL might be what you need.

http://www.pythonware.com/products/pil/index.htm

This page of the tutorial talks about pasting images and transparency

http://www.pythonware.com/library/pil/handbook/intro01.htm

You may want to look up the function 'composite'.


-------------------------------------------------------------
Where can I find some kind of helping material (tutorials etc) about how to use XML and Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3468
-------------------------------------------------------------
Fiona Czuczman
Boudewijn Rempt, Harry George

Have a look here:

http://www.valdyas.org/python/index.html

dbobj - the repository is XML based
kura  - webserver that takes XML templates and spews out HTML, and an
        XML data import utility.
kxml  - a small PyKDE xml file browser (unfinished!)

A really simplistic reader/writer for Abiword.  Available at:
  http://www.seanet.com/~hgg9140/comp/index.html

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

Install PyXML, then try this.  Read the xml.dom.core for element names
and the basic functions.  If you are not familiar with xml itself, see
the XML Bible.  Try this routine on some known-good .xml file.

#!/usr/bin/env python
#/* helloxml.py 

"""
Usage: helloxml.py [options] filename
E.g.,  helloxml.py myfile
Options
-h,--help        this message
-v,--version     version

Revisions:
2000-02-23  Harry George   initial version
"""

import sys,os,getopt,time,re,string,exceptions

modname="helloxml"
__version__="0.1"

def debug(ftn,txt):
        sys.stdout.write(modname+'.'+ftn+':'+txt+'\n')
        sys.stdout.flush()

def fatal(ftn,txt):
        sys.stdout.write(modname+'.'+ftn+':FATAL:'+txt+'\n')
        sys.stdout.flush()
        sys.exit(1)

def usage():
        print __doc__

#====================================
from xml.dom import core, utils

def main():
        ftn = "main"
        pargs=[]    #positional args, default is empty
        opts,pargs=getopt.getopt(sys.argv[1:],'hv',['help','version'])
        for opt in opts:
                if opt[0]=='-h' or opt[0]=='--help':
                        usage()
                        sys.exit(0)
                elif opt[0]=='-v' or opt[0]=='--version':
                        print modname+": version="+__version__
                        sys.exit(0)

        #---get the DOM "doc"---
        if len(pargs)==0:
                        usage()
                        sys.exit(0)             
        filename = pargs[0]
        doc = utils.FileReader( filename ).document

        #---dump it out---
        doc.dump();

        #---convert DOM tree back to XML, and dump that
        xml = doc.toxml()
        print xml

        #---dump the doc manually---
        root=doc.get_documentElement()  
        mydump(root)

def mydump(node):
        ftn = "mydump"
        debug(ftn,"name=" +node.get_nodeName())
        #---dump attributes---
        attrs=node.get_attributes()
        keys=attrs.keys()
        for attr in keys:
                debug(ftn,"attr="+attr+" 
value="+attrs[attr].get_nodeValue())
        #---dump children---
        children=node.get_childNodes()
        for child in children:
        #--report only elements
                if child.get_nodeType()==core.ELEMENT: 
                        mydump(child)
                         
if __name__ == '__main__': main()






From dalke at acm.org  Tue Jun 20 13:52:40 2000
From: dalke at acm.org (Andrew Dalke)
Date: Tue, 20 Jun 2000 11:52:40 -0600
Subject: Roman Number Representation
References: <S9Q1YMA7B5T5Ew1U@jessikat.demon.co.uk> <394FA7BE.F9AE2509@home.com>
Message-ID: <8ioarl$6mp$1@slb7.atl.mindspring.net>

Ivan Van Laningham wrote:
>I'm sure if you search DejaNews, you can find [a discussion of Roman
>numerals in Python].


There was a discussion about 2 to 2.5 years ago in c.l.py.  However,
according to this Salon.com article:
  http://salon.com/tech/log/2000/06/20/deja/index.html

> at the end of May, Deja.com moved to new servers and took down the
> Usenet archives that pre-dated May 1999.

                    Andrew
                    dalke at acm.org





From eq3pvl at eq.uc.pt  Mon Jun 12 11:47:14 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Mon, 12 Jun 2000 16:47:14 +0100
Subject: IDLE  can be rewrited using wxPython?
References: <8i25jd$1u8g$1@news.cz.js.cn>
Message-ID: <39450602.87388B98@eq.uc.pt>

mrfx wrote:

> The current IDLE uses TCL/TK as GUI widget.
> IDLE  can be rewrited using wxPython as GUI widget?

Sure. But not an easy task, replacing a GUI toolkit is always
a pain. Also it seems that the text widget is better in tk than
in wx.

pvl



From rumjuggler at cryptarchy.org  Tue Jun  6 23:49:31 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Wed, 07 Jun 2000 03:49:31 GMT
Subject: abuse of lambda question
Message-ID: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com>

I'm attempting to abuse Python's lambdas, and I've come across a
problem I can't figure out.  Well, the problem is actually that I'm
attempting to abuse lambdas, but aside from that I don't know why the
abuse is failing.  Here's what I've got:

def __everyother(): #factory to return lambdas.
	return lambda l, evens, odds, app=lambda a,e: lambda
a=a,e=e:a.append(e): Compose(apply(map, (lambda e, evens=evens,
odds=odds, l=l,app=app: If(l.index(e)%2==0, app(evens,e), app(odds,e))
,l)), Return(evens, odds))

The functions it calls are:

def Return(*args):
	return lambda args=args: args

def Compose(*args):
	return lambda fs=args: For(len(fs), lambda c, fs=fs:
If(lambda c=c,fs=fs: c==len(fs)-1, Return(fs[-1]()), fs[c]))()

def For(count, fct, *args, **kwargs):
	for i in range(count):
		fct(i,*args,**kwargs)

I can do this:
>>> import lambdautils
>>> a = lambdautils.__everyother()
>>> evens, odds = [], []
>>> a(range(20),evens, odds)
<function <lambda> at 1510c00>
>>> lam = a(range(20),evens, odds)
>>> evens
[0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18]
>>> odds
[1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19]
>>> lam()
Traceback (innermost last):
  File "<pyshell#14>", line 1, in ?
    lam()
  File "C:\PYTHON16\lambdautils.py", line 16, in <lambda>
    return lambda fs=args:\
  File "C:\PYTHON16\lambdautils.py", line 4, in For
    fct(i,*args,**kwargs)
  File "C:\PYTHON16\lambdautils.py", line 17, in <lambda>
    For(len(fs), lambda c, fs=fs: \
  File "C:\PYTHON16\lambdautils.py", line 44, in If
    return f()
TypeError: call of non-function (type list)

Can't quite figure out where the error occurs.

-- 
Barnabas T. Rumjuggler

Policeman: So sir, did the suspect have four arms, each one holding a
sword, and was she wearing a girdle adorned with severed hands, her
whole body covered in the blood of past victims?
Mr Flynn:  Well, it was overcast and everything, so it was hard to say
for certain, you know?
 -- John Flynn on apihna


From adams_s at lab.eng.usyd.edu.au  Mon Jun 12 05:59:17 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Mon, 12 Jun 2000 19:59:17 +1000
Subject: string.atoi() doesn't do negative base16 - why??
Message-ID: <8i2c9l$b0r$1@news1.wire.net.au>

Hi,

my quick hack for going back and forth between binary and base 10, 16 is
backfiring a bit..

I can form the values as a string, BUT, string.atoi() can't convert negative
base hexadecimal strings to integers.

does that make sense?
e.g
>>> import string
>>> hex(-20)
'0xffffffec'
>>> string.atoi('0xffffffec',16)
Traceback (innermost last):
  File "<pyshell#28>", line 1, in ?
    string.atoi('0xffffffec',16)
ValueError: atoi() literal too large: 0xffffffec

why is this?

because after looking at it for too long - I found exec() would do the trick
e.g.
>>> myHexString='0xffffffec'
>>> exec('l='+myHexString)

why does this work, but the conversion by string.atoi() not?? curiouser and
curiouser

confused,

Steven








From cjc26 at nospam.cornell.edu  Sat Jun  3 20:31:27 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sun, 04 Jun 2000 00:31:27 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <3937D345.CC0935D5@uab.edu> <39385DC2.BC0B4853@stud.ntnu.no> <slrn8jj4ot.om.thor@localhost.localdomain> <20000603234710.Q469@xs4all.nl>
Message-ID: <slrn8jj8ub.6eb.cjc26@localhost.mindriot.net>

* Thomas Wouters <thomas at xs4all.net> menulis:
| 
| > Seguro que si, quien sabe, a lo mejor mi perro sabe el holandes. :P
| 
| Eso es divertido, porque mi pescado preferido habla espa?ol :)

Maaf, tapi saya bisa berbahasa hanya dua bahasa: Bahasa Inggris dan
Bahasa Indonesia.  Dan saya tidak punya anjing atau akan yang bisa
berbicara bahasa lagi.  (Walaupun teman rumah saya bisa berbahasa
Jerman, dan dia punya banyak hewan..:)

and-Indonesia-used-to-be-a-Dutch-colony-ly y'rs,


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From moshez at math.huji.ac.il  Fri Jun  9 16:43:03 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 23:43:03 +0300 (IDT)
Subject: readlines() question
In-Reply-To: <8457258D741DD411BD3D0050DA62365907A1FA@huina.oceanic.com>
Message-ID: <Pine.GSO.4.10.10006092339580.29560-100000@sundial>

On Fri, 9 Jun 2000, Doug Stanfield wrote:

<a very good explanation of file objects, concluding with an example>
> Your example would then have to be:
> 
> my_file = open('file')
> contents = my_file.readlines()
> 
> # Use the file object in any other way you need here.
> 
> my_file.close()

Ummmmm...I'm known as a professional paranoid in many places, so I usually
code like:

my_file = open("blah")
try:
	...my_file.read, my_file.readlines, etc.
finally:
	my_file.close()

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From rsteelandt at bucodi.com  Tue Jun 20 10:56:15 2000
From: rsteelandt at bucodi.com (BuCoDI-Rony Steelandt)
Date: Tue, 20 Jun 2000 16:56:15 +0200
Subject: Tkinter and setting focus
Message-ID: <394F860E.EF6042D6@bucodi.com>

I have created a class wich expands the standard entry widget.
In that class i did a override of the return key to do some validation
of my entry

self.entry.bind(<return>,self.validate)

The validation routine works fine, but now comes my problem

I want at the end of the validation, if validation is ok set the focus
to the next entry widget in the form.
I searched documentation but didn't find anything on this.
The problem is since my class is a general class, i don't know in the
class wich widget is the next to put the
focus on.
Here is some code to show what i'm doing:
class numentry:
 def
__init__(self,parent,rw,clmn,wdt,relf=SUNKEN,mask='$',min=0,max=999999999):

  self.w = wdt
  self.m = mask
  self.min = min
  self.max = max
  self.field = Entry(parent,relief=SUNKEN,width=wdt)
  self.field.grid(row=rw,column=clmn)
  self.field.bind("<Return>",self.validate)
  self.returnval=''
  self.ind = 0
  self.pointindex = 0


 def validate(self,event):
    # do some validation
===> here i want to set the focus to the next widget


#main
frame = Frame(root,width=100,height=100)
# frame.bind("<Up>",callback)
frame.pack()
f=numentry(frame,1,1,10,SUNKEN,'0000.00',5)
f2=numentry(frame,2,1,10,SUNKEN,'0000.00')
root.mainloop()


I'm realy sticked on that ?????
Anyone has a idea how to do that

Rony Steelandt
--
======================================================
Bureau de Conseil et de D?veloppement Informatique
2, Rue Henri Avril
22480 Saint - Nicolas du Pelem

T?l : +33.2.96.29.79.08
Fax : +33.2.96.29.79.09

e-mail : bucodi at worldnet.fr
======================================================




From db3l at fitlinxx.com  Wed Jun 14 22:07:13 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 22:07:13 -0400
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu> <8F51D9DB3gmcmhypernetcom@199.171.54.154> <jl7lbtm60s.fsf@endor.bbb.caltech.edu> <8F52DC18Agmcmhypernetcom@199.171.54.194> <jl1z1zaffg.fsf@endor.bbb.caltech.edu>
Message-ID: <uaegnlmsu.fsf@ctwd0143.fitlinxx.com>

Michael Vanier <mvanier at endor.bbb.caltech.edu> writes:

> def callback(event):
>     global thread_on
>     key = event.char
> 
>     if key == "s":
>         if not thread_on:
>             stop_flag = 0
              ^^^^^^^^^^^^^
>             thread_on = 1
>             thread.start_new_thread(run, ())
>         else:
>             print "halting thread"
>             stop_flag = 1
              ^^^^^^^^^^^^^
>             thread_on = 0
>     elif key == "q":
>         raise SystemExit

You haven't stop_flag as global ("global stop_flag"), so the
highlighted operations have created a local (to the callback function)
variable, and thus are not changing the global variable, which the
thread function is correctly seeing.

This is definitely a somewhat subtle issue with Python scoping -
scopes are used dynamically but determined statically.  See near the
end of section 9.2 of the tutorial for some info.  See also sections
4.36 and 4.57 of the FAQ.

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


From snowcat98 at my-deja.com  Sun Jun  4 23:45:23 2000
From: snowcat98 at my-deja.com (snowcat98 at my-deja.com)
Date: Mon, 05 Jun 2000 03:45:23 GMT
Subject: Pointrel Data Repository System for Python
References: <392AA43E.B3014C3B@kurtz-fernhout.com> <392D4BE1.1A9AB90C@uab.edu> <392DF7D5.58F632DB@kurtz-fernhout.com>
Message-ID: <8hf7ok$kfn$1@nnrp2.deja.com>

TheBrain is not ideal. But its integration with
desktop & web is the best till now. The better
news is it has a SDK. Thus, more advanced
concept, say, arbitrary relationships, can be
implemented with a easy-to-use user interface.

In article <392DF7D5.58F632DB at kurtz-fernhout.com>,
  Paul Fernhout <pdfernhout at kurtz-fernhout.com>
wrote:
> Shae-
>
> Thanks for the great comments!
>
> Shae Erisson wrote:
> >
> > Paul Fernhout wrote:
> > >
> > > This is the very first (rough!) version of
the Pointrel Data Repository
> > > System.
> >
> > This sounds like a three-way linked list with
indexes. Am I close?
> > The only difference I see is that a linked
list is generally thought of
> > as a linear construct whereas this looks more
like spaghetti.
>
> Very insightful and concise, with one minor
quibble that "list" might
> more accurately be "lists".
>
> Yes, I guess that might be in a way a
technically accurate description,
> depending on how broadly one
interprets "lists". Obviously Lisp heavily
> uses lists, and it can do a lot!
>
> Like you said, "list" would to me commonly mean
something that was
> intended to be primarily sequential. The
Pointrel system allows one to
> support arbitrary structures. Obviously Lisp
supports structures too,
> made of list elements, and in practice Lisp
lists made of two-element
> pairs are often used to build trees to define
structures.
>
> Also, "list" implies to me perhaps one list.
When using the Pointrel
> system, you can have multiple structures
defined by triads which are not
> connected in any way, but which you can
navigate through if you know the
> location of an item in the structure. In
practice you might anchor these
> to a common root with some symbolic names to
find what you stored again
> using a text key; so in that way most data
might be seen as connected,
> and in that sense a "list" if you allow "list"
to include something that
> is more like a tree or a web.
>
> Also, "indexes" is probably appropriate, but
doesn't quite catch the
> flavor of being able to easily find all the
connections to any triad in
> the system and the implications in terms of
transparency of the system.
>
> This implementation of the Pointrel system
supports creating and
> traversing webs of nodes (called triads). The
nodes can have with from
> zero up to three ordered (A,B,C) links each.
Each node can have an
> arbitrary binary string associated with it. One
can always easily find
> all the nodes that have linked to any specific
node from any particular
> ordered slot (A,B,C).
>
> So, could this be called more or less
supporting "three-way linked
> list(s) with indexes"? I guess so.
>
> Could it be called "spaghetti"? I guess so too
(depending on how you
> link up the triads in your repository)!
>
> > How is this different from something like
Gadfly[1]?
> > [1] http://www.chordate.com/gadfly.html
>
> Here's an analogy. Imagine having a hundred and
twenty-two thousand
> miles of string. :-)
>
> Having it all in one piece is like having a
relational database
> management system (RDBMS), like Gadfly. You can
cut that string into
> many strings of whatever length you want, and
then use those lengths of
> string for tying up all sorts of things, using
as a leash for your pet
> cat Eric (if you've got the proper license from
the man in the van), or
> representing weights and measures, and so on.
Everyone knows the value
> of a long continuous piece of string. It's a
bit dangerous to keep
> scissors around for cutting the string of
course, but the latest
> scissors the nice people give out :-) have
blunt edges so it's not too
> bad. It is a pain to keep hauling around a
mountain of string though,
> but you do it because once you cut off a
length, it's hard to reuse it,
> and you want to get the most efficiency out of
all that precious string.
>
> Now imagine chopping up that string into three-
inch lengths.
> That's the Pointrel Data Repository System. :-)
>
> Now, before one calls this "bad planning",
consider the following:
>
> Fold over the string and tie a knot in the
middle.
>
> This wil produce something that looks vaguely
like this asciii art:
>
>      ------
> ====K
>      ------
>
> These super water absorbent
triadal "stringettes" can now be easily tied
> together to make all sorts of interesting webs
without requiring any
> additional cutting. So, the nice people don't
have to worry about
> providing dangerous sharp devices. You never
has to make a big
> commitment to cutting a specific length of
string, and you can easily
> carry around just a small pile of stringettes
and get more if you run
> out. You never have to worry about keeping
track of unusual lengths of
> string cut to useless lengths due to bad
planning -- since all your
> stringettes are a uniform size, which makes
life a lot easier all
> around! :-)
>
> What's more, with a microprint pen, one can
write on the knot, thus
> saving the trouble of encoding text via more
stringettes tied to the
> knot.
>
> Further, if you assume a non-euclidian infinite-
dimentional space where
> all string knots are at most apparently less
than 1.5 inches apart, you
> can tie the end of any stringette to the knot
of any other stringette,
> allowing your stringettes to represent
arbitrary structures of
> information.
>
> Stringettes for everyone!
>
> And now for something completely different....
>
> Gadfly is a great system. I had used it briefly
as part of learning to
> use SQL under Zope.
>
> As I understand it, Gadfly is basically a
relational database
> implemented in Python with support for a large
subset of SQL. It loads
> tables into memory before operating on them,
which has various
> implications.
>
> So, the question is perhaps more, "how is the
Pointrel system different
> from a typical relational database"? William
Kent discusses issues like
> this at length in his book "Data & Reality".
> It is in large part an issue of conceptual
emphasis.
>
> Let me try to explain this difference by giving
my take on what types of
> activities are assumed to
be "common", "occasional", or "rare" in the
> two types of systems.
>
> Relational databases are designed for storing
large numbers of similar
> "relations" (table rows). Changing the types of
things stored in a
> relational database is a rare
event. "Introspection" of database
> elements referring to other database elements
is also rare. Changing a
> value in a relation is common event, and adding
new data is an
> occasional event [might be common in some
cases]. Processes that need to
> analyze a significant portion of the data in
the database are rare.
> Deleting data is done occasionally.
>
> The Pointrel Data Repository System is designed
for storing large
> numbers of ad-hoc relations, where changing the
types of things stored
> in the database is a common event (but you
still want to get at legacy
> data), "introspection" of database elements
referring to other elements
> is also common, and changing a value in a
relation is an occasional
> event, and adding new data is a common event.
Processes that need to
> analyze a significant portion of the data in
the database are common.
> Deleting data is done rarely.
>
> Summary chart:
>
> Activity            RDBMS       Pointrel System
> ===============================================
> Changing Types      Rare        Common
> Introspection       Rare        Common
> Changing Values     Common      Occasional
> Adding data         Occasional  Common
> Analyze Lots        Rare        Common
> Deletion            Occasional  Rare
>
> Some other differences:
> * The Pointrel system puts all data in one file
(well, really two);
> typically a RDBMS use one file per table.
> * The Pointrel system never deletes; a typical
RDBMS allows deletion
> (which may require packing). [Deletion can be
approximated in the
> Pointrel System, but only by marking triads
deleted and checking for
> that mark at the application level, creating an
alternate structure
> including only the non-deleted data and working
exclusively with the
> alternate, or cloning the repository without
the deleted data].
> * The Pointrel system has an absolute minimum
of base concepts (triad
> and string); a RDBMS typically has many base
concepts to learn (like
> table restructuring, allowable data types,
indexing methods, etc.).
> Succesfully using the Pointrel system does
require learning many new
> concepts, but those are more at the application
level (i.e. how to
> represent and retrieve application specific
data in useful ways.)
> * The Pointrel system uses triads to allow
arbitrary higher relational
> layers to be built; a RDBMS typically
implements relations as arrays of
> values in files.
> * The Pointrel database makes it easy and fast
to dive recursively into
> relations; a RDBMS with SQL typically is not
designed to support queries
> where one is taking a value from a row
(relation) and using that value
> to drill down into rows (relations) of the same
table (type). [You can
> do it, but I not easily.]
> * The Pointrel system leans towards supporting
information retrieval
> where the information retrieved may change what
additional information
> is retrieved and what it means (by encouraging
fine grained traversal);
> a RDBMs with SQL leans towards supporting
information retrieval where
> the major variable is how many records might
come back and you know what
> the structure of the query result will be (by
what SQL's syntax makes
> convenient).
> * The Pointrel system could easily (not
implemented yet) support doing
> reports on a consistent view of the database by
ignoring all triads
> after a certain location; a RDBMS has to work
hard to provide a
> consistent view (if it can) given all the
various tables and such which
> may be undergoing changes while the report is
running.
>
> The core Python source file for the Pointrel
System is only 20K, but I
> hope you'll agree it can do quite a bit for
such a small size and thus
> has a high usefulness per byte ratio. Of
course, it can't quite match
> the ratio of the Forth "idea", which can fit a
whole
> OS/Compiler/Linker/Editor in about 1K.
>
> I'm sure Pointrel in its current initial state
will suffer from the same
> tradeoffs you get when you use a 1K Forth,
compared to a 1MB Forth with
> lots of bells and whistles. Pointrel will be
far more useful when it has
> lots of Python scripts for importing and
exporting data, for searching
> the data, and for doing other interesting
things (versioning, supporting
> a reliable file system, etc.). But the fact
that you can go anywhere
> with a 1K Forth says there is an elegant idea
there. I hope that someday
> one might say that about the Pointrel Data
Repository System vs. a
> typical RDBMS.
>
> It may always be possible to say a RDBMS does
things you want to do
> better than the Pointrel Data Repository
System, just like I could say
> Python does many things I want to do more
easily than Forth. If I was
> fitting an interpreter into 8K, I'd probably
pick Forth over Python. If
> I was fitting an intepreter into 1MB, I'd
probably pick Python. So, my
> hope is to shape the Pointrel system so that it
fits some niche very
> well. Right now I think that niche is as a
flexible data repository.
> Maybe some other niche would be better?
>
> In any case, I should look a lot more at
Gadfly. It might be possible to
> merge the Pointrel system with the Gadfly SQL
parser and related search
> code, to make a version of Gadfly that didn't
have to load tables in
> memory. This version would have many of the
performance tradeoffs
> Pointrel has (flexible but much slower for
routine things), but it might
> be interesting...
>
> > The idea somewhat reminds me of natrificial's
TheBrain software. I guess
> > arbitrary relationships is one of its strong
points?
>
> I looked at a few pages on the site
http://www.thebrain.com but it
> wasn't immediately obvious to me how the
implemented what they have.
> Their intent does sound similar: "TheBrain
gives people the ability to
> organize disparate pieces of information into
one meaningful structure
> that conveys valuable relationships in an
easily understandable
> display."
>
> Yes, handling arbitrary relationships is
(hopefully) a strong point of
> the Pointrel Data Repository System. I have
implemented versions of the
> Pointrel system focused on n-ary (tuple)
relations (not triads), but I
> feel the triads provide more flexibility,
simplicity, and speed, while
> not loosing that much of the power arbitrary
length tuples provide
> (since you can build more complex relations
from triads).
>
> There are many systems people have developed to
store arbitrary
> relationships.
> Many have been done in the past by AI
developers (typically with Lisp).
>
> Things really clicked for me though when I was
sitting in a tutorial
> about Lisp (around 1982) and I realized "Aha!
Lisp culturally is about
> lists and structures; what I want to do is more
about pointers and
> relationships (hence, "Pointrel".)
>
> > No disrespect meant, I'm just trying to
understand more about Pointrel.
>
> None taken. Thanks again for the great
questions.
>
> -Paul Fernhout
> Kurtz-Fernhout Software
>
==================================================
=======
> Developers of custom software and educational
simulations
> Creators of the Garden with Insight(TM) garden
simulator
> http://www.kurtz-fernhout.com
>



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


From ge at nowhere.none  Fri Jun 30 11:38:02 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 30 Jun 2000 15:38:02 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com>     <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com>     <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com>     <geT65.2262$iN5.511399@ptah.visi.com> <m2lmznzk2t.fsf@mycroft.actrix.gen.nz> <slrn8lp673.u8u.thomas@centurion.xs4all.nl> <8jidt1$2k5s$1@noao.edu>
Message-ID: <u9375.2324$iN5.533105@ptah.visi.com>

In article <8jidt1$2k5s$1 at noao.edu>, Richard Wolff wrote:
>thomas at xs4all.nl (Thomas Wouters) writes:
>
>>On 30 Jun 2000 19:47:38 +1200, Paul Foley <see at below> wrote:
>>>On Fri, 30 Jun 2000 02:03:56 GMT, Grant Edwards wrote:
>
>>>> C predates the PDP-11 so we'll chalk it up to convergent
>>>> evolution.
>
>>>Oh no it doesn't.
>
>>Oh yes it does ;) The first bits of C, including, for instance, the
>>auto-increment operators, were designed and used on the PDP-7, not the
>>PDP-11.
>
>>> C was written for the Unix port to PDP-11.
>
>>More like 'during' than 'for'. UNIX and C both, together, evolved on for
>>quite a number of years (witness the new C & UNIX 'standards')
>
>"C was originally written for the PDP-11 under UNIX ... "
>  The C Programming Language, by D. M. Ritchie, S. C. Johnson, M. E. Lesk,
>  and B. W. Kernighan, Bell System Technical Journal, Vol 57, #6, 1978, Pg 1991

OK, weren't most of the C operators (which match so well PDP-11
addressing modes) first present in C's ancestors that did
predate the PDP-11?

-- 
Grant Edwards                   grante             Yow!  ... My pants just
                                  at               went on a wild rampage
                               visi.com            through a Long Island
                                                   Bowling Alley!!


From johann at physics.berkeley.edu  Mon Jun 12 12:21:21 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 12 Jun 2000 09:21:21 -0700
Subject: lambda and if
References: <8i2c9l$b0r$1@news1.wire.net.au> <0006122328441W.01005@localhost.localdomain> <07d501bfd461$ced14280$1906a8c0@fc.fenx.com> <0006122345501X.01005@localhost.localdomain>
Message-ID: <mtk8fuooou.fsf@astron.berkeley.edu>

matt  writes:

> sure would ... doh! 

Oh, don't do that.  Please?  Just out of respect for anyone who has to
maintain your code.

Functions are easy to define in python, and you don't have to do it
all in one line.  I'd recommend:

def make_name(x):
  suffix = ""
  if x[1] in ("big", "little", "whatever"):
    suffix = x[1]
  name = "%s_%s_%s.txt" % (x[2], x[0], suffix)
  return name
filenames = map(make_name, items)

And, yes, you can just plug this inside any function.

It's longer, but it's both more extensible and easier to read.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From heath7952 at my-deja.com  Fri Jun 30 12:22:27 2000
From: heath7952 at my-deja.com (heath7952 at my-deja.com)
Date: Fri, 30 Jun 2000 16:22:27 GMT
Subject: Embeded in GUI,Python output ?
Message-ID: <8jihfg$et5$1@nnrp1.deja.com>

I'm attempting to embed Python inside a GUI. Python will serve as a
command interpreter. I can type commands in the GUI command window and
they are being executed correctly, I'm using PyRun_String. The only
thing I need now is the output from the commands , so I can post them in
the GUI command window.

i.e. a = 10
     print a

I'd like to be able to get at the output string "10" which should be
returned, and display it.

I've read a a lot of Python embedding documentation and can't quite
figure it out.

Any help would be greatly appreciated. Thanx




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


From bhat at cs.mu.OZ.AU  Wed Jun 14 22:46:55 2000
From: bhat at cs.mu.OZ.AU (Tim Bell)
Date: 15 Jun 2000 12:46:55 +1000
Subject: Why does time.gmtime() use Monday=0?
Message-ID: <8i9g2v$ecj$1@vife.cs.mu.OZ.AU>

Hi,

Can anyone tell me why time.gmtime() (and time.localtime()) use the
convention that Monday=0, rather than Sunday=0?  The Sunday=0 convention
is used by C's gmtime(3), as well as both C and Python's strftime().
Oh, and Python's strptime() uses Monday=0 also, but C's uses Sunday=0.

Here's the relevant code from tmtotuple() in python-1.5.2/Modules/timemodule.c,
line 262:

	(p->tm_wday + 6) % 7, /* Want Monday == 0 */

Why does Python go out of its way to be inconsistant, both with itself
and with C?  (Or have I missed something here?)

Any enlightenment would be appreciated.

Tim.
-- 
Tim Bell - bhat at cs.mu.oz.au - Dept of Comp Sci & SE - Uni of Melbourne, Aust.


From bparsia at email.unc.edu  Fri Jun 16 23:30:09 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Fri, 16 Jun 2000 23:30:09 -0400
Subject: Learning curves (was: Python so far)
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu> <39498084.972E36D2@san.rr.com> <1ecb5va.1nch8nwz2t5siN%bparsia@email.unc.edu> <bhahn-ya02408000R1606000950220001@news.corp.hp.com> <Fw9E6w.JKC@world.std.com>
Message-ID: <1ecbw3m.1qe5j2s12fuueN%bparsia@email.unc.edu>

Will Ware <wware at world.std.com> wrote:

> Brendan Hahn (bhahn at spam-spam.g0-away.com) wrote:
> > [learning curve axes]
> > X is time elapsed, Y is knowledge acquired.  So the common usage of "steep"
> > with regard to learning curves is in fact exactly backwards.  Too late now,
> > though.
> 
> There is a pretty clear analogy to climbing a hill (or a vertical cliff
> face). When climbing a hill the Y axis represents energy or effort.
> Maybe X would make sense as knowledge acquired; then a shallow learning
> curve means you can gain a lot of knowledge with little exhaustion. It's
> not clear that time belongs in there at all (except insofar as your
> effort-per-time has some upper bound).

Yes, this was my point. If I understand things correct, the term
"learning curve", as it was coined by, oh, *psychologists* probably, is
as Brendan put it. Your interpretation is--albeit clever--post facto :)


-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From dsavitsk at e-coli.net  Mon Jun  5 13:27:01 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 05 Jun 2000 17:27:01 GMT
Subject: Pausing Loop
Message-ID: <FpR_4.6466$gr4.252142@news1.rdc1.il.home.com>

is there a way to conditionally pause a loop until some event (button press)
takes place?  opening a modal window is not my first choice, but i don't
know another way of doing it.
(this needs to be using tkinter on win32 if that makes a difference)
thanks again
doug




From pauldubois at home.com  Mon Jun 26 12:20:27 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Mon, 26 Jun 2000 16:20:27 GMT
Subject: Array Memory Allocation Size
References: <39540628.B284C15F@bioeng.ucsd.edu>
Message-ID: <fpL55.4836$i5.75596@news1.frmt1.sfba.home.com>

print len(ravel(x))*x.itemsize()

Be sure to also check out the new spacesaver features.

Curtis Jensen <cjensen at bioeng.ucsd.edu> wrote in message
news:39540628.B284C15F at bioeng.ucsd.edu...
> I'm using NumPy to allocate heaps of arrays.  Is there a command to find
> out how much space each array is using in memory?
>
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451




From rwolff at noao.edu  Fri Jun 30 11:21:05 2000
From: rwolff at noao.edu (Richard Wolff)
Date: 30 Jun 2000 15:21:05 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com>
     <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com>
     <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com>
     <geT65.2262$iN5.511399@ptah.visi.com> <m2lmznzk2t.fsf@mycroft.actrix.gen.nz> <slrn8lp673.u8u.thomas@centurion.xs4all.nl>
Message-ID: <8jidt1$2k5s$1@noao.edu>

thomas at xs4all.nl (Thomas Wouters) writes:

>On 30 Jun 2000 19:47:38 +1200, Paul Foley <see at below> wrote:
>>On Fri, 30 Jun 2000 02:03:56 GMT, Grant Edwards wrote:

>>> C predates the PDP-11 so we'll chalk it up to convergent
>>> evolution.

>>Oh no it doesn't.

>Oh yes it does ;) The first bits of C, including, for instance, the
>auto-increment operators, were designed and used on the PDP-7, not the
>PDP-11.

>> C was written for the Unix port to PDP-11.

>More like 'during' than 'for'. UNIX and C both, together, evolved on for
>quite a number of years (witness the new C & UNIX 'standards')

"C was originally written for the PDP-11 under UNIX ... "
  The C Programming Language, by D. M. Ritchie, S. C. Johnson, M. E. Lesk,
  and B. W. Kernighan, Bell System Technical Journal, Vol 57, #6, 1978, Pg 1991

(Interpretation is left to the reader.)
-- 
Richard Wolff, National Optical Astronomy Observatories, Tucson, AZ
Internet: rwolff at noao.edu         Phonenet:     +1 520 318 8392


From bill at replywillfail.com  Wed Jun 28 10:49:46 2000
From: bill at replywillfail.com (=?Windows-1252?Q?Bill_de_h=D3ra?=)
Date: Wed, 28 Jun 2000 15:49:46 +0100
Subject: more python evangelising...
References: <200006271406.AAA06218@mbuna.arbhome.com.au> <D9465.71358$7o1.1704294@news2.rdc1.on.home.com>
Message-ID: <8jd3af$3cl$1@taliesin2.netcom.net.uk>

: But you're falling into the old trap of thinking that one language is best
: for everything.

Also falling into the trap that the best languages win out.

-Bill de h?ra




From dgrisby at uk.research.att.com  Tue Jun 20 05:42:02 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 20 Jun 2000 09:42:02 GMT
Subject: simple thread problem
References: <394E2F9C.6BCD96B5@iue.tuwien.ac.at> <8ilbfp$6a8$1@nntp9.atl.mindspring.net> <newscache$2quewf$nzd$1@jenpc07.jennic.com>
Message-ID: <8ine9a$ju0$1@pegasus.csx.cam.ac.uk>

In article <newscache$2quewf$nzd$1 at jenpc07.jennic.com>,
 Robert Cragie <rcc at nospamthanks_jennic.com> wrote:

>The mysterious message you get is when 'threading' encounters threads
>created in 'thread', so it sounds to me that you are mixing 'threading' and
>'thread' objects - are you using 'threading.Semaphore' or such like? As Aahz
>says, try to do it all using 'threading' objects.
>
>BTW, it's not actually a problem as such - it will probably still work.
>'threading' calls these 'Dummy threads' - look at threading.py for more
>info.

The difficulty is that the threading module keeps a dictionary mapping
thread ids to threading.Thread objects. When it encounters a thread id
which doesn't exist in the dictionary, it creates a DummyThread object
and puts it in the dictionary. The problem is that there is now a
reference to the DummyThread in the dictionary, but no way of getting
rid of it again. If you regularly create and delete threads, these
DummyThread objects accumulate.

This isn't really a problem if you are writing pure Python code, since
you can easily make sure that you create all the threads you need with
the threading module. It _is_ a problem if you need to interface with
some C code which creates its own threads. In that case, you have to
mess with the dictionary in the threading module to remove entries
from it. It would be nice if threading had an API for adding and
removing thread ids from its dictionary.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From jlj at cfdrc.com  Thu Jun 29 17:12:40 2000
From: jlj at cfdrc.com (Lyle Johnson)
Date: Thu, 29 Jun 2000 16:12:40 -0500
Subject: simple question ... exec in W32?
References: <Nxu65.20928$Ip.511269@cac1.rdr.news.psi.ca>
Message-ID: <395BBBC8.7BCA2DCC@cfdrc.com>

Depending on the complexity of what you want to do with the application
once it's running, you might look into Python's COM interface
capabilities. There is a nice introduction to this in "Learning Python"
and of course a much more detailed account in "Win32 Programming With
Python".

Brian Alexander wrote:
> 
> Hello;
> 
> How to I start a W32 application, like Word etc, from a Python script? I'm
> actually trying to rewrite a backup script, which, additionally, does a lot
> of non-backup processing overnight.
> 
> Many thanks,
> 
> Brian.


From gmcm at hypernet.com  Thu Jun  1 14:54:17 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 01 Jun 2000 18:54:17 GMT
Subject: str(None) == '' ?
References: <8h5rpb$fl3$14$1@news.t-online.com>
Message-ID: <3936b159$0$9534@wodc7nh0.news.uu.net>

Eduard Hiti <eduard.hiti at t-online.de> wrote:

>I would like to suggest that
>
>    str(None)
>
>should NOT return the string 'None', since most of the time, it 
complicates
>things.
>
>Then code like
>
>    value = str(get_stuff())
>    if value:
>        # Do something to value
>
>would not stumble over the occasional None.

I think the complication is really in what get_stuff() returns (or in 
what you *want* it to return). Unless badly coded, get_stuff() 
probably means something different when it returns None rather 
than ''.

It's not really str's job to hide that difference. If you want to do so, 
then you could use:

 value = str(get_stuff() or '')

>
>At least one of the string converting functions repr|str should 
return a
>empty string on None.

Definitely not repr, since that would voilate eval(repr(x)) == x.

>Is there a reason for str(None) == 'None' that I'm missing?

For those situations where None has a contextual meaning, and you 
don't want conversion to a string to hide that.

- Gordon



From darrell at dorb.com  Wed Jun 28 20:13:53 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 28 Jun 2000 17:13:53 -0700
Subject: file/re search and sub problem
References: <8jdl4t$r58$1@nnrp1.deja.com>
Message-ID: <0b2401bfe15e$e8cd0930$6401a8c0@home>

Here's two solutions.

>>> s='11ebitda</td>'
>>> re.sub('(?i)[^A-Z]ebitda[^A-Z]','EBITDA', s)
'1EBITDA/td>'
>>> re.sub('(?i)(?![^A-Z])ebitda(?=[^A-Z])','EBITDA', s)
'11EBITDA</td>'
>>> re.sub('(?i)([^A-Z])ebitda([^A-Z])','\g<1>EBITDA\g<2>', s)
'11EBITDA</td>'

----- Original Message ----- 
From: <tiddlerdeja at my-deja.com>
> 
> I have a re:
> wilf_re        = re.compile('[^A-Z]ebitda[^A-Z]', re.IGNORECASE)
> 
> And I do a search and replace
> if wilf_re.search(file):
> 
>      # perform the substitution
>      new_file = wilf_re.sub(" EBITDA", file)
> 
> This isn't really what I want. If I have
>  ebitda</td>
> It replaces it with
>  EBITDA/td>
> 
> 
> Can someone tell me how I go about getting:
>  EBITDA</td>





From nickgeo at merle.acns.nwu.edu  Sun Jun 18 13:45:02 2000
From: nickgeo at merle.acns.nwu.edu (Nicholas Geovanis)
Date: Sun, 18 Jun 2000 12:45:02 -0500
Subject: Rude question about BeOpen.com
In-Reply-To: <20000615010758.D26436@xs4all.nl>
References: <Pine.HPX.4.10.10006141744570.3588-100000@merle.acns.nwu.edu> <20000615010758.D26436@xs4all.nl>
Message-ID: <Pine.HPX.4.10.10006181239180.21364-100000@merle.acns.nwu.edu>

On Thu, 15 Jun 2000, Thomas Wouters wrote:

> On Wed, Jun 14, 2000 at 05:47:15PM -0500, Nicholas Geovanis wrote:
> > How will the move of the Python crew to a commercial venture affect the
> > Computer Programming For Everybody (CP4E) project?
> 
> Read the FAQ:
> 
> Q. 7. What is happening to CP4E (Computer Programming for Everybody)?
> 
> A. We love this project and are looking for industry sponsors to help us
>    expand this multi-disciplinary work to include educators and usability
>    experts. Contact sponsor at beopen.com if you are interested.
> 
> (http://www.pythonlabs.com/faq.html)

OK, but that doesn't answer my question.
It seems then that the attempt to obtain US government funding for CP4E
has been abandoned, and therefore that there is no upcoming funding
decision to be made (possibly could have elicited an answer of "no" from
them). True or false? 

I need to know because I need to forward the information to a local
school.

> Thomas Wouters <thomas at xs4all.net>

+-------------------->
| Nick Geovanis            Optimists tend to be promoted, so the
| IT Computing Svcs          higher up in the organization you are,
| Northwestern Univ          the more optimistic you tend to be.
V n-geovanis at nwu.edu         - "Davis's Law" (Tom Davis, SGI)



From dana at mmi.oz.net  Tue Jun 20 13:59:41 2000
From: dana at mmi.oz.net (Dana Booth)
Date: 20 Jun 2000 17:59:41 GMT
Subject: Python and mSQL
References: <8io97i$d6u$1@newsg1.svr.pol.co.uk>
Message-ID: <8iobed$jov$0@216.39.141.36>

Jim Ford <jaford at watford53.freeserve.co.uk> wrote:

JF: I use mSQL version 1.0.16 on Linux and want to write a script to access the
JF: database. I'm currently fiddling around with Perl and am starting to get
JF: somewhere, but what little I know about Python attracts me and I'm wondering
JF: whether it would be a better choice.

Of course, my opinion is bias, and python is a better choice. :)

The msql module worked just fine for me when I fiddled with it... That was
with msql verion 2.


-- 
-----
Dana Booth <dana at mmi.oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From ge at nowhere.none  Wed Jun 28 14:45:26 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 28 Jun 2000 18:45:26 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <aJr65.2046$iN5.445612@ptah.visi.com>

In article <Pine.SOL.3.96.1000628105812.1967C-100000 at condor.ee.washington.edu>, Steve Juranich wrote:

>I am *brand* new to python (as of Sunday), and I just got to
>the section in the tutorial about classes.  I was wondering why
>there really isn't such an idea as a "private" member of
>classes?  I understand that members can be hidden, but not
>really protected from the user.

We Python programmers are much more polite than C++
programmers. Just give us a hint that we should leave something
alone, and we will. No need to lock things up in a box if
nobody's trying to steal them.  Apparently C++ programmers are
an untrustworthy lot, and you've got to keep everything under
lock and key.  ;)

>I was wondering what the benefit of this would be?  Wouldn't it
>open the possibility of security holes?

Security holes?  A program is always vulnerable to being proken
by the person writing it.  Private members can't protect a
program from the programmer.

-- 
Grant Edwards                   grante             Yow!  NANCY!! Why is
                                  at               everything RED?!
                               visi.com            


From c.kotsokalis at noc.ntua.gr  Sun Jun 18 07:58:49 2000
From: c.kotsokalis at noc.ntua.gr (Constantinos A. Kotsokalis)
Date: 18 Jun 2000 11:58:49 GMT
Subject: yahoo! webmail written in python?
Message-ID: <slrn8kpebo.l3r.c.kotsokalis@ajax.noc.ntua.gr>

This is advoc-stuff, but it seems like yahoo! webmail service has
been written in our favourite language. A sample url is:

http://us.f52.mail.yahoo.com/py/ymTop.py?......(misc parameters).......

-- 
Constantinos A. Kotsokalis || c.kotsokalis at noc.ntua.gr
National Technical University of Athens
Network Management Center
Tel. No: +30 1 7721861


From aa8vb at yahoo.com  Mon Jun 12 10:09:24 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 12 Jun 2000 10:09:24 -0400
Subject: where is PyOpenGL?
In-Reply-To: <39409F0C.8935C54F@inf.bme.hu>; from rod@inf.bme.hu on Fri,
 Jun 09, 2000 at 09:38:52AM +0200
References: <39409F0C.8935C54F@inf.bme.hu>
Message-ID: <20000612100924.A742284@vislab.epa.gov>

Gabor Papp:
 |Where can I download PyOpenGL for win32 from? I've tried the
 |following:
 |http://starship.python.net/~da		- the link seems to be broken
 |http://PyOpenGL.sourceforge.net/	- it's empty

   http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=626358133

is David's last post on the topic.  On the PyOpenGL mailing list, Robin
<rjh at cita.utoronto.ca> mentions you can fetch the latest PyOpenGL from CVS
using the instructions at:

   http://www.pythonpros.com/cvs.html

See the "ViewCVS Page" link at the bottom as well.  Or if you'd like a
tarball of the latest release (PyOpenGL-1.5.5), let me know via and I'll
send you one.

-- 
Randall Hopper
aa8vb at yahoo.com



From niall.smart at ebeon.com  Thu Jun  8 16:19:51 2000
From: niall.smart at ebeon.com (Niall Smart)
Date: Thu, 08 Jun 2000 21:19:51 +0100
Subject: Can this be stopped from leaking memory?
Message-ID: <393FFFE7.1887276C@ebeon.com>

Hi,

I'm wondering if this can be stopped from leaking memory?

import imp
import time

for i in range(0, 20000):
        fn = "mod.pyc"
        m = imp.load_module(str(time.time()), open(fn, "rb"), fn, (".pyc", "rb", imp.PY_COMPILED))
        m.main()

Where mod.py is:

def main():
	print "foo"

or something else trivial.

--
Niall Smart

email:  niall.smart at ebeon.com
phone:  (087) 8052390


From tim_one at email.msn.com  Sun Jun  4 00:46:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 4 Jun 2000 00:46:17 -0400
Subject: Um? *boggle* Silly question about listobject.c
In-Reply-To: <3939D5A3.2B56E309@san.rr.com>
Message-ID: <000e01bfcddf$d2e09d40$1d2d153f@tim>

[Courageous]
> So why is it that most/all the list operations
> (for example, list_item) are implemented with
> int and not long? Was it forseen that nobody
> would ever desire to index a list with more than
> maxint entries?

Python was written before ANSI C, and eventually all memory is gotten from
malloc, and that's where the int comes from -- the notion that malloc takes
a size_t argument was introduced with ANSI C.

> Am I missing something or is this a critical
> oversight in the definition of sequences in python?

Critical?  No more than that your experience with more-recent C caused you
to make the similar "critical oversight" of suggesting "long" is adequate.
Guido is on record as favoring changing PyObject_VAR_HEAD to declare ob_size
of type size_t, but that won't happen in time for 1.6.  In the meantime, if
you really need lists with more than 2 billion elements, your hardware
budget is so large you can afford to pay someone to patch you up a custom
Python <wink>.

mildly-in-favor-of-reality-some-days-ly y'rs  - tim





From matthewm at zebrasoft.co.nz  Sat Jun  3 23:56:20 2000
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Sun, 04 Jun 2000 15:56:20 +1200
Subject: win32com and Excel
Message-ID: <3939D364.2EF22B13@zebrasoft.co.nz>

hi everyone,

A couple o0f questions:-
1)
can you access an excel (.xls) file via com without having excel loaded
on that machine?
2)
once you've done:-
 blah = win32com.client.Dispatch("Excel.Application")
can anyone give details of
	a) howto open a particular s/s in the current directory
and
	b) howto access particular cells in that particular s/s
and
	c) howto close that s/s

I have to admit that I'm hoping some kind soul will help before I'm
consigned to grappling with what looks like grooling documentation!
thanks alot matthew.


From dworkin at ccs.neu.edu  Thu Jun  8 16:39:03 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 08 Jun 2000 16:39:03 -0400
Subject: function and library listing
In-Reply-To: "Blu Dragoon"'s message of "Thu, 8 Jun 2000 02:08:41 -0000"
References: <393f4787$0$25327@wodc7nh0.news.uu.net>
Message-ID: <vndk8fzzz4o.fsf@camelot-new.ccs.neu.edu>

"Blu Dragoon" <eddthompson at ssi.parlorcity.com> writes:

> Is there a site that list what functions and or libraries  are availible.
> Not nesseccarily all of them just most of them.

Someone already posted about Parnassus, but you may also not have been 
aware of the very nice online documentation for Python itself, at:

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

Current library references are found at:

http://www.python.org/doc/current/lib/

-Justin

 



From richard_chamberlain at ntlworld.com  Thu Jun 22 16:33:47 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Thu, 22 Jun 2000 21:33:47 +0100
Subject: trouble with tkinter
References: <3951D431.B0F20B4@structurex.net>
Message-ID: <gBu45.7678$fw6.147767@news6-win.server.ntlworld.com>

Yep, it would be better to show some code.

Richard

walter hanagriff <walter111 at structurex.net> wrote in message
news:3951D431.B0F20B4 at structurex.net...
> hi all, i havent been programming for long so i started a simple program
> to practice, the python part is easy, but tkinter is nothing but trouble
>
> first question, and plenty more to come, lol, i hope yall can help, i
> stumped the people at tutor with earlier questions that i thought were
> simple, but just in case i really do ask a simple question, is it
> allowed here?
> i have a class, and some def statements in it, each def is a frame, how
> can i call one frame from another def?
>
> if thats not enough info for an answer, should i just show the source
> code for that class or is this enough for an answer?
>
>




From mhammond at skippinet.com.au  Fri Jun  9 10:20:29 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 09 Jun 2000 14:20:29 GMT
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3940D192.97E4ACC3@colconsulting.com>
Message-ID: <N2705.447$cQ.7514@news-server.bigpond.net.au>

"Frank V. Castellucci" <frankc at colconsulting.com> wrote in message
news:3940D192.97E4ACC3 at colconsulting.com...

> Define 'orthodox canonical class form'

I believe this is from "Advanced C++ Programming Styles and Idioms", by
James Coplien (whom Ive actually had the pleasure of dinner with - I was
one of a few - Im sure he wont remember me :-)

Mark.





From root at rainerdeyke.com  Wed Jun 14 17:06:18 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 14 Jun 2000 21:06:18 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB333@ALTNET> <8hr8fj$fu$1@sunnews.cern.ch> <B9e05.1145$vp3.81025@news1.rdc2.tx.home.com> <8i7frt$o02$1@sunnews.cern.ch>
Message-ID: <etS15.1090$OZ.300179@news1.rdc2.tx.home.com>

Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote in message
news:8i7frt$o02$1 at sunnews.cern.ch...
> "Rainer Deyke" <root at rainerdeyke.com> writes:
> >This statement is both incorrect and misleading.
>
> Bold words.
>
> >Incorrect because there
> >may be other requirements: for example, std::set (along with several
others)
> >requires that operator< and operator== are defined for the object.
>
> Nonsense. You can specialize std::set with your own version of comparison
> class instead of the standard less and you can make sure that your version
> of it does not require said operators.

True, but writing your own comparison class is at least as much work as
implementing operator <.  Besides, IIRC you still need operator ==.

> >Misleading because for many classes the compiler generated functions
cause
> >incorrect behavior.
>
> As I have explained in another post, 'many' is only true for buggy code
> that is not exception safe.

For some values of 'many'.  Smart pointers can help, but the smart pointer
classes themselves almost always need explicit copy contructors and operator
=, and if you deal with a variety of objects and resources you'll need a
variety of smart pointers.  For example, COM objects would need their own
smart pointer class.  Then there are classes for which copy and assignment
operations simply don't make sense - nodes in a linked list for example (and
don't tell me to use STL for linked lists - this was only an example).

> Of course, your statements only underline the original point, that is,
that
> the STL is to difficult to use for C++ novices. However, although it has
> its flaws, the STL is well designed and all the simple things have been
> taken care of real good.

I actually agree with this.


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




From aahz at netcom.com  Wed Jun 14 11:39:59 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 15:39:59 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB334@ALTNET>
Message-ID: <8i890f$4fj$1@slb7.atl.mindspring.net>

In article <65118AEEFF5AD3118E8300508B1248774CB334 at ALTNET>,
Mike Steed  <MSteed at altiris.com> wrote:
>
>ctor 

Okay, does anyone else think this is a misspelling of "Chtorr"?




Q: What should you make when you invite a Chtorran to dinner?
A: Your will.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From yawyi at gmx.de  Mon Jun 26 07:09:40 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Mon, 26 Jun 2000 13:09:40 +0200
Subject: SuSE Linux, python and setgid
References: <8j72mh$7q0$1@sun27.hrz.tu-darmstadt.de>
Message-ID: <39575360$0$1860@personalnews.de.uu.net>

Thomas Leonhardt <leonhardt at informatik.tu-darmstadt.de> wrote:
>I have got the problem that I should install a script on our server
>which was written in python. The problem is that the script checks
>wether the group id was set as specified by the file permissions.
>
>But somehow the gid is not set when I execute the script. Is there
>something like suidperl for python or smoethingelse special with
>SuSE linux which causes the gid not beeing set??

set[ug]id scripts are unsafe and for that reason disabled by the
kernel in most UNIXes (reason: race condition between when the
kernel reads the #! line and the the called interpreter reopens
the script for execution; you could exchange the script with
something else in the meantime).

You can use a wrapper like Misc/setuid-prog.c (from the python
source distribution; if you need the single file only you can get
it via the CVSweb interface from www.python.org). REad the
instructions inside setuid-prog.c.

Compile with:
cc -DFULL_PATH='"/full/path/of/script"' setuid-prog.c -o myscriptname

It might still be a security risk to use that. You must trust
setuid-prog.c and Python to have no exploitable security holes
(the same way you would trust Perl and setuidperl...).

Johannes



From olczyk at interaccess.com  Sat Jun 17 03:49:43 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 17 Jun 2000 07:49:43 GMT
Subject: Why whitespace denotation of blocks doesn't work.
Message-ID: <394ac7ea.609024015@nntp.interaccess.com>

After using Python for a while I've come to the conclusion that
using whitespaces to determine the begining and ending of blocks
is bad for three reasons.

Before I list those three reasons let me make two points. People say
that the indentation requirements don't matter because you generally
indent the same way. That may be true, but do you think it is a good
idea to make conventions requirements ( remember that conventions
start flame wars )? The second is that modern editors with good
indentation rules ( like emac's python mode ) make indentation based
coding easy. The thing is that by doing this you are abrogating
control flow to the editor. When it was the editor indenting code
which was well defined then there was no problem. By having
the code depend on indentation, and the editor do the indentation
for you, you give the editor responsiblity in how the code will flow.

Now on to my points.
1) The first is rather stylistic and may not appeal to most, but I
have often found it useful. When debugging code I often put in
code to help me figure out whats happening. I also avoid indenting
such code so that it easy to remember where it is later when I want to
take it out. An example ( assuming I use begin and end to denote
blocks ):

def Average():
begin 
      sum=0
print "number of numbers=",len(numbers) #illegal
i=0
      for n in  numbers:
      begin
print "n=", n #illegal
            sum=sum+n
print "sum=",sum #illegal
 i=i+1                     #illegal
if sum>MaxNumber*I:#illegal
begin                     #illegal    
  print "sum is too large"#illegal
end                        #illegal
     end
     if len(numbers)!=0:
     begin
        return sum/len(numbers) 
     end
     return 0
end

You can immediately see code that you used for debugging and remove
it. However right now writing code that way makes it illegal.

2) You can easily insert a tab introducing a bug.  Consider the
following code ( similar to a bug I recently fixed ):

def FindInvoice(idnum): 
      for invoice in invoices:
            if invoice.id=idnum:
	      return invoice
            return None # should be indented one less tab.

....

for lines in file.readline():
     idnum=lines[:6]
     invoice=FindInvoice(idnum)
     if not invoice:
              invoice=InvoiceClass(idnum)
	 invoices.append(invoice)
....
Initially the line in error was indented correctly, but someplace
along the line the extra tab was introduced. A lot of effort then had
to go in to finding this bug. With begin...end or braces this is less
likely to happen.

3) Finally. Python is a language where it is almost easy to write
extensible programs. To illistrate what I mean consider this example
of a python program I recently wrote:
There is a C++ project which involves many project files and workspace
files ( the project files are roughly equivalent to makefiles, the
workspace files primarily are used to couple projects and define
dependencies ). Version 1 simply read all the project files and 
generated a call to compile each one with all different configuartions
(debug , release, alpha, beta etc..) . Since this built everything
even if it was already built, and since this had to be run several
times ( sometimes compiles would fail because a project that the
present project was dependent on wasn't compiled ), I made version
two. Version 2 read the workspace file and organised the projects
into workspaces. Version 3 passed the string to a second script so
that the second script could figure out if this was something that
should be built ( for example whether a standard Win32 build or a
WinCE build should be done ). Version 4 tracked dependencies more
carefully ( it did a topological sort and orgainsied the build
according to it ). Since the core stuff should be built first, I had
to find out a way to figure out which was core. I used a definition of
the core stuff is the stuff with the longest chain of dependencies.
That was Version 5.

As you can see I started with a simple program and made it more
complex as I went on.  That's what I mean by extendible. Python is
almost good at it. Almost. But that damned code indentation dependence
backs its ugly head. In the process of extending code, I have to move
blocks of code around. Sometimes put it in a loop, sometimes add an if
clause to the top. Sometimes extract code into a subroutine. That
means that I have to reindent the block. So as I reindent I have to
try to remember how the indenting went. Something I stink at. 
Emacs also doesn't reindent correctly. So I have to go through the
code, and try to figure out how it works. This distracts from my 
rewrite ( this also applies to refactorings ). I would rather think
about how new code is supposed to work rather than how old code
is supposed to work.

Anyway I hope this causes TPTB to think about how blocks should be
handled in Pyhton 3000.







From gansevle at cs.utwente.nl  Wed Jun 21 07:00:44 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Wed, 21 Jun 2000 11:00:44 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <8iq78m$u5m$1@nnrp1.deja.com>

In article <ShZ35.11603$Uw3.753134 at bgtnsc04-news.ops.worldnet.att.net>,
  "William Dandreta" <wjdandreta at worldnet.att.net> wrote:
> Hi Cameron,
>
> What I am doing is reading a comma delimited text file like this
>
> line = AB12345,20000621,3.56,...
>
> fields = splitfields(line,',')
>
> recognizing that
>
> fields[Date][:5]  is the year
>
> is more obvious and less cryptic than
>
> fields[1][:5]
>

If you use enums only as a 'conveniant way to represent values', the
next (quick-hack, untested) might be usefull:

class Enum:
    def __init__(self, **kw):
        self.__dict__.update(kw)

enum = Enum(Tag = 0, Date = 1, Ammount = 2)

print "Date-field:", enum.Date

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From johnm at magnet.com  Tue Jun 27 10:45:12 2000
From: johnm at magnet.com (John Mitchell)
Date: Tue, 27 Jun 2000 10:45:12 -0400
Subject: Palm & Python
In-Reply-To: <Pine.LNX.4.21.0006261638270.1767-100000@startrek.codeit.com>
Message-ID: <Pine.SGI.4.05.10006271042590.13130-100000@lemur.magnet.com>

pilot-link allows you to install/backup things from your palm pilot --
it's *not* Python for the Palm.

(However, it's great -- go grab it.)

- j

On Mon, 26 Jun 2000, Grakka wrote:

> 
> I saw this morning someone that was requesting any information regarding
> python & palmtops, check this link out:
> 
> http://pda.tucows.com/palm/dev_unix.html
> 
> Item:
> 
> PalmPython:  An extended replacement for the Python interface provided 
> with Pilot-Link...
> 
> Regards,
> Julio
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 




From darrell at dorb.com  Sat Jun  3 02:24:50 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 2 Jun 2000 23:24:50 -0700
Subject: Escapes???
Message-ID: <03de01bfcd24$6cc01e70$6401a8c0@home>

Arint? wrote:
> I am trying to write this escape seq to a device from python
> "H\x1b\x23\x12art\x1b\x23\x01\n"
> The \x1b\x23\x12 is supposed to be a command tell the device to do
something
> special with "art", but python is making it look like this \x1b\x23\x12a
> "rt".  Am I missing something on how to work with escape sequences?  Is
> there another way to write hex values in a string.  I rather us \x00
because
> that is what are testers are familiar with.
>
Hope this is useful. I've been meaning to write this sort of thing for a
while.
--Darrell Gallion

import re, string, struct

hexData1='''
0016 4210 0100 0004 0000 0101 c042 0100
0242 0500 000f
'''
hexData2='''
0x0016 0x4210 0x0100 0x0004 0x0000 0x0101 0xc042 0x0100
0x0242 0x0500 0x000f
'''
hexData3=r'''
\x0016 \x4210 \x0100 \x0004 \x0000 \x0101 \xc042 \x0100
\x0242 \x0500 \x000f
'''
hexData3=r'''
\x00\x16 \x42\x10 \x01\x00 \x00\x04 \x00\x00 \x01\x01 \xc0\x42 \x01\x00
\x02\x42 \x05\x00 \x00\x0f
'''
hexData3=r'''
\x00\x16 \x42\x10 \x01\x00 \x00\x04 \x00\x00 \x01\x01 \xc0\x42 \x01\x00
\x02\x42 \x05\x00 \x00\x0f
'''
hexData4=r'''
0x0016 \x42\x10 0100 \x0004 \x00\x00 \x01\x01 \xc0\x42 \x01\x00
\x02\x42 \x05\x00 \x00\x0f
'''

bigEndian={1:"c", 2:"c", 3:">h", 4:">h", 5:">l", 6:">l"}
def binConvert(s, sizeMap={1:"c", 2:"c", 3:"h", 4:"h", 5:"l", 6:"l"}):
    """
    Convert s to binary
    s can be formated like:
    0xabcd \x11 \x9999 abcd

    sizeMap maps the size of each number to a struct format string
    for it's conversion. bigEndian was used for regression so that
    r"\x00\x16" == r"\x0016"
    """
    res=[]
    temp = re.findall(r"(?i)[^ \\x\n\t]+(?!x)",s)
    fmt=sizeMap[len(temp[0])]
    for v in temp:
        if fmt != 'c':
            res.append(struct.pack(fmt, eval("0x"+v)))
        else:
            res.append(struct.pack(fmt, eval(r'"\x%s"'%v)))

    return string.join(res,'')

def hexConvert(s, size=1, outPutStyle="0x%s "):
    """
    Convert binary string s to a hex format
    size        ==> how large should each hex value be
    outPutStyle ==> how should each value be formated
    """
    res=[]
    while s:
        temp=s[:size]
        s=s[size:]
        l=[]
        for i in temp:
            l.append("%x"%ord(i))
        val=string.join(l,'')
        res.append(outPutStyle%val)

    return string.join(res,'')

def regress(show=0):
    lastB=None
    for k,v in globals().items():
        if re.match("hexData", k):
            print 'Testing:', k
            b=binConvert(v, bigEndian)
            h=hexConvert(b)
            if show:
                print b
                print h
                print '-'*50
            assert(b==binConvert(h, bigEndian))
            if lastB:
                assert(b==LastB)
                lastB=b

regress(1)



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000602/80d91fad/attachment.html>

From timr at probo.com  Thu Jun 22 01:40:20 2000
From: timr at probo.com (Tim Roberts)
Date: Thr, 22 Jun 2000 05:40:20 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au> <39505457.5223ADC5@san.rr.com> <20000621155338.C11308@fangorn.csse.monash.edu.au>
Message-ID: <DJ93LSK9JGLN6534NMGQNL8M8SQ4ANQUNT@4AX.COM>

Toby J Sargeant <tjs at cs.monash.edu.au> wrote:

>set lcs=tab:?\
>set list

Very cool.  But as long as you are patiently explaining things, can you
tell me how you generated the "?" character inside gvim?  The "Alt-0187"
trick did not seem to work.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From pinard at iro.umontreal.ca  Tue Jun  6 11:06:57 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 11:06:57 -0400
Subject: operator overloading
In-Reply-To: david_ullrich@my-deja.com's message of "Tue, 06 Jun 2000 14:39:40 GMT"
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com>
	<393C2386.307A160D@sightreader.com> <20000605235010.C469@xs4all.nl>
	<8hj2f9$1rn$1@nnrp1.deja.com>
Message-ID: <oqwvk2df1q.fsf@titan.progiciels-bpi.ca>

david_ullrich at my-deja.com writes:

> This reminds me of a question that I've never got around to asking: Are
> there implicit requirements on __cmp__, like the order it defines has
> to be transitive, __cmp(a,b)__ has to be the opposite of __cmp(b,a)__, etc?

If one wants tools like `sort()' to work properly, `__cmp__' should establish
a proper order relation, yes.  Otherwise, rather cryptic bugs might pop out.

However, to give a special meaning to `<=', say, one might surely play
tricks and not use `__cmp__' as it was meant to be.  A butterfly is surely
free to travel nearby a fire :-).

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





From jand at easics.be  Thu Jun  8 12:57:38 2000
From: jand at easics.be (Jan Decaluwe)
Date: Thu, 08 Jun 2000 16:57:38 GMT
Subject: ANNOUNCE: early alpha Quixote Web development tool
References: <200006080135.VAA04781@mira.erols.com>
Message-ID: <393FD07E.20EAB1B2@easics.be>

"A.M. Kuchling" wrote:
 
> The basic idea is that a Web application is written as a Python
> package, and URLs are then treated as paths walking down into the
> package.  Objects, functions, and methods are accessible through URLs
> if they're listed in an __exports__ attribute of the object.
> (Obviously this is inspired by Zope's object publishing.)
> 
> Templating is done in Python Templating Language (PTL), which tries to
> stick close to Python's statements and semantics.

For some time now, I was getting frustrated with my attempts
to use Zope in "bobo-mode", so if this is the new bobo, that's
great news. For our applications, we need something that focusses
on giving a web interface to programs, instead of dynamic web
site development, to do things such as:
 
  http://www.easics.com/webtools/crctool

I had a quick look at the alpha distribution:

- Didn't find the doc subdirectory mentioned in the README file.
  I was looking for more info on PTL and on the details of
  publishing the python module on the web server. I'm currently
  using Zope's pcgi while quixote seems to rely on FastCGI.

- It seems the top-level package should be placed in a 'quixote'
  directory, but this is not explicitly documented. Very good
  that all the code is put into a package (unlike the Zope components).

One thing I'm using from Zope is Persistent objects - objects
that are automatically written and restored from disk in response
to requests. (However, I'm completely confused about the continuing
renames and functional changes of the Zope DBs, so I am still using
quite an old version from the good old bobo era). Is there anything
planned for this in  quixote, or will the user have to implement 
this explicitly? 

Thanks and regards, Jan

-- 
Jan Decaluwe	       Easics              
Design Manager         System-on-Chip design services  
+32-16-395 600	       Interleuvenlaan 86, B-3001 Leuven, Belgium
mailto:jand at easics.be  http://www.easics.com


From xklbr at hotmail.com  Tue Jun 13 20:44:00 2000
From: xklbr at hotmail.com (Dirk Xklbr)
Date: Tue, 13 Jun 2000 19:44:00 -0500
Subject: Session
References: <394106F9.A58F87CE@micron.net> <8hr23k$bvk$1@nntp9.atl.mindspring.net> <8i2e0e01rrn@news2.newsguy.com>
Message-ID: <8i6kav$hpi$1@news.mty.itesm.mx>

I need to make a program that makes a session in web (like a email web
session), where do I start from?, do I need cookies, how does it work?
Thanks




From andre at UltraShell.Net  Fri Jun 30 04:34:07 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Fri, 30 Jun 2000 08:34:07 GMT
Subject: Problem with redirection, buffering and chaining scripts.
References: <395c3b52.183760656@nntp.interaccess.com> <gnF65.8680$Q7.225830@news-east.usenetserver.com> <395b8570.268270515@nntp.interaccess.com>
Message-ID: <3YY65.25715$Q7.555799@news-east.usenetserver.com>

As you read from a file to print out and I haven't seen a 
.close() statement, it was a suggestion that not closing the file is
retaining or delaying the data read together with the associated
print.
I have no further clue if this didn't help. Sorry.

-- avs


Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
> On Thu, 29 Jun 2000 10:17:48 GMT, Andre van Straaten
> <andre at UltraShell.Net> wrote:

>>I never called a Python script from another Python script, but probably
>>it might be helpful to flush the buffer with file.close().
>>This is even necessary if you read and write to the same file within the
>>same script.
>>The buffers are flushed always when the script terminates, but .close()
>>is sometimes necessary explicitely.
>>
>>-avs
> And since the buffers already get flushed when the script closes, how
> does this help?

-- 

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From Sara.Fowler at usa.alcatel.com  Wed Jun  7 16:13:02 2000
From: Sara.Fowler at usa.alcatel.com (Sara Fowler)
Date: Wed, 07 Jun 2000 15:13:02 -0500
Subject: Solaris 2.6 Build Fails
Message-ID: <393EACCE.495F8421@usa.alcatel.com>

I downloaded the source for the Python Interpreter from
http://www.python.org/download/download_source.html.  When trying to
build it on Solaris 2.6 (SunOS 5.6),
using gcc for Solaris 2.6, I get the following errors:

---------------------------------------------------------------------
./posixmodule.c: In function `posix_listdir':
./posixmodule.c:944: `DIR' undeclared (first use in this function)
./posixmodule.c:944: (Each undeclared identifier is reported only once
./posixmodule.c:944: for each function it appears in.)
./posixmodule.c:944: `dirp' undeclared (first use in this function)
./posixmodule.c:945: parse error before `struct'
./posixmodule.c:958: `ep' undeclared (first use in this function)
make: *** [posixmodule.o] Error 1
----------------------------------------------------------------------

Any ideas what the problem is?  Where is DIR defined?  I can't find it
in either the system include files or the source code I downloaded.    

Thanks for the help!

Sara
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Sara.Fowler.vcf
Type: text/x-vcard
Size: 330 bytes
Desc: Card for Sara Fowler
URL: <http://mail.python.org/pipermail/python-list/attachments/20000607/65ff1cc0/attachment.vcf>

From bgroeger at hrzpub.tu-darmstadt.de  Mon Jun 12 06:40:40 2000
From: bgroeger at hrzpub.tu-darmstadt.de (Brigitte =?iso-8859-1?Q?Gr=F6ger?=)
Date: Mon, 12 Jun 2000 12:40:40 +0200
Subject: Successful Solaris 2.6 Compile???
References: <393F9E65.80E9C383@usa.alcatel.com>
Message-ID: <3944BE28.7972943B@hrzpub.tu-darmstadt.de>

you may also download Solaris 2.6 binaries from www.sunfreeware.com

Sara Fowler wrote:

> Has anyone successfully compiled the Python Interpreter on a Solaris
> 2.6?  If so, do you know where DIR is declared?
>
> Thanks!
>
> Sara



From bvdpoel at uniserve.com  Mon Jun 19 14:34:54 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Mon, 19 Jun 2000 11:34:54 -0700
Subject: nest class name problem
Message-ID: <394E67CE.9E29D310@uniserve.com>

I'm having a problem accessing objects created in a nested class. What I
have (extracted) is:

class Goption:
        __name = ""

        def __init__(self, name):
                self.__default = default
                .. more ..

        def change(self):
                class MyDialog(tkSimpleDialog.Dialog):

                        def body(self, master):
                                print self.__name
                                ... more ...


The print 'self.__name' doesn't work. How do I access the variable
__name???

Thanks!

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



From paper_rod at yahoo.com  Sun Jun  4 14:23:07 2000
From: paper_rod at yahoo.com (paper_rod at yahoo.com)
Date: Sun, 04 Jun 2000 18:23:07 GMT
Subject: Awsome Site
Message-ID: <XLKT46P3.I08YV3B5@yahoo.com>

I came across this site and wondered if anyone else has seen it.  Its awsome!


 http://216.15.169.227































If You Received this message in error we apologize.
To Be REMOVED from this list please email us
NOTE: Type REMOVE in the Subject Line when requesting
To be removed. We apologize for any inconvenience



From bragib at my-deja.com  Thu Jun 22 09:07:24 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Thu, 22 Jun 2000 13:07:24 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand>
Message-ID: <8it320$g9$1@nnrp1.deja.com>

This goes using re:

>>> p = '\,'
>>> pat = re.compile(p, re.IGNORECASE)
>>> t = 'uid=joe, ou=orgunit, o=some\\org'
>>> pat.split(t)
['uid=joe', ' ou=orgunit', ' o=some\\org']



In article <m3k8fiosky.fsf at gondolin.beleriand>,
  Steffen Ries <steffen.ries at sympatico.ca> wrote:
> Hi,
>
> I'm trying use re.split to separate an LDAP Distinguished Name into
> its naming components. The naming components are of the form
> "attr=value" and they are separated by ",".
>
> The part, which I can't solve is, that "value" may contain the
> sequence "\,", which obviously does not separate a naming component.
>
> e.g.
> "uid=joe, ou=orgunit, o=some org"
> should be split into ["uid=joe", "ou=orgunit", "o=some org"]
>
> "uid=joe, ou=org\\, unit, o=some org"
> -> ["uid=joe", "ou=org\\, unit", "o=some org"]
>
> Is there a way to use re.split to achieve this? (I solved it searching
> [^\\], and splitting the string in a loop, but I was wondering if it
> can be done easier.)
>
> TIA,
> /steffen
> --
> steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!
>


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


From bparsia at email.unc.edu  Fri Jun 16 07:48:38 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Fri, 16 Jun 2000 07:48:38 -0400
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu> <39498084.972E36D2@san.rr.com>
Message-ID: <1ecb5va.1nch8nwz2t5siN%bparsia@email.unc.edu>

Courageous <jkraska1 at san.rr.com> wrote:

> > > Python may be too slow for some tasks, but in the area
> > > of programmer productivity, very little beats it, and
> > > arguably the environments that do (e.g., lisp) have too
> > > many other negatives associated with them.
> > 
> > Care to explain?
> 
> Very high learning curve.

Er...the learning curve claims always confuse me since I never know what
the axes measure. But, while I personally am intimidated a bit by Common
Lisp, I will say that there seems to be an "initial barrier" (for
"getting into it"), then a lot of things that come more or less
naturally after that (where the chief barrier is really programmatic,
i.e, how to take *advantage* of the language, learning idioms, etc.) and
then some things that strike me as intrinsically difficult (but for
which CL doens't seem to *add* difficulty): macros, the MOP, etc.

The disvalue of such a learning pattern depends highly on where you want
to end up.

> Syntactically heavy weight.

Heh. That's what I say about Python (vs. Smalltalk) :) Given that Common
Lisp has a kind of syntactic *flexibility* that Python lacks, I guess
that although it's base notion of syntax is lighter, the general
experience is going to be more complex. (At least, in the sense that CL
*permits* you to think about syntax in certain ways, and that people are
more likely to *extend or change* the syntax, leading to the plethera of
special forms).

> High
> current cost of maintenance.

Don't understand this one.

> Continually deteriorating
> support and availability of technical expertise.

This requires substantial and substantive justification. My impression
is exactly the contrary (deteriorating from what level? if it's
*continual*, when's the zero point going to hit?).

(And surely the nod here goes to Smalltalk :) Python's is exploding, of
course.)

> Expensive
> license fees from Franz. 

Though with non-commerical varients. Plus, MCL is quite reasonable (even
if you count in a mac), LispWorks has a personal version, Cornman Lisp
is shareware, and there are several free beer&speech versions.

> Paucity of recent commercial publication,
> indicating failure of advancement of the state of the art
> vice recent trends.

To the degree that I can disintangle this, both your data and your
reasoning are highly tendentious, and your conclusion simply false.

[snip]

> > Even the syntax (parentheses) became an asset after I learnt to use
> > Emacs effectively.
> 
> Arguably, it's a very bad sign when a language is so complex
> that it requires an editor tuned to it in particular in order
> to be effective in the language.
[snip]

Er...I strongly doubt that the statment you quoted and the one you made
are even vaguely about the same universe.

I *believe* the original poster is referring to the "icky need to count
parenthesis" "problem".

(For those interested in Common Lisp, I personally found the recent
"AllegroServe" web/application server to be quite promising as a
learning vehicle. It's pretty simple and clear, and if you have
experience, as many do, with network applications, it helps with the
comfort level. This is what I've found from perusing the quite decent
documentation. AllegroServe is intended to be or become portable to any
ANSI CL (I believe) so I don't think it "hurts" to learn it using
Franz's NC offerings, pace whatever pain Allegro Lisp intrinically
offers :))

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From dgoodger at bigfoot.com  Sun Jun 18 16:42:34 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 18 Jun 2000 16:42:34 -0400
Subject: Getting Password from User
In-Reply-To: <8ij9kr$qa5$1@bob.news.rcn.net>
References: <8ij9kr$qa5$1@bob.news.rcn.net>
Message-ID: <B572AC79.517B%dgoodger@bigfoot.com>

on 2000-06-18 15:49, Tom Santos (tomsantos at email.com) wrote:
> I'm trying to get a username and password from a user so that I can create
> an FTP connection.  I'm using raw_input() to get the username but it's not
> acceptable to get their password.  If I use raw_input() to get the password,
> the user's password is echoed to the screen.  Is there any way to make it so
> that their password is not shown entirely, or characters like '*' are
> displayed instead of the characters in their password?

Use module getpass:

    import getpass
    password = getpass.getpass()

If you still see the password echoed to the screen, then you're missing some
functionality, probably termios. Check out the getpass.py code for the
decision tree that is followed.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come! real soon now!)



From dalke at acm.org  Thu Jun 29 01:10:24 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 28 Jun 2000 23:10:24 -0600
Subject: ?Does Python have sorting via "Dictionary Order"?
References: <395A92D0.65A205B1@email.sps.mot.com>
Message-ID: <8jellm$k4n$1@slb2.atl.mindspring.net>

Norman Shelley wrote in message <395A92D0.65A205B1 at email.sps.mot.com>...
>Does Python have sorting via "Dictionary Order"?  This would be a very
>handy addition as an option to list's sort() function.


No it doesn't that option, and no it won't be a handy addition to the
built-in sort() function.
string
There are many other sorts you may want, like case-insensitive
order or reversed order.  They can't all go into Python's core.
Instead, you can pass a comparison function into sort to make it
work the way you want it to, so the better solution is to build
up a library of sort functions.

For example, here is a way to get what you want.  The 'normalize'
function converts the string into a list of words in normal term.
Letters are converted to lower(letter)+letter.  This forces letters
to be put in the same place, but with caps first.  Numbers are pulled
out into integers.

import string

def normalize(s):
  terms = []
  tmp = ""
  flg = -1  # 0 for string, 1 for integer, -1 for undecided
  for c in s:
    if c in "0123456789":
        if flg == 0:
           terms.append(tmp)
           tmp = c
        else:
           tmp = tmp + c
        flg = 1
    else:
        if flg == 1:
           terms.append(int(tmp))
           tmp = string.lower(c) + c
        else:
           tmp = tmp + string.lower(c) + c
        flg = 0
  if flg == 1:
      terms.append(int(tmp))
  else:
      terms.append(tmp)
  return terms

For examples:
>>> normalize("abc")
aabbcc
>>> dictionary_sort.normalize("Abc")
aAbbcc
>>> dictionary_sort.normalize("x10y")
['xx', 10, 'yy']
>>> dictionary_sort.normalize("x9y")
['xx', 9, 'yy']

So normalize converts the word into something which can be compared
with Python's normal comparison method.

Then to compare two words using the dictionary comparison:

def dictionary_compare(x, y):
  return cmp(normalize(x), normalize(y))

This comparison function can be passed into the list's sort() method

>>> words = ["abc", "Abc", "aBc", "x10y", "x9y"]
>>> words.sort(dictionary_compare)
>>> words
['Abc', 'aBc', 'abc', 'x9y', 'x10y']

For long lists there is a lot of overhead because words are normalized
multiple times.  You can get around that with:

def dictionary_sort(data):  # untested!
  terms = []
  for element in data:
   terms.append( (normalize(element), element) )
  terms.sort()
  data = []
  for element in terms:
    data.append(element[1])
  return data

For more info on this trick, see
  http://python.org/doc/howto/sorting/sorting.html

                    Andrew
                    dalke at acm.org





From neilh at scintilla.org  Mon Jun 19 21:21:49 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 20 Jun 2000 01:21:49 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <l03102800b573e58c0e58@[193.78.237.139]>
Message-ID: <NGz35.3149$qz4.24314@news-server.bigpond.net.au>

> ... (In my MacPython IDE the user can choose
> between the spaces and pixels as a visual size unit when setting the
> indentation level...)

   What has been the feedback on defining space in pixel terms?

   The thing that stopped me doing it for Scintilla was that it would not
work very well with the text magnification command. Maybe it should be
defined in terms of x-width which doesn't change as much as the width of a
space which varies a lot between fonts.

   Neil




From philippe.possemiers at the-ecorp.com  Mon Jun 26 03:52:46 2000
From: philippe.possemiers at the-ecorp.com (Philippe Possemiers)
Date: Mon, 26 Jun 2000 09:52:46 +0200
Subject: ppp-script
Message-ID: <39570BCE.5E5DEF97@the-ecorp.com>

Hello all,

I'm sure this question is rather simple to solve, but I haven't found a
good solution yet. I'm writing a small application in python which needs
Internet access at a certain moment.
My platform is Linux and I use a modem to dial my ISP. Therefore, I use
os.exec* in my script to activate /etc/ppp/ppp-on. The question is : how
do I return to my python script? From the moment I call 'ppp-on', I seem
to lose all control and cannot return to the python interpreter.
Apparently, what I need (I think) is something like os.spawnv to spawn a
new process, and execute the ppp script in that process but this is only
available in Windows....
I would appreciate any help or pointers to a solution.

Philippe Possemiers



From trentm at activestate.com  Wed Jun 14 23:06:09 2000
From: trentm at activestate.com (Trent Mick)
Date: Wed, 14 Jun 2000 20:06:09 -0700
Subject: 8 queen/game of life
In-Reply-To: <20000615104403.A1240@sirius.ftsm.ukm.my>
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <20000614200609.A323@activestate.com>

On Thu, Jun 15, 2000 at 10:44:03AM +0800, root wrote:
> newbie question.
> 
> has anyone wrote the 8 non-attacking queen problem and conway's game of life
> in python? i try but face a few newbie problems.
> 
> 1. conway's game of life:
>    how to have a matrix in python? 
>    the following is not good;
>    
>            r0=[0,0,0]
> 	   r1=[0,0,0]
> 	   ........
> 	   
> 	   m=[r0,r1,....]
> 
>    the matrix is fixed, which is not good. my solution use dictionary (i
>    know this is not the right way). the key is a tuple of the matrix
>    indices. the value is '0'. then base on game of life rule, update the
>    dictionary and print for next generation.
> 
> 	   m={}
> 	   m[(0,0)] = 0, m[(0,1)] = 1,....


How about this:

>>> m =[]
>>> for i in range(5):
...     m.append([])
...     for j in range(5):
...             m[i].append(i*j)
...
>>> m
[[0, 0, 0, 0, 0], [0, 1, 2, 3, 4], [0, 2, 4, 6, 8], [0, 3, 6, 9, 12], [0, 4,
8, 12, 16]]
>>> m[0][0]
0
>>> m[1][4]
4
>>> m[2][4]
8


However, if you are doing heavy linear algebra I believe you should check out
NumPy.


Trent

-- 
Trent Mick
trentm at activestate.com



From neilh at scintilla.org  Thu Jun 15 10:25:45 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 15 Jun 2000 14:25:45 GMT
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com> <dCL15.4673$cQ.38267@news-server.bigpond.net.au> <8i8m11$bni$1@nnrp1.deja.com> <8_W15.4949$cQ.40252@news-server.bigpond.net.au> <042601bfd6cc$2fe59230$4500a8c0@thomasnb>
Message-ID: <JH525.5613$cQ.44208@news-server.bigpond.net.au>

> >    All that is really needed is for Sam Rushing's calldll/windll to be
> > included in the standard distribution. Then any required OS calls could
be
> > implemented in Python rather than requiring an extension written in C. I
> > know it opens Python up to the world of untyped, unchecked hardware
> pointers
> > but the increase in capability is large.
> This would not help for the popen case, because on win95/98 (not on NT)
> an additional executable helper program is needed.

   It works for me without the shim. The problem as stated in MSDN is
"""On Windows 95, however, this design causes the parent process to hang
because the redirection pipe does not close when a redirected MS-DOS
application or batch file exits. """
   So what I do is poll the pipe. If there is no data available, check to
see if the child process is dead. If the process is dead then a short delay
is executed and then any remaining output grabbed.

   Neil





From thomas at xs4all.net  Wed Jun 14 14:00:12 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 14 Jun 2000 20:00:12 +0200
Subject: Which linux distros have Python by default?
In-Reply-To: <3947C5E8.4CE1032E@estinc.com>; from eric@estinc.com on Wed, Jun 14, 2000 at 05:48:50PM +0000
References: <uKs15.437$i06.133733@ptah.visi.com> <8i86sf$e7c$0@216.39.141.36> <1VO15.607$i06.193222@ptah.visi.com> <3947C5E8.4CE1032E@estinc.com>
Message-ID: <20000614200011.C26436@xs4all.nl>

On Wed, Jun 14, 2000 at 05:48:50PM +0000, Eric Lee Green wrote:

> But I would strongly suggest using TCL/Tk instead if it is a simple little
> configuration program, because I don't know of any Linux distribution that
> does not come with TCL/Tk -- amongst other things, 'make xconfig' needs TCL/Tk
> in order to work. TCL is brain-dead, but for simple stuff it's usable. 

Actually, Eric Raymond is (or at least was) working on a replacement for the
current kernel config tools (make (old|x|menu|)config) and he is writing it
in Python. Not that it hasn't spawned a huge thread on the linux-kernel
list, though ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hinsen at cnrs-orleans.fr  Wed Jun  7 05:29:43 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 11:29:43 +0200
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <da1eh8.r4g.ln@demonseed.net>
Message-ID: <m3zooxhm9k.fsf@chinon.cnrs-orleans.fr>

Jason Petrone <jp at NOSPAMdemonseed.net> writes:

> I've found the Scheme implementation LispMe(http://www.lispme.de) to be
> usable onboard a palm.  It is very complete and has plenty of support for
> things such as graphics and sound.  However, as much as I hate typing all 
> those parentheses in Scheme on a keyboard, it is 10x worse in graffiti.

In my experience LispMe is the best on-board development system as far
as system integration is concerned. The only drawback is that it's
Lisp! I'd prefer Forth as a language, but development with
QuartusForth is just too messy; every time you start the interpreter
you have to reload all your definitions. And of course I'd *really*
prefer Python...

Which brings me back to Palm Python. Would it be possible to borrow
some of the Palm interface code from LispMe, which recently became
Open Source? That could save time and lead to a more uniform API among
Palm languages.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From hzhu at knowledgetrack.com  Wed Jun 28 17:56:27 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 28 Jun 2000 21:56:27 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com>
Message-ID: <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com>

On Wed, 28 Jun 2000 14:57:41 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:
>Note that all the problems you describe completely go away if you do:
>
>  import AAA
>  import BBB.AAA
>
>IMHO you deserve any/all complications you get when you do "from foo
>import *", and I strongly believe that we shouldn't do _anything_ to
>encourage this use...
>

Noop.  The issue is not what names are imported from a package, but from
which package/module files they are imported.

Perhaps I shouldn't have "abstracted" my example so much as to sidetrack the
reader's minds.  In fact I was doing

from AAA import aaa
from BBB.AAA import aaa

but the first statement does the wrong thing in the directory python/BBB
because it imports from the file python/BBB/AAA.py instead of the intended
file python/AAA.py.

Huaiyu


>Huaiyu Zhu wrote:
>> 
>> Given the following directory structure,
>> 
>> python/
>>    AAA.py
>>    BBB/
>>       __init__.py
>>       AAA.py
>> 
>> if sys.path contains 'python', then
>> 
>> from AAA import *
>> from BBB.AAA import *
>> 
>> would do the right thing at most locations, except when pwd is python/ and
>> when '.' is before 'python' in sys.path, in which case they both import from
>> python/AAA.py.
>> 
>> Is this a fair description of current situation?  If so this makes package
>> name spaces essentially flat. That is, in order to avoid name clashes one
>> has to avoid identical module names even if they are buried deep inside
>> packages.  This is even worse on Windows - its file names are caseless - as
>> one has to avoid filenames with different cases as well.
>> 
>> This situation occurs quite often in practice.  For example, AAA could be a
>> module for doing something.  BBB could be a wrapper package for doing things
>> somewhat differently.  So within BBB/AAA.py one calls 'import AAA' to get
>> the real job done, and oops, it imports itself!
>> 
>> Although it is not advisable to put '.' in path, especially at front, a lot
>> of people actually do. And the functionality of a package should not depend
>> on this detail.  Otherwise what happens if one imports several packages each
>> expecting a different ordering of sys.path?
>> 
>> In Unix shells, however, this can always be overcome by using absolute
>> paths, like /the/absolute/path/command, whatever the PATH is.
>> 
>> So now to my proposed solution.  Is it possible to have something like this?
>> 
>> sys.rootpath  = ['/usr/lib/python1.5', '/usr/lib/python1.5/site-packages']
>> from rootpath.AAA import *
>> from rootpath.BBB.AAA.import *
>> 
>> This guarantees that module names within packages are well separated.  One
>> can still use
>> from AAA import *
>> to import from any module named AAA that happens to be in the path. IMHO,
>> this would not break any existing code (unless it happens to contain the new
>> keyword 'rootpath').  It would certainly facilitate much easier use of
>> packages. The only real problem I can see of is that developers might (for
>> what reason?) rush to occupy real estates under rootpath.  But since the
>> standard distribution is centrally controlled and add-ons are user
>> controlled, this shouldn't be a big problem.
>> 
>> --
>> Huaiyu Zhu                       hzhu at users.sourceforge.net
>> Matrix for Python Project        http://MatPy.sourceforge.net
>> --
>> http://www.python.org/mailman/listinfo/python-list
>


-- 
Huaiyu Zhu                               hzhu at knowledgetrack.com
KnowledgeTrack Corporation               Tel: 925 738 1907
7020 Koll Center Parkway, Suite 110      Fax: 925 738 1039
Pleasanton, CA 94566

Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From dalke at acm.org  Mon Jun 26 21:25:40 2000
From: dalke at acm.org (Andrew Dalke)
Date: Mon, 26 Jun 2000 19:25:40 -0600
Subject: unexpected behaviour in class variables
References: <8j83ml$2s7$1@slb3.atl.mindspring.net> <8F5FCA994gmcmhypernetcom@199.171.54.194>
Message-ID: <8j8vjr$v63$1@nntp9.atl.mindspring.net>

Gordon McMillan wrote in message
<8F5FCA994gmcmhypernetcom at 199.171.54.194>...
>Andrew Dalke talks himself into a circle:


Circle?  I was thinking more of a drunk^Wrandom walk :)

>An attribute found in the instance dict will never get wrapped. Only stuff
>found in the class __dict__ (or in a base class __dict__) qualifies for
>wrapping.


But the documentation at http://python.org/doc/current/ref/types.html says
(with some spacing changes to enforce better alignment)

: User-defined method objects are created in two ways: when getting
:  an attribute  of a class          that is a user-defined function object,
: or when getting
:  an attributes of a class instance that is a user-defined function object.

Isn't an "instance dict" in your terminology the same as "attributes of
a class instance" in the quoted documentation?  In which case, the docs
do not agree you or with the observed behaviour.  I have no problem with
that, in which case I'll ask the docs people to fix the docs.  My question
was to make sure that this really is a doc problem and not my
misunderstanding of terminology or behaviour.

>Why the wrapping? To hide away "self" - so it's implicit to the method
>object but explicit to the function object.
>
>Make sense now?


No.  Why does it only occur for user-defined function objects added to
the class namespace and not C-based functions or callable objects?  That
limits how I design my code if I want the ability to change between a
C-based implementation, a Python-based one or to a function object, which
is exactly the problem I have.  (I replaced a call to a C extension
library to a Python based one used for regression tests.)

It suggests that Python methods are defined as "user-defined functions
which are accessed in class namespace" rather than as I thought "functions
which were tagged by the compiler/importer to be methods" (ie, user-defined
functions which were created in a class scope or C extension methods which
told the run-time that they were methods.)

I did not expect the behaviour and now, even knowing that it occurs,
I cannot give a good explanation for the difference other than perhaps
it simplified implementation.

                    Andrew
                    dalke at acm.org






From info at pythonware.com  Wed Jun 21 07:00:33 2000
From: info at pythonware.com (PythonWare)
Date: Wed, 21 Jun 2000 13:00:33 +0200
Subject: ANNOUNCE: soaplib.py version 0.8
Message-ID: <004e01bfdb6f$edad3eb0$0900a8c0@SPIFF>

more interoperability stuff from the people who brought you
xmlrpclib.py:

    Secret Labs' soaplib.py module implements the SOAP 1.1
    protocol for Python 1.5.2 and 1.6.

    Get your copy from:

        http://www.pythonware.com/products/soap

    This version is somewhat experimental, and has a couple
    of known shortcomings that we plan to fix asap.  It has
    been tested against Userland's SOAP implementation, and
    it's likely that it will work with IBM's Java library
    (we'll know more about that later this week).  However,
    it will most likely not work with Microsoft's current
    SOAP toolkit (but hopefully, they'll fix that soon ;-).

    For more information, see our site.

    Note that this is unsupported software.  For commercial
    support,or if you're interested in sponsoring further
    development, contact us at info at pythonware.com.

enjoy,
the pythonware team

    "Secret Labs -- makers of fine pythonware since 1997."




From matthias.huening at univie.ac.at  Mon Jun  5 10:50:58 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 5 Jun 2000 16:50:58 +0200
Subject: Python's RE
References: <393B7A18.D0F1AA78@barak-online.net>
Message-ID: <8hgetj$h54$1@www.univie.ac.at>

>How exectly do I something that reassembles perl's :
>
>$string = "qwerwe kkljlksdfg (first string) asdkjflkjg
>(secondstring)jlskjlfdkg(one more string)";
>@words = ($string =~ /\((.+?)\)/smg);
>foreach(@words){
>        print "$_\n";
>}


I know there must be a (much) easier way to do this, but the following
code will do the job:

######
import string, re

the_string = r"""aaa(foo)bbb(bar)asdfsadf
ee(g)(adfsadf)ddd xxxx"""

elements = string.split(the_string, ')')

x = re.compile(r'\((.+)$', re.M | re.S)

for element in elements:
   z = x.search(element)
   if z: print z.group(1)
######

On the other hand, if you do it like this, you don't even need Regular
Expressions at all... The string module will be sufficient:

for element in elements:
    z = string.find(element, "(")
    if z != -1:
        print element[z+1:]


Hope this helps,
Matthias

- - - - -
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/






From meng.engineering at bluewin.ch  Wed Jun 28 11:05:35 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Wed, 28 Jun 2000 17:05:35 +0200
Subject: PythonQt Binding Is is available?
Message-ID: <8jd4hk$nda$1@bw107zhb.bluewin.ch>

best regards

markus meng

--
********************************************************************
** Meng Engineering        Telefon    056 222 44 10               **
** Markus Meng             Natel      079 230 93 86               **
** Bruggerstr. 21          Telefax    056 222 44 10               **
** CH-5400 Baden           Email      meng.engineering at bluewin.ch **
********************************************************************
** Theory may inform, but Practice convinces. -- George Bain      **








From alexandre.ferrieux at rd.francetelecom.fr  Thu Jun 29 11:37:21 2000
From: alexandre.ferrieux at rd.francetelecom.fr (Alexandre Ferrieux)
Date: Thu, 29 Jun 2000 17:37:21 +0200
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <3958C648.1D76@rd.francetelecom.fr> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com>
Message-ID: <395B6D30.2486@rd.francetelecom.fr>

Paul Duffin wrote:
> 
> Alexandre Ferrieux wrote:
> >
> > Paul Duffin wrote:
> > >
> > > Apparently there is a port of Expect for Python so why not use that
> > > rather than try and reimplement it all over again.
> > > Expect has over 7 years of development and centuries of testing on many
> > > different platforms, do you really want to reproduce that. If so why,
> > > what exactly is wrong with Expect.
> >
> > The same as with TkInter: it needs a full Tcl inside. Again I've got
> > next to nothing against Tcl, it's just that the separation of orthogonal
> > things is such a natural thing for me.
> >
> 
> ...why not do as I suggested
> and work with Don Libes on seperating the pseudo tty control stuff into
> a seperate library which both Tcl and Python can use.

Perfect. For some reason I didn't realize you were suggesting this.
What next ? You're a native English speaker, you call Don on the phone
;-) ?

-Alex


From Norman_Shelley-RRDN60 at email.sps.mot.com  Wed Jun 28 20:05:36 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Wed, 28 Jun 2000 17:05:36 -0700
Subject: ?Does Python have sorting via "Dictionary Order"?
Message-ID: <395A92D0.65A205B1@email.sps.mot.com>

Does Python have sorting via "Dictionary Order"?  This would be a very
handy addition as an option to list's sort() function.

This is defined by D. Richard Hipp as:
"The case of alphabetic characters is ignored, except to break ties.
Thus "B"
comes before "b" but after "a".  Also, integers embedded in the strings
compare
in numerical order.  In other words, "x10y" comes after "x9y", not
before it as
it would when using strcmp().



From shapr at uab.edu  Thu Jun  1 15:52:18 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 01 Jun 2000 19:52:18 GMT
Subject: Styling Temporary Code
References: <3936A9BB.FBEF9201@visionart.com>
Message-ID: <3936C52F.520B04FD@uab.edu>

Pete Shinners wrote:

> int myfunc(int nullargs)
> {
>     int newval = do_fancy_stuff(nullargs);
> cout << "newval = " << newval << endl;
>     return more_fancy(newval);
> }
> 
> ok, in this 5 line example it doesn't really sparkly,
> but when i use it in big sections of code, it can't
> be beat. call it the mark of the beast, chase me with
> pitchforks, i simply love it and i don't want to change!
> 
> now in python this obviously won't work, but i haven't
> been able to settle on a style that works for me yet.
> what do other people use for this "working" code?

The two things I use are assert statements and print

def myfunc(invar)
    assert invar == type(''),'invar in myfunc is not a string!'
    newval = do_fancy_stuff(nullargs)
    print newval
    return more_fanc(newval)

I don't do C, so this may not be an exact translation....

On the other hand, I also use PyUnit, and that often renders this kind
of code moot.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From schneiker at jump.net  Sun Jun 18 21:37:39 2000
From: schneiker at jump.net (Conrad Schneiker)
Date: Sun, 18 Jun 2000 20:37:39 -0500
Subject: Python 3000 -> No, but there's 'R***'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <8iib7d$73f$1@news.jump.net> <394D26D5.E69282A7@uab.edu>
Message-ID: <8ijtc2$2es$1@news.jump.net>

Hi,

"Shae Erisson" writes

<snip>

> You posted the first part of the Ruby FAQ to comp.lang.python at some
> point in the last month or two.

(Minor aside: that was first part of the Newsgroup FAQ for comp.lang.ruby,
which includes pointers to the Ruby FAQ URL; we don't have a postable ASCII
version of the Ruby FAQ yet.)

> Would you be offended if I posted part
> of the Python FAQ to comp.lang.ruby?

So long as you mentioned Python in the subject line, then I wouldn't be
offended at all. And for previously mentioned reasons, I certainly hope that
no one else on comp.lang.ruby would be offended either.

[This is also applicable to the neighboring post on this thread:] I also
would not take offense if you or others additionally stated your (I presume)
belief that Python is a generally superior language to Ruby, nor (when
expressed a possibly shared belief versus as an indisputable fact of nature)
would I regard that as rude or inflamatory, although I would likely
disagree. Obviously opinons about what various people regard as rude or
inflamatory vary widely here.

To recapitulate, I think Ruby is "the latest and greatest"--for what I am
_personally_ interested in--but I certainly don't regard Ruby as _the_
perfect tool for everyone or everything--for many things Python or Perl
would be more suitable (for various reasons of personal taste, institutional
compatibility, existing module support, and so on). So I don't think usage
of, let alone (newsgroup) interest in, Python, Perl, or Ruby is inherently
mutually exclusive for everyone, although it certainly is for many diehard
purists. But "many" is not "everyone". Likewise, I certainly don't think
that it is out of line for others to provide comp.lang.ruby readers who fall
into this category pointers to other relevant resources from time to time.

Since Ruby has borrowed heavily from other languages and I believed that its
further development will benefit from knowing what people happened to like
better in related languages and knowing what developments were underway in
related languages (among other concerns), I deliberately minimized the
restrictiveness of the comp.lang.ruby charter so as to *not* encourage the
sort of the insular and puritanical mind-set that takes hair-trigger offense
at the mention of other languages or of other language preferences--which I
had too-often observed on comp.lang.perl.misc. The only thing I ask people
who post to comp.lang.ruby is to make their subject lines appropriate, so
people can readily skip stuff that doesn't interest them.

> However you answer, it doesn't seem
> polite to me.

<Stuff about various other dislikes snipped.>

> There are times when it's completely natural to
> talk about Ruby on comp.lang.python. But, small irritations add up, and
> repeated off-topic Ruby advocacy, even of a very small caliber, is not
> what I want to read on comp.lang.python.

So I've noticed. :-)

I'm also crossposting this back to comp.lang.ruby to make sure others do as
well, and I'll leave it at that.

Conrad









From paul at prescod.net  Wed Jun 14 05:15:09 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 14 Jun 2000 11:15:09 +0200
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com> <01bfd2ed$42b67880$29cf8ad1@daves-dell> <39463BA7.43A3CD65@gssec.bt.co.uk>
Message-ID: <39474D1D.F959969C@prescod.net>

Alan Gauld wrote:
> 
> So what? Unix has supported both for a long long time. You
> just escape the spaces or quote the names.... The only
> programs that seem to get really confused are Win16 or
> old DOS programs.


Try double clicking a PDF document with a space in its path. Adobe
Acrobat 4 will not find it. It's annoying as hell because Internet
Explorer puts files from the Web in the "Temporary Internet Files"
directory.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's





From neurocrat at one.net.au  Mon Jun 12 02:48:06 2000
From: neurocrat at one.net.au (Neurocrat)
Date: 12 Jun 2000 16:48:06 +1000
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <85zoor5ra1.fsf@one.net.au>

mkx at excite.com writes:

[snip re Perl roots in Unix]

> So my question is this: Is Python a more "Windows-friendly" language?
> Is it's base structure less related to any one specific operating
> system? Is learning Python an exercise in learning the language, and
> not a Unix tutorial? Can I be successful even if I do not use Unix ;)
> ?

IMO, yes to all.

Although Perl runs everywhere, it is steeped in Unix tradition. (Roots
in sed, awk, etc). For Mac and Windows users, its historical roots
make it something of a "foreign" language. Takes a long time to learn
it effectively.

As far as I can see, Python is remarkably free from the trappings of
any local culture, and although I mostly use Unix I hope it stays
that way. (I seem to recall reading that Guido van R does a lot of
work on the Mac, but I may be mistaken).



From heiland at ncsa.uiuc.edu  Tue Jun  6 14:49:45 2000
From: heiland at ncsa.uiuc.edu (Randy Heiland)
Date: Tue, 06 Jun 2000 13:49:45 -0500
Subject: PyRun_SimpleString error handling
References: <393D062D.8413AC36@ncsa.uiuc.edu> <etdem6a4v8k.fsf@m56-129-16.mit.edu>
Message-ID: <393D47C9.763628AF@ncsa.uiuc.edu>

Alex wrote:

> > Here's my simple pgm:
>
> Sorry I can't help, I've never done any embedding before.  But how do
> you get it to compile?
>
> gcc tst.c -I ../include/python1.5/ -L ../lib/python1.5/config \
>         -lpython1.5
>
> gives complaints about millions of undefined references.
>
> Alex.

rf. to the Makefile in your Python dist under /Demo/embed

--Randy



From digitig at cix.co.uk  Fri Jun 16 19:34:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 17 Jun 2000 00:34 +0100 (BST)
Subject: = vs. == in Py3k
References: <39480966.C69FFDB7@prescod.net>
Message-ID: <memo.20000617003439.42607A@digitig.compulink.co.uk>

In article <39480966.C69FFDB7 at prescod.net>, paul at prescod.net (Paul 
Prescod) wrote:

> Tim Rowe wrote:
> > 
> > The construct
> >         a = a + 1
> > can confuse newbies too, provoking a response along the lines of "not 
> > for
> > finite 'a' it doesn't!"
> > 
> > I reckon that's sufficient ground for a distinction between 
> > assignment and
> > equality. '=' possibly isn't the best choice for assignment, though I 
> > much
> > prefer it to "ADD 1 TO a GIVING a".
> 
> How about:
> 
> a <- a+1
> 
> pronounced "a gets a plus 1"

Yes, I'd come across that, and have forgotten where. Like you, I see no 
reason for a change -- none that is worth the hassle, anyway.


From rwallace at esatclear.ie  Fri Jun  2 11:21:04 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Fri, 02 Jun 2000 16:21:04 +0100
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no>
Message-ID: <3937D0E0.2EA7@esatclear.ie>

Joseph Santaniello wrote:
> 
> Hi,
> 
> I have an SQL query which I can format to give me a nice value like
> 23,342.40 which Python squashes into 23342.4 when I stick it into a
> list element. Is there a nice way to format floats so they look like
> (US) currency?
> 
> Thanks,
> 
> Joe

I've written a formatting routine that does this.  (For physical
quantities, it can also round to a specified number of significant
figures irrespective of where the decimal point is.)  Let me know if you
want a copy.

Is there a central repository for these sort of things?

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From aahz at netcom.com  Tue Jun 20 22:21:35 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jun 2000 02:21:35 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <8ip8rf$1ig$1@slb7.atl.mindspring.net>

In article <QoU35.15352$Xx5.796205 at bgtnsc06-news.ops.worldnet.att.net>,
William Dandreta <wjdandreta at worldnet.att.net> wrote:
>
>Is there a Python type like the C enum type?

I generally prefer using a dict to simulate it.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From jkraska1 at san.rr.com  Sat Jun  3 19:07:04 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 03 Jun 2000 23:07:04 GMT
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com> <39398EEE.C18EC987@tismer.com>
Message-ID: <3939903D.79BD8D2D@san.rr.com>

> As per unusual, I will incorporate this into Stackless
> Python immediately if it is written as clear as you described
> it.

It actually *is* pretty clear, I discovered that I needed to be
very careful with head inserts. I ended up with 3 seperate critical
conditions (and am not sure if there might be some more clever way
of expressing them):

    if ((dlist->vfront < 1) && (dlist->len < (2*dlist->alloclen/3)))
    {
        DlistReposition(dlist);
        dlist->vfront--;
        dlist->items[dlist->vfront]=object;
    }
    else if (dlist->vfront < 1)
    {
        DlistGrow(dlist);
        dlist->vfront--;
        dlist->items[dlist->vfront]=object;
    }
    else
    {
        dlist->vfront--;
        dlist->items[dlist->vfront]=object;
    }

Condition number 1 occurs when the virtual index is underflowing
the actual allocated memory, but there in fact is plenty of memory
to hold the current array. Therefore it is reposition using the
"1/3 rule".

Condition number 2 occurs when the virtual index underflows the
allocated memory, but the memory space is too small. The allocated
region is doubled and then repositioned using the 1/3rd rule.

Condition number 3 occurs normally. The virtual index is
decremented and the item is inserted, cost = 1.

I'm not certain exactly, but I believe that the first two conditions
amortize away over all inserts at the head to O(1)+k.

The rule of 2/3 availabile memory being "enough" was an entirely
arbitrary pick. There has to be some rule, however, otherwise
condition number two will occur on every underflow, resulting in
growth at every insert at the head. Ran into that one by mistake.

Oopsie. :)-



C/


From erno at iki.fi  Sat Jun 24 14:11:11 2000
From: erno at iki.fi (Erno Kuusela)
Date: 24 Jun 2000 21:11:11 +0300
Subject: EMail Monitoring in Linux
References: <8iph0p$ekl$1@nnrp1.deja.com>
Message-ID: <kubt0rvtj4.fsf@lasipalatsi.fi>

>>>>> "buttnerw" == buttnerw  <buttnerw at my-deja.com> writes:

    buttnerw> I need to know the easiest way to do the
    buttnerw> following.  Whenever the mailing program (sendmail) on a
    buttnerw> mail server (linux) receives a mail for a user, a
    buttnerw> "program" needs to be notified about the incoming mail
    buttnerw> and a UDP message has to be send with the user's
    buttnerw> information and the amount of mails in his inbox.  All I
    buttnerw> need is the "program" which will listen to incoming
    buttnerw> mails, and send the UDP packet.

chances are you already have such a program on your system, namely
comsat(8) (or in.comsat). it writes some information to the users
terminal about incoming mails if the user so wishes (see biff(1)).

you can easily write a replacement for the comsat deamon if you want
something else to happen. briefly, a udp message of the format
username at offset is sent to the biff udp port, and then comsat seeks to
that offset in the users mailbox and writes some information to the
users terminal.

the udp packets are sent by the mail delivery program (procmail is
used for this on many systems), not sendmail.  (this detail is
probably irrelevant to you...)

  -- erno


From jajs at NOSPAMretemail.es  Thu Jun 15 23:37:13 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Fri, 16 Jun 2000 05:37:13 +0200
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <394928A2.BEA2BC1@sage.att.com>
Message-ID: <slrn8kj879.7fg.jajs@Pascual.es>

Garry Hodgson dijo sobre Re: Waffling between Python and Ruby,
en fecha: Thu, 15 Jun 2000 19:04:02 GMT

|Mark Jackson wrote:
|> 
|> "Warren Postma" <embed at NOSPAM.geocities.com> writes:
|> 
|> > Having something
|> > look correct on the screen, but not work because of misuse of tabs versus
|> > spaces, or incorrect indentation, is something that still continues to bug
|> > me.
|> 
|> Indulge my curiosity - how many times has this actually happened in
|> your six months of using Python?
|
|well, i'm not warren, but over the last several years of 
|python use this has happened to me maybe twice.  each time i got
|a syntax error, looked at it, said "oh.", and fixed it.
|lost productive time: perhaps a minute and a half.

And now the (almost retoric) question (for Warren) is:

When you program in C/C++/Java/etc, How many times do you forgot to 
type the closing brace/END/etc?


From barryhart at my-deja.com  Tue Jun 20 00:01:03 2000
From: barryhart at my-deja.com (barryhart at my-deja.com)
Date: Tue, 20 Jun 2000 04:01:03 GMT
Subject: Strange problem with PythonCOM
Message-ID: <8imq9s$i0u$1@nnrp1.deja.com>

I'm trying to use PythonCOM to call into TLBINF32.dll (the Microsoft
type library info DLL), and I'm getting a strange error. My call to
create the object - Dispatch("TLI.TLIApplication") - fails with "No
such interface supported."

On the surface, this error would appears to indicate the TLIApplication
class does not support IDispatch. This makes no sense, however, because
I am able to use the class successfully using late bound calls (i.e.
using an Object variable) from VB.

The code is posted below - the currently non-working Python code and
the working VB code.

Any ideas?

Thanks,
Barry Hart

---Python code---

import win32com.client

def GetTypeLibPath(sGuid, nMajorVersion, nMinorVersion):
    objTypeLibApplication = win32com.client.Dispatch
("TLI.TLIApplication")
    objTypeLibInfo = objTypeLibApplication.TypeLibInfoFromRegistry(
        sGuid,
        nMajorVersion,
        nMinorVersion,
        0)
    return objTypeLibInfo.ContainingFile

print GetTypeLibPath("{00000600-0000-0010-8000-00AA006D2EA4}", 2, 1)

---equivalent VB code---

Function GetTypeLibPath(ByVal sGuid As String, ByVal nMajorVersion As
Integer, ByVal nMinorVersion As Integer) As String
    Set objTypeLibApplication = CreateObject("TLI.TLIApplication")
    Set objTypeLibInfo = objTypeLibApplication.TypeLibInfoFromRegistry(
_
        sGuid, _
        nMajorVersion, _
        nMinorVersion, _
        0)
    GetTypeLibPath = objTypeLibInfo.ContainingFile
End Function

Sub test()
    MsgBox GetTypeLibPath("{00000600-0000-0010-8000-00AA006D2EA4}", 2,
1)
End Sub


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


From pinard at iro.umontreal.ca  Wed Jun 21 09:06:27 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 21 Jun 2000 09:06:27 -0400
Subject: Current date and time
In-Reply-To: William Park's message of "Tue, 20 Jun 2000 01:03:56 -0400"
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com>
	<20000620010356.A892@better.net>
Message-ID: <oq4s6n89po.fsf@titan.progiciels-bpi.ca>

[William Park]

> >>> import time
> >>> time.ctime(time.time())
> 'Tue Jun 20 01:02:03 2000'

Let me advocate (or rather, let Markus Kuhn) for ISO 8601.  I always found a
bit miserable that, by default, dates are not international.  Let me see...
Yes, this was sent on 2000-02-24:


From: Markus Kuhn <Markus.Kuhn at cl.cam.ac.uk>
Subject: [tar 1.13.17] Problems with USE_OLD_CTIME 
To: ro at TechFak.Uni-Bielefeld.DE, bug-tar at gnu.org
Cc: pinard at IRO.UMontreal.CA, karl at cs.umb.edu


Rainer Orth <ro at TechFak.Uni-Bielefeld.DE> write 2000-02-03:
> 
>    I've never understood why GNU tar changed to use the ISO time
>    format instead of the traditional ctime-based format.

The two arguments brought forwards at the time that I remember were:

  - ISO 8601 is considered to be acceptable in practically all Gregorian
    locales, while the old format isn't, as it contains English words. A
    locale-dependent output format would make its automatic post processing
    horribly difficult, which is why locale-dependent date formatting
    should be considered with caution.

  - The all-numeric bigendian nature of ISO 8601 actually makes
    comparing dates in columns by humans significantly faster and
    easier, so we considered ISO 8601 not only more practical for
    i18n purposes, but also more user-friendly for English users.
    (Experiment: try to find the oldest and newest date/time in a
    printed 100-entry column of dates/times both when written
    in ISO 8601 and in the old ctime format.)

http://www.cl.cam.ac.uk/~mgk25/iso-time.html

Markus

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





From dworkin at ccs.neu.edu  Tue Jun 20 22:30:35 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 20 Jun 2000 22:30:35 -0400
Subject: <OT> New filemanager needs people with files
In-Reply-To: etoffi's message of "Mon, 19 Jun 2000 02:40:56 -0400"
References: <394DC078.86C1FB02@softhome.net>
Message-ID: <vndr99rvk8k.fsf@camelot-new.ccs.neu.edu>

etoffi <etoffi at softhome.net> writes:

> I am going to be hosting a project on sourceforge called Newfie. 

I wonder how many Newfies will end up using it.

Spent much time in northeastern Canada?

-Justin

 



From moshez at math.huji.ac.il  Sat Jun 10 09:20:07 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 16:20:07 +0300 (IDT)
Subject: Checking socket type?
In-Reply-To: <eviltofu-36614B.07382410062000@news.earthlink.net>
Message-ID: <Pine.GSO.4.10.10006101618110.5325-100000@sundial>

On Sat, 10 Jun 2000, Jerome Chan wrote:

> How would I check to see if an object is an instance of a socket?

Use the type() function:

>>> type(socket.socket(socket.AF_INET, socket.SOCK_STREAM))
<type 'socket'>

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From Norman_Shelley-RRDN60 at email.sps.mot.com  Fri Jun 30 19:57:34 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Fri, 30 Jun 2000 16:57:34 -0700
Subject: ?Combination of a Dictionary and a List?
Message-ID: <395D33EE.75F1E36D@email.sps.mot.com>

Sometime  in the past I wrote a simple combination of a dictionary and a
list, where __getitem__() would first look in a hidden dictionary and
then if not found it would try looking in a hidden list.  I've seen this
type of code out there somewhere also but I can not find my code or
anyone elses. I guess I haven't figure our the "right" search criteria
on www.python.org.  Any pointers?



From db3l at fitlinxx.com  Fri Jun 16 22:43:29 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 16 Jun 2000 22:43:29 -0400
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com>
Message-ID: <uitv9qb72.fsf@ctwd0143.fitlinxx.com>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> OK, so I agree that the (setq-default tab-width 4) in my .emacs is evil (it
> sits there since way back when I copied my first .emacs from somebody else),
> I'm wondering what's the easiest way to revert this on multiple files,
> either within or without emacs.  If I'm going to change my default emacs
> behavior I want to do this on all the files, and safely.

One fairly simple way is to untabify your file, change the tab width,
and then if you want re-tabify.  Emacs will use a mixture of TABs
(using the new tab width definition) and spaces to achieve the same
spacing you already have.

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


From hzhu at knowledgetrack.com  Fri Jun 16 21:02:10 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Sat, 17 Jun 2000 01:02:10 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com>
Message-ID: <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com>

On 16 Jun 2000 17:16:50 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>hzhu at knowledgetrack.com (Huaiyu Zhu) writes:
>
>> Oops!  You're right, as several people also pointed out to me in email.
>> Actually I had this in .emacs and didn't realize it for years:
>> 
>> (setq-default tab-width 4)
>
>Eeek - that's evil.  Don't do that :-)
>
[snip]

>> Wondering what other people will see when reading my programs ...
>
>They'll see a TAB character expand to get to the nearest 8-character
[snip]


OK, so I agree that the (setq-default tab-width 4) in my .emacs is evil (it
sits there since way back when I copied my first .emacs from somebody else),
I'm wondering what's the easiest way to revert this on multiple files,
either within or without emacs.  If I'm going to change my default emacs
behavior I want to do this on all the files, and safely.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From soper at delanet.com  Fri Jun  2 18:43:08 2000
From: soper at delanet.com (Paul Soper)
Date: Fri, 2 Jun 2000 18:43:08 -0400
Subject: rld failure on IRIX
Message-ID: <393839a7_4@corp.newsfeeds.com>

After building Numerical Python 15.3 and MMTK 2.0 on IRIX 6.46 without any
errors during the 'build' or 'install' steps, both packages generate error
messages like the following;

ImportError: 135737:python: rld: Fatal Error: Cannot Successfully map soname
'/usr/local/lib/python1.5/site-packages/MMTK/irix646/MMTK_universe.so' under
any of
the filenames
/usr/local/lib/python1.5/site-packages/MMTK/irix646/MMTK_universe.so

The file referred to does exist and has all of its permissions set
correctly.  Why is the dynamic linker upset?






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


From aruth at intercation.com  Thu Jun 29 09:42:36 2000
From: aruth at intercation.com (Adam Ruth)
Date: Thu, 29 Jun 2000 07:42:36 -0600
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com>
Message-ID: <8jfjs1$2aq3$1@news.aros.net>

Check out www.aviddiver.com or www.efranchiseguide.com.  Pure Python and
PostgreSQL.

--
Adam Ruth
InterCation, Inc.
www.intercation.com


<robin.escalation at ACM.org> wrote in message
news:8isjlssf3j3ivrg7bn4ln567pobgo0eb3l at 4ax.com...
> Does anyone have any success stories (or gotchas) related to
> implementing a PostgreSQL database with Python that they'd like to
> share? I am hoping to use it as a back-end to a b2b web site.
>
> Technical info or just evangelism fine!
>
> E-mail if you don't wish to share with all here.
>
> -----
> robin                          robin at illusionsexeculink.com
> media artist /                 remove illusions to reply
> information architect          www.execulink.com/~robin/
>
> "We told it about general relativity and gravity, the kind of people
> that overcome hardships."




From thomas at xs4all.nl  Tue Jun 27 03:55:44 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 27 Jun 2000 07:55:44 GMT
Subject: Getting double* from NumPy array?
References: <ted.drain-2606001323500001@mjordan.jpl.nasa.gov>
Message-ID: <slrn8lgnfi.ke4.thomas@centurion.xs4all.nl>

On Mon, 26 Jun 2000 13:23:50 -0700, Ted Drain <ted.drain at jpl.nasa.gov> wrote:
>I have some existing C/C++ code that can accept a double array ( int len,
>and double* ptr) to do some processing.  I want to allow users to create
>arrays in python using NumPy and then provide an interface to pass these
>arrays to my C/C++ code.  

>Assuming a 1-dimensional NumPy array, is there an easy way to extract a
>pointer to a double and length from the python object?  The C/C++ isn't
>going to change the object so I'd prefer not to make a copy of it.

NumPy not withstanding (I don't really know it) but doesn't the 'array'
module do what you're looking for ? It won't allow you to store the length
in the same structure, so you'll have to pass that seperately (or convert it
to a double and store it as the first element) but it does give direct
access to the stored values, using the ob_items member.

But I'm sure NumPy has a similar method ;)


From gavrilov at iname.com  Tue Jun 27 19:46:27 2000
From: gavrilov at iname.com (Alexander Gavrilov)
Date: Tue, 27 Jun 2000 23:46:27 GMT
Subject: BXXP - Blocks eXtensible eXchange Protocol 
Message-ID: <n1b65.7188$i5.105827@news1.frmt1.sfba.home.com>

Hi All!

What do you think about BXXP? http://http://www.invisible.net

They have Blocks SpaceKit for Java and Perl. May be, now it's time to build
the same kit for Python.




From kens at sightreader.com  Fri Jun 16 02:38:42 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 23:38:42 -0700
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <3949CB72.7CFFF0E0@sightreader.com>

Oops I hit "Send" too soon.  My code in the previous message is way bogus.

Your homework assignment, Zamri, is:  What's wrong with [[0]*8]*8 ?

Hint: what does the multiply operator really do with a list?

root wrote:

> newbie question.
>
> has anyone wrote the 8 non-attacking queen problem and conway's game of life
> in python? i try but face a few newbie problems.
>
> 1. conway's game of life:
>    how to have a matrix in python?
>    the following is not good;
>
>            r0=[0,0,0]
>            r1=[0,0,0]
>            ........
>
>            m=[r0,r1,....]
>
>    the matrix is fixed, which is not good. my solution use dictionary (i
>    know this is not the right way). the key is a tuple of the matrix
>    indices. the value is '0'. then base on game of life rule, update the
>    dictionary and print for next generation.
>
>            m={}
>            m[(0,0)] = 0, m[(0,1)] = 1,....
>
> 2. 8 non-attacking queen.
>
>    use backtracking. the function call itself back. how to do recursive call
>    in python? roughly, the C program I study look like this;
>
>    queen_count = -1
>
>    main()
>    {
>      addqueen();
>    }
>
>    addqueen()
>    {
>    queen_count++
>    if queen_count != 8
>         addqueen()
>    else
>         print result
>    }
>
>    how to pass a global variable (queen_count) to a function?
>
> many thanks in advance.
>
> zamri
>

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/4feee546/attachment.vcf>

From db3l at fitlinxx.com  Wed Jun 21 15:04:50 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Jun 2000 15:04:50 -0400
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <39504E63.691C86FB@san.rr.com>
Message-ID: <uu2emhn3h.fsf@ctwd0143.fitlinxx.com>

Courageous <jkraska1 at san.rr.com> writes:

> It would be nice if an editor had a vertical guide, much as some
> of the drawing programs do. You can click and drag the guide around,
> leaving a vertical line drawn across your editor, allowing you to
> move up and down in the file to see what matches that indentation
> level.

It's not really a visual guide, but Emacs can set set a goal column
(C-x C-n, C-u C-x C-n to clear), then just moving up and down the file
will try to keep you in the column (unless its past the end of the
current line), so you can pretty easily move up and down and see what
lines up with that column.

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


From a.rottmann at gmx.at  Thu Jun 22 09:13:52 2000
From: a.rottmann at gmx.at (Andreas Rottmann)
Date: 22 Jun 2000 15:13:52 +0200
Subject: Python interface to Oracle
References: <3951C636.2E985CE@ust.hk>
Message-ID: <87og4t26zz.fsf@ns.f7.ph4.lan>

David YEUNG <dyeung at ust.hk> writes:

> Does anyone know if any python module has been developed for
> interfacing with Oracle8i database on Linux?
> 
That would be interesting for me, too. I'm likely to develop one, if
it doesn't exist, since I have the infrastructure ready - see
GQL.sourceforge.net.

All it would take is writing an Oracle driver for GQL, which can be
done within a few days. Anyway, I would appreciate if someone could
help me with that, since I don't have Oracle and am not willing to pay
the $$$.

Andy
-- 
Andreas Rottmann   | Dru at ICQ     | 54523380 at ICQ | a.rottmann at gmx.at
Pfeilgasse 4-6/725 | A-1080 Wien | Austria      | Europe
http://www.8ung.at/rotty         | Key is there | gpg (GnuPG) 1.0.1
Fingerprint        | 3E9A C485 49A4 1D17 2EA7  2BA7 22AE C9BF 8173 6279
[one of 78,35% Austrians who didn?t vote for Haider!]


From jajs at NOSPAMretemail.es  Mon Jun 12 20:52:05 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Tue, 13 Jun 2000 00:52:05 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au> <394484C0.A5E412D9@san.rr.com>
Message-ID: <slrn8ka8kq.1r0.jajs@Pascual.es>

Courageous dijo sobre Re: Python Productivity over C++,
en fecha: Mon, 12 Jun 2000 06:33:05 GMT

|Python is a simply wonderful prototyping environment, IMO.
|Now only if it had an optimizing compiler...

I think every Python fan dream with that... 

It's really so difficult?


From rsteelandt at bucodi.com  Tue Jun 13 12:11:39 2000
From: rsteelandt at bucodi.com (BuCoDI-Rony Steelandt)
Date: Tue, 13 Jun 2000 18:11:39 +0200
Subject: Two newbie questions
References: <20000613165659.A7828@basajaun>
Message-ID: <39465D3B.3356DCC4@bucodi.com>

Don't know exactly what you mean, could it be that??
st = "asdfasfasdf"
l = []
for c in st:
    l.append(c)
#process here
#....
#now get string back
s=''
for c in l:
    s= s+ c


Arkaitz a ?crit :

> Hi all,
>
> I'm learning Python, as it seems everybody around here :-), and I found two
> problems that I can't find the solution to:
> 1. As the strings are inmutable, if I want to modify the order of the characters
> in the string(e.g. sort them), what I do is:
>
> st = "asdfasfasdf"
> l = []
> for c in st:
>     l.append(c)
>
> Then I process the list, and the problem is that I have no idea of how I can get
> the string back, any help?
> 2. The second question is about random numbers, how can I get a random int between the minimum and maximum integers of the machine?
> The next code doesn't worK
> import sys, random
>
> i = random.randint(-sys.maxint-1, sys.maxint)
>
> Thanks in advance,
>
> Arkaitz.

--
======================================================
Bureau de Conseil et de D?veloppement Informatique
2, Rue Henri Avril
22480 Saint - Nicolas du Pelem

T?l : +33.2.96.29.79.08
Fax : +33.2.96.29.79.09

e-mail : bucodi at worldnet.fr
======================================================




From petersc at stud.ntnu.no  Tue Jun  6 12:42:06 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 06 Jun 2000 18:42:06 +0200
Subject: Small bug in zlib module
References: <001601bfcfca$22907860$56a21c3f@versa>
Message-ID: <393D29DE.FE761725@stud.ntnu.no>

Greg Chapman wrote:
> 
> Using Python 1.5.2 (under Windows 98), this script causes python to enter a
> loop in which it progressively allocates larger and larger blocks of memory:
> 
>   import zlib
>   s = zlib.decompress("")
> I assume PyZlib_decompress should have a check for an empty input string and
> (if found) should generate an appropriate exception.

On the current CVS version it immediately (as far as I can see without
looping) returns with an error.

appropriateness-is-another-question-though-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From alex at e-cryption.com  Fri Jun 23 16:50:41 2000
From: alex at e-cryption.com (Alexandre A. Rodioukov)
Date: Fri, 23 Jun 2000 20:50:41 GMT
Subject: python email system help
References: <3953944C.1934AD89@tcgfinancial.com>
Message-ID: <39527C72.E23205BB@e-cryption.com>

I think if you will use Linux there is no real need to write anything,
since there are already available packages that will do the trick for
you. You might want to check 'fetchmail' (this will fetch e-mail from
remote POP3/IMAP account and store it locally) and 'diald' which will
establish an internet connection on demand. Check
http://www.freshmeat.net
to figure out where to get this stuff from if it's not already
installed.

The Commonwealth Group wrote:
> 
> Hi, Im an 18 year old network kid, who knows nothing about programming.
> I accidently made a suggestion at a meeting and now I must implement
> that suggestion, and I know nothing about progamming.  I need to write
> this program by Monday.  I started learning python about a month ago,
> but I've been too busy to make real progress. I need to write a program
> that:
> 1st, dial internet,(if busy, or someone else is using, dial again later)
> 
> 2nd, get email for several accounts and put on local file server
> 
> I'm not sure If I will be dialing from winNT or Linux yet(not sure if
> what OS I'm dialing from makes a difference).  Also, I need this program
> to also be instantly executable if someone needs to get their mail that
> instant.  Also, can I write the email program in a fahion so that after
> retrieving the mail, there are still full email copies on the server.  I
> would aappreciate if someone that isn't busy this weekend could sort of
> guide me in writing this, maybe via irc, multiple news postings, emails,
> or others.  Or if someone is real nice, they could write it for me.
> Thanks for the attention and help.  Regardless of if I get help or not,
> I have to have this by Monday and I'd rather not pull my hair out trying
> to write something in a language I don't know.  Its like trying to have
> a conversation in a foreign language you don't know.
> Jason Toy
> gee308 at mediaone.net
> btw, the name you see for this posting is the work name, so you can gind
> me as toy or through email.  Thanks again.

Regards,
Sasha.


From ge at nowhere.none  Fri Jun 30 11:06:41 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 30 Jun 2000 15:06:41 GMT
Subject: MimeWriter (how to use/find boundaries)
References: <Pine.LNX.4.21.0006301138390.15363-100000@lade.src.no>
Message-ID: <5I275.2321$iN5.531647@ptah.visi.com>

In article <Pine.LNX.4.21.0006301138390.15363-100000 at lade.src.no>, Morten W. Petersen wrote:
>I've built a mail-sending function, which, when sending mail with
>attachments; does send the mail, with the encoded attachments; but it
>doesn't make the mail readable on the other side. I.e., it does not
>include the boundary thing, and I don't have a clue as to how to go about
>to do it. (...)

Take a look at 

 ftp://ftp.visi.com/users/grante/stuff/mfetch.py
 
There's a class in there called SmtpWriter (or something like
that) that builds e-mails with attachments using MimeWriter,
and then sends them using smtplib.

Look for the method named Message()

    def Message(self,sender="", subject="", recipients=[],body="", attachments=[]):
        if self.__debugLevel > 2:
            sys.stderr.write("SmtpWriter: Building RFC822 ....

-- 
Grant Edwards                   grante             Yow!  I would like to
                                  at               urinate in an OVULAR,
                               visi.com            porcelain pool --


From paulb at infercor.no  Thu Jun  8 06:29:14 2000
From: paulb at infercor.no (Paul Boddie)
Date: Thu, 08 Jun 2000 12:29:14 +0200
Subject: SQL Variable Substitution Question
References: <8hmacu$ia9$1@news-int.gatech.edu> <uf8ujs0siug4cfsacqcuvrl37gimb82bel@4ax.com>
Message-ID: <393F757A.26D489AE@infercor.no>

Tim Roberts wrote:
> 
> liang at cad.gatech.edu (Shengquan Liang) wrote:
> >
> >say i have yyyy = 1000
> >
> >in the documentation of DC Oracle, it says that
> >i shoul d use 'select * from attt'
> >               'where bbbb = :yyyy'
> >
> >to make it work.
> 
> No, SQL doesn't have any idea what your Python variables are.  To use the
> ":" feature, you have to have set an _SQL_ variable called "yyyy".  How you
> do that varies with the backend.

Guessing, with the help of the DB-API specification, I would try:

  cursor.execute('select * from attt where bbbb = :yyyy', {"yyyy" : yyyy})

Or:

  cursor.execute('select * from attt where bbbb = :yyyy', yyyy=yyyy)

See the specification at:

  http://www.python.org/topics/database/DatabaseAPI-2.0.html

> In your case, it's probably easier to build the complete string in Python
> and pass that to the database:
> 
>   'select * from attt where bbbb = %d' % yyyy

I would advise against this, and given the numerous recent occasions when such
advice has been given on comp.lang.python, I rather suspect that most Python
database application writers who make their opinions known either don't know
about bind variables or don't trust them. As for being easier, take a look at
this suggestion:

Scott Barron wrote:
> 
> Perhaps you are looking for something like this:
> 
> query_string = "SELECT AAA FROM BBB WHERE cc = %s"
> 
> then you could do:
> 
> query_function (query_string % `pp`)  # if pp is an integer type
> 
> or
> query_function (query_string % pp)   # if pp is a string type
> 
> or change %s to %d or whatever,  depending on what pp is.

Will the second case actually work? I don't see the % operator adding '
characters to make the query string correct, or dealing appropriately with '
characters inside any string literal. It seems easier to use bind variables, in
my opinion.

Tim Roberts wrote:
> 
> liang at cad.gatech.edu (Shengquan Liang) wrote:
> >
> >BUT, when i tried it this way, the
> >iDC oracle reponded by:
> >
> >  oci.err:(1008,'ORA-1008: not all variables bound\012')
> >
> >what's the problem with that?
> 
> You haven't bound the SQL variable yyyy to any value.  As I said, the
> method for doing that varies with the backend.

Indeed. If the above methods don't work, try using numeric variable names:

  cursor.execute('select * from attt where bbbb = :1', (yyyy,))

  cursor.execute('select * from attt where bbbb = :1 and cccc = :2',
    (yyyy, zzzz))

This definitely worked in previous Oracle modules. Note that a tuple must be
supplied as the second argument.

Paul


From matt at null.net  Mon Jun 19 20:39:50 2000
From: matt at null.net (Matthew Schinckel)
Date: Mon, 19 Jun 2000 18:39:50 CST
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <20000619184000.639516@210.9.22.17>

"William Dandreta" <wjdandreta at worldnet.att.net> wrote:
> I recently read a message that suggested that using regular
> expressions might be faster than the find function. I've heard the
> term regular expression used from time to time but I don't know what
> it means. Could someone give me a simple explanation?
> 
> Bill

I have come across one example where re was much faster than using
string.find: I was stripping out html tags, and doing a one line
re.sub('<*?>','') was much faster than using string.find() and slice
notation.

Try both, use the profiler, and compare. (I'll guess someone else will
give an explanation about them).

Matt.


From dgoodger at bigfoot.com  Sat Jun  3 22:08:22 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sun, 4 Jun 2000 02:08:22 GMT
Subject: exit(); halt; in pyhton
In-Reply-To: <3939B4C1.3FACC11D@tin.it>
References: <3939B4C1.3FACC11D@tin.it>
Message-ID: <B55F3255.458E%DGOODGER@BIGFOOT.COM>

on 2000-06-03 21:45, Jilani Khaldi (jilanik at tin.it) wrote:
> How can I write this Pascal code in python?
> var
> x: double;
> begin
> if x = 0 then halt
> else
> do others things
> end;
>
> Freeing everything.

Does this do the job?

    if x == 0 then:
        import sys  # only if it's not imported by the module itself
        sys.exit()

--
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From ge at nowhere.none  Tue Jun 13 11:49:46 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 13 Jun 2000 15:49:46 GMT
Subject: Which linux distros have Python by default?
Message-ID: <uKs15.437$i06.133733@ptah.visi.com>

I've got a Tkinter-based GUI interface used to do
administrative stuff for a device driver under Linux.  I know
RedHat installs Python and Tkinter by default.  Is there a
handy list somewhere of which Linux distributions can be
expected to have Tkinter installed?

I know it's available on almost all of the distributions, but
I'd like to know which ones install it in the "normal"
configuration that joe-windows-user is going to get if he does
a default workstation or server install.

Had I the spare time and disk space I'd love to install a half
dozen distributions, but resources don't permit that at the
moment.

-- 
Grant Edwards                   grante             Yow!  Hand me a pair of
                                  at               leather pants and a CASIO
                               visi.com            keyboard -- I'm living
                                                   for today!


From dgoodger at bigfoot.com  Sat Jun 10 14:23:21 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 10 Jun 2000 14:23:21 -0400
Subject: namespace question
In-Reply-To: <eviltofu-8E2F45.12073510062000@news.earthlink.net>
References: <eviltofu-8E2F45.12073510062000@news.earthlink.net>
Message-ID: <B567FFD8.4AD2%dgoodger@bigfoot.com>

on 2000-06-10 12:07, Jerome Chan (eviltofu at rocketmail.com) wrote:
> If I do the following:
> 
> import socket
> 
> socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
> 
> how do I call a function from the socket module later on?

Answer: Don't do that. Use another name instead.

However, if you insist on doing that, here's the workaround:

>>> import socket
>>> socket
<module 'socket' (built-in)>
>>> socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
>>> socket
<socket object, fd=4, family=2, type=1, protocol=0>
>>> import sys
>>> sys.modules['socket']
<module 'socket' (built-in)>
>>> sys.modules['socket'].gethostbyname('www.python.org')
'132.151.1.90'



From stephen at cerebralmaelstrom.com  Tue Jun  6 05:08:53 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Tue, 06 Jun 2000 09:08:53 GMT
Subject: Local Databases with Python
References: <iT0%4.36661$WS3.441157@typhoon.we.rr.com> <jg1%4.7291$gr4.288442@news1.rdc1.il.home.com>
Message-ID: <Fc3%4.37636$WS3.443234@typhoon.we.rr.com>

    I will not be using tkinter :) I've a personal bias against it. It is
largely aesthetic, but I do have other issues with it. The only python GUI
that I like is wxPython -- but i'm still going to be using the RAD
capabilities of C++Builder, to get my application to look and behave just as
I want :)

    As for accessing MSAccess with ADO and DAO -- is that a COM derivation?
To distribute an application that uses this technology, what all do I need
to include in it to go about doing this? I'm not sure if its safe to assume
that all machins have the nessecary files, drivers, and so on installed --
and, if not, is it easily obtainable on the internet? For free? And what
about redistribution of said nessecary files? (Beyond win32all and the
python part)

--S

dsavitsk <dsavitsk at e-coli.net> wrote in message
news:jg1%4.7291$gr4.288442 at news1.rdc1.il.home.com...
> I use tkinter for the interface and ms access for the db to run on
win2000.
> there is a tutorial at the python site on using both DAO and ADO from
python
> to connect to access (also in mhammond's book with a typo or two). it is
> quite easy.
> doug
>
> "Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
> news:iT0%4.36661$WS3.441157 at typhoon.we.rr.com...
> >     I'm writing my current project, i'm writting a program primarily in
> > Python, but am embedding it in C++ for a couple reasons. Primarily, I
want
> > the native, fast and flexible GUI that I can get out of using a RAD tool
> > like C++Builder, but also, I am not aware of any easyish ways to have a
> > local database with Python, so will provide the data-access needed via
BDE
> > or some other such tool in C++Builder.
> >
> >     A majority of my actual functionality will be done in Python, and
i'm
> > wondering if I really need C++Builder to get the database functionality
I
> > want. Being Cross-platform doesn't concern me at the moment -- but
> > working -good- and -prettily- (heh) on Win98 is very important to me.
Most
> > of the database interfaces I saw on the Vaults, deal with various
> > Client/Server SQL database engines that are available... but I don't
need
> > that. There will only ever be one connection -- my program -- to this
one
> > database.
> >
> >     Are there any options for a local database which is accessible via
> > Python? Gadfly seems like one, except the idea of storing it all in
memory
> > terrifies me. There may be a LOT of information in this database as time
> > goes on :)
> >
> > Thanks,
> >
> > --S
> >
> >
>
>




From godzilla at netmeg.net  Fri Jun  9 11:05:51 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Fri, 09 Jun 2000 15:05:51 GMT
Subject: HTMLParser: getting one extra space around &gt;/&lt;/^M tags
References: <m21z2grsut.fsf@netmeg.net>
Message-ID: <m266rigaa8.fsf@netmeg.net>

problem solved. i was adding an extra space in by accident using
string.join(XXX)

les schaffer


From shapr at uab.edu  Thu Jun 15 08:12:35 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 15 Jun 2000 12:12:35 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <1ec8zim.1jmu7c7qqgdkwN%bparsia@email.unc.edu>
Message-ID: <3948C878.E97ED299@uab.edu>

Bijan Parsia wrote:

> I haven't *per se*, mostly because I'm not set up to really do XP of any
> kind (being all on my lonesome :)). I have done a bit of *remote* pair
> programming, and that works pretty well in Python (best if you can share
> remote buffers).

I'm looking for people to try VirtualPairProgramming (Python) with me.
The best setup I've come up with so far is:
SpeakFreely
Emacs with make-frame-on-display or screen with -x

How are you doing VirtualPairProgramming?
If you have other solutions, I'd love to see them on the
VirtualPairProgramming page on the Original Wiki.

> With the advent of a complete, accessible PyUnit (on sourceforge), I
> think eXtreme "Pythoning" is ready to take off.

Another shameless plug for the Python Refactoring Browser, which will
have a minimal webpage up in the next few hours: 
http://bicyclerepair.sourceforge.net

> Heck if you can XP in Java, XPing in Python *has* to be good! :)

I completely agree.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From jar at mminternet.com  Tue Jun 13 14:53:19 2000
From: jar at mminternet.com (jim roush)
Date: Tue, 13 Jun 2000 11:53:19 -0700
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com> <etdn1kpn3v3.fsf@w20-575-109.mit.edu>
Message-ID: <qi0dkscdn3fpbtc9kfk11d1t2kem5dhpnf@4ax.com>

On 13 Jun 2000 08:48:48 -0400, Alex <cut_me_out at hotmail.com> wrote:

>
>> pi = 3.141592654
>> radius = input("Enter is the radius: ")
>> circ = 2 * pi * radius
>> print circ
>
>Perhaps the third line should be 'circ = 2 * pi * float (radius)'
>
>As it is, you are multiplying a float by a string, I think.  That should
>be giving you a type error.
>
No, pi is a variable.

>By the way, you can get pi from the math module:
>import math
>print math.pi
>
>It's good not to have constants encoded directly in to your program, in
>case they ever change. :)

I'm aware of that but I'm following a set of exercises from a book as
a means to learn Python.



jar at mminternet.com
James A Roush


From MarkH at ActiveState.com  Tue Jun 27 20:19:16 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 28 Jun 2000 00:19:16 GMT
Subject: PythonWin:: Restart of Interactive Window??
References: <8jau20$lnv$1@news2.iabg.de>
Message-ID: <8wb65.44$Tb7.380@news-server.bigpond.net.au>

Sorry - no way of reloading everything in Pythonwin like you need for
this.

> is there a possibility??? And the reuse of the history (I miss this in
IDLE
> too)

Pythonwin (Ctrl+Up/Down) and IDLE (Alt+N/Alt+P) both have history, but
AFAIK neither are saved across sessions.

> BonusQuestion: Why doesn't the source-tree-navigator show nested classes
> (only parent classes)? Is it difficult to activate it?

Im not sure what you mean by "nested classes".  Note that the class
browser is now showing instances, just the class definition.

Once you have created a class instance, use the "normal" Pythonwin browser
on it - this scavenges inside the instance attributes - it will probably
show you more than the class or modules browsers.

Mark.





From just at letterror.com  Wed Jun 14 04:41:20 2000
From: just at letterror.com (Just van Rossum)
Date: Wed, 14 Jun 2000 09:41:20 +0100
Subject: tuple creation in C extensions
In-Reply-To:  <1011EAC254FBE96B.1C7D23CE0AAFAF2E.34C32DAFEA9F1880@lp.airnews.net>
References: <l03102801b569016fbd23@[193.78.237.174]>
Message-ID: <l03102800b56cf504d1d6@[193.78.237.168]>

At 6:10 PM -0400 13-06-2000, Jeremy Hylton wrote:
>Just van Rossum <just at letterror.com> writes:
>
>> At 11:15 PM +0200 10-06-2000, Peter Schneider-Kamp wrote:
>> >t = PyTuple_New(2);
>> >PyTuple_SET_ITEM(t,0,PyInt_FromLong(a));
>> >PyTuple_SET_ITEM(t,1,PyInt_FromLong(b));
>> >PyErr_SetObject(PyExc_IndexError,t);
>> >
>> >Is there "A Better Way(TM)"?
>>
>> t = Py_BuildValue("(ii)", a, b);
>>
>> Just
>
>This is a shorter-to-write, slower-to-run version.  If you use
>Py_BuildValue, the code has to parse the format string in order to
>build the object.  If you are coding in C because you're trying to
>speed up a speed-critical part of your program, the code you posted is
>the better way.

On the other hand, creating tuples is probably not often something one does
inside an inner loop...

Just





From damonNOdaSPAM at unispace.com.invalid  Thu Jun 15 00:13:53 2000
From: damonNOdaSPAM at unispace.com.invalid (damonr)
Date: Wed, 14 Jun 2000 21:13:53 -0700
Subject: Python??
Message-ID: <061bd950.e89c1b89@usw-ex0106-045.remarq.com>

Hi,

I am running a new site(http://www.devcritic.com) and i was
wondering if you people knew some good Python resource sites
that i could add to the Python part of the site directory:
http://www.devcritic.com/sites/Python/. Feel free to add any
sites by clicking "Add a site" on the side menu or tell me here
and i will add em to there.
Thanks in advance.


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



From embed at NOSPAM.geocities.com  Wed Jun 14 11:56:13 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 14 Jun 2000 11:56:13 -0400
Subject: Waffling between Python and Ruby
Message-ID: <FVN15.3523$227.83704@nnrp1.uunet.ca>

I was looking at the Ruby editor/IDE for windows and was disappointed with
RubyWin, although it is definitely coming along. I noticed the smalltalk
like immediate mode (Type code, hit ctrl-J, results inserted into the
buffer) but did not see any debugging features. Are such features
(debugging, syntax highlight) in the works for this? Is there going to be a
cross-platform gui other than Tcl/TK for Ruby?  I love Python and I love the
wxPython effort, but I also like a few things about Ruby better than python.
They are:

Chief Irritants In Python:

1.  After using Python seriously for 6 months, I must confess I still
dislike the "whitespace as syntax".  I like the fact that Ruby uses block
markers (begin, end) instead of tabs. This means cheifly to me that I don't
need to worry about how my editor handles tabs and spaces.  Having something
look correct on the screen, but not work because of misuse of tabs versus
spaces, or incorrect indentation, is something that still continues to bug
me.

2. Python uses ":" to introduce a new scope, something I find extraneous,
and more irritating than begin/end or Ruby's def and end for functions, for
example.

3. Python is much slower than Ruby, in my informal tests.

So here are the things I'm still wondering, and perhaps there is a "Python
to Ruby FAQ" somewhere to answer it:

1. What is the Ruby equivalent of Python's ability to import a module and
view its contents through code:

            import somemodule
            print dir(somemodule)

2. What is the Ruby equivalent of eval, and exec?

3. Is there a decent Language Reference guide similar to Pythons?

4. Is there a reference guide to writing extensions in C for Ruby similar to
Pythons?

I am by no means going to abandon Python, as it is more mature, it solves
many problems, and Python has extremely advanced Windows integration (the
windows extensions give more COM and DCOM and WinAPI functionality than
Ruby) and a more thorough set of GUI toolkits, and a better set of database
toolkits, and I can live with the whitespace-as-syntax as long as I have the
right kind of editor. Python is truly a joy to work with. However,  I forsee
room for both Python and Ruby in my repertoire in the future, and I see that
Ruby will be a success as both a scripting and an application development
language.

You want a killer Ruby feature?  I suggest static typing and a Ruby to C
converter that accelerates 2x or more over ruby.

Ruby is a great language. If I had to make it even, by criticizing something
in Ruby, it would be that I am not sold on some of the Perl idioms ($_
anyone? Blech). But I like Ruby's syntax, and I don't have to use the perl
headspace if I don't want to. :-)

Warren




From digitig at cix.co.uk  Wed Jun 14 17:31:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 14 Jun 2000 22:31 +0100 (BST)
Subject: How to find out height of a Canvas
References: <393da9e6.79611437@news1.on.sympatico.ca>
Message-ID: <memo.20000614223153.31901A@digitig.compulink.co.uk>

In article <393da9e6.79611437 at news1.on.sympatico.ca>, 
rjroy at takingcontrol.com (Robert Roy) wrote:

> On Sun, 4 Jun 2000 13:53 +0100 (BST), digitig at cix.co.uk (Tim Rowe)
> wrote:
> 
> >In article <eaq_4.3871$JO2.84543 at news2-win.server.ntlworld.com>, 
> >richard_chamberlain at ntlworld.com (richard_chamberlain) wrote:
> >
> >> Hi Tim,
> >> 
> >> from Tkinter import *
> >> root=Tk()
> >> canvas=Canvas(root,height=250,width=100)
> >> canvas.pack()
> >> print canvas ['height']
> >> 
> >> canvas.itemcget has two arguments, the first is the tag or id of a
> >> particular widget on the canvas and the second is the option you 
> > want >back.
> 
> You might want to look at the winfo_height method. Getting the height
> configuration value is not always accurate.
> 

Ah! Thanks. I was wondering why I was occasionally a pixel out here and 
there. I take it there's a winfo_width method, too.


From richard_chamberlain at ntlworld.com  Tue Jun  6 07:57:29 2000
From: richard_chamberlain at ntlworld.com (RS Solutions)
Date: Tue, 6 Jun 2000 12:57:29 +0100
Subject: Magazine Article
Message-ID: <8hiosq$87j$1@gxsn.com>

There is a five page python article in this month's Visual Systems Journal
 a uk based magazine ). Also a link to it from their web site www.vsj.co.uk

Richard






From jerome.quelin at insalien.org  Fri Jun 30 04:15:47 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 10:15:47 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
Message-ID: <962352947.103057778@news.libertysurf.fr>

Hi all,

I was wondering how to pass variable-length keyworded arguments list
to my base-class constructor? Look at the example code below:

-- File test.py
from Tkinter import *
class F(Frame):
    def __init__(self, **args):
        Frame.__init__(self, args)
        # Here my initialisation stuff
F(bg='blue').mainloop()
-- End of file

When doing this, I got an error in the Tkinter module, because it does
not expect a dictionnary for an argument, but rather a list of
keyworded arguments.

If I write:
    def __init__(self, *args):

it doesn't work because I create the class instance with keyworded
arguments:
    F(bg='blue')

and the * (one star) expects just arguments. Therefore, I must use the
** (double star) construct to fetch a dictionnary.
But how to pass this dictionnary to the parent class constructor?

Thanks,
Jerome
--
jerome.quelin at insalien.org



From stevena at neosynapse.net  Sun Jun 18 18:52:44 2000
From: stevena at neosynapse.net (Steven D. Arnold)
Date: Sun, 18 Jun 2000 18:52:44 -0400
Subject: Nth digit of PI
In-Reply-To: <8idoj0$lea$1@nntp.itservices.ubc.ca>
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com>
 <394a5109.80394991@10.1.1.28>
 <394A5795.F4DE98FB@san.rr.com>
 <394a5ed7.83928743@10.1.1.28>
Message-ID: <4.3.1.2.20000618184856.015dbef8@phear.dementian.com>

At 05:36 PM 6/16/2000 +0000, Robert Israel wrote:
>In article <394a5ed7.83928743 at 10.1.1.28>,
>Tim Dixon <tdixon.no at spam.fwi.com> wrote:
> >>
> >>This is a collosally bad idea, IMO.
>
> >Just for curiosity, why?
>
>Because the real key is the number of the digit that you start at, and
>that's pitifully small by cryptographic standards.  The time to generate
>the n'th digit is not much less than the time to generate all the digits
>up to the n'th, given that you have lots of memory available.  So the
>cryptanalyst with a computer somewhat bigger than yours could try all
>the keys you could be using.

Perhaps we could specify a key; the key would be a sequence which we would 
find in pi.  The next bit in pi would be the start of our one-time 
pad.  For example, the key might be the first ten words in today's London 
Times.

This would likely give us a very large n and yet it is still an easy 
requirement to transmit.

Course, this idea too might have holes the size of a Mack truck -- I liked 
the previous idea too until you shot it down. ;-)


--
Steven D. Arnold           Que quiero sera         stevena at neosynapse.net
"If you have  built castles in the air, your work  need not be lost; that
is where they should be.  Now put the foundations under them!" -- Thoreau




From dsavitsk at e-coli.net  Wed Jun 14 11:36:56 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Wed, 14 Jun 2000 15:36:56 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com> <8i6u39$re3$1@nnrp2.deja.com>
Message-ID: <sEN15.1243$dF.71051@news1.rdc1.il.home.com>

the best site i have found for asp help is
http://www.aspmessageboard.com

doug


<millja at my-deja.com> wrote in message news:8i6u39$re3$1 at nnrp2.deja.com...
> Note: I'm reposting because those messed up
> commas are ugly.
>
> I am having a bit of an issue.  My current
> situation is that I am able to use Python with
> ASP and I am able to use VBScript with ASP.
> However, I am still trying to figure out how to
> use the two together.
>
> What is occurring is that I am working on a
> website, most of which is already written with
> VBScript.  I have a rather large function that I
> would like to integrate.  Is there some way to
> call this function from within the VBScript?
>  Or, it would work if I can get numbers,
> generated by the VBScript, into the Python
> function and the results of this function back to
> the VBScript.
>
> Thanks for you time.
>
> jason milliron
>
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From jkraska1 at san.rr.com  Sat Jun 10 01:58:43 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 10 Jun 2000 05:58:43 GMT
Subject: windll
References: <3941ACC5.305B8830@videotron.ca> <3941B6B4.264FA7A1@san.rr.com> <l8j05.833$cQ.10319@news-server.bigpond.net.au>
Message-ID: <3941D9BD.46BBEF0C@san.rr.com>

> > > Is there someting like to "windll" to use
> > > dynamic librairy under Linux/Unix plateform.
> >
> > What does windll do? "Using" dynamic libraries
> > under Unix is as easy as linking with them and
> > then making certain that they are in your
> > LD_LIBRARY_PATH.
> 
> Using them from Python is that easy?  Really?
> 
> windll tries to make using arbitrary DLLs, *from Python* that simple!

Yeah, I had a discussion off line with someone else.
For some reason I didn't consider that the person meant
a python module. Anyway, what you're looking for on unix
is "dl", as in:

import dl



C/


From wjdandreta at worldnet.att.net  Wed Jun 21 02:30:09 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 06:30:09 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <20000620184723.A26354@activestate.com>
Message-ID: <RhZ35.11602$Uw3.753381@bgtnsc04-news.ops.worldnet.att.net>

Thanks Trent that works.

Bill
Trent Mick wrote in message <20000620184723.A26354 at activestate.com>...
>
>(elem0, elem1, elem2, elem3) = range(4)
>





From istevens at calum.csclub.uwaterloo.ca  Tue Jun  6 20:07:38 2000
From: istevens at calum.csclub.uwaterloo.ca (ian stevens)
Date: 7 Jun 2000 00:07:38 GMT
Subject: Network byte ordering question ...
References: <CC59868A0A76D311B1F50090277B321C1F3394@VRTMAIL>
Message-ID: <8hk3oa$pht$1@watserv3.uwaterloo.ca>

In article <CC59868A0A76D311B1F50090277B321C1F3394 at VRTMAIL>,
Mike Fletcher  <mfletch at tpresence.com> wrote:

>( val1, val2, val3, val4, val5 ) = unpack( ">16sBBBi", header )

Thanks!  Actually, the following works better:

    ( val1, val2, val3, val4, val5 ) = unpack( "<16sBBBI", header )

I tried this both on an Intel and a Sun machine and they both return
what I expect.  I also forgot to mention that the last value was
unsigned.

>Yay! another Waterloovian Pythonista!

Actually, ex-Waterloo ... graduated last year.


thanks again,
ian.


From alwalker at mars.ark.com  Tue Jun 13 11:32:15 2000
From: alwalker at mars.ark.com (Alex Walker)
Date: Tue, 13 Jun 2000 15:32:15 GMT
Subject: Paring down Python (physically)
Message-ID: <39465328.2352344@news.ark.com>

Hi All

I've been tasked with a question of how to use Python on a WinNT
system where isn't installed.  Of the options available (and I don't
thing I've discovered them all yet,) installing with our app just
those Python files absolutely required to run a script is the
front-runner.  I've looked all over the Python web site, the docs, and
my books but haven't come across any info the help me pare down Python
to a size, other than just hacking off modules I'm not calling.   Any
suggestions on a better way or where next to look for helpful info is
appreciated ?   Thanks!



From petersc at stud.ntnu.no  Mon Jun 12 10:43:37 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Mon, 12 Jun 2000 16:43:37 +0200
Subject: Sorts
References: <20000612142535.957.qmail@web1505.mail.yahoo.com>
Message-ID: <3944F719.639A0472@stud.ntnu.no>

Robin Porter wrote:
> 
> Does anyone know what type of sort is used for the internal sort
> function in Python (quick, merge, insertion, etc.)?

The following assumes that we talk about lists in the standard
(C) Python edition.

binary insertion sort is used for small lists (<100 elements) whereas
a quicksort derived sorting algorithm called "samplesort" is used for
larger lists. A description can be found in the list object
implementation:

   samplesort is basically quicksort on steroids:  a power of 2 close
   to n/ln(n) is computed, and that many elements (less 1) are picked at
   random from the array and sorted.  These 2**k - 1 elements are then
   used as preselected pivots for an equal number of quicksort
   partitioning steps, partitioning the slice into 2**k chunks each of
   size about ln(n).  These small final chunks are then usually handled
   by binarysort.  Note that when k=1, this is roughly the same as an
   ordinary quicksort using a random pivot, and when k=2 this is roughly
   a median-of-3 quicksort.  From that view, using k ~= lg(n/ln(n))
makes
   this a "median of n/ln(n)" quicksort.  You can also view it as a kind
   of bucket sort, where 2**k-1 bucket boundaries are picked
dynamically.

   The large number of samples makes a quadratic-time case almost
   impossible, and asymptotically drives the average-case number of
   compares from quicksort's 2 N ln N (or 12/7 N ln N for the median-of-
   3 variant) down to N lg N.

quite-a-hack-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From cg at gaia.cdg.acriter.nl  Sat Jun 17 06:02:55 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 17 Jun 2000 12:02:55 +0200
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>
Message-ID: <8ificf$tke$1@gaia.cdg.acriter.nl>

John Schmitt <jschmitt at vmlabs.com> said:
>I'd love to take the diversion of writing Make using Python.  The things
>that are keeping me from doing this are, in order:
>
Check out Ant (at jakarta.apache.org). It's a Make replacement, in Java,
that has XML syntax and lets you build Java stuff reasonably
platform-indepentend. Now, I think XML syntax for a makefile is an
astonishingly stupid idea, but as it is wildly succesful you might as well
follow the lead. At the very least, you may get some good ideas from the
architecture, it seems well done.

On the other hand, check Perl. The makemake script is quite a nice idea 
and deals with Makefiles as an output format (which is very satisfying
for lots of people :-)). Similarly, automake, which has an input syntax
closer to makefiles but similarly hides platform differences from you
(although I'm not sure that automake has been ported beyond Unix).

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From wtanksle at dolphin.openprojects.net  Tue Jun  6 15:45:03 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 06 Jun 2000 19:45:03 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net> <393C81D0.36732896@Lugoj.Com>
Message-ID: <slrn8jql5v.6d4.wtanksle@dolphin.openprojects.net>

On Mon, 05 Jun 2000 21:45:04 -0700, James Logajan wrote:
>William Tanksley wrote:
>> And for case sensitivity it _just doesn't matter_.  The only people who
>> are strongly helped or hurt by case sensitivity appear to be newbies, and
>> they're being hurt.

>Physically hurt? Emotionally hurt? Psychologically scarred?

You know, this is frustrating.

>After thousands of programmers have learned Python doesn't anyone find it
>strange that this issue should suddenly become THE hot issue that prevents
>Pythonic world domination? Is this a plot by Perl advocates?

I'm not the one that's claiming that removing case-sensitivity will ruin
Python.  My claim is very simple: first, it's a minor change for us to
make, and second, it helps newbies.

For some reason, there are a lot of people here who seem to believe that
case insensitivity is truly horrible, and would result in job loss.

-- 
-William "Billy" Tanksley


From sdhyok at email.unc.edu  Sat Jun  3 15:56:17 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Sat, 3 Jun 2000 15:56:17 -0400
Subject: Automatic reload()?
Message-ID: <8hbnt1$62i$1@news2.isis.unc.edu>

Is there no program to reload all the changed modules automatically?

Peter




From digitig at cix.co.uk  Sun Jun  4 08:53:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 4 Jun 2000 13:53 +0100 (BST)
Subject: How to find out height of a Canvas
References: <eaq_4.3871$JO2.84543@news2-win.server.ntlworld.com>
Message-ID: <memo.20000604135335.59489B@digitig.compulink.co.uk>

In article <eaq_4.3871$JO2.84543 at news2-win.server.ntlworld.com>, 
richard_chamberlain at ntlworld.com (richard_chamberlain) wrote:

> Hi Tim,
> 
> from Tkinter import *
> root=Tk()
> canvas=Canvas(root,height=250,width=100)
> canvas.pack()
> print canvas ['height']
> 
> canvas.itemcget has two arguments, the first is the tag or id of a
> particular widget on the canvas and the second is the option you want 
> back.

Thanks -- it looks as if I was making things much more complicated than 
they need to be. Situation normal!


From victor at idaccr.org  Tue Jun 27 10:46:05 2000
From: victor at idaccr.org (Victor S. Miller)
Date: Tue, 27 Jun 2000 14:46:05 GMT
Subject: float literal too large
Message-ID: <ulsntz2kra.fsf@runner.princeton.idaccr.org>

I just wrote a python program to read output produced by a Fortran
program that somebody had written.  Much to my surprise I got the
exception:

float() literal too large.

The literal in question was '0.23489798479873E-2325' (this was
generated on a Cray computer).  I couldn't find anything in the
documentation about this restriction.  Even more puzzling was the fact
that eval() worked just fine (it gave the answer 0.0 which is
perfectly acceptable to me, given that the platform on which I was
running had less precision than the platform that generated it).  Now,
on general grounds I would rather convert this using float, because I
had a statement like

a[i,j] = float(s)

where a is a numerical python array.  So, is there some way around
this?  How do I find out what the restrictions on literals are?

I'm running:

Python 1.5.2 (#4, Nov 23 1999, 14:57:03)  [GCC 2.7.2.2] on sunos5
-- 
Victor S. Miller     | " ... Meanwhile, those of us who can compute can hardly
victor at idaccr.org    | be expected to keep writing papers saying 'I can do the
CCR, Princeton, NJ   | following useless calculation in 2 seconds', and indeed
    08540 USA        | what editor would publish them?"  -- Oliver Atkin


From ajs at ix.netcom.com  Thu Jun  1 21:14:53 2000
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Thu, 1 Jun 2000 21:14:53 -0400
Subject: *blink* That's neat.
References: <GXJY4.28606$WS3.320758@typhoon.we.rr.com> <SZKY4.6213$wYl.227247104@newsb.telia.net> <8h07fv$bnc$3@newshost.accu.uu.nl> <UINY4.6227$wYl.223307264@newsb.telia.net> <8h0ne2$70i$2@newshost.accu.uu.nl> <3934242B.ECFDBAAE@endea.demon.nl> <8h1k0q$h04$2@newshost.accu.uu.nl> <wUfZ4.6380$wYl.229799424@newsb.telia.net> <8h4djp$pkc$1@newshost.accu.uu.nl> <jXnZ4.6415$wYl.227858944@newsb.telia.net> <8h6p4n$ecq$1@newshost.accu.uu.nl>
Message-ID: <8h72ji$32e$1@slb0.atl.mindspring.net>

> Martijn Faassen wrote:
>I was  imagining some controlled circumstances where there > aren't any
> helpers around when the newbies get down to the actual business of
> programming.

Why  the insistence on the fiction that the Alice studies were
conducted in the context of learning programming?

Talk about virtual worlds.

comp.lang.python has become one, IMO.

Congratulations to all.

Off  to build joystick IDE.

Bye-bye.





From nospam at bitbucket.com  Tue Jun 20 01:20:46 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Mon, 19 Jun 2000 22:20:46 -0700
Subject: Access violation when no memory
Message-ID: <olD35.82$_%.4481@newsfeed.avtel.net>

The following program progressively allocates all memory:
    size = 1 << 20
    list = [None,]
    while size:
        try:
            mem = [None] * size
            mem[0] = list
            list = mem
        except:
            size = size / 2

It fails with an access violation.  (Turn off virtual memory
to provoke it.)  This is not a theoretical situation for me:
I have a beta release in the field that is crashing like this.
The immediate cause is this:

1. we are handling a MemoryError
2. the following call cannot allocate a traceback object
     ceval.c line 1800:    PyErr_Fetch(&exc, &val, &tb);
3. the NULL pointer at tb is pushed here:
     ceval.c line 1816:    PUSH(tb);
     ceval.c line 1817:    PUSH(val);
     ceval.c line 1818:    PUSH(exc);
4. later it is popped here and its ref count is decremented:
     ceval.c line 682:    case POP_TOP:
     ceval.c line 683:        v = POP();
     ceval.c line 684:        Py_DECREF(v);
     ceval.c line 685:        continue;

This is with 1.5.2 but the crash still happens with 1.6a2.
At this stage I am way out of my depth, and  the only way I
see to fix this is to change Py_DECREF to Py_XDECREF, which
is a (minor?) inefficiency. Can anyone recommend a better way?

And should I report this at http://www.python.org/python-bugs?
I do not have access to the current CVS tree.

TIA, Phil Mayes
--
pmayes AT olivebr DOT com





From jerry_spicklemire at my-deja.com  Fri Jun 23 18:50:48 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Fri, 23 Jun 2000 22:50:48 GMT
Subject: List without duplicates?
References: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net> <394FFA1A.FBF8F81D@roguewave.com>
Message-ID: <8j0pk3$ot9$1@nnrp1.deja.com>

Greetings Gurus and Fellow Wannabes,

Just a minor "heads up" for a possible "gotcha".

It looks to me as if this method, though speedy, will return the
cleaned list in a much different order than the original, since the

  dict.keys()

method outputs in the (seemingly random) order of the dicts internal
hash table.

The moral is, if the order of the list members matters, be sure to wait
until after the cleaning step to sort, or at least remember to re-sort!

Later,
Jerry S.

In article <394FFA1A.FBF8F81D at roguewave.com>,
  Bjorn Pettersen <bjorn at roguewave.com> wrote:
> William Dandreta wrote:
> >
> > Is there a built in Python function that eliminates duplicates in a
list?
> >
> > I have been sorting the list and compairing adjacent elements and
if equal
> > deleting one:
> > -------------------------------
> > mylist.sort()
> >     j = len(mylist)-1
> >     while j > 0:
> >       if mylist[j] == mylist[j-1]:
> >         del mylist[j]
> >       j = j - 1
> > ------------------------------------
> > I recently tried using a dictionary. The list is the keys and I put
and
> > empty string for the entry
> > --------------------
> > mydict = {}
> > for e in mylist:
> >   if mydict.has_key(e): continue
> >   else: mydict[e] = ''
> > ------------------------------------
> > Is ther a better way?
> >
> > Bill
>
> This should probably be a faq by now...:
>
> def unique(lst):
>     d = {}
>     for item in lst:
>         d[item] = None
>     return d.keys()
>
> -- bjorn
>
>


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


From hungjunglu at hotmail.com  Thu Jun 15 16:59:22 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Thu, 15 Jun 2000 13:59:22 PDT
Subject: Python Productivity over C++ (Encapsulation)
Message-ID: <20000615205922.99304.qmail@hotmail.com>

>Then why does Python support encapsulation that is just as strict as C++? 
>Guido
>van Rossum - exposed as a traitor to the people's revolution all along!

Ha. Then why does Guido explicitly say in the manual that these variables 
are still accessible? Just to be nice, I guess? :)

>I can easily subvert encapsulation in many C++ classes simply with the line
>"#define private public" before I include any header files. Perhaps this 
>means
>that C++ will in fact lead to the overthrow of evil and oppression in the 
>world?

Sure. What do you think I did when it was 3:00 AM I needed to ship a product 
the next day and the engine programmers weren't available to change and 
recompile the libraries? (Well, it was not 3:00 AM nor the next day, but you 
get the idea.) Subversion, of course. I'll let you figure out how to do 
things without using "#define private public" yet still be able to access 
private data members. In my world there are no variables that can't be made 
public, ha! The thing is, if you have to constantly recurr to subversion to 
add functionalities, there is a problem with the language.

Yeah yeah, of course people later were shocked (and one actually screamed) 
at how I operated. But sorry, do you guys want the product shipped, or miss 
the season and lose millions of dollars?

There have been just way too many times when I needed extra functionality 
and the engine team just couldn't respond in time. Why? Because some of the 
codes were written a long time ago, and it takes them (engine programers) a 
long time to warm up and remember all the details. If I have to wait a few 
weeks for the changes to come, we'd've missed the deadlines. By nature you 
don't want to release a new version of the engine per each tiny 
change/request, right? But then, how do I add functionalities when I need 
them urgently? This kind of situations would've NEVER happened with Python, 
where your hands are free to add more functionalities.

Then again, I guess you've never been handcuffed. :) Have you ever worked in 
a corporate world? (Or am I again talking to a newbie without real-world 
experience?)

>Also, remember that Java's encapsulation isn't exactly inviolable. It can't 
>be.
>Think about it.

Done that thinking. :)

regards,

Hung Jung


________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From daniel.dittmar at sap.com  Wed Jun  7 11:34:42 2000
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Wed, 07 Jun 2000 17:34:42 +0200
Subject: Embedding + extending without DLLs
References: <wkbt1da7w2.fsf@turangalila.harmonixmusic.com>
Message-ID: <393E6B92.8D4CE23C@sap.com>

Have you tried calling the initxxx function of your extension directly?
This should register your extension as a module. Afterwards, the Python
interpreter knows it's already loaded.

-- 
Daniel Dittmar
SAP AG, Basisentwicklung Berlin
daniel.dittmar at sap.com


From ychoi at pipeline.com  Sun Jun  4 00:22:58 2000
From: ychoi at pipeline.com (Yoon S. Choi)
Date: Sun, 04 Jun 2000 04:22:58 GMT
Subject: instantiate multiple automation objects
Message-ID: <3939d663.1392772@nntp.pipeline.com>

I am trying (without much luck) to create multiple instances of an
automation object.  I can create 1 instance correctly using the
function call win32com.client.Dispatch().  However, when I use the
same function to try and create another automation object, it just
refers to the first instance.

Here is an example using Microsoft Word as the automation object:

import win32com.client
w1 = win32com.client.Dispatch( "Word.Application")
w2 = win32com.client.Dispatch( "Word.Application")
w1.Visible = 1  # Microsoft Word becomes visible
w2.Visible = 1  # Nothing happens because w2 is referring to the same
object as w1

How can I for example create another instance of Microsoft Word?  In
other words, w1 and w2 would refer to two separate automation objects
of the same type.
Can I do this without having to resort directly to the pythoncom
module and its associated CoCreateInstance and CoGetClassObject
functions?                        


From PClaerhout at CREO.BE  Thu Jun  1 08:02:32 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 1 Jun 2000 14:02:32 +0200 
Subject: Getting the directory size
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130118F@msgeuro1.creo.be>

Hi all,

I want to be able to find the total size of a directory on my NT machine,
and I started with writing it using the os.path.walk function, and it looks
something like this:

<code>
import os
import sys

def calcDirSize(arg, dir, files):
	for file in files:
		stats = os.stat(os.path.join(dir, file))
		size = stats[6]
		arg.append(size)

def getDirSize(dir):
	
	sizes = []
	os.path.walk(dir, calcDirSize, sizes)
	total = 0
	for size in sizes:
		total = total + size
	if total > 1073741824:
		return (round(total/1073741824.0, 2), 'GB')
	if total > 1048576:
		return (round(total/1048576.0, 2), 'MB')
	if total > 1024:
		return (round(total/1024.0, 2), 'KB')
	return (total, 'bytes')

def main():
	
	dir = sys.argv[1]
	
	print "Testing directorySize..."
	print "Directory: %s" %(dir)
	print "Size: %s %s" %(getDirSize(dir)[0], getDirSize(dir)[1])

if __name__ == '__main__':
	main()
</code>

Is there any way to fasten up this beauty, because depending on the
number of directories it has to walk through, it takes a long time. Is
there maybe a function in the win32 modules who does this for me,
but lots faster?

Kind regards,


Pieter Claerhout



From shecter at darmstadt.gmd.de  Fri Jun 16 08:52:10 2000
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Fri, 16 Jun 2000 14:52:10 +0200
Subject: Suggestion: "Completing" HTTP support in Python.
References: <3949028F.3B512AF9@darmstadt.gmd.de> <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru> <20000615213621.F30050@exeter.exeter.org>
Message-ID: <394A22FA.28E28EAB@darmstadt.gmd.de>

"Karl G. Ulbrich" wrote:

> Note that POST support in urllib lacks support for posting form contents
> that include files to be uploaded.  That means you're back to down and
> dirty with the HTTPlib and some ugly http and mime boundary syntax.

This would be a cool feature, too.

- Robb



From dalke at acm.org  Thu Jun  8 12:19:18 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 8 Jun 2000 10:19:18 -0600
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <20000608122619.I469@xs4all.nl>
Message-ID: <8hogq2$cce$1@slb6.atl.mindspring.net>

Thomas Wouters :
>On Thu, Jun 08, 2000 at 01:47:14AM +0000, Huaiyu Zhu wrote:
>> My wishlist for Python3K (This is not to start a flame war):
>
>> 2. Augmentation operators: +=, *=, etc.  Dot operators .*, ./, etc (for
>>    numerical computation to distinguish matrix and elementwise
operations).
>
>This needn't wait for Py3K, and in fact I'm writing code for this right
now,


Would there be a problem with the ambiguity of "3./2" being parsed
as either
  '3.' '/' '2' == 1.5
  '3' './' '2' == whatever ./ does to scalar integers, probably
                  make it be regular division == 1
  ?

The parser is greedy so will make it be the first option, 1.5, which is
the expected behaviour.  But if the '2' is actually the variable 'x'
which is an array, there could be some confusion.   Eg
  x = [1. 2.]
      [3. 4.]
  3./x     ---> should yield [3. 1.5]  or (3.) * (invert_matrix(x)) ?
                             [1. .75]

                    Andrew
                    dalke at acm.org





From arta at NOSPAMx-stream.nl  Mon Jun 26 12:27:06 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Mon, 26 Jun 2000 18:27:06 +0200
Subject: socket-module
Message-ID: <eOU#4853$GA.336@net003s>

Hi,

I'm having a question. Maybe a strange one, but don't blame me. Just
a newbie or something like that :-)

If you're looking at the source from SOCKET.PY, you see a
two classes. One for the socket-object and one for the file-descriptor.
But, this module (in Windows) doesn't import anything.
And par example:

def accept(self):
    sock, addr = self._sock.accept()
    return _socketobject(sock), addr

This is the accept method. Where can I see the lowest-level-Python-code?
I can't see the real code from HOW it's accepting the connection.
Is this possible and how? It's basically for me to see the TCP/IP-packages
Python is creating....(very low level)

A[r]TA

--
-Life can kick your ass, but all you got to do is stand up and move on...





From stephen at cerebralmaelstrom.com  Thu Jun 22 05:58:57 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 22 Jun 2000 09:58:57 GMT
Subject: Manipulating specific bits
Message-ID: <Brl45.80022$WS3.716680@typhoon.we.rr.com>

    Okay, this is a Very Newbie-ish question.

    What I need to do is manipulate specific bits in a byte, for
transmission. I have absolutely *no* idea how to do this. :)
    For example, say I have my 1 byte: 0 0 0 0 0 0 0 0. Say I want to set
the 2nd and 5th bit to 1.. how the heck do I do that? :)
    I've never actually even /blinked/ in the direction of bit-ish
manipulation, so am utterly clueless. Next i'm gonna have to stick it into a
struct I think, before transmission.. I think. dunno. Will look into that
afterwards.

Thanks. :)

--S




From cba at mediaone.net  Fri Jun 16 19:04:54 2000
From: cba at mediaone.net (Craig Allen)
Date: Fri, 16 Jun 2000 19:04:54 -0400
Subject: free python hosting?
References: <8geo5g$3rh$1@bob.news.rcn.net> <39487022.5C2D0F11@stud.ntnu.no>
Message-ID: <394AB296.26209FA3@mediaone.net>

Peter Schneider-Kamp wrote:

> now-if-I-could-get-free-zope-hosting-ly y'rs

Check out http://www.zope.org/Members/NIP/FreeZope

NIP also operates searchable archives of Zope mailing lists that more usable
than the one at eGroups.  It's Domino-based, which means that unlike most search
engines on the web, a string of words is treated as a phrase.

Craig
-- 
Craig Allen  - Managing Partner - Mutual Alchemy
Web Architecture - http://alchemy.nu



From rdejuana at my-deja.com  Thu Jun 29 18:09:30 2000
From: rdejuana at my-deja.com (rdejuana at my-deja.com)
Date: Thu, 29 Jun 2000 22:09:30 GMT
Subject: running a python function from c
Message-ID: <8jghed$17u$1@nnrp1.deja.com>

I'm working right now on try  to get a python function run from c.  I
looked at the tutorial (and it said it was easy), but I'm still not
sure how i run a function i've defined.  Where does it get loaded into
my c code.  Where do i tell it what .py script to hit?  If someone has
a simple example this working.  Can i borrow the code to see how it is
done?

thanks
-rod

rdejuan at us.ibm.com


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


From aahz at netcom.com  Mon Jun 12 17:22:58 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jun 2000 21:22:58 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu>
Message-ID: <8i3kbi$3a0$1@slb1.atl.mindspring.net>

In article <Pine.GSO.4.21.0006121559430.13331-100000 at y.glue.umd.edu>,
Roy Katz  <katz at Glue.umd.edu> wrote:
>
>I wanted to test this out, so I built a project in Python.  It is a small
>CPU emulator, implementing twelve or so basic instructions (push,
>pop, mov, add, interrupt-execute, etc...).  The time invested in its
>Python build amounted to two hours.  Mostly of the time was spent writing
>new code. 
>
>Then I moved it to C++.  In a nutshell, this required re-writing
>everything using STL.  This took me about twice as long to implement (not
>counting nitpicking over the syntax errors).
>
>Then I move to C.  The process took about twice as long as the C++
>attempt; not counting the time it took for me to write miscellaneous
>libraries (stack, memory allocation), I noticed that I was spending far
>more time hunting pointer bugs and syntax errors than writing new
>code. oThis was the longest port; it took me about four times longer than 
>it did in Python. 
>
>At the same time, C was faster than (straight) Python by a factor of
>ten.  Whatever, I now have three implementations of a virtual CPU to play 
>around with :)

The real moral is that, had you invested the same amount of time
speeding up your Python code (possibly by writing part of it in straight
C), you almost certainly would have received a speedup of somewhere
between two and a hundred times -- so at worst your C program would have
been five times faster and at best your C program would have been ten
times slower.  (Best and worst being relative to the Python program, of
course.  ;-)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From paraguana at my-deja.com  Wed Jun 14 09:02:03 2000
From: paraguana at my-deja.com (paraguana at my-deja.com)
Date: Wed, 14 Jun 2000 13:02:03 GMT
Subject: Wondering about Domingo's Rational Mean book"
References: <pn5djsceirmrdhdd2ncht9c26hv0q5gkdq@4ax.com> <ralyh3rlx4xa@forum.mathforum.com> <8hp1g8$47o$1@barcode.tesco.net> <8i475e$48n$1@nnrp1.deja.com> <8i5vsn$llr$1@epos.tesco.net>
Message-ID: <8i7vo0$p8v$1@nnrp1.deja.com>

In article <8i5vsn$llr$71 at epos.tesco.net>,
  "Iain Davidson" <Sttscitrans at tesco.net> wrote:
>
> > "Iain Davidson" <Sttscitrans at tesco.net> replied:
> >
>
> It just seems to be a variant on the method used in schools to find
square

That _very particular_ example of rational process for approximating
the square root has been analyzed by using the Mediant, by many
mathematicians as I previously pointed out, i.e: Chuquet, J. Ortega,
Heath, etc.

> The same idea can obviously be extended to
> higher degree polynomials but conditions for
> convergence may be complicated.

Of course!, it is pretty obvious how to extend such very particular
case of the new general and fundamental rational process concept.
WORST!,  it is pretty obvious how to find Bernoulli?s, Newton and
Halley?s method (among many other new algorithms) by agency of
this fundamental and general principle of Arithmetic: The Rational Mean.
No Geometry, no decimals, no derivatives, no Cartesian system.

Of course, that?s pretty obvious!

So,
If you have some doubts, something to say,  about the rational mean and
the rational process, then your arguments are straightaway directed to
the CFs, Bernoulli?s, Newton and Halley?s method, because all of them
are ruled by the rational mean as has been trivially and
obviously stated in my book and summarized in my home page:
www.etheron.net/usuarios/dgomez/

All of them are just rational processes ruled by the rational mean.

Of course it is fairly obvious!. That?s why I find of so much
importance to consider the striking fact that western culture needed
more than two or three thousands years to find Bernoulli?s, Newton?s
and Halley?s methods by means of the infinitesimal calculus and the
cartesian system, also that all  their most outstanding mathematicians
never realized that such methods could have been  easily developed by
using simple arithmetic.

Just striking, indeed, specially for a humble purchaser
of "rigurous" math books containing so much rigorous talk about best
approximations and roots solving.

Now, considering that this not a one-way discussion I?m sure you
will be so kind and take some time to formulate here those CFs method
you mentioned for finding a sequence of best approximations for the
cube root of any number, I?m sure it will be amusing and very useful
not only for me but for the whole audience of this newsgroup.


Also I would like to get any response about the other topics
I have mentioned on the precedents and history of developing
Bernoulli?s, Newton?s, and specially Halley?s method
by means of simple arithmetic, if I?m wrong on this then
there must be, at least, two or three thousands years, sorry,
I meant to say "two of three thousands references" (books, articles,
etc) on such a very simple arithmetical matter.
:-)


Domingo










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


From cjc26 at nospam.cornell.edu  Sat Jun  3 10:51:16 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sat, 03 Jun 2000 14:51:16 GMT
Subject: Marking hyperlinks in a Text widget
References: <8h04jo$lgh$1@zingo.tninet.se> <oq7lcc41zr.fsf@titan.progiciels-bpi.ca> <8h0kn5$6pv$1@cubacola.tninet.se> <slrn8j94er.ebv.cjc26@localhost.mindriot.net> <8h4d2e$ap4$1@cubacola.tninet.se> <slrn8jblag.jua.cjc26@localhost.mindriot.net> <cjmbjsoask73rledo0kefavcrf60cangeq@4ax.com>
Message-ID: <slrn8ji6u7.5b1.cjc26@localhost.mindriot.net>

* Ben Wolfson <rumjuggler at cryptarchy.org> menulis:
| 
| On windows you can just do 
| 
| netscape www.python.org
| 
| from the command line.  If you do

Great.  If you do this on unix it puts up an obnoxious error message
saying that another instance of netscape is already running.


| netscape -remote 'openURL(www.python.org)'
| 
| it will look for "www.remoteopenurlwww.python.org"

let's-hear-it-for-cross-platform-compatibility-ly y'rs,


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From akuchlin at mems-exchange.org  Thu Jun 22 12:12:33 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 22 Jun 2000 09:12:33 -0700
Subject: curses input with threads
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <394AD545.AC791A2B@san.rr.com> <394AF6C8.FFCBF182@houston.rr.com> <Pine.GHP.4.21.0006211541260.6090-100000@don.et.byu.edu>
Message-ID: <3d8zvxelu6.fsf@mems-exchange.org>

"C. Porter Bassett" <porter at et.byu.edu> writes:
> screen.  I want to add a pause function.  I tried creating a separate
> thread that runs stdscr.getch() to catch the keyboard entries.  But when I
> do that, the entire program seems to pause to wait on that
> getch() function.  

The problem is that the curses module doesn't contain Py_BEGIN_THREAD
/ Py_END_THREAD calls around the call to getch(), to release the
global interpreter lock and let other threads run.  This is a bug
that's worth fixing for 1.6.

Which curses calls should release the interpreter lock?  All of them?
The ones which can actually pause for a long time?  getch() and
refresh() are the two obvious ones.  Adding BEGIN_THREAD/END_THREAD
refs to the various macros would take care of lots of other functions,
but I'm doubtful that it's worth the effort.

I'll try to check in this fix tonight; in the meantime you can add the
BEGIN_THREAD / END_THREAD to cursesmodule.c yourself.

--amk


From aruth at intercation.com  Tue Jun 20 20:34:28 2000
From: aruth at intercation.com (Adam Ruth)
Date: Tue, 20 Jun 2000 18:34:28 -0600
Subject: "TypeError: read-only buffer, None" returned from file.write...
References: <MIS35.664$PQ4.13634@news1.online.no>
Message-ID: <8ip2q3$2jrk$1@news.aros.net>

I get this all the time too, simple error.  Somewhere you assigned sBuffer
to None.  It happens to me a lot when I call functions that return strings
or return None when there is a problem.  You can confirm this by writing:

f.write(str(sBuffer))

The file then should contain the word 'None'.

--
Adam Ruth
InterCation, Inc.
www.intercation.com


"Dag Sunde" <dag at orion.no> wrote in message
news:MIS35.664$PQ4.13634 at news1.online.no...
> I have the following code-snippet that retuns a string:
>
>    result = m_XMLDoc.transformNode(xslDoc)
>
> ...and the string shows the expected result when printed,
> but when I try to save the string to a file with the following
> code:
>
>   f = open( outFile, 'w')
>   f.write(sBuffer)
>   f.close()
>
> I get this error:
>
> Traceback (innermost last):
>   File "f:/apache/cgi-bin/getfinalconfirmation.py", line 142, in
> saveConfirmedData
>     f.write(sBuffer)
> TypeError: read-only buffer, None
>
> Can anyone give me a hint?
>
> TIA...
>
> Dag Sunde
>
>
>




From tim_one at email.msn.com  Sun Jun 18 17:20:05 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 18 Jun 2000 17:20:05 -0400
Subject: Rude question about BeOpen.com
In-Reply-To: <Pine.HPX.4.10.10006181239180.21364-100000@merle.acns.nwu.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEOEGEAA.tim_one@email.msn.com>

[posted & mailed]

[Nicholas Geovanis]
> It seems then that the attempt to obtain US government funding for
> CP4E has been abandoned,

It did get some govt (DARPA) funding, but not enough to get very far.  Last
I heard from Guido, he figured it had a better chance of attracting adequate
funding by expanding the net beyond govt sources.  The move to BeOpen can't
be viewed as negative from that viewpoint.

> and therefore that there is no upcoming funding decision to be made
>(possibly could have elicited an answer of "no" from them). True or
> false?

Yes or no <wink -- but I can't parse your question, although perhaps that's
because it's based on a faulty assumption:  all funding decisions that were
to be made have already been made>.

> I need to know because I need to forward the information to a local
> school.

Can only echo Thomas's suggestion that you ask on the Edu-SIG instead.
Guido is recovering from his honeymoon now, and answering a CP4E question
may get him back in tune with reality <wink>.

honeymoons-for-everyone-would-have-been-more-popular-ly y'rs  - tim





From Johannes.Nix at mail.uni-oldenburg.de  Fri Jun  2 15:05:15 2000
From: Johannes.Nix at mail.uni-oldenburg.de (Johannes Nix)
Date: 02 Jun 2000 21:05:15 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <20000522183552.A29224@mudge> <3929D3F6.DE9CCCBC@prescod.net> <3929F215.166D@esatclear.ie>
Message-ID: <8h90hb$5e1@news.Informatik.Uni-Oldenburg.DE>

Russell Wallace <rwallace at esatclear.ie> writes:

> It seems to me that the important argument in this context is "case
> insensitivity will break millions of lines of existing code, and perhaps
> split the user base if a significant number of existing users refuse to

It is a trivial task to convert case-sensitive identifiers in unique
case-insensitive identifiers in case of a name clash.

Johannes



From thomas at xs4all.net  Mon Jun 12 15:24:52 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 12 Jun 2000 21:24:52 +0200
Subject: Does Python help with the no-Unix handicap?
In-Reply-To: <hs815.297$i06.89565@ptah.visi.com>; from ge@nowhere.none on Mon, Jun 12, 2000 at 04:45:01PM +0000
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <hs815.297$i06.89565@ptah.visi.com>
Message-ID: <20000612212452.P20700@xs4all.nl>

On Mon, Jun 12, 2000 at 04:45:01PM +0000, Grant Edwards wrote:

> >Is it's base structure less related to any one specific operating
> >system?

> Perl isn't so much related to an OS, as it is to some other
> utilities (awk,sed,grep) and languages (bourne shell).

Yeah, right, and a chair isn't geared as much to a human as to a one-to-two
meter tall, bi-legged creature with two arms, a flat back and bends in all
the right places ;)

I guess it depends on your definition of OS, and of UNIX ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From shouldbe at message.com  Fri Jun 30 08:49:42 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 30 Jun 2000 12:49:42 GMT
Subject: Is this a leaker?
Message-ID: <GH075.17918$NP5.455672@newsread2.prod.itd.earthlink.net>

 comm =PyObject_GetItem(args, Py_BuildValue("i",2) );
Is the Py_BuildValue leaking, because I don't save a pointer to the return
value and decref it?  You should Decref comm too right?

Thanks,
Maybe I should change my name to major payne.





From chafik at nevrax.com  Tue Jun  6 05:09:24 2000
From: chafik at nevrax.com (Sameh chafik pro)
Date: Tue, 6 Jun 2000 11:09:24 +0200
Subject: How do I create extensions to Python in C?
References: <BlAW4.3473$DQ3.245148@news1.atlantic.net> <392BE0ED.12E4DA0E@cfdrc.com> <00ea01bfc594$02d2e2f0$4500a8c0@thomasnb> <393C95BD.77108EE@san.rr.com>
Message-ID: <8hieq9$k1m$1@gw.localdomain>

It is okay, but why coud we dot to translate a class write in C++. The
problem that the standard way is to use the PyTypeObject and define the
constructor as a function, but python don't recognize the translation as
class but as type object.morality if i do the translation of a class "X" it
is'nt posible in python to use it as a derivate class:
class A(X):
    def dosomething(self):
    .....
    ....

is not possible.


"Courageous" <jkraska1 at san.rr.com> a ?crit dans le message news:
393C95BD.77108EE at san.rr.com...
>
> > http://starship.python.net/crew/arcege/extwriting/pyext.html
>
> _Python Essential Reference_ also has a very nice section which
> summarizes many of the available primitives. If you're going
> to be doing this yourself, I recommend you take a look at
> xxobject.c in the python source. It's an empty template for
> creating a python extension module.
>
>
>
> C/




From rod at inf.bme.hu  Fri Jun  9 04:10:27 2000
From: rod at inf.bme.hu (Gabor Papp)
Date: Fri, 09 Jun 2000 10:10:27 +0200
Subject: tkinter -- why doesn't this work?
References: <jl66rj3cqa.fsf@endor.bbb.caltech.edu>
Message-ID: <3940A673.20DFC895@inf.bme.hu>

use frame.bind_all("<Key>", callback) instead of .bind
i don't know an exact explanation why it is needed,
but i think that a frame widget does not handle key events,
if you use bind on an entry widget which deals with keypresses
it works.
bye,
Gabor Papp


From emile at fenx.com  Thu Jun  1 13:56:40 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 1 Jun 2000 10:56:40 -0700
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and others" is this possible?
References: <393692FB.8610EBCA@dallas.net>
Message-ID: <007901bfcbf2$c3b4e680$1906a8c0@fc.fenx.com>

Jim,

Take a look at os.popen, as in:

import os
mylist = os.popen('locate myname').readlines()
for eachfile in mylist: do something with eachfile

HTH,

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message ----- 
From: Jim 
To: python-list at python.org 
Sent: Thursday, June 01, 2000 9:44 AM
Subject: Python to call commands, e.g. "find , perl scripts , nmap , and others" is this possible?


  
  
Dear list, 
I'm new to Python, and I'm sorry if I'm asking about anything that has been 
already discussed. 

I would like to use Python to call commands, e.g. "find , perl scripts  , nmap , and others" is this possible? 

I do not need GUI. All I need is to link the out put and work with it. Is perl or shell scripts my only 
answer? 

jim 

-- 
Proprietary & Confidential.  The information transmitted is intended only for the person or entity to whom it is addressed and may
contain confidential and/or privileged material.  Any review, retransmission, dissemination, or other use of, or taking of any
action in reliance upon, this information by persons or entities other than the intended recipient is prohibited.  If you received this in error, please
contact the sender and delete the material from any computer.
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000601/a781f954/attachment.html>

From maa at research.bell-labs.com  Thu Jun  1 11:37:43 2000
From: maa at research.bell-labs.com (Mark Ardis)
Date: Thu, 01 Jun 2000 10:37:43 -0500
Subject: Linux build failure when enabling tkinter
References: <39353E67.71DAC05@research.bell-labs.com> <x7puq2e6nr.fsf@lindm.dm>
Message-ID: <39368347.E55E4C23@research.bell-labs.com>

Dieter Maurer wrote:
> 
> Mark Ardis <maa at research.bell-labs.com> writes on Wed, 31 May 2000 11:31:35 -0500:
> > I am trying to build Python 1.5.2 on a Linux system (Mandrake 7.0).
> > I followed the directions in the Modules/Setup file for enabling
> > tkinter, but I still get errors when I run make test:
> >
> >   PYTHONPATH= ./python ./Lib/test/regrtest.py
> >   ./python: error in loading shared libraries: libtk8.3.so: cannot open
> > shared object file:    No such file or directory
> >   make: [test] Error 127 (ignored)
> >
> > Here is what I used in Modules/Setup:
> >
> >   TKPATH=:lib-tk
> >   _tkinter _tkinter.c tkappinit.c -DWITH_APPINIT \
> >   -I/usr/local/include \
> >   -I/usr/X11R6/include/ \
> >   -L/usr/local/lib \
> >   -ltk8.3 -ltcl8.3 \
> >   -L/usr/X11R6/lib \
> >   -lX11
> >
> > I verified that libtk8.3.so is in /usr/local/lib/
> > I have recently installed Tcl/Tk 8.3 on my Linux system (HP Kayak XU),
> > and it
> > seems fine.
> Seems that "/usr/local/lib" is not in your "/etc/ld.so.conf", the
> configuration file of the dynamic linker.
> Try "ldconfig -p" to verify, whether "libtk8.3.so" is listed.
> 
> If not, you can either set "LD_LIBRARY_PATH" to include "/usr/local/lib"
> or (as root) edit "/etc/ld.so.conf" and then run "ldconfig".
> 
> You can use "ldd python" to get a list of shared objects used
> by python and determine whether some cannot be found.
> 
> Dieter

That did it.  Thanks!
Mark


From gmcm at hypernet.com  Mon Jun 12 15:22:04 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 12 Jun 2000 19:22:04 GMT
Subject: Embedded Python and Restricted Execution
References: <8i35cl$ch8$1@nnrp1.deja.com>
Message-ID: <8F5198BEAgmcmhypernetcom@199.171.54.194>

cgfandrich at my-deja.com wrote:

>I'm embedding Python in an application and I'd like to be able to run
>Python code (including callable PyObjects) in a restricted environment -
> much like rexec - but I'd like to control the environment in C/C++.
>I've tried the following and had some success:
>
>1. Create a new module.
>2. Add a "__builtins__" dictionary to the module.
>3. Copy references from existing "__builtins__" to new "__builtins__"
>(except for "__import__" and "open" - use my functions for those 2)
>4. Use the new "__builtins__" dictionary whenever I want to run code in
>restricted environment.
>
>When importing modules, I've also had to make sure that the module
>being imported gets the correct "__builtins__" dictionary.  The only
>problem is ... if I have 2 modules "test1" and "test2" that use the
>same Python source code, I can't run them in the different
>environments.  It seems like they both run with
>whichever "__builtins__" was used to import the first module.

I think you're saying that test1 and test2 both import some other module? 
If that's the case, the import in test2 will find the module already in 
sys.modules. So if test1 did an unrestricted import, test2 will find the 
other module to have an unrestricted builtins.

No easy answers for you. I assume you've checked out what rexec and bastion 
do (in pure Python). You might consider having 2 separate interpreters (as 
in Py_NewInterpreter, not 2 instances of python). Or you could completely 
take over the management of namespaces (as in using the equivalent of "mod 
= __import__(...)" which doesn't automatically put mod in sys.modules).

- Gordon


From andyl at azaal.dircon.co.uk  Fri Jun 23 14:41:34 2000
From: andyl at azaal.dircon.co.uk (Andy Leighton)
Date: 23 Jun 2000 18:41:34 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <8F5B6C0A8gmcmhypernetcom@199.171.54.154>
Message-ID: <slrn8l7bqu.cmq.andyl@azaal.dircon.co.uk>

On 22 Jun 2000 13:45:58 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:
>William Dandreta wrote: 
>
>[looking for dBase module]
>>I am running uner DOS. Any version will be fine, even older ones as long
>>as I can index it.
>
>You're kind of SOL. You need a pure Python solution (because of DOS). 
>
>There's a couple of DBF readers on the Vaults DB page: 
>http://www.vex.net/parnassus/apyllo.py/973100124
>
>but you'll have to hack your own writing.

It might be worth looking at the xbase library from startech which is
GPL.  It does reading and writing (including memos) as well as ndx and
ntx indices.  The dbf access routines from the harbour project might
also be worth borrowing.

http://www.startech.keller.tx.us/xbase.html
http://www.harbour-project.org/

-- 
Andy Leighton => andyl at azaal.dircon.co.uk
"... January is your third most common month for madness" - _Sarah Canary_


From stephen at cerebralmaelstrom.com  Thu Jun  1 08:09:44 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 01 Jun 2000 12:09:44 GMT
Subject: ANNOUNCEMENT: PythonWorks 1.0 now available (may 31, 2000)
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com> <0CmZ4.30723$WS3.355848@typhoon.we.rr.com> <39362f52$0$10880@senator-bedfellow.mit.edu>
Message-ID: <cosZ4.31355$WS3.360373@typhoon.we.rr.com>

    Wellll.. Picture a RAD -- C++Builder, Visual Basic, etc.. You have your
entire form, the menu bar, everything all together in one pretty little
place where you can Rapidly Develop your environment.
    You can do some 'visual' editing in Visual C++, but its not all together
on one form.. you have your menu off in one place, your 'bars' off in
another, the dialogs off in another.. not 'windows', really...

--S

Robert <soundhack at hotmail.com> wrote in message
news:39362f52$0$10880 at senator-bedfellow.mit.edu...
>
> Stephen Hansen wrote in message
> <0CmZ4.30723$WS3.355848 at typhoon.we.rr.com>...
> >shop.microsoft.com, but the Learning edition is also about $100) [ MS
> VC++'s
> >standard is about $100 I think, also, but i'm not including it in this
> >because I have a hard time calilng it a 'RAD' environment... that
> >splitting-of-a-window thing it makes you do is irritating, I want to be
> able
> >to draw a whole form :P ]
>
>
> I have my own ideas about pricing, but I'm curious about this. could
> you please explain to me what you mean by the 'splitting-of-a-window
thing'?
> I agree that VC++ is not a RAD environment btw.
>
> Robert
>
>
>




From borlak at home.com  Wed Jun  7 22:20:54 2000
From: borlak at home.com (Michael Morrison)
Date: Thu, 08 Jun 2000 02:20:54 GMT
Subject: Question about Jpegs
References: <mKi%4.8088$r7.1635010@news1.rdc2.tx.home.com> <393DDEE9.F0F3C4F5@sightreader.com>
Message-ID: <aqD%4.8753$r7.1817270@news1.rdc2.tx.home.com>

Hrm, I'll try this again, it gave me an error last time.....

Is there something.....simpler?  more....Python like?

import simplemodule

newjpeg = simplemodule.shrink_jpeg(file,320,240)


"Ken Seehof" <12klat at sightreader.com> wrote in message
news:393DDEE9.F0F3C4F5 at sightreader.com...
> Take a look at PIL (Python Imaging Library).
>
> Michael Morrison wrote:
>
> > This may not be the appropriate place for this question, but it's for a
> > Python program and I know there are a lot of experienced programmers
here,
> > so....
> >
> > Does anyone know how to "resize" a .jpg or similar file?  My script is
> > dealing with uploaded pictures, and I want it to automatically create a
> > thumbnail for said pictures.  Does python by chance have a module that
can
> > do this?
> >
> > Thanks for any input
>
> --
> Ken Seehof
> kens at sightreader.com
> starship.python.net/crew/seehof
> Hi! I'm a .signature virus! copy me into your .signature file to help me
> spread!
>
>




From kstar18 at hotmail.com  Fri Jun 23 16:11:04 2000
From: kstar18 at hotmail.com (kstar18 at hotmail.com)
Date: Fri, 23 Jun 2000 20:11:04 GMT
Subject: Event Binding and Variable Passing
Message-ID: <8j0g88$htm$1@nnrp1.deja.com>

hi,

I'm relatively new to python and am experimenting with some Tk binding.
I was wondering how I would pass a variable to a function that is called
when a Tk event occurs.

For example,
b = Button(root, text = 'hi')
b.pack()
x = 5
b.bind('<Button-1>', self.fun(x))

def fun(self, event, x):
   print x

This doesn't seem to work, and the compiler gives me complaints about
'self' and the variable being passed.

Any help would be greatly appreciated.

Thanks!

Kapil

This method doesn't seem to work.


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


From thelocust at bigfoot.com  Tue Jun 27 13:41:06 2000
From: thelocust at bigfoot.com (Timothy Allen)
Date: Tue, 27 Jun 2000 12:41:06 -0500
Subject: e-mail problem
References: <8j9knn+dmcu@eGroups.com>
Message-ID: <8jalf2$gjj$1@lendl.cc.emory.edu>

Mutt is very cool, does that. Is that the sort of thing you're looking for?

In article <8j9knn+dmcu at eGroups.com>, hohage at muenster.de wrote:
> Hello, I get the python-list by digest. My problem is that I'd like to
> have a program which sorts the e-mails by thread or date etc. Is there
> something like this or do I have to write it by myself?
>             Thanks
> 
> 


-- 
Timothy Allen 
thelocust at bigfoot.com       >>        tim at locust.za.net 
Somewhere, just out of sight, the unicorns are gathering.



From shapr at uab.edu  Thu Jun 15 11:46:11 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 15 Jun 2000 15:46:11 GMT
Subject: incremental parsing of Python
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no> <8hrhqa$afm$1@news1.xs4all.nl> <G2g05.688$cQ.9254@news-server.bigpond.net.au> <39441CDB.8A6E05EA@uab.edu>
Message-ID: <3948FA8D.2CF7DF3F@uab.edu>

Shae Erisson wrote:
> 
> Mark Hammond wrote:
> [snip explanation of the difficulties of correct code completion]
> 
> > So if anyone can come up with generic code that can do this, all current
> > and future IDEs for Python could benefit.
> >
> > Mark.
> 
> If the python parser could be hacked up into an incremental parser, that
> would solve all of the problems.
> 
> I have no idea how to do it, but Ka-Ping Yee's IPC8 Developers Day Notes
> mention that Guido would like to have such a thing. You can find the
> notes in the 'purpose statement' of the compiler-sig. The compiler-sig
> is the place for this kind of thing to happen.

I read the parser module C source for 1.6, it's structured specifically
to be extended into an incremental parser. I guess it's time for me to
learn C.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
 squirrels furnished by Paul E. Black, you're safe now.


From gmol at my-deja.com  Mon Jun 12 23:33:56 2000
From: gmol at my-deja.com (gmol at my-deja.com)
Date: Mon, 12 Jun 2000 23:33:56 -0400
Subject: dynamic classes/objects and the like
Message-ID: <642A954DD517D411B20C00508BCF23B001296891@mail.sauder.com>

Hi all,
I am actually planning on using some dynamic classes and objects in my
application..looking at

http://starship.python.net/crew/gandalf/DNET/classes/DynaObj.html

it seems just a tad dated(1.4)...although I can't think of how different
something else would be , there isn't a more recent generally accepted
way to do this kind of thing is there?

Are there any examples of more large applications that rely on on
dynamic classes / objects that I could look at to avoid a possible
hitherto unseen pitfalls?

Thanks a tonne for help if you can offer any...



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



From tom-main at REMOVEME.home.com  Wed Jun  7 08:59:13 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Wed, 07 Jun 2000 12:59:13 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com> <UUg%4.7105$Hz.52820@news-server.bigpond.net.au>
Message-ID: <BGr%4.171660$55.3606399@news2.rdc1.on.home.com>

True, but that future - WebForms - isn't even here yet.

And the 'web interfaces' that I see today seem kind of primitive compared to
the old fashioned interfaces.  (And I have no idea how I would create a web
interface w/o getting into yucky HTML.)

Tom.

"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:UUg%4.7105$Hz.52820 at news-server.bigpond.net.au...
> > I agree that MFC seems to be on the way out ...
> > 2) What's next?
> > - STL?  Yes, but that's not a GUI.
> > - ATL & WTL?  Too MS only.
> > - wxWindows?  Only has a tiny fraction of the number of users & code
that
> > the other options have.
> > - TCL/TK?  I've heard this is slow, and it can't create native windows'
> apps
> > (users want the windows look & feel).
> > - Java?  Again, too proprietary.
>
>    It looks to me like the future of UIs (and I think this is the MS
> direction as well) is HTML and related standards. MS has put a lot of
effort
> into HTML generation and bridges between HTML and code.
>
>    Neil
>
>
>




From mvanier at endor.bbb.caltech.edu  Mon Jun 12 18:36:28 2000
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 12 Jun 2000 15:36:28 -0700
Subject: stupid thread question
Message-ID: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu>

Python gurus,

Apologies if this has come up before...

I wrote a little python script using the thread module the other day.  I
wanted to be able to set a flag in the parent thread that would be read in
the child thread.  So I set a flag variable (an integer in this case) in the
parent, created the child thread, and then checked the value of the flag
every time through a loop in the child thread to see if it had changed.  It
turns out that the child thread apparently made a copy of the flag, because
when it changed in the parent it didn't change in the child.

To fix this, I changed the flag to be a list with one integer element.  Now,
when I change the list element in the parent, the child sees it too.  I
gather that what this means is that there is a fundamental distinction
between atomic types like integers and reference types like lists or
dictionaries.  However, since python treats everything as a reference this
seems inconsistent to me.  So what I want to know is: what is the cause of
this behavior, and what is the justification for it?  Or could it be a bug?

FWIW this is on a RH Linux 6.1 system running python 1.6a2 with pthreads (I
assume).  

TIA,

Mike

--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.



From gmcm at hypernet.com  Thu Jun 29 12:44:19 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 16:44:19 GMT
Subject: pythonservice on NT
References: <8jfe0g$3pl$1@nnrp1.deja.com>
Message-ID: <8F628CD32gmcmhypernetcom@199.171.54.194>

Dale Burnett wrote:

>Do all services run under one pythonservice process?  

No.

>If not then is
>there a way to change the name pythonservice in NT TaskManager to
>something more descriptive?


class MailFetchService(win32serviceutil.ServiceFramework):
    _svc_name_ = "MailFetchService"
    _svc_display_name_ = "Fetch SMTP mail on regular schedule"

- Gordon


From bhat at cs.mu.OZ.AU  Wed Jun 14 22:46:55 2000
From: bhat at cs.mu.OZ.AU (Tim Bell)
Date: Wed, 14 Jun 2000 22:46:55 -0400
Subject: Why does time.gmtime() use Monday=0?
Message-ID: <642A954DD517D411B20C00508BCF23B001296B46@mail.sauder.com>

Hi,

Can anyone tell me why time.gmtime() (and time.localtime()) use the
convention that Monday=0, rather than Sunday=0?  The Sunday=0 convention
is used by C's gmtime(3), as well as both C and Python's strftime().
Oh, and Python's strptime() uses Monday=0 also, but C's uses Sunday=0.

Here's the relevant code from tmtotuple() in python-1.5.2/Modules/timemodule.c,
line 262:

	(p->tm_wday + 6) % 7, /* Want Monday == 0 */

Why does Python go out of its way to be inconsistant, both with itself
and with C?  (Or have I missed something here?)

Any enlightenment would be appreciated.

Tim.
-- 
Tim Bell - bhat at cs.mu.oz.au - Dept of Comp Sci & SE - Uni of Melbourne, Aust.



From mlauer at trollinger-fe.rz.uni-frankfurt.de  Wed Jun  7 14:56:15 2000
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 7 Jun 2000 20:56:15 +0200
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com> <8hluop$mlo$1@news1.xs4all.nl>
Message-ID: <393e9acf@nntp.server.uni-frankfurt.de>

Boudewijn Rempt (boud at rempt.xs4all.nl) wrote:
> Tom <tom-main at removeme.home.com> wrote:

> > I agree that MFC seems to be on the way out, but there are two things that
> > prevent me (and some other MFC developers) from switching:
> > 1) The CodeTools & CodeGuru sites contains TONS of peer reviewed code (in
> > the form of messages at the end of every article) that is free, even for
> > commercial use.  80% of this code is in the form of MFC extensions.
> > 2) What's next?
> > - STL?  Yes, but that's not a GUI.
> > - ATL & WTL?  Too MS only.
> > - wxWindows?  Only has a tiny fraction of the number of users & code that
> > the other options have.
> > - TCL/TK?  I've heard this is slow, and it can't create native windows' apps
> > (users want the windows look & feel).
> > - Java?  Again, too proprietary.

> There's another option, of course. You could use Qt (and PyQt), which
> works on Windows and Unix/X11, offers a great GUI and a good design.
> And there are tons of programs using it you can look at & learn from.
> (Not to mention the great documentation.)

Not to forget PyGnome and PyGTK, which runs quite good on Unix/X11 and
is evolving on Windows, BeOS and MacOS.

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




From andre at beta.telenordia.se  Sun Jun  4 11:36:11 2000
From: andre at beta.telenordia.se (André Dahlqvist)
Date: Sun, 04 Jun 2000 16:36:11 +0100
Subject: Marking hyperlinks in a Text widget
References: <8h04jo$lgh$1@zingo.tninet.se> <oq7lcc41zr.fsf@titan.progiciels-bpi.ca> <8h0kn5$6pv$1@cubacola.tninet.se> <slrn8j94er.ebv.cjc26@localhost.mindriot.net> <8h4d2e$ap4$1@cubacola.tninet.se> <slrn8jblag.jua.cjc26@localhost.mindriot.net> <8h5hvi$djt$1@cubacola.tninet.se> <slrn8ji7dm.5b1.cjc26@localhost.mindriot.net>
Message-ID: <8hdphb$l03$1@zingo.tninet.se>

Hi again Cliff, and thanks for helping me with this.

> Running netscape this way will fail if the user is already running it;
> to get around this you can check if it is running (using the output of
> os.popen("ps ax")) and use the openURL command if it is.

I will have to do something like this, yes. Debian GNU/Linux has a nice
wrapper which smartly handles these cases automatically. If you 
try to start netscape when it's already started it will use the already
started browser to open the link. Unfortunately not everyone has such 
a wrapper, and I will therefore have to check for running processes.

> Right, you'll have to read the text directly from the widget itself. You
> won't even need to use tag_names in this case.

Okay, it would be great if you could give a small example of how to do
this. Here is the code I am currently using, which is ugly and doesn't
handle all cases:

First I create a 'url' tag, and bind appropriate callbacks to this tag:

textwidget.tag_config('url', foreground='blue', underline=1)
textwidget.tag_bind('url', '<Enter>', hand_cursor)
textwidget.tag_bind('url', '<Leave>', normal_cursor)
textwidget.tag_bind('url', '<Button-1>', browser)

Below is the code that finds hyperlinks in the string called text. This 
string has also been inserted into the Text widget earlier in the code. 
http:// and ftp:// are the only protocols that needs to be found, and it 
doesn't have to find URLs like "www.address.com" (although that
wouldn't hurt.) What it would need to find, besides the url type
described above, are links that have a character right in front of it, like 
an opening pharentesis: (http://www.link.com). While almost all of the 
links in the text that I am searching for are separated by a space from 
the rest of the case, a few links are enclosed in pharentesis like this. 
I could add a "(http" and "(ftp" to the list below, but then I would have 
to remove the parenhetis in front of http:// and ftp:// in the links that I 
have found. So instead of doing this, which would be pretty ugly, I 
would prefer to use regular expressions for this. How would I use
regular expressions to only find the links themselves, and not the
stuff just in front of it?

Anyway, here's the code that I am not proud of which finds the URLs:

for word in string.split(text):
    if string.split(word, "://")[0] in ("http", "ftp"):
        pos = helpwin.textwidget.search(word, 1.0, END)
        # Add a tag from the start to the end of the URL
        textwidget.tag_add('url', pos, + pos + " + " + `len(word)`+ " chars")
        
Then I have the functions that handle the events that occur when you 
put your cursor over the link:

def hand_cursor(event):
    event.widget.configure(cursor="hand1")

def normal_cursor(event):
    event.widget.configure(cursor="xterm")

And finally the code that has to find what URL we clicked on, and start
the browser. I have not yet changed this to check for already running 
browsers, but I will do that later. If there is an easier way of finding 
these URLs I would love to hear it.

def browser(event):
    import bisect
    common_browsers = ['mozilla', 'netscape', 'lynx', 'w3m']

    # Find the ranges that have been tagged with the 'url' tag
    ranges = event.widget.tag_ranges('url')
    cursor_position = event.widget.index("@%d %d", event.x, event.y)
    # Find out in range the cursor was clicked
    slot = binsect.binsect(ranges, cursor_position)
    # Get the url
    hyperlink = event.widget.get(ranges[n-1], ranges[n])
    
    for browser in common_browsers:
        if not os.system(browser + hyperlink):
            # Found browser, leave
            break

> You could even post it so that everyone could look at it :)

I found the courage and did that now:-)

// Andr?


From jovecka at ethicalhackerz.com  Thu Jun 22 20:17:34 2000
From: jovecka at ethicalhackerz.com (Jim Ovecka)
Date: Thu, 22 Jun 2000 19:17:34 -0500
Subject: Python interface to Oracle
References: <3951C636.2E985CE@ust.hk> <87og4t26zz.fsf@ns.f7.ph4.lan>
Message-ID: <3952AC9E.E01ED3A6@ethicalhackerz.com>

You can get Oracle8i from RedHat for free.

Andreas Rottmann wrote:

> David YEUNG <dyeung at ust.hk> writes:
>
> > Does anyone know if any python module has been developed for
> > interfacing with Oracle8i database on Linux?
> >
> That would be interesting for me, too. I'm likely to develop one, if
> it doesn't exist, since I have the infrastructure ready - see
> GQL.sourceforge.net.
>
> All it would take is writing an Oracle driver for GQL, which can be
> done within a few days. Anyway, I would appreciate if someone could
> help me with that, since I don't have Oracle and am not willing to pay
> the $$$.
>
> Andy
> --
> Andreas Rottmann   | Dru at ICQ     | 54523380 at ICQ | a.rottmann at gmx.at
> Pfeilgasse 4-6/725 | A-1080 Wien | Austria      | Europe
> http://www.8ung.at/rotty         | Key is there | gpg (GnuPG) 1.0.1
> Fingerprint        | 3E9A C485 49A4 1D17 2EA7  2BA7 22AE C9BF 8173 6279
> [one of 78,35% Austrians who didn?t vote for Haider!]

--
Proprietary & Confidential.  The information transmitted is intended only
for the person or entity to whom it is addressed and may contain
confidential and/or privileged material.  Any review, retransmission,
dissemination, or other use of, or taking of any action in reliance upon,
this information by persons or entities other than the intended recipient
is prohibited.  If you received this in error, please contact the sender
and delete the material from any computer.




From weiwa592 at student.liu.se  Fri Jun 16 20:35:28 2000
From: weiwa592 at student.liu.se (Charles W.W.)
Date: Sat, 17 Jun 2000 02:35:28 +0200
Subject: Monitor on unix
Message-ID: <394AC7D0.EE82B737@student.liu.se>

I want to write a monitor program to monitor the users running on unix
system. I
planned to implement it in python/TK GUI and perhaps Jpython, who
can give me some pieces of advice so that I can compare and start to
work.
Thanks.


From jvickroy at sec.noaa.gov  Thu Jun 15 12:17:40 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Thu, 15 Jun 2000 10:17:40 -0600
Subject: how to pause a forked process
Message-ID: <3949019E.1D9B53B3@sec.noaa.gov>

Hello,
Path: news!uunet!ffx.uu.net!newsfeed.easynews.net!easynews.net!easynet-tele!easynet.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!logbridge.uoregon.edu!den-news-02.qwest.net!qwest!news.boulder.noaa.gov!not-for-mail
Xref: news comp.lang.python:99777

How can execution be paused in a child process?

This is what I tried:

import os
import time

pid = os .fork()

if pid == 0: # child
    # do some work
    time .sleep (5)

else: # parent
    # do some work

The sleep() call is flagged as:
    AttributeError: sleep

The above code is part of a Tkinter application.  Could this error have
anything to do with that?

Thanks for your time.



From geoffm at nojunkmailqsiuk.com  Fri Jun 23 06:18:00 2000
From: geoffm at nojunkmailqsiuk.com (geoff munday)
Date: Fri, 23 Jun 2000 11:18:00 +0100
Subject: adding mail attachments
Message-ID: <8ivdl4$90k$1@supernews.com>

OK, I  can send mail via the smtplib but am at a loss as how to attach files
to the Email.
Can anyone advise where to look for instructions or examples??

Geoff




From jepler.lnk at lnk.ispi.net  Fri Jun  9 09:22:45 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Fri, 09 Jun 2000 13:22:45 GMT
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl> <393D1155.D067A821@decrc.abb.de> <m33dmqygdj.fsf@atrus.jesus.cam.ac.uk> <slrn8jt5ck.779.jepler.lnk@potty.housenet> <m3bt1dvzq4.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <slrn8jv9mc.8v2.jepler.lnk@potty.housenet>

On 08 Jun 2000 00:27:31 +0100, Michael Hudson
 <mwh21 at cam.ac.uk> wrote:
>$ env LC_NUMERIC=de_DE python -c 'import gnome.ui; print float("1.1")'
>
>Gdk-WARNING **: shmget failed!
>1.1
[Which failed w/traceback for me---incidentally, with LC_NUMERIC=de_DE,
float("1,1") succeeds and returns the float 1.1]

>Versions?

I was able to just observe the problem with a RedHat 6.2 machine, with
the following rpm versions of some possibly relevant packages:
	pygnome-1.0.51-1
	pygtk-0.6.4-1
	gtk+-1.2.6-7
	gnome-libs-1.0.55-12
	gnome-core-1.0.55-12
	glibc-2.1.3-15

I do *not* observe the problem on this RedHat 6.1 machine, with the
following rpm versions of some possibly relevant packages:
	pygnome-1.0.50-2
	pygtk-0.6.3-2
	gtk+-1.2.6-1
	gnome-libs-1.0.54-1
	gnome-core-1.0.54-2
	glibc-2.1.2-11
Unfortunately, a little bit of everything is different here.  :(

Jeff


From mwh21 at cam.ac.uk  Fri Jun  9 08:28:50 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 13:28:50 +0100
Subject: Obscure Exception message
References: <75F7304BB41CD411B06600A0C98414FC0A418B@ORSMSX54> <3dr9a7yi91.fsf@amarok.cnri.reston.va.us> <20000609010518.C20700@xs4all.nl>
Message-ID: <m3wvjzujgd.fsf@atrus.jesus.cam.ac.uk>

Thomas Wouters <thomas at xs4all.net> writes:

> On Thu, Jun 08, 2000 at 05:28:58PM -0400, Andrew M. Kuchling wrote:
> 
> > "Daley, MarkX" <markx.daley at intel.com> writes:
> > > TypeError: string member test needs char left operand
> > > I have no idea what this is telling me, although the 'char' comment makes me
> > > wish I had a flamethrower!
> 
> > Good point.  Hmm... how about:
> > TypeError: in <string> requires character as left operand
> 
> Put quotes around 'in <string>':
> 
> TypeError: 'in <string>' requires character as left operand
> 

The Rapid-Fire-Python-Error-Message-Improver (me) says:

Index: stringobject.c
===================================================================
RCS file: /cvsroot/python/python/dist/src/Objects/stringobject.c,v
retrieving revision 2.66
diff -u -r2.66 stringobject.c
--- stringobject.c	2000/06/01 03:12:13	2.66
+++ stringobject.c	2000/06/09 12:24:54
@@ -393,7 +393,7 @@
 		return PyUnicode_Contains(a, el);
 	if (!PyString_Check(el) || PyString_Size(el) != 1) {
 		PyErr_SetString(PyExc_TypeError,
-				"string member test needs char left operand");
+		    "'in <string>' requires character as left operand");
 		return -1;
 	}
 	c = PyString_AsString(el)[0];
Index: unicodeobject.c
===================================================================
RCS file: /cvsroot/python/python/dist/src/Objects/unicodeobject.c,v
retrieving revision 2.21
diff -u -r2.21 unicodeobject.c
--- unicodeobject.c	2000/05/09 19:54:43	2.21
+++ unicodeobject.c	2000/06/09 12:25:01
@@ -2991,7 +2991,7 @@
     /* Check v in u */
     if (PyUnicode_GET_SIZE(v) != 1) {
 	PyErr_SetString(PyExc_TypeError,
-			"string member test needs char left operand");
+	    "'in <string>' requires character as left operand");
 	goto onError;
     }
     ch = *PyUnicode_AS_UNICODE(v);

and:

I confirm that, to the best of my knowledge and belief, this
contribution is free of any claims of third parties under
copyright, patent or other rights or interests ("claims").  To
the extent that I have any such claims, I hereby grant to CNRI a
nonexclusive, irrevocable, royalty-free, worldwide license to
reproduce, distribute, perform and/or display publicly, prepare
derivative versions, and otherwise use this contribution as part
of the Python software and its related documentation, or any
derivative versions thereof, at no cost to CNRI or its licensed
users, and to authorize others to do so.

I acknowledge that CNRI may, at its sole discretion, decide
whether or not to incorporate this contribution in the Python
software and its related documentation.  I further grant CNRI
permission to use my name and other identifying information
provided to CNRI by me for use in connection with the Python
software and its related documentation.

Cheers,
M.

-- 
93. When someone says "I want a programming language in which I
    need only say what I wish done," give him a lollipop.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From godzilla at netmeg.net  Tue Jun  6 10:49:33 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Tue, 6 Jun 2000 10:49:33 -0400 (EDT)
Subject: author of htmllib/formatter?
In-Reply-To: <Pine.GSO.4.10.10006061736170.26389-100000@sundial>
References: <m2og5ec2q9.fsf@netmeg.net>
	<Pine.GSO.4.10.10006061736170.26389-100000@sundial>
Message-ID: <14653.3965.113400.816770@gargle.gargle.HOWL>

> If you'd mention the problems, people here would be able to help you
> (and quite likely, use the answers given to enhance the
> documentation)

i did already, no one answered....  ;-)

see my posts under 

Subject: HTMLParser: getting one extra space around &gt;/&lt;/^M tags

i thought i would change tack to see if anyone knew the authors and
work from there.

I'd be happy to help enhance the docs ONCE i understand the problem
myself. But i am looking to get a little help with how htmllib and
formatter work together to parse HTML prior to data being shipped off
to the formatter.... (see original post noted above)

les schaffer



From tanzer at swing.co.at  Wed Jun 14 02:10:00 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 14 Jun 2000 08:10:00 +0200
Subject: help formatting number 
In-Reply-To: Your message of "Tue, 13 Jun 2000 22:01:15 +0200."
             <8i645i$lht$1@nslave2.tin.it> 
Message-ID: <m1326NE-000wcDC@swing.co.at>

"Mario" <mtbros at tin.it> wrote :

> I have an integer variable and I want it to become a string variable
> formatted so that there is a dot between each three digits of the original
> number. An example:
> var =3456789
> It must become:
> "3.456.789"
> Ideas ?

-------------------------------------------------------------------------------
Python 1.5.2 (#5, Jan  4 2000, 11:37:02)  [GCC 2.7.2.1] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import re
>>> var =3456789
>>> sep_1000 = "."
>>> sep_1000_pat = re.compile ("(\d{1,3}) (?= (?: \d\d\d)+ (?! \d) )", re.X)
>>> print sep_1000_pat.sub ( r"\g<1>%s" % sep_1000, "%s" % var)
3.456.789
>>> 
-------------------------------------------------------------------------------

PS: If you want to understand the regular expression, don't ask me,
    get Jeffrey Friedl's book `Mastering Regular Expressions'

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From gmcm at hypernet.com  Thu Jun  8 21:59:47 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 09 Jun 2000 01:59:47 GMT
Subject: How to tell if I can open a file or not
References: <Pine.GSO.4.10.10006081936380.15963-100000@sundial> <8hokva$2rqa$1@nntp6.u.washington.edu>
Message-ID: <8F4DDEE56gmcmhypernetcom@199.171.54.195>

Donn Cave wrote: 


>One reason it's better than the C way:  sooner or later, you will check
>whether a file can be opened, go ahead to open it on the strength of
>this check, and find that it can't be opened anyway.  Your program
>crashes, the air traffic control system goes out to lunch and hundreds
>die. 
>
>After you check, there's a period of time before you actually open
>the file.  It's a brief window, but it's real, and many things can 
>happen in that time if the system is busy, including changes to the
>file's permission bits or even whether it exists.  Might as well
>just open the file, and if it works then you have a green light.

And hundreds of people die, anyway.

God-am-I-glad-I'm-off-air-traffic-control-ly y'rs

- Gordon


From ehagemann at home.com  Sun Jun 18 19:20:11 2000
From: ehagemann at home.com (Eric Hagemann)
Date: Sun, 18 Jun 2000 23:20:11 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <01bfd93c$66c03180$2ace8ad1@daves-dell> <oO935.12054$Xx5.557215@bgtnsc06-news.ops.worldnet.att.net> <01bfd96b$f7ddcea0$85a38ad1@daves-dell>
Message-ID: <LOc35.5656$_54.105800@news1.rdc1.md.home.com>

[snip]
Also you might have better luck with the regular expression stuff (rather
than the find command) and precompiling the string
[snip]

I think I am the somebody who said this ;)

My mind was racing faster than my fingers.  The re stuff _is_ slower than
find. I have measured the slowdown to be a factor of 2 or 3. As also proven
below.

My real thoughts were that most grep's (or findstr in NT) do more than
simple string matching and thus the re module might be desired or a better
comparison.  If all that is desired is simple matching however, then
string.find is your way to go.

Cheers
Eric




"David C. Ullrich" <ullrich at math.okstate.edu> wrote in message
news:01bfd96b$f7ddcea0$85a38ad1 at daves-dell...
> William Dandreta <wjdandreta at worldnet.att.net> wrote in article
> <oO935.12054$Xx5.557215 at bgtnsc06-news.ops.worldnet.att.net>...
> > Hi David,
> >
> > The message is titled Python GREP(... you can take a look if you like.
It
> > was suggested that reg expr might be faster so give it a try.
>
> Sure enough, someone said "Also you might have better luck
> with the regular expression stuff (rather than the find command) and
> precompiling the string" and it seems clear from the context that
> "better luck" means "faster". I doubt it's so. I could be wrong, it's
> happened before, not that I can ever think of any examples.
> (Note that comparing string.find() to re.search() is not the same
> as comparing string.find() to the system's grep function...)
>
> Ok, try this. I spent a few minutes just now looking up how
> a trivial regex works, not gonna look up the docs on profiler.py
> today. Instead I made a huge example that I can "time" by counting
> "one, two, three...". You may want to start with less ado at
> first and then crank the numbers back up if it goes by too fast:
>
> import string
> import re
>
> r=re.compile('dog')
>
> s='ado'*10000
>
> l=[]
> for j in range(1000): l.append(s[:])
> l.append(s+'g')
>
> def testre():
> localr = r
> for str in l:
> res = localr.search(str)
> if res:
> print res.start()
> break
>
>
> def testfind():
> find=string.find
> for str in l:
> res = find(str,'dog')
> if res > -1:
> print res
> break
>
> I don't _think_ I'm cheating here. On the machine I
> tested this on calling testre() seems to take about
> six seconds while testfind() seems to be less than
> three.
>
> DU
>
> > Bill
> >
> > David C. Ullrich wrote in message
> <01bfd93c$66c03180$2ace8ad1 at daves-dell>...
> > >
> > >
> > >William Dandreta <wjdandreta at worldnet.att.net> wrote in article
> > ><sy535.11434$Xx5.530580 at bgtnsc06-news.ops.worldnet.att.net>...
> > >> I recently read a message that suggested that using regular
> expressions
> > >> might be faster than the find function.
> > >
> > > I would have guessed that if anything find would be faster, although
> > >I certainly could be wrong. Did someone say that using a regular
> expression
> > >would be _faster_, or did they actually say it was more _powerful_ or
> some
> > >such?
> > >
>




From alan.gauld at gssec.bt.co.uk  Thu Jun 29 08:17:08 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Thu, 29 Jun 2000 13:17:08 +0100
Subject: Tutorials
References: <8ite7j$pv5$1@lola.ctv.es>
Message-ID: <395B3E44.86860520@gssec.bt.co.uk>

The python web sirte and the vaults of parnassus list 
several at varying levels of difficulty.

mine is at:

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

Plugging like mad,

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


From gmcm at hypernet.com  Thu Jun 15 17:06:53 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jun 2000 21:06:53 GMT
Subject: Python Productivity over C++ (Encapsulation)
References: <20000615183521.59984.qmail@hotmail.com>
Message-ID: <8F54AAD9Cgmcmhypernetcom@199.171.54.194>

Hung Jung Lu wrote: 

>Type strictness is fine. It has all to do with performance: small, fast,
>efficient machine code. Encapsulation is a totally different story.
>
>One of the biggest lies in modern computer language theory is that 
>encapsulation is there for a good technical reason.

Whew! That's quite a rant <omitted>.

But I can think of at least 3 different meanings for "encapsulation".

One is that the object is well designed, so I don't have to worry about 
it's internals. It just does what I want. Nothing to do with language, just 
a good API.

Another is they've made it so I can't (nicely) override the way the thing 
works. Like C++'s "private". Unless the object is *also* an example of the 
first kind of encapsulation, this sucks. Personally, "protected" is the 
strongest I use in C++.

Then there's binary only (including C++ libs where the headers define only 
virtual classes and no source included). Unless it's free and works the 
first time, I don't bother.

But there's nothing wrong with the *goal* of encapsulation. There's just a 
lot of stuff with bad API's and no sensible way to fix it.

- Gordon


From kens at sightreader.com  Thu Jun 15 17:07:26 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 14:07:26 -0700
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <1ecb5k1.1s3tce6xg2hfxN%bparsia@email.unc.edu>
Message-ID: <3949458D.9164FA5C@sightreader.com>



Bijan Parsia wrote:

> Russell Wallace <rwallace at esatclear.ie> wrote:
>
> > > stevemul at ozemail.com.au (Steve Mullarkey) wrote:
> > >
> > > > I have read in several places productivity claims of 5 to 10 times over
> > > > 'C' and "C++". I would like to ask for some feedback from "C++"
> > > > programmers who have moved to Python as to whether these estimates are
> > > > realistic.
> >
> > As someone who uses and likes both C++ and Python, I think estimates
> > like that are excessively simplistic.
> [snip]
>
> Actually, the "simplistic" part is not taking into account things like
> problem domain, tools, and, most importantly, programmer "type" (skill,
> experience, etc.). Strangely enough, I would bet that *temperment*
> counts for oodles: If you *really hate* Python, I doubt you'll be wildly
> productive in it. On the other hand, if you're like me and there's no
> way in this lifetime that you'll learn C++ (granted I'm not a programmer
> *per se*), then Python makes me *way* more productive. (Er...ok,
> Smalltalk/Squeak makes me way more productive, but actually having
> written Python code, and actually being willing to read Python code,
> still makes Python a significant increaser of my programming
> productivity over C++ code, the mere thought of which makes me more
> likely to work on my disseration, or clean the toilet).
>
> --
> Bijan Parsia
> http://monkeyfist.com/
> ...among many things.

Good point.  Actually C++ makes people way more productive that python does.
Ever since I learned python, I haven't cleaned the toilet.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/a60ce8eb/attachment.vcf>

From gherman at darwin.in-berlin.de  Thu Jun 29 17:47:24 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Thu, 29 Jun 2000 23:47:24 +0200
Subject: RC4.py?
Message-ID: <395BC3EC.C8418EF@darwin.in-berlin.de>

Hello,

does anybody have a working, free (or not free) implementa-
tion of RC4, a symmetric encryption algorithm developped by
RSA, Inc., written in pure Python?

Thanks,

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From jkraska1 at san.rr.com  Wed Jun 28 00:04:05 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 28 Jun 2000 04:04:05 GMT
Subject: boolean true and false values.
References: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl> <memo.20000626221037.43263C@digitig.compulink.co.uk> <slrn8lgsjq.us.hat@se-46.wpa.wtb.tue.nl> <39596A3C.D3A1A824@cosc.canterbury.ac.nz>
Message-ID: <395979D9.40A533A9@san.rr.com>

> > Until now, I like the basic idea of Python of variables not having a type, but
> > I don't really see the advantage.

The advantage would be in not having to describe the object
before you use it. Instead you just use it. This makes the
code blissfully terse.



C/


From jerome.quelin at insalien.org  Sun Jun 25 15:45:42 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Sun, 25 Jun 2000 21:45:42 +0200
Subject: How to change class instances behavior via reload?
Message-ID: <961962342.297785615@news.libertysurf.fr>

Hi all,

I would like to change the behavior of all the existing instances of a
class when using the reload statement. I wonder if this is possible?

Consider the following example:

-- file test.py
class Test:
    def func(self):
        print "first version"
-- end file
>>> import test
>>> t1=test.Test()
>>> t1.func()
first version
>>>
-- Editing file test.py, without leaving interpreter
class Test:
    def func(self):
        print "second version"
--
>>> reload(test)
<module 'test' from 'test.py'>
>>> t2=test.Test()
>>> t2.func()
second version
>>> t1.func()
first version
>>>

This seems correct, because t1 holds a reference to the code of the
first version of test.py. But when reloading, I would like to change
the behavior of all existing instances of the class Test, and not only
the behavior of the new instances that I will create. That is, I want
t1.func() to output "second version". I wonder if this is possible and
how?

Thanks,
Jerome
--
jerome.quelin at insalien.org


From wlav at hpatl26.cern.ch  Thu Jun  8 10:15:52 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 8 Jun 2000 14:15:52 GMT
Subject: How to tell if I can open a file or not
References: <8ho4to0m8n@enews2.newsguy.com>
Message-ID: <8ho9qo$ejm$1@sunnews.cern.ch>

"Skip Hollowell" <thollowe at opentext.com> writes:

>Is there a way I can check the attrbutes of the file before I open it?

>>> os.access( fileName, os.R_OK )

Similarly os.W_OK, os.X_OK (as expected) and os.F_OK to test the existence.
Returns '1' for success, '0' for failure. Works on files and paths.

Best regards,
       Wim Lavrijsen


From shapr at uab.edu  Fri Jun 23 19:48:53 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 23 Jun 2000 23:48:53 GMT
Subject: Python and HTML forms?
References: <2iL45.19443$Xx5.1074706@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <3953F888.955BE4F2@uab.edu>

William Dandreta wrote:
> 
> We use a web based DB that is updated with HTML forms. If a large number of
> updates are needed it takes a long time to type all the data into the forms.
> 
> Can Python be used to automate the updating?
> 
> If so, what do I need to know to do it?

check out urllib in the Python reference documentation., I vaguely
remember that it will allow you to use GET and POST to automate sending
form content. I don't think you can POST a file with it though.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From spex66 at my-deja.com  Tue Jun 20 13:14:17 2000
From: spex66 at my-deja.com (spex66 at my-deja.com)
Date: Tue, 20 Jun 2000 17:14:17 GMT
Subject: Kjbuckets-Module, Python 1.6 and NT
References: <394E4B02.5F4A0AB4@szs.ira.uka.de> <394F6E92.42AD72B9@interet.com>
Message-ID: <8io8oi$hh3$1@nnrp1.deja.com>

Hi James,

----- Original Message -----
In article <394F6E92.42AD72B9 at interet.com>,
  "James C. Ahlstrom" <jim at interet.com> wrote:
From: James C. Ahlstrom <jim at interet.com>
Newsgroups: comp.lang.python
Sent: Tuesday, June 20, 2000 3:16 PM
Subject: Re: Kjbuckets-Module, Python 1.6 and NT


>
> Use the compiler switch to compile as a C++ module.  Be sure to
> declare initkjbuckets() as an exported symbol with declspec().
>
this was the missing puzzle tile!!!

this code works for me (with the one I posted as answer to Robert Kern
today):
__declspec(dllexport) void initkjbuckets()
{
  initmodule("kjbuckets", kjbuckets_methods);
}


permanent working with SWIG isn't helpful if you are faced with such
code :> (but I have copied the decl-piece out of it)

thanks
Peter Arwanitis
(=PA=)


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


From stephen at cerebralmaelstrom.com  Thu Jun  1 05:03:55 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 01 Jun 2000 09:03:55 GMT
Subject: PyRun_String
Message-ID: <%FpZ4.31001$WS3.359957@typhoon.we.rr.com>

    What do I enter into PyRun_String's "globals" and "locals" dictionary
position if I want it to be in __main__?

    E.g: I'm embedding Python into a program. I call in prior places
PyModule_Import("languages"), and execute an arbitrary string or two
(namely, atm, PyRun_SimpleString("sys.path.append('a-path')"))

    I want to call "languages.Translate()" from __main__...

Thanks,

--S




From dsavitsk at e-coli.net  Fri Jun 30 19:28:02 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 30 Jun 2000 23:28:02 GMT
Subject: Web server executing Python scripts
References: <4.3.1.2.20000630151218.021a7858@phear.dementian.com>
Message-ID: <62a75.28853$dF.1122319@news1.rdc1.il.home.com>

just to give a different point of view :-)
while iis may be insecure, ms is on to something with the asp thing (or did
a good job copying php/coldfusion/jsp etc).  if you are on an intranet and
security is less of an issue, asp is really easy and runs a bit faster than
cgi.  there are also asp plug-ins for apache (www.halcyionsoft.com and
www.chilisoft.com) that are sometimes free, sometimes not depending on what
you are doing etc. i'm not sure if these will allow PythonScript, but it
might be worth looking into.
doug

"Steven D. Arnold" <stevena at neosynapse.net> wrote in message
news:4.3.1.2.20000630151218.021a7858 at phear.dementian.com...
> At 08:24 PM 6/30/2000 +0200, Pieter Claerhout wrote:
>
> >What do you think would be the best idea?
> >- Use whatever webbrowser with CGI-scripting
>
> I would use apache and either CGI or mod-python, more likely the latter.
>
>
> >- Use MS IIS with ASP (and Python)
>
> IIS is quite probably the most insecure server on the Internet.  In
> addition, I don't like the closed nature of the system, I don't trust MS
to
> introduce the features I want/need (where I can introduce them myself in
> Apache), and I don't want to depend on MS's support services to solve
> problems that only they can fix because only they have the source.
>
>
> >- Use Zope to build the whole system?
>
> I don't know enough about Zope to give an answer from personal experience
> here.  I've heard many good things about Zope; I've also heard that the
> docs are poor and the learning curve steep.  I have faith in the speed and
> flexibility of Apache, where (to me) Zope is an unknown in these regards;
> are there any benchmarks?
>
>
> steve
>
>
> --
> Steven D. Arnold           Que quiero sera         stevena at neosynapse.net
> "If you have  built castles in the air, your work  need not be lost; that
> is where they should be.  Now put the foundations under them!" -- Thoreau
>
>




From jamarijr at my-deja.com  Mon Jun 12 12:40:55 2000
From: jamarijr at my-deja.com (Arinté)
Date: Mon, 12 Jun 2000 16:40:55 GMT
Subject: Python Issues.
Message-ID: <8i33qk$bhl$1@nnrp1.deja.com>

What is the webpage that shows all the issues python has with certain
OS'es?
Someone posted it a while ago when I had trouble with python and NT,
but I
couldn't find it on deja.  Post the link again please...


thanx



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


From englim at pc.jaring.my  Thu Jun  8 03:46:40 2000
From: englim at pc.jaring.my (Thomas)
Date: Thu, 08 Jun 2000 15:46:40 +0800
Subject: Pyrite
Message-ID: <393F4F60.494DA736@pc.jaring.my>

Has anyone used Pyrite?  Are there any sample programs?

Thomas.




From mfletch at tpresence.com  Sat Jun  3 15:49:16 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sat, 3 Jun 2000 15:49:16 -0400 
Subject: A Mountain of Perl Books + Python Advocacy
Message-ID: <CC59868A0A76D311B1F50090277B321C1F338F@VRTMAIL>

http://www2.linuxjournal.com/lj-issues/issue73/3882.html

Enjoy,
Mike

-----Original Message-----
From: digitig at cix.co.uk [mailto:digitig at cix.co.uk]
Sent: Saturday, June 03, 2000 3:12 PM
To: python-list at python.org
Subject: Re: A Mountain of Perl Books + Python Advocacy


In article <3937035C.EA99CB2B at uab.edu>, shapr at uab.edu (Shae Erisson) 
wrote:

> Chris Stromberger wrote:
...
> I've seen it in the special issue of the Linux Journal that is dedicated
> to Python, I don't know if it's online.
...



From shahn at kits.cs.vu.nl  Thu Jun 15 14:17:08 2000
From: shahn at kits.cs.vu.nl (Hahn S)
Date: Thu, 15 Jun 2000 18:17:08 GMT
Subject: apply for objects
Message-ID: <8ib6j4$dq5@cs.vu.nl>

Hello everyone,

I have a problem and i can't figure out how to solve it in Python.
I to call a method on an object with a uncertain number of keyword
parameters.

For example:
class TestHandler:
	def start(self, **attrs):
		print "%s %s" % (name, attrs)

	def dispatch(self, style, attrs):
		if style == 'begin':
			apply(self.start, attrs)

Now this won't work, because the self.start doesn't have a __call__
property. Any ideas how to solve this?

The **attrs in start is necessary, because there is other code calling
this function.

Am i overlooking something?

Thank you for your assistance!
Sander Hahn




From bgriggs at pobox.com  Wed Jun 28 01:15:40 2000
From: bgriggs at pobox.com (B Griggs)
Date: Wed, 28 Jun 2000 05:15:40 GMT
Subject: Python in home automation
References: <fsteele-E5789E.13533715062000@news.inet-systems.net> <cBd25.38645$cH1.355549@typhoon.austin.rr.com> <m3r99swt4x.fsf@hare.demon.co.uk> <fsteele-DDF9A7.16453020062000@news.inet-systems.net> <slrn8kvrup.ld1.kain@cableadmin.com>
Message-ID: <0Sf65.49355$Gh.511576@typhoon.austin.rr.com>

Frank, all,

I responded to your initial message  (name was munged up - sorry.)  I am
very interested in Python home automation.  I have received a CM11A
controller and should be receiving a CM17A any day now.  I read the Mr.
House article in The Perl Journal and was intrigued.  You can do some pretty
neat stuff with X10.

I did some research and found Scott's XAL.  The thing that bothered me was
it is written in C and the X10 lib needs to be compiled natively before the
Python wrappers can be used.  The source is currently written to compile
under Unix, not Windows.  Not a problem in and of itself, but useless to me.

The question is- is it worthwhile to write this in Python.  Does it become
more portable.  I think so.  The only real challenge I see is serial access.
It seems as though each platform implements its' own serial support under
Python.  I believe 2 packages exist for Windows, for example.

Has anyone done any cross-platform dev in areas such as this?  In C we would
use IFDEFs.  I assume the best construct to handle the serial problem would
be simple conditional statements in Python?   More to the point, does anyone
know of a Python serial package that supports all Python platforms
magically?

Les - would love to have a look at your work.  If possible, email it.

I would like to have a go at writing a CM17A control class.  It would be the
simplest since the firecracker unit is not bi-directional.   The problem is
that once I am done I will have a class that supports Windows only.

I think I need to investigate how Perl has solved the serial support
cross-platform.  Am I missing something here?  Does Python provide some
magical serial support I am unaware of?



<kain at cableadmin.com> wrote in message
news:slrn8kvrup.ld1.kain at cableadmin.com...
> I recently took over maintenance of XAL which is the X10 Abstraction
> Library, as well as began writing Python modules for it.  The XAL
> site went up at sourceforge today (http://xal.sourceforge.net) and
> the pyX10 page will go up in a couple days at http://pyx10.sourceforge.net
> I would be very interested in working with anyone on python home
automation.
>
> -Scott
> sbarron @ ansic.net
>
> In article <fsteele-DDF9A7.16453020062000 at news.inet-systems.net>,
> Frank Steele wrote:
> >In article <m3r99swt4x.fsf at hare.demon.co.uk>, Les Smithson
> ><lsmithso at hare.demon.co.uk> wrote:
> >
> >> I have some Python scripts that control a TWSER & TW523. This includes
> >> low-level drivers, and a higher-level server that lets you control X10
> >> over a network. I also wrote a simple TkInter GUI for controlling my
> >> X10 devices.
> >>
> >> A while ago I spoke to someone who hacked my code to work with a
> >> CM12U. We planned to rework this code to properly support both TW523
> >> and CM12U, but nothing ever came of it.
> >>
> >> Is it time to resurrect this?
> >>
> >
> >
> >I would certainly be interested in having a look and modifying these for
> >my own (sinister) purposes, after which I would be happy to share
> >changes back into the community.
> >
> >I had one other response, so I suppose we could actually try to work
> >together in an organized way to move the project forward....
> >
> >Any other takers?
> >
> >please-take-one-step-forward'ly y'rs,
> >
> >Frank
> >fsteele at mindspring.com
> >
> ><snip>
> >
> >>     >> So, is anyone aware of Python modules or extensions to do X10
> >>     >> home automation? I'm interested hearing about them on _any_
> >>     >> platform, but I intend to implement on Linux and MacOS....
> >>     >>
> >>     >> Thanks, Frank fsteele at mindspring.com




From moshez at math.huji.ac.il  Sat Jun  3 04:08:06 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 3 Jun 2000 11:08:06 +0300 (IDT)
Subject: search for a string?
In-Reply-To: <8h6211$ark$1@news-int.gatech.edu>
Message-ID: <Pine.GSO.4.10.10006031107390.14749-100000@sundial>

On 1 Jun 2000, Shengquan Liang wrote:

> how do i search for a string in a document
> 
> and inseart content of another file before
> that string?

Read the documentation of the string module, and the builtin function
"open".

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From david_ullrich at my-deja.com  Thu Jun  8 12:16:54 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 08 Jun 2000 16:16:54 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl>
Message-ID: <8hogt2$57t$1@nnrp1.deja.com>

In article <FvuAy4.51q at cwi.nl>,
  "Dik T. Winter" <Dik.Winter at cwi.nl> wrote:
> In article <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta at 4ax.com>
urner at alumni.princeton.edu writes:
>  > Recent number theory breakthrough:  you can get the
>  > nth digit of PI without doing all the work needed
>  > to compute all the digits up to the nth digit (same
>  > with some other numbers -- not sure if we know all
>  > the relevant criteria as to which).
>
> I checked the pages and have to make a correction here.  There is no
such
> claim.  The only claim is that it can be done without using much
memory.
> As one of the referred pages says:
>   "we have not found a faster algorithm, nor have we proven that one
does
>   not exist."
> Se essentially the amount of work is still the same, but the memory
> requirents are smaller.

    ??? In one of the references cited the authors say

"The computation can
be achieved without having to compute the preceeding digits. We claim
that the algorithm has a more theoretical rather
than practical interest, we have not found a faster algorithm, nor have
we proven that one does not exist."

The statement "The computation can be achieved without having to
compute the preceeding digits." could be a lie, I suppose, but
it's certainly at least a _claim_ that they can calculate the
n-th digit without finding the preceding ones.

    Regarding the "have not found a faster algorithm": Are
you certain they're saying that none of this is any faster
than finding the first n digits, or do they mean that they
have not found a faster "digit-extraction" algorithm than
the original Bailey-Borwein-Plouffe algorithm?

    The other reference begins

"David Bailey, Peter Borwein and Simon Plouffe have recently computed
the ten billionth digit in the hexadecimal
 expansion of pi. They utilized an astonishing formula:

[astonishing formula goes here]

 which enables one to calculate the dth digit of pi without being forced
to calculate all the preceding d-1 digits."

    If these algorithms _really_ take as much time as finding
all the digits, as you seem to be saying, it's hard to see why
they'd have everybody so excited (or rather why they _did_ have
everybody so excited back when this was news). It's also hard to
see why people would suggest that one point to the digit-extraction
algorithms would be to be able to check the accuracy of calculations
of _all_ the digits.

> --
> dik t. winter, cwi, kruislaan 413, 1098 sj  amsterdam, nederland,
+31205924131
> home: bovenover 215, 1025 jn  amsterdam, nederland;
http://www.cwi.nl/~dik/
>


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


From aahz at netcom.com  Tue Jun 27 13:56:55 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 17:56:55 GMT
Subject: C++ style const's & types in Python
References: <v3465.71261$7o1.1704126@news2.rdc1.on.home.com>
Message-ID: <8japt7$su7$1@slb2.atl.mindspring.net>

In article <v3465.71261$7o1.1704126 at news2.rdc1.on.home.com>,
Tom <tom-main at REMOVEME.home.com> wrote:
>
>I'm a C++ developer learning Python.

I'm not sure I'd agree with that statement, given your insistence on
following C++ paradigms in Python.

>One is const types.  Particularly given that values are passed to
>methods by reference.  But am I missing something?  Is there any way
>in Python to specify that a value (other than a tuple) must remain
>constant?

Not really.  All-caps (e.g. DEFAULT_TIMEOUT) are sometimes used as an
indicator.  Note, however, that function defaults are pretty close to
constant.

>Also, type declarations, while a nuisance in short programs, can be
>valuable in larger programs.  It would be nice if Python names could
>optionally have their types fixed.  As a substitute, I think I will
>start using Hungarian notation to prefix my Python names in cases where
>I don't intend to dynamically change their type.  Is this stupid?  Is
>there a better alternative?

Hungarian notation is a Bad Idea even in C++ programs.  Type checking
(other than for optimization) just doesn't seem to be an issue for
experienced Python programmers.  There's a belief around here that the
reason for this is that type checking is a crutch for a class of
problems that doesn't occur naturally if you're programming "the Python
Way".

Python is all about interfaces and algorithms, not types or inheritance
(despite Python's excellent class system).  For example, if you write a
specialized disk-based merge sort function, you *know* that it will work
with any input (including user-defined classes) as long as the supports
pickle and cmp().

Here's my standard quote on the subject:  The way to build large Python
applications is to componentize and loosely-couple the hell out of
everything.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From pshinners at mediaone.net  Sat Jun 17 14:22:56 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Sat, 17 Jun 2000 18:22:56 GMT
Subject: block comments
Message-ID: <4mP25.63531$WS3.629201@typhoon.we.rr.com>

as i get working with python more i start to miss the
block commenting i had in C/C++

i'm just doublechecking.. is there a block comment
symbol in python? i'd like to commentout a large
chunk of code with a simple start/end comment.

thinking about it, i suppose i could use the triple
quotes to simply turn a large chunk of code into a
string. is there a better way?

should python have block comments? (py3k) and if it
does there must be an answer to the eternal question...
Are the block comments nestable?






From cut_me_out at hotmail.com  Tue Jun  6 17:53:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 17:53:54 -0400
Subject: Why Do We Use This?
References: <8hjogr$d14@autodesk.autodesk.com>
Message-ID: <etdvgzma32l.fsf@w20-575-109.mit.edu>

> Okay, what is the difference between these?
> 
> If these are same, whey do we bother using the "long" version such as
> __len__(a)??

As far as I know, there is no reason to use __len__ (a), except to
access an overriden method of a subclass.  If a class A defines a __len__
method, that is what ends up being called when the function len is
called with an instance of A.

Sorry for the clumsy sentence.  Perhaps this example will help:

>>> class eg:
...  def __len__ (self):
...   print 'In the __len__ method, now'
...   return 1729
... 
>>> t = eg ()
>>> print len (t)
In the __len__ method, now
1729
>>> 

Alex.


From phd at phd.russ.ru  Mon Jun 26 11:54:57 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 26 Jun 2000 15:54:57 +0000 (GMT)
Subject: Pickle Question
In-Reply-To: <eviltofu-9E08E0.21184121062000@news.earthlink.net>
Message-ID: <Pine.LNX.4.21.0006261553420.26469-100000@fep132.fep.ru>

On Thu, 22 Jun 2000, Jerome Chan wrote:
> If I unpickle from a non - pickled file or byte stream what error does 
> Pickle throw? I've gotten both KeyError and EOFError, does it throw 
> anything else?

   Sometimes it "throws" segmentation fault, and you cannot catch this.
There is a bug in cPcikle (which I beleieve fixed in 1.6).

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From eduard.hiti at t-online.de  Tue Jun 13 17:03:16 2000
From: eduard.hiti at t-online.de (Eduard Hiti)
Date: Tue, 13 Jun 2000 23:03:16 +0200
Subject: WIN32API GetLastError()
References: <8i3460$ppm@news.or.intel.com> <8i5vjl$bh9$1@nnrp1.deja.com>
Message-ID: <8i67nm$l37$12$1@news.t-online.com>

>
> Note that many functions DO NOT support GetLastError() - they return
> the error code in a result, so it is pointless.  GetLastError() is
> mainly used for functions that return bools, or (say) handles, where
> only failure is indicated, and not the specific failure.
>

I had problems with win32api.GetLastError, too. Try this code:

    >>> a = win32event.CreateEvent(None, 0, 0, "Test")
    >>> b = win32event.CreateEvent(None, 0, 0, "Test")
    >>> win32api.GetLastError()
    0
    >>>

This is said in MSDN about CreateEvent:
If the function succeeds, the return value is a handle to the event object.
If the named event object existed before the function call, the function
returns a handle to the existing object and GetLastError returns
ERROR_ALREADY_EXISTS.

But obviously win32api.GetLastError does not return ERROR_ALREADY_EXISTS.

I helped myself by using OpenEvent and catching exceptions, but it seems
that GetLastError is problematic.

Bye,

    Eduard






From m.faassen at vet.uu.nl  Mon Jun  5 08:03:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Jun 2000 12:03:59 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <memo.20000604004935.14417B@digitig.compulink.co.uk> <slrn8jlm1j.bkc.wtanksle@dolphin.openprojects.net>
Message-ID: <8hg4vf$6eo$2@newshost.accu.uu.nl>

William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>Lesson taken. I won't give any credence to those who complain about case 
>>sensitivity on the same grounds.

> But this mythical person complaining about case-sensitivity is *correct*;
> Python truly *is* case-sensitive.

Doesn't mean they'd be right any more than if they complained about Python's
indentation.

Let's-leave-the-mythical-persons-out-of-this-ly yours,

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


From david_ullrich at my-deja.com  Sat Jun 10 15:28:56 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sat, 10 Jun 2000 19:28:56 GMT
Subject: No strptime in Windows?
Message-ID: <8hu4tc$6m0$1@nnrp1.deja.com>

    The docs say that time.strptime may not be available
on all platforms. It doesn't seem to be available under MS
Windows. I don't see why not - there's nothing Windological
about strptime but the same applies to the other time
functions that _are_ included.

    (I'm getting the time from somewhere on the internet.
What I get is something like "Sat Jun 10 18:34:10 2000"
and I have to subtract five hours. Ended up more or less
writing my own strptime, at least I _think_ I did...)

    Also: Is Monday really day 0? I woulda thought
Sunday was day 0. (Doesn't seem to matter in the
subtract-five-hours thing, I get the same results
either way. Evidently mktime() ignores the day-of-week
if it's inconsistent with the day-of-month field?)

DU.


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


From m.faassen at vet.uu.nl  Mon Jun  5 08:01:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Jun 2000 12:01:04 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net>
Message-ID: <8hg4q0$6eo$1@newshost.accu.uu.nl>

William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
> On 30 May 2000 10:59:24 GMT, Martijn Faassen wrote:
>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
[snip]
>>> Your rhetorical questions answer themselves.  Why should a programmer
>>> enforce these things?  Because it's stupid to NOT enforce them.  So
>>> enforce them already!

>>That's nice to say, but you don't agree with that yourself if you think
>>Python's use of indentation is a good thing. It can help a lot if the
>>language makes it impossible to be inconsistent. Unless you think the
>>_only_ advantage of Python's indentation strategy is the lesser line 
>>count.

> This is a classical example of false dichotomy ;-).

It's not.

> I liked Python's
> indentation not because it enforced consistency, but rather because it
> makes code look good.  In fact, it could be argued that Python does _not_
> enforce consistency; the programmer is free to choose any indentation for
> each block, so long as each block can be parsed (which does require
> internal consistency).

"""
I liked Python's
case sensitivity not because it enforced consistency, but rather because it
makes code look good. In fact, it could be argued that Python does _not_
enforce consistency; the programmer is free to choose any case-spelling for
each variable name, as long as each variable name can be parsed.
"""

Consistency in style happens to one of the most important factors that
makes code look good (on the syntactic level). Your indentation looks
better in Python because Python complains loudly if you don't indent properly.
That's not the most important thing though; you can indent properly in
any language. It's just that everybody indents properly in Python, and
more or less in the same way.

> More recently I've come to realise that the reason Python indentation
> looks good is that Python, like most modern languages, is a
> tree-structured language, and so is best represented in two dimensions.
> Most modern languages, however, attempt to represent themselves in one
> dimension: an ASCII stream which ignores lines and whitespace.

Well, why shouldn't a programmer enforce such indentation in any
language? Indentation looks good in any language, not just Python, so
that's not really an argument for Python's indentation style.

Besides, if you ignore case you ignore that words in Python are spelled
consistently in case, which is *also* a property of lots of modern languages.
Sometimes not enforced, but still a desired property of the programmer.

> So I don't really care about languages enforcing consistency; if I need
> it, I'll enforce it myself.

A computer language would be worth less if it brought no consistency at
all. I want my language to enforce the amount of parameters I pass to
a function, unless I overrule it myself. I want my language to consistently
enforce namespaces. I want my language to language to be consistent in
the way I can call functions (foo(1, 2) and not also foo 1, 2). I want
my language to be consistent in the way I spell keywords (I don't
want to see code with IF and ELIF all over the place while other modules
use 'if' and 'elif'). I want the language to be consistent about rejecting
things that make no direct sense ( 1 + "foo"..see also the discussion on
1/2).

According to your argument, you will just enforce this consistency
yourself. The problem is that, *even* if everybody enforces self
consistency, people will enforce *different* consistency and their
coding style will vary over time. From a maintenance and communication
point of view, this is bad. 

Any consistency that *can* be enforced by the language *should* be enforced
by the language, with the following caveats:

  * I explicitly overrule something

  * the enforcement of consistency makes the language significantly *less*
    readable/writable.

> Take a lesson from the people complaining
> about Python's indentation: one of the leading false accusations is that
> it's inflexible, that it won't allow people to choose thier own
> indentation.

I don't see how this maps to the case sensitivity argument at all. 
Python's indentation *is* less flexible in what code-layouts are
allowed, and it *is* a silly accusation. So I don't see why I should
take a lesson from this complaint or those people. Besides, people who make
this silly accusastion generally use languages which *are* case
sensitive. So I don't see how this relates to anything.

Regards,

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


From beat.bolli at dware.ch  Fri Jun 30 09:20:14 2000
From: beat.bolli at dware.ch (Beat Bolli)
Date: Fri, 30 Jun 2000 15:20:14 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8jhum101vku@news2.newsguy.com> <962367281.1198008296@news.libertysurf.fr>
Message-ID: <8ji6qg$8oq$1@news1.sunrise.ch>

Jerome Quelin <jerome.quelin at insalien.org> schrieb in im Newsbeitrag:
962367281.1198008296 at news.libertysurf.fr...
> "Alex Martelli" <alex at magenta.com> wrote:
> >> I was wondering how to pass variable-length keyworded arguments list
> >> to my base-class constructor? Look at the example code below:
> >Doesn't apply work for __init__ just as for any other function?
> Sure it works for __init__ too, I've already tried this approch, but it
doesn't
> work either. In fact, I'm asking the question for __init__ because I run
into
> this problem when calling constructors, but the question is the same for
every
> inherited and overloaded method.
>
> When using the *args syntax, I get an error when calling my constructor
because
> I'm using keywords arguments.
>
> When using the **args syntax, python fetches all keyworded arguments in
_one_
> dictionnary in my constructor, and I want to keep the arguments in
multiple
> key/value pairs (and not flattened in one dictionnary) since the parent
> constructor is waiting for keyworded arguments too.
> If I use the **args syntax and then call the base constructor with:
> Parent.__init__(self, args)
> I got an error since it doesn't wait for a dictionnary as an argument (and
args
> is a dictionnary in this case) but for key=value arguments, with key and
value
> taken from the args dictionnary.
>
> Any idea?

Starting with Python 1.6a[12], you can call Parent.__init__(self, **args).

Beat Bolli
--
dware design & software GmbH
Mattenstrasse 11, CH-2555 Br?gg b. Biel
Telefon: +41 (32) 374 27 00, Telefax: +41 (32) 374 27 01
E-Mail: beat.bolli at dware.ch





From ngps at madcap.dyndns.org  Tue Jun 13 10:50:45 2000
From: ngps at madcap.dyndns.org (Ng Pheng Siong)
Date: 13 Jun 2000 14:50:45 GMT
Subject: Python Extensions with Borland's Free Compiler Tools
Message-ID: <8i5ho5$40d$1@violet.singnet.com.sg>

Hi,

Here's a followup to Gordon Williams's document 'Creating Python 
extensions using Borland's free compiler', available at the following
URLs:

    http://www.deja.com/=dnc/getdoc.xp?AN=619698470 
    http://www.cyberus.ca/~g_will/pyExtenDL.shtml


I've found a problem with the above in my environment. Using Gordon's 
example, if I change example_add() to simply return Py_None (after 
Py_INCREF()'ing it, of course), instead of a Python integer, I get a crash.


Py_None is declared thusly in Python's Include/object.h:

  extern DL_IMPORT(PyObject) _Py_NoneStruct; /* Don't use this directly */
  #define Py_None (&_Py_NoneStruct)


And in Python's config.h, DL_IMPORT is defined thusly:

  #ifdef _M_IX86 (this for MS C compiler)
  [ other stuff ... ]
  #define DL_IMPORT(RTYPE) __declspec(dllimport) RTYPE
  #endif
  
  #ifdef __BORLANDC__
  [ other stuff ... ]
  #define DL_IMPORT(RTYPE) RTYPE __import
  #endif


DL_IMPORT appears to be defined incompatibly. Compiling an extension with 
BC and linking it with Python's MSVC-built core generates a runtime crash 
when Py_None is [de]referenced.


Here's how to work around it:

1. Add the following to Python's config.h, after the Windows 
compiler-specific stuff:

  #ifdef MSC_CORE_BC_EXT
  #define DL_IMPORT(RTYPE) __declspec(dllimport) RTYPE
  #define DL_EXPORT(RTYPE) __declspec(dllexport) RTYPE
  #endif

  /* End of compilers - finish up */

Note the above comment line already exists. 


2. When compiling an extension, add "-DMSC_CORE_BC_EXT" to CFLAGS.

MSC_CORE_BC_EXT informs the compiler it is compiling an extension with 
BC for linkage with MSVC-built Python core.


I've used the above technique with M2Crypto, where (in the latest snapshot,
to be posted) the SWIG-generated extension is compiled by BC and then linked 
with MSVC-built Python and OpenSSL DLLs. It seems to work. In fact, at one 
troublesome spot where the MSVC-built package crashes, the BC-built one works! 
Go figure...


Cheers.

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



From stephen at cerebralmaelstrom.com  Thu Jun 22 14:34:26 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 22 Jun 2000 18:34:26 GMT
Subject: strings and integers?
References: <8ir75b$qfi$1@news2atm.raleigh.ibm.com> <394FF466.51A5B77B@sightreader.com> <8it0j0$1du6$1@news2atm.raleigh.ibm.com>
Message-ID: <S_s45.81332$WS3.720017@typhoon.we.rr.com>

Arint? <shouldbe at message.com> wrote in message
news:8it0j0$1du6$1 at news2atm.raleigh.ibm.com...
> Sorry,  I wasn't clear in last post.  I was in a hurry and frustrated with
> embedding python in my app.  If I had time I would probably go with COM
for
> the application and us python to talk to it.  Anyhow I don't even know
COM.
> > > Is
> > > x = "txt %(somevar)"%vars()
> > > the only way to append text to a string?
> What I meant to ask here was, Is this the only way to append integers to a
> string?

    My *personal* opinion is that it's the best way. I've said it before, I
*love* the modulo-printf-ish-operator... but, you can also use the '+'

    e.g. text = "Some String" + str(YourNumber)

    You have to stick 'str' in there, otherwise it'll raise a TypeError

> > > Also, why do I get a len() of unsized object errors?  Do I need to
> rebuild
> > > the python dlls? Download 1.6?
> >
> > No.  It is very unlikely that you have found a python bug.  Show me an
> > example.  When asking a question, always show an example.
>
> This question was my example.
> xrt = 8
> print'ioctl %d <---' %xrt
> Python will print "ioctl 8 <----", but it will still give an error of
len()
> of unsized object on the corresponding  line of code.

    This works perfectly for me. What version are you using? Do you have any
nasty 'from module import *'s laying aroudn? Dunno, it seems to me that
whenever something behaves totally /oddly/ that one of those is to blame. :)

--S





From bvdpoel at uniserve.com  Tue Jun 13 20:39:32 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Tue, 13 Jun 2000 17:39:32 -0700
Subject: PATCH: Augmented assignment
References: <m130q7t-000wcDC@swing.co.at> <3946C3A3.619B@esatclear.ie>
Message-ID: <3946D444.8B1AEF64@uniserve.com>

Russell Wallace wrote:
> 
> > invalid.address at 127.0.0.1 (David) wrote:
> >
> > > On Fri, 9 Jun 2000 16:39:38 +0200, Thomas Wouters <thomas at xs4all.net>
> > > wrote:
> > >
> > > >As promised, here is my patch to add augmented assignment (+=, -= and
> > > >family) to Python. It's still Work In Progress, and probably requires a
> > > >rewrite of some magnitude before Guido will accept it.
> > >
> > >
> > > Why would the Python community want this?  It obfuscates code, in as much
> > > that newbies aren't likely to intuite what the symbols mean.
> 
> I disagree.  Indeed, I put it to you that
> 
>         foo += 1
> 
> is *more* intuitive than
> 
>         foo = foo + 1
> 
> It's not a huge issue, but it makes things that little bit simpler and
> more convenient; and "simpler and more convenient" is Python's raison
> d'etre, IMO.

It certainly saves more than one or two keystrokes to 'lazy typists',
expecially if using longish variable names. In addition, this syntax can
help in eliminating errors when you might:

	foobar = fooobar + 1

when, of course, you meant

	foobar =foobar + 1

I'm sure that folks who use more creative names will see more creative
examples.

I'm all for including += and -=. *= and /= would be nice, but not nearly
as useful. Auto increment/decrement is proably best left to C-guys.


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


From m.faassen at vet.uu.nl  Wed Jun 14 07:12:27 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: Wed, 14 Jun 2000 07:12:27 -0400
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <642A954DD517D411B20C00508BCF23B001296A31@mail.sauder.com>

jin choung <jin.choung at gte.net> wrote:
> howdy all,

> i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
> tried to pick up python before moving on to c++.  however, i was trying to
> work my way through the book LEARNING PYTHON and while it says that there is
> no need to have prior experience in languages like c or c++, i really found
> that that was not true.

> anyone else feel this way?

I think that this would be more a problem of the book (Learning Python)
than a problem with the language. "Learning Python", while it's fine for
people who have significant experience with another language, really should've
gotten a different title, in my opinion. It's *not* good for people with
little or no experience with programming. I've heard both "The Quick
Python Book" and "Python in 24 hours" are much better in that respect.

"Learning Python" assumes too much that a newbie programmer really doesn't
know. I've tried to teach Python to someone with that book, but the book
was just too much for her. Too many concepts assumed that she didn't
know about. That said, it has worked for others.

> i'm currently working my way through a 24 hour c++ book and i think this is
> going to work for me.

Oh, you are so wrong. :) I mean, you may pick up some simple C++, but C++
is definitely a way more complicated language than Python. You just picked
up a book that works better for you than the Python book, but the language
is far far more complicated. It takes many people a year to get proficient
in C++; even experienced programmers. It rarely takes an experienced
programmer longer than a month to get proficient with Python, especially if
that programmer has experience with object oriented concepts.

> oh, well, will be back to contribute here in a couple
> of months... in the meantime, i'll just leech valuable nuggets of wisdom
> from you guys.

I do suggest you take a look at some other Python books first, and then
take a look at some other C++ books. C++ is really not an easy language
to pick up. And it's really not necessary to learn C or C++ before you
can learn Python -- the other way around is indeed an easier path to take.

Regards,

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



From boncelet at udel.edu  Tue Jun 13 20:20:45 2000
From: boncelet at udel.edu (Charles Boncelet)
Date: Wed, 14 Jun 2000 10:20:45 +1000
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net>
Message-ID: <3946CFDD.62D0A032@udel.edu>

David wrote:
> 
> On Fri, 9 Jun 2000 16:39:38 +0200, Thomas Wouters <thomas at xs4all.net>
> wrote:
> 
> >As promised, here is my patch to add augmented assignment (+=, -= and
> >family) to Python. It's still Work In Progress, and probably requires a
> >rewrite of some magnitude before Guido will accept it.
> 
> Why would the Python community want this?  It obfuscates code, in as much
> that newbies aren't likely to intuite what the symbols mean.

I, for one, think the patch idea is great and hope it gets added 
immediately.

It is important that Python be accessible to newcomers, but it also 
must be convenient to experts as well.  It does little good if 
newcomers learn Python, then move on to a more convenient/powerful
language.

For instance, I just finished some code that did a lot of
>>>self.array[index] = self.array[index] + 1

I would have loved to be able to type
>>>self.array[index] += 1
instead.  My code wasn't causing any side effects, so that's not the
issue.  The issue is I didn't like typing all those symbols.  (I'm
sorry,
but I don't like typing and I expect I'm not alone. I like that I don't
have to type those silly braces other languages require. :-)

I know there are plenty of languages that allow the "+=" construct. 
That Python lacks it is not enough to cause me to change, but it may
cause
others to, or, more likely, cause them to never pick up the language.
 
>
> And it's not like it adds new functionality nor improves code readibility.
> It just saves the lazy programmer a whopping one or two keystrokes.

As others have pointed out, it does add functionality.  For example, 
awhile back Tim Peters pointed out that x = x+1 can be much slower than
x += 1 for large matrices.  And, in my opinion, it does improve
readability.

------
Charles Boncelet, University of Delaware, <boncelet at udel.edu>
On sabbatical at ADFA, Canberra Australia, <boncelet at ee.adfa.edu.au>
Home Page: http://www.ece.udel.edu/~boncelet/


From wlfraed at ix.netcom.com  Thu Jun 15 23:45:54 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 15 Jun 2000 20:45:54 -0700
Subject: OS-Memories (was: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <85em60nozb.fsf@one.net.au> <8i8o8b$83i$0@216.39.141.36> <crigkss2m7ick2imqdctubg5d4lqjpvtd1@4ax.com> <8iar91$gt6$0@216.39.141.36>
Message-ID: <7q6jkscmulnv1f1o4voq09abte261n88np@4ax.com>

On 15 Jun 2000 15:04:01 GMT, Dana Booth <dana at mmi.oz.net> declaimed the
following in comp.lang.python:

> 
> Good old Tandy/Radio Shack original that came with my TRS-80 III, complete
> with 48k ram. :)
>
	LDOS (which later became LSDOS, and then licensed as TRSDOS for
the model 4) was my first major purchase -- it could handle some
operations on a single drive system...
 
> Ever try out CP/M on those old Osborne's or Kaypro's?

	Worse -- CP/M on a Zentec! CP/M required RAM at low addresses
(was, as I recall, how it got back to the command prompt after a program
exited). Zentec's had ROM down there, so the entire CP/M kernal had to
be customized. I think the copy of Magic Wand was also customized --
they had "Lockheed Missles and Space" <G>.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From ssthapa at midway.uchicago.edu  Fri Jun 23 10:08:40 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Fri, 23 Jun 2000 14:08:40 GMT
Subject: How to convert a string to hex?
References: <3952EC0D.D5CF9066@esec.com.au>
Message-ID: <slrn8l6q3d.h4i.ssthapa@ntcs-ip48.uchicago.edu>

Sam Wun <swun at esec.com.au> wrote:
>Suppose here is:
>
>b = "6b584"
>
>I would like to b to hex. Does anyone know how to do that in python?
>

    Try this:
	
    def atoh(s, hexdigits = string.hexdigits):
	value = 0
	for char in s:
	    index = string.find(string.hexdigits, char)
	    if index >= 16:
		index = index - 6
	    value = value*16 + index
	return value
    
    This may not work in general but it works for the case you gave and
some other test case I tried out.  It should work for any string including
those where the hex digits are in caps or in mixed case just make sure
you strip thr leading 0x if its present and that the string passed in is
really a hex number.

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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From johngrayson at home.com  Mon Jun  5 10:21:32 2000
From: johngrayson at home.com (John Grayson)
Date: Mon, 05 Jun 2000 14:21:32 GMT
Subject: Scrolled canvas confusion
References: <8hg4a3$a09$1@zingo.tninet.se>
Message-ID: <8hgd0r$1p8$1@nnrp1.deja.com>

In article <8hg4a3$a09$1 at zingo.tninet.se>,
  "Andr? Dahlqvist" <andre at beta.telenordia.se> wrote:
> Hi
>
> I have an application (using Tkinter) which currently stacks a bunch
of
> buttons on top of each other in a Frame. The number of these buttons
> change dynamically, which sometimes causes them to not fit on the
> screen. For this reason I'd like to use a scrollbar. Since the Frame
class
> doesn't have a yscrollcommand I think I need to pack this frame into a
> Canvas, and then somehow use the create_window method with
> the frame as an argument. What confuses me though are the various
> coordinates, like the scrollregion of the Canvas and the x and y
> coordinates to the create_window method. Also I don't know what
> sizes to give the Frame and Canvas to make them behave as I want
> them to. If possible I'd like to do this with plain Tkinter, no Pmw.
>
> I would greatly appreciate any help. A small example would be
> great.
>
> // Andr?
>

If you don't want to use Pmw (which is only Tkinter code anyway),
then look at PmwScrolledFrame.py and extract the bits you want and
manage resources directly.

If you want an example of putting buttons inside a scrolled frame,
see page 69 of Python and Tkinter Programming.


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


From pduffin at hursley.ibm.com  Fri Jun 30 06:11:57 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 30 Jun 2000 11:11:57 +0100
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <3958C648.1D76@rd.francetelecom.fr> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com> <395B6D30.2486@rd.francetelecom.fr>
Message-ID: <395C726D.D6B17A@hursley.ibm.com>

Alexandre Ferrieux wrote:
> 
> Paul Duffin wrote:
> >
> > Alexandre Ferrieux wrote:
> > >
> > > Paul Duffin wrote:
> > > >
> > > > Apparently there is a port of Expect for Python so why not use that
> > > > rather than try and reimplement it all over again.
> > > > Expect has over 7 years of development and centuries of testing on many
> > > > different platforms, do you really want to reproduce that. If so why,
> > > > what exactly is wrong with Expect.
> > >
> > > The same as with TkInter: it needs a full Tcl inside. Again I've got
> > > next to nothing against Tcl, it's just that the separation of orthogonal
> > > things is such a natural thing for me.
> > >
> >
> > ...why not do as I suggested
> > and work with Don Libes on seperating the pseudo tty control stuff into
> > a seperate library which both Tcl and Python can use.
> 
> Perfect. For some reason I didn't realize you were suggesting this.
> What next ? You're a native English speaker, you call Don on the phone
> ;-) ?
> 

The thing is that I am not interested enough in this area to invest my
limited amount of time doing the work. What is needed is a group of
people from the different languages who want to provide really good
Expect like support to get together and sort it out.


From matt at nightrealms.com  Mon Jun 26 03:44:56 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Sun, 25 Jun 2000 23:44:56 -0800
Subject: Control cmp used with 'in'?
Message-ID: <52D55.2660$Gh6.168003@news-west.usenetserver.com>

Is there any way to control the function that is used during
an 'in' search?

Thanks in advance.





From ullrich at math.okstate.edu  Mon Jun 19 11:29:08 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 19 Jun 2000 15:29:08 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <20000619184000.639516@210.9.22.17>
Message-ID: <01bfda03$3bc22700$55a28ad1@daves-dell>


Matthew Schinckel <matt at null.net> wrote in article
<20000619184000.639516 at 210.9.22.17>...
> "William Dandreta" <wjdandreta at worldnet.att.net> wrote:
> > I recently read a message that suggested that using regular
> > expressions might be faster than the find function. I've heard the
> > term regular expression used from time to time but I don't know what
> > it means. Could someone give me a simple explanation?
> > 
> > Bill
> 
> I have come across one example where re was much faster than using
> string.find: I was stripping out html tags, and doing a one line
> re.sub('<*?>','') was much faster than using string.find() and slice
> notation.

	??? You're not saying that the re was faster than find, you're saying
that it was faster than find _plus_ something else. Of course it is - if
regular expressions were not faster than equivalent blocks of code
they wouldn't be there.

	(For that matter the search itself would require some code, not
just one find - if you say find(s, '<*?>') you not going to find what
you're
looking for.)

> Try both, use the profiler, and compare. (I'll guess someone else will
> give an explanation about them).
> 
> Matt.
> 


From bparsia at email.unc.edu  Tue Jun  6 22:00:37 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Tue, 6 Jun 2000 22:00:37 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <m3wvk2wy1u.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <1ebtr5d.ucu7i31i9yjgmN%bparsia@email.unc.edu>

Michael Hudson <mwh21 at cam.ac.uk> wrote:

> "Richard P. Muller" <rpm at wag.caltech.edu> writes:
> 
> > Does Programming in Python Marginalize You?
> 
> I hope not.
>  
> > I love Python. Of the 10-15 programming languages I know, Python is
> > the only language which I actually look forward to using. 
> 
> Then you should learn Common Lisp, Haskell, ocaml, scheme and
> Smalltalk (? I don't know smalltalk, but it *looks* like fun).

Ooo, ooo. With a straight line like that, how can I keep silent?

(Duct tape, suppose. A lot. Though, I'm pretty good at chewing, so it
might not suffice.)

Smalltalk is a lot of fun. Observing someone going from Python (his
first language) to Smalltalk (his second), as I am observing, is
enlightening.

Coming soon (er...for some value of #soon which is near "sometime when I
get around to it"), a Smalltalk <--> Python snippet mapping. (Strange
brain glitch, I forgot what these are standardly called :))

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From owen at astro.washington.edu  Mon Jun 12 13:01:57 2000
From: owen at astro.washington.edu (Russell E. Owen)
Date: Mon, 12 Jun 2000 10:01:57 -0700
Subject: Where are the math functions?
References: <slrn8k8dfv.tu.hzhu@mars.localdomain>
Message-ID: <owen-97F32C.10015712062000@news.u.washington.edu>

In article <slrn8k8dfv.tu.hzhu at mars.localdomain>, 
hzhu at users.sourceforge.net wrote:

>I'm quite puzzled by not being able to find most of the common special
>functions, such as (incomplete) gamma, beta and bessel and erf.
>
>This arrises from effort to provide statistical functions for the MatPy
>package.  Many stat functions require incomplete gamma/beta and some
>variants.
>...

The random module (which is available on all platforms) offers beta and 
gamma functions. I can't comment on incomplete gamma, bessel and erf. I 
found the information about the random module in "Python Essential 
Reference" by David Beazley; the index lists beta and gamma, but not the 
others.

-- Russell


From tbradshaw at orchestream.com  Thu Jun 15 08:48:08 2000
From: tbradshaw at orchestream.com (Tobe)
Date: Thu, 15 Jun 2000 13:48:08 +0100
Subject: Python and the PIL...
Message-ID: <8iajbv$m1m$1@hunter.orchestream.com>

For reasons too desperately boring to go into.. I am missing my Linux box at
the moment..

I have some code that relies on the Python Imaging Library which I'm trying
to use on my NT box but I'm finding a few probs.. namely Image.thumbnail(())
produces garbage..

Is this a known problem or do I just have the thing misinstalled (or
other)..

Cheers,

t o b e
Orchestream
London.




From wlav at hpatl26.cern.ch  Fri Jun  9 05:04:59 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 9 Jun 2000 09:04:59 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <8hqbvr$fgr$1@sunnews.cern.ch>

stevemul at ozemail.com.au (Steve Mullarkey) writes:

>I have read in several places productivity claims of 5 to 10 times over
>'C' and "C++". I would like to ask for some feedback from "C++"
>programmers who have moved to Python as to whether these estimates are
>realistic.

I'm a C++ programmer (leave out the quotes ;) ).

First off, any claim of an order of magnitude increase in productivity is
pure nonsense. See Brooks for an explanation or Boehms and McConnell for
hard data.

<snip enumeration of tools>

>I don't want to start a flame war but, given the above, I just can't see
>where the productivity increase is generated. I am genuinely interested
>to ask "C++" programmers :-

>1. What productivity increase do you achieve ?

>2. How long did you use Python before you achieved increased
>productivity ?

I want to use Python for scripting tasks. I dislike 'ordinary' scripting
languages as they lack the ability to recover from errors; a problem that
Python, being a real programming language, does not have.

Up to now, I've been using C++ for scripting. I'm not proficient in Python
yet, so writing a Python script takes far longer than writing a C++ utility
and the Python code itself is larger than the equivalent C++ code.

What I expect from Python, is that I can reuse scripting utilities,
something that I do not do for C++ utilities. This could, perhaps, yield
a 10% increase in productivity, but I'm not writing scripts all day, so
my overall productivity increase will be negligible.

Most of all, however, I enjoy the incremental coding (I come originally
from a Smalltalk background). I *like* Python, which is all excuse I need
to use it.

Best regards,
         Wim Lavrijsen



From stephen at cerebralmaelstrom.com  Wed Jun  7 01:26:45 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 07 Jun 2000 05:26:45 GMT
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com>
Message-ID: <p2l%4.40729$WS3.454680@typhoon.we.rr.com>

Ack. _Why_? Ick, yick, sick!

I know how you can solve your problem, though.

C:> del lambdaabuse.py

Assuming you are in UNIX, replace the 'del' with 'rm'.

Please? :)

--Stephen

Ben Wolfson <rumjuggler at cryptarchy.org> wrote in message
news:stgrjs0sbt74kb3m4b7dbkkc87910fl1r6 at 4ax.com...
> I'm attempting to abuse Python's lambdas, and I've come across a
> problem I can't figure out.  Well, the problem is actually that I'm
> attempting to abuse lambdas, but aside from that I don't know why the
> abuse is failing.  Here's what I've got:
>
> def __everyother(): #factory to return lambdas.
> return lambda l, evens, odds, app=lambda a,e: lambda
> a=a,e=e:a.append(e): Compose(apply(map, (lambda e, evens=evens,
> odds=odds, l=l,app=app: If(l.index(e)%2==0, app(evens,e), app(odds,e))
> ,l)), Return(evens, odds))
>
> The functions it calls are:
>
> def Return(*args):
> return lambda args=args: args
>
> def Compose(*args):
> return lambda fs=args: For(len(fs), lambda c, fs=fs:
> If(lambda c=c,fs=fs: c==len(fs)-1, Return(fs[-1]()), fs[c]))()
>
> def For(count, fct, *args, **kwargs):
> for i in range(count):
> fct(i,*args,**kwargs)
>
> I can do this:
> >>> import lambdautils
> >>> a = lambdautils.__everyother()
> >>> evens, odds = [], []
> >>> a(range(20),evens, odds)
> <function <lambda> at 1510c00>
> >>> lam = a(range(20),evens, odds)
> >>> evens
> [0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18]
> >>> odds
> [1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19]
> >>> lam()
> Traceback (innermost last):
>   File "<pyshell#14>", line 1, in ?
>     lam()
>   File "C:\PYTHON16\lambdautils.py", line 16, in <lambda>
>     return lambda fs=args:\
>   File "C:\PYTHON16\lambdautils.py", line 4, in For
>     fct(i,*args,**kwargs)
>   File "C:\PYTHON16\lambdautils.py", line 17, in <lambda>
>     For(len(fs), lambda c, fs=fs: \
>   File "C:\PYTHON16\lambdautils.py", line 44, in If
>     return f()
> TypeError: call of non-function (type list)
>
> Can't quite figure out where the error occurs.
>
> --
> Barnabas T. Rumjuggler
>
> Policeman: So sir, did the suspect have four arms, each one holding a
> sword, and was she wearing a girdle adorned with severed hands, her
> whole body covered in the blood of past victims?
> Mr Flynn:  Well, it was overcast and everything, so it was hard to say
> for certain, you know?
>  -- John Flynn on apihna




From thomas at xs4all.nl  Sat Jun 24 12:02:12 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 24 Jun 2000 16:02:12 GMT
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu> <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk> <1ecjb1r.27wv4iz2rfkpN%bparsia@email.unc.edu> <m34s6l4flf.fsf@atrus.jesus.cam.ac.uk> <1ecn1bz.17wl5nd1s63kc3N%bparsia@email.unc.edu>
Message-ID: <slrn8l9mrp.elg.thomas@centurion.xs4all.nl>

On Thu, 22 Jun 2000 17:20:10 -0400, Bijan Parsia <bparsia at email.unc.edu> wrote:
>Michael Hudson <mwh21 at cam.ac.uk> wrote:

>> Certainly at some point there was no LOAD_FAST opcode; every name
>> lookup would first look into a local dict, then the global one, then
>> the builtins.  Now references to locals are just indexes into (C)
>> vectors; this is much quicker.

>Ah! Thanks! Am I correct in surmising from this that the gap between a
>local access success and a miss (but hitting somewhere up the chain) is
>even greater (than before LOAD_FAST)? Hmm.

'Optimized' local accesses are very unlikely to fail, because the compiler
has already assured the name exists, locally, or it wouldn't have turned it
into a LOAD_FAST :) What can happen is that an UnboundLocalError occurs;
this happens when you use a variable before you assign to it.

What happens is this: When you use a variable (by name) in Python code, the
compiler compiles it into a LOAD_NAME bytecode. The name you try to load
gets added to the 'co_names' tuple in the func_code code-block of the
function, and the LOAD_NAME bytecode takes an index into that tuple as
argument. This is the same for all name lookups.

Then, after a function is fully compiled, the optimizer walks past all
bytecodes, taking note of all bytecodes that can introduce a name.
(STORE_NAME and IMPORT_FROM). If it sees an 'EXEC' or 'IMPORT_FROM' with '*'
as argument, it notes that too.

The list of variable names is turned into two tuples, one to hold the names
and one to hold the values, and all bytecodes that refer to one of these
local variables are translated from '(LOAD|STORE|DELETE)_NAME <index in
namelist>' into '*_FAST <index in locals-values list>'. If no 'exec' or
'from foo import *' was found, all other '*_NAME' bytecodes are translated
into '*_GLOBAL' bytecodes -- they aren't referring to local variables, so
they must be refering to global ones.

I'm not sure if the difference in speed is ever measured (I assume so, 'twas
probably before I was born anyway ;) but I'm fairly certain it's worth the
(actually pretty small!) impact on code-simplicity: the *_FAST bytecodes can
access the desired variable almost directly, using a simple index &
pointer-dereference operation. The *_GLOBAL bytecodes are slightly more
expensive, being a normal dictionary lookup. The *_NAME bytecodes are
definately more expensive, as they do a dictionary lookup in the 'locals'
dictionary, as well as the gobals dictionary if the locals fail. (Which is
fairly likely, because most local acceses will already have been translated
to *_FAST)

I'm-a-newbie-though-and-probably-incorrect--don't-take-my-word-for-it-ly yr's,
Thomas.


From ge at nowhere.none  Thu Jun 15 11:56:30 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 15 Jun 2000 15:56:30 GMT
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <O0725.1084$i06.239312@ptah.visi.com>

In article <20000615104403.A1240 at sirius.ftsm.ukm.my>, root at sirius.ftsm.ukm.my wrote:

>has anyone wrote the 8 non-attacking queen problem and conway's game of life
>in python? i try but face a few newbie problems.

I did the 8-queen program in STk (Scheme w/ Tk).  It had a
little chessboard and would show you the states as it searched. 
Alternatively, it would let you place some of the queens and
then search for a solution from there.  I should translate it
into Python.

It used continuations, but I think it was only to have a simple
way to stop it in mid-search.

-- 
Grant Edwards                   grante             Yow!  I have no actual
                                  at               hairline...
                               visi.com            


From Stuart.Ford at Boeckh.com  Wed Jun 28 16:13:22 2000
From: Stuart.Ford at Boeckh.com (Stuart Ford)
Date: Wed, 28 Jun 2000 15:13:22 -0500
Subject: Why won't exec accept this?
Message-ID: <DD768E0BEA95D211B27A0008C74C86250A6B91@BEKEXCH1>


Gene <gchiaramonte at mail.com> wrote in message
news:<RCs65.11832$_b3.285418 at newsread1.prod.itd.earthlink.net>...
> I am trying to store scripts in a dictionary, then exec the scripts as
> necessary. I have stored the following text in a dictionary:
> 
> # test ...
> def FooBar(x):
>     return x + x
> 
> # here is the dictionary...
> {'script': '# test ...\015\012def FooBar(x):\015\012    return x + x',
> 'version': 1.0, 'type': 'script'}
> 


This version works for me...

I had to also add the newline after the return statement, 
or exec returned a syntax error.

{'script': '#test ...\ndef FooBar(x):\n  return x+x\n', 
'version':1.0, 'type':'script'}

After the exec, the new function will be created in 
the namespace.

Stuart



From david_ullrich at my-deja.com  Wed Jun 14 13:11:53 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 14 Jun 2000 17:11:53 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i34pv$2ruu$1@nntp6.u.washington.edu> <8i5p73$6eo$1@nnrp1.deja.com> <3dr9a046wy.fsf@amarok.cnri.reston.va.us>
Message-ID: <8i8ec8$5f8$1@nnrp1.deja.com>

In article <3dr9a046wy.fsf at amarok.cnri.reston.va.us>,
  akuchlin at mems-exchange.org (Andrew M. Kuchling) wrote:
> david_ullrich at my-deja.com writes:
> >     One of the very few things that bugs me about the Python
> > docs... let's say one of the _incredibly_ few things that
> > bugs me about the Python docs is where the explanation for
> > how something works refers to a UNIX manpage.
>
> The problem is that the semantics of some operations can be quite
> complicated, and sometimes they'll vary across different Unix
> implementations.  It would be a big job to document all this in
> Python, so it's easiest to just point people to the man pages for
> their platform.

   Um, the thread was about no-UNIX "handicaps". If Python were
supposed to be UNIX-only then referring to man pages would be
fine - the problem is when you're running Python on Windows
and the docs say that the meaning of this parameter is just
like the UNIX function of the same name...

> Perhaps the best compromise would be to give a brief
> explanation of the most important points, and refer to the platform's
> docs for the fine points of semantics.
>
> --
> A.M. Kuchling			http://starship.python.net/crew/amk/
> On a scale of one to ten I'd give it a tim.
>   -- William Tanksley, 13 Sep 1999
>
>


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


From aahz at netcom.com  Tue Jun 27 17:39:21 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 21:39:21 GMT
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com>
Message-ID: <8jb6u9$ddr$1@nntp9.atl.mindspring.net>

In article <8jb6jr$2uov$1 at news.enteract.com>,
Greg Gallagher  <ggallag at red-bean.com> wrote:
>
>I'm a little disheartened here... I know how to do this in Perl, but I'm
>lost in Python and 
>searching for 'complex data structures' is landing me nowhere. 
>Basically,  I want
>a dictionary whosevalues are lists of dictionaries.  Simple.
>I want to basically loop through some data and so something like this:
>
>while loop:
>   new_forward = {'ip' : ip_address,
>                  'hostname' : hostname,
>                  'mac_address' : mac_address}
>
>   forward[domain].append(new_forward)

Try this:

  if forward.has_key(domain):
    forward[domain].append(new_forward)
  else:
    forward[domain] = [new_forward]
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From thomas at xs4all.nl  Wed Jun 28 02:02:09 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 28 Jun 2000 06:02:09 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <395a7440.132798875@nntp.interaccess.com> <39597A25.8BF553C@san.rr.com>
Message-ID: <slrn8lj56i.l52.thomas@centurion.xs4all.nl>

On Wed, 28 Jun 2000 04:05:21 GMT, Courageous <jkraska1 at san.rr.com> wrote:

>> On top of that, very likely that if SUN sues they win.

>Ridiculous.

Very much so. But it wouldn't be suprising just the same, given the American
patenting/registration practices and the average american courthouse ;-)

Overseas-ly y'rs,
	Thomas


From jkraska1 at san.rr.com  Sat Jun 10 17:35:12 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 10 Jun 2000 21:35:12 GMT
Subject: Stackless Python in Commercial Game Project!
References: <392E7C07.B456D9F@tismer.com> <8htqhp$a0n$1@kopp.stud.ntnu.no>
Message-ID: <3942B525.686F63CF@san.rr.com>

Magnus Lie Hetland wrote:
> 
> Cool!
> 
> Just a question: What is the motivation for using Stackless Python as
> opposed
> to standard Python here? Threading? (Just curious...)

Stacklessness allows you to context switch at a very rapid
pace, basically. An operating system thread has very high
overhead; this is particularly true when preemptive multi-
tasking doesn't buy you anything, in which case you are
paying context switch overhead costs for context switches
you don't even desire to make.

The usefulness of first class continuations is proven in
massively parallel operating system compilers like CILK,
where none of the procedures make use of the C stack at
all.

(The continuation is an obvious and natural thing to
 follow out of a stackless environment: an object which
 encapsulates the immediate run state of your thread,
 and if restored, restores the state of the thread).


C/


From nobody at nowhere.nohow  Wed Jun 14 22:09:53 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 15 Jun 2000 02:09:53 GMT
Subject: Which linux distros have Python by default?
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial> <%TS15.655$i06.206156@ptah.visi.com> <20000614175918.A5625@dophnic.yi.org>
Message-ID: <RVW15.694$i06.216628@ptah.visi.com>

On Wed, 14 Jun 2000 17:59:18 -0500, Derek Moeller <moeller at dophnic.yi.org> wrote:
>
>> >Pretty soon, everyone will want to have Python to use ESR's CLM2, a
>> >configuration utility for the the kernel written in Python/Curses. 
>> >Tkinter is a different matter though
>> 
>> Well, people who compile kernels will, anyway.  There are
>> getting to be a lot of users who never do a kernel build...
>
>Yeah, but the distribution maintainers will be forced to include it as
>a default to be considered seriously.

Do most of the "consumer" distros even include kernel sources
by default?  I'm not positive, but I don't think RH includes
kernel sources in the standard "workstation" configuration.

-- 
Grant Edwards                   grante             Yow!  World War Three can
                                  at               be averted by adherence
                               visi.com            to a strictly enforced
                                                   dress code!


From bragib at my-deja.com  Thu Jun 15 09:44:56 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Thu, 15 Jun 2000 13:44:56 GMT
Subject: telnetlib strange behavior?
References: <8i39kg$fla$1@nnrp1.deja.com> <3946bb29.508639203@news1.on.sympatico.ca>
Message-ID: <8iamki$q45$1@nnrp1.deja.com>

[snip]
>
> The documentation for telnetlib is confusing. What is does not state
> is that if you supply a hostname when you instantiate an instance of
> Telnet() that it will open the instance. What it does tell you however
> is not to try to re-open a connected instance... (see the source code
> for __init__ and the docstring for open.
>
> You are doing :
> >tn = telnetlib.Telnet(HOST)
> >tn.open(HOST)

This does not seem to cause any problems.
I also tried tn = telnetlib.Telnet(), tn.open(HOST)

Thanks, bragi


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


From olczyk at interaccess.com  Tue Jun 27 23:48:30 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 28 Jun 2000 03:48:30 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <395a7440.132798875@nntp.interaccess.com>

On Tue, 27 Jun 2000 13:10:46 GMT, "Don Tuttle" <tuttledon at hotmail.com>
wrote:

>Consider this a heads up NOT a promotion!!
>
No I consider it a troll.
But to tqackle the fundamentals.
Every object a COM object. Shades of Don Box
we get what we finally need a String class that's a COM
object!
Of course we can say goodbye to all those wonderful 
lightweight classes that we have in C++. Such as the nodes
of a tree. 
Not to speak of the added registry corruption not only for customers 
but for all the people that buy stuff written with this shit.
On top of that, very likely that if SUN sues they win. Even a
noncomputer guy who does technical work can see they are virtually the
same.


From richard_chamberlain at ntlworld.com  Wed Jun 14 12:52:26 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Wed, 14 Jun 2000 17:52:26 +0100
Subject: tcl/tk vs. python/Tkinter - Example attached
References: <3946B170.17A2630@visionart.com>
Message-ID: <DCO15.563$fb5.9835@news2-win.server.ntlworld.com>

Hi Daniel,

What platform are you working on?

On my win98 machine I don't see any difference in speed (except python
starts up very slightly slower).

Other than that the readability of the python is the real clincher.

Richard
Daniel Kramer <daniel at visionart.com> wrote in message
news:3946B170.17A2630 at visionart.com...
>
> I have just started to use python this last week.. I wanted to port a
> tcl/tk app to it and see how I like working with it.  As far as
> programming with it I like it more.. but just as I got started on my app
> I noticed some speed differences in TK.  The tcl/tk app is much more
> responsive then the python/Tkinter app.
>
> I have put together a simple example done both ways.. it's attached to
> this mail.
>
> Within the tar you will find:
>
> canvas.tcl - This is just a simple tcl/tk prog.. 2 rectangles are added
> to a canvas and you can move them around with the mouse.
>
> canvas.py - This is the same thing written in python/Tkinter.
> nodeUI.py - This contains most of the classes/functions to drive the
> canvas.py.. it should be in the same dir as canvas.py
>
>
> Since I am new to this lang I was just wondering if there was a way to
> get better python/TK performance... if anyone could try out the examples
> and let me know.. for me the tcl/tk one feels much more responsive but I
> don't have any hard numbers to prove that.
>
> I have been reading that wxWindows is the much faster solution.. but I
> am very familiar with TK and it's plenty fast for what I need.. but I
> worry that the Tkinter ver won't be fast enough once I add alot more to
> it.
>
>
> -Daniel
>
>
>


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


>
>
> --
> --------------------------------------------------------------------------
---
> Daniel Kramer
> http://www.visionart.com/~daniel/
> daniel at visionart.com:           VisionArt Design and Animation
>
>
>




From jbauer at rubic.com  Fri Jun  9 09:48:10 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 09 Jun 2000 08:48:10 -0500
Subject: Bypass Sytem Output?
References: <8hp3u1$5dg$1@news-int.gatech.edu>
Message-ID: <3940F59A.2D106F79@rubic.com>

Shengquan Liang wrote:
> i 'm running a Python script that generates
> an HTML file under Solaris 7.  The script also
> generates a PDF file and puts a link to it.
> 
> on thing i don't like is that Solaris gives
> a message like "saved /usr/local/apache/htdocs/PDF/1234602.pdf "
> and printed on the generated HTML file.
> 
> Q: how do i filter out this message?


It's ReportLab's pdfgen library -- not Solaris -- that
prints the message.  I hope this particular message will 
be removed in subsequent releases.  You can excise it 
yourself temporarily, by commenting out one line from 
the library source.

Alternatively, you can redirect stdout.

    >>> print "spam"
    spam
    >>> import sys
    >>> old_stdout = sys.stdout
    >>> sys.stdout=open('/dev/null','w')
    >>> print "spam"
    >>> 
    ... # perform some stdout emitting operations ...

    >>> sys.stdout = old_stdout
    >>> print "spam"
    spam
    >>> 

You may also want to sign up for reportlab's mailing
list on egroups to follow future discussions.

Jeff Bauer
Rubicon Research



From anders.eriksson at morateknikutveckling.se  Wed Jun 14 04:01:14 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 14 Jun 2000 10:01:14 +0200
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 2000)
References: <20000613212401.48393.qmail@hotmail.com>
Message-ID: <sidekssjcodoffd8eqd9e98t1qaquumd6l@4ax.com>

On Tue, 13 Jun 2000 14:24:01 PDT, "Hung Jung Lu"
<hungjunglu at hotmail.com> wrote:

>To put in some good words for PythonWare. I still remember from 3 years ago 
<snip>
>download it, click, and done. Back then, PythonWare saved the day for many a 
>newbie.
>

I totally agree! PythonWare have saved many hours for me and PIL is
the best thing since sliced bread!!

>Just as PythonWare, I am not a strong registry believer. I would like to be 
>able to fully control my software programs, so I can build things, and move 
>it to another computer (or even another operating system) and be sure that 
>it will work. So I would not put things in the registry. 

If you don't put it in the registry you need to put it somewhere else,
don't you? You need to define a PYTHONPATH, if you do as a environment
variable (.or whatever it's called in Unix/Mac) or in the registry,
same thing. If you move something out of the path you need to change
the path...

// Anders
Different-people-will-think-differently



From eric at estinc.com  Wed Jun 28 15:28:37 2000
From: eric at estinc.com (Eric Lee Green)
Date: Wed, 28 Jun 2000 19:28:37 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <395A521F.4374B144@estinc.com>

Steve Juranich wrote:
> I am *brand* new to python (as of Sunday), and I just got to the section
> in the tutorial about classes.  I was wondering why there really isn't
> such an idea as a "private" member of classes? 

Mostly because Python doesn't "really" have classes. What Python "really" has
are hash tables whose entries are objects (either methods or other data
objects), and some syntactic sugar for instantiating, calling methods within,
and zapping these guys. Thus a class instance 'foo' with methods 'bar' and
'grr' and data values 'ugh' and 'umm' would actually be a hash table with four
items in it, { 'bar': <some method> , 'grr': <some method>, 'ugh': 0, 'foo':
555 } and then 'foo.bar(x)' is syntactic sugar for foo['bar'](foo,x) . Though
the interpreter of course will not accept the latter syntax (it enforces the
class syntax). 

This is one reason why inheritance is so efficient in Python, BTW. All
inheritence does is add into the hash table for the overall class object those
entries that were in the base class object. Then when you create a new
instance, you actually are just creating a copy of the overall class object,
without having to refer back to the base object(s). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From rmartin at objectmentor.com  Fri Jun 30 19:56:52 2000
From: rmartin at objectmentor.com (Robert C. Martin)
Date: Fri, 30 Jun 2000 18:56:52 -0500
Subject: OmPyUnit -- A python unit test framework.
References: <slnttnejjev80@news.supernews.com> <8jhr2t$j3k$1@pollux.ip-plus.net>
Message-ID: <slqd02qqop349@news.supernews.com>

"J?rgen Hermann" <jhe at webde-ag.de> wrote in message
news:8jhr2t$j3k$1 at pollux.ip-plus.net...
>
> "Robert C. Martin" <rmartin at objectmentor.com> schrieb im Newsbeitrag
> news:slnttnejjev80 at news.supernews.com...
> > I have made our OmPyUnit unit test framework for python available in the
> > 'downloads' section of http://www.objectmentor.com.  This framework is
> very
> > similar in appearance and operation to JUnit.
>
> Is it in any way related to PyUnit (http://pyunit.sourceforge.net/),
besides
> the common JUnit heritage?

No, that's a completely separate implementation.  I haven't put OmPyUnit up
on sourceforge yet.


--

Robert C. Martin    | "Uncle Bob"              | Training Courses:
Object Mentor Inc.  | rmartin at objectmentor.com | OOD, Patterns, C++, Java,
PO Box 85           | Tel: (800) 338-6716      | Extreme Programming.
Grayslake IL 60030  | Fax: (847) 548-6853      | http://www.objectmentor.com

"One of the great commandments of science is:
    'Mistrust arguments from authority.'" -- Carl Sagan




From rjroy at takingcontrol.com  Tue Jun  6 21:55:17 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 07 Jun 2000 01:55:17 GMT
Subject: How to find out height of a Canvas
References: <eaq_4.3871$JO2.84543@news2-win.server.ntlworld.com> <memo.20000604135335.59489B@digitig.compulink.co.uk>
Message-ID: <393da9e6.79611437@news1.on.sympatico.ca>

On Sun, 4 Jun 2000 13:53 +0100 (BST), digitig at cix.co.uk (Tim Rowe)
wrote:

>In article <eaq_4.3871$JO2.84543 at news2-win.server.ntlworld.com>, 
>richard_chamberlain at ntlworld.com (richard_chamberlain) wrote:
>
>> Hi Tim,
>> 
>> from Tkinter import *
>> root=Tk()
>> canvas=Canvas(root,height=250,width=100)
>> canvas.pack()
>> print canvas ['height']
>> 
>> canvas.itemcget has two arguments, the first is the tag or id of a
>> particular widget on the canvas and the second is the option you want 
>> back.

You might want to look at the winfo_height method. Getting the height
configuration value is not always accurate.


import Tkinter
root=Tkinter.Tk()
root.geometry('400x200+40+40')
canvas=Tkinter.Canvas(root, height=100, background='green')
canvas.pack(fill='both', expand=1)
# canvas.winfo_height will return 1 when the window is first created
# the update command sets the geometry
root.update()
print canvas["height"]
print canvas.winfo_height()
root.mainloop()

Bob
 
>
>Thanks -- it looks as if I was making things much more complicated than 
>they need to be. Situation normal!



From jkraska1 at san.rr.com  Fri Jun 16 21:30:20 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 17 Jun 2000 01:30:20 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com>
Message-ID: <394AD545.AC791A2B@san.rr.com>

Well, speaking of tabs, in vim (.vimrc), I have:

set tabstop=4
set expandtab

This sets up vim to "on tab, take the cursor to the
next appropriate tabstop, but to insert spaces instead
of tabs when doing so". This is always just what I
want.


C/


From just at letterror.com  Mon Jun 19 03:08:02 2000
From: just at letterror.com (Just van Rossum)
Date: Mon, 19 Jun 2000 08:08:02 +0100
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <u1z1ua7qc.fsf@ctwd0143.fitlinxx.com>
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
 <8i8bi2$1s8$1@news.wrc.xerox.com>
 <slrn8kg937.328.hzhu@rocket.knowledgetrack.com>
 <ud7ljln92.fsf@ctwd0143.fitlinxx.com>
 <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com>
 <l03102800b570d8042c72@[193.78.237.143]>
Message-ID: <l03102800b57372d82b38@[193.78.237.140]>

[me]
>> The fact that the "tab == 8 spaces" rule is a) not generally accepted and
>> b) not generally enforcable across editors makes it IMHO a pretty useless
>> definition. Tools that break when using anything but 8 are badly designed:
>> it's like using two digits to store a year...

[David Bolen wrote]
>In this particular case, it's not so much that anything is actually
>breaking (since the use in question was solely using TABs for
>indentation), but that the visual indentation being enforced by
>setting the editor to use 4 instead of the more common 8 just makes it
>harder to get the same visual representation in other tools - without
>somehow also adjusting each of those tools.

I know, but to me that reinforces the Y2K problem analogy: most of these
problems didn't make apps crash, they just gave wrong results silently...

>But I do agree with you that the fact that this sort of discussion
>takes place over what a TAB really should be defined to do makes it
>largely unsuitable for use in code that is meant to be distributed and
>shared.

But that's not at all what I wanted to say. I'm perfectly happy with using
tabs (I use them all the time). It's only the mixing of tabs & spaces
that's bad.

The thing I really wanted to say (but somehow didn't ;-) is that using a
mix of tabs and spaces to get an indentation that's not the same as the
current tab stop comes from the assumption that "a tab is X spaces". By
saying "a tab setting of anything else but 8 is evil" you're guilty of
_encouraging_ the evil mixing of tabs and spaces. In other words, if noone
ever assumed anything about tab size, noone in their right mind would've
even started mixing tabs with spaces.

Using spaces-only is a safe solution, but only if you are absolutely sure
you never want anyone to use a proportional font to view your code... (For
those of you who have never used proportional fonts before <wink>: the
width of the space character varies from font to font (as it should); this
means spaces are pretty much useless as indentation!)

Just





From petersc at stud.ntnu.no  Tue Jun 20 01:32:13 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 20 Jun 2000 07:32:13 +0200
Subject: serial I/O
Message-ID: <394F01DD.E0CF3EB@stud.ntnu.no>

Sorry if I am missing the obvious, but what module
would I use to make serial I/O?

In this case I need 2400 Baud 8N1 serial in- and
output on a Win98 platform (sorry, the USB camera
only works there) but a platform independent solution
would be even better (or at least partly independent).

thanks
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From stephen at cerebralmaelstrom.com  Sat Jun 17 11:51:55 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 17 Jun 2000 11:51:55 -0400
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E8C@mail.sauder.com>

    I discovered Python a bit over a year ago, and have been using it
relatively constantly in idle projects. Some are simple, some are not.
Currently i'm working on one that's not.

    Initially, the indentation thing struck me as weird. My main complaint
about it is not that it's difficult to maintain proper indentation, but that
if you're scrolled far down in nested 'if', for instance, it's sometimes
hard to see which indentation-level is for which 'if'... but then i'll just
split-my-screen and scroll up.

    I've never encountered any of your problems. Everything that you've
described, though, i've done with C++'s braces.. forgot one here, added one
here.. etc...etc... The difference is that in Python, it's actually EASY to
just scan your code and see something indented improperly.

    As for your first point.. isn't it just as easy to scan for '__debug__'?
Heck, make your editor hilight it specially. Then just do your:

if __debug__:
    print Whatever

--S

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
news:394ac7ea.609024015 at nntp.interaccess.com...
> After using Python for a while I've come to the conclusion that
> using whitespaces to determine the begining and ending of blocks
> is bad for three reasons.
>
> Before I list those three reasons let me make two points. People say
> that the indentation requirements don't matter because you generally
> indent the same way. That may be true, but do you think it is a good
> idea to make conventions requirements ( remember that conventions
> start flame wars )? The second is that modern editors with good
> indentation rules ( like emac's python mode ) make indentation based
> coding easy. The thing is that by doing this you are abrogating
> control flow to the editor. When it was the editor indenting code
> which was well defined then there was no problem. By having
> the code depend on indentation, and the editor do the indentation
> for you, you give the editor responsiblity in how the code will flow.
>
> Now on to my points.
> 1) The first is rather stylistic and may not appeal to most, but I
> have often found it useful. When debugging code I often put in
> code to help me figure out whats happening. I also avoid indenting
> such code so that it easy to remember where it is later when I want to
> take it out. An example ( assuming I use begin and end to denote
> blocks ):
>
> def Average():
> begin
>       sum=0
> print "number of numbers=",len(numbers) #illegal
> i=0
>       for n in  numbers:
>       begin
> print "n=", n #illegal
>             sum=sum+n
> print "sum=",sum #illegal
>  i=i+1                     #illegal
> if sum>MaxNumber*I:#illegal
>


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


>
> You can immediately see code that you used for debugging and remove
> it. However right now writing code that way makes it illegal.
>
> 2) You can easily insert a tab introducing a bug.  Consider the
> following code ( similar to a bug I recently fixed ):
>
> def FindInvoice(idnum):
>       for invoice in invoices:
>             if invoice.id=idnum:
>       return invoice
>             return None # should be indented one less tab.
>
> ....
>
> for lines in file.readline():
>      idnum=lines[:6]
>      invoice=FindInvoice(idnum)
>      if not invoice:
>               invoice=InvoiceClass(idnum)
> invoices.append(invoice)
> ....
> Initially the line in error was indented correctly, but someplace
> along the line the extra tab was introduced. A lot of effort then had
> to go in to finding this bug. With begin...end or braces this is less
> likely to happen.
>
> 3) Finally. Python is a language where it is almost easy to write
> extensible programs. To illistrate what I mean consider this example
> of a python program I recently wrote:
> There is a C++ project which involves many project files and workspace
> files ( the project files are roughly equivalent to makefiles, the
> workspace files primarily are used to couple projects and define
> dependencies ). Version 1 simply read all the project files and
> generated a call to compile each one with all different configuartions
> (debug , release, alpha, beta etc..) . Since this built everything
> even if it was already built, and since this had to be run several
> times ( sometimes compiles would fail because a project that the
> present project was dependent on wasn't compiled ), I made version
> two. Version 2 read the workspace file and organised the projects
> into workspaces. Version 3 passed the string to a second script so
> that the second script could figure out if this was something that
> should be built ( for example whether a standard Win32 build or a
> WinCE build should be done ). Version 4 tracked dependencies more
> carefully ( it did a topological sort and orgainsied the build
> according to it ). Since the core stuff should be built first, I had
> to find out a way to figure out which was core. I used a definition of
> the core stuff is the stuff with the longest chain of dependencies.
> That was Version 5.
>
> As you can see I started with a simple program and made it more
> complex as I went on.  That's what I mean by extendible. Python is
> almost good at it. Almost. But that damned code indentation dependence
> backs its ugly head. In the process of extending code, I have to move
> blocks of code around. Sometimes put it in a loop, sometimes add an if
> clause to the top. Sometimes extract code into a subroutine. That
> means that I have to reindent the block. So as I reindent I have to
> try to remember how the indenting went. Something I stink at.
> Emacs also doesn't reindent correctly. So I have to go through the
> code, and try to figure out how it works. This distracts from my
> rewrite ( this also applies to refactorings ). I would rather think
> about how new code is supposed to work rather than how old code
> is supposed to work.
>
> Anyway I hope this causes TPTB to think about how blocks should be
> handled in Pyhton 3000.
>
>
>
>
>




From cpu at bigfoot.com  Sat Jun 24 00:04:06 2000
From: cpu at bigfoot.com (S. Hoon Yoon)
Date: Sat, 24 Jun 2000 04:04:06 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se>
Message-ID: <WqW45.20500$lU5.157840@news1.rdc1.nj.home.com>

I had same problem myself. Untill some dork messed up my registry and I had
to reinstall everything. Some how it was fixed. I don't know how. I wish I
can be more help, but I can only tell you how it was fixed.
On my NT 300Mhz Pentium II, it take about a second first time and takes no
time afterward. I do have a  lot of memory. On my 650Mhz pentium, I can't
even tell you how fast it is. Hit the enter key and I am done. So, it may be
something else entirely.

Hoon,

"Niklas Frykholm" <r2d2 at mao.acc.umu.se> wrote in message
news:slrn8l3ifj.hv1.r2d2 at mao.acc.umu.se...
> I've been noticing a significant difference in the start-up speed of
> Python on my NT machine and on a Linux system where I have an account.
>
> On the Linux system
>
> python -c "print 'hello'"
>
> times to about 0.1s.
>
> However, on the NT system, the very same command takes 4.0 s.
>
> Can someone explain this difference. (The processor on the Linux system is
> certainly not 40 times faster.)
>
> The startup speed on the NT system is a real problem. If you write a
> simple system utility (such as wc) you don't want to wait 4 s for it
> to start each time you run it.
>
> // Niklas




From davidw at linuxcare.com  Thu Jun  8 11:23:31 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 08 Jun 2000 17:23:31 +0200
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com> <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us> <393F8ACC.C49E0116@uab.edu>
Message-ID: <87ya4g19jg.fsf@eugene.prosa.it>

Shae Erisson <shapr at uab.edu> writes:

> Andrew Kuchling wrote:
> 
> > So, what would people want on top of distutils?  A Debian-style "do
> > apt-get, upgrade" to update currently installed packages?  A
> > CPAN-style "fetch and install this module for me, please"?  Simple
> > version checking?

It would also be nice if it were compatible, or easy to integrate,
with various packaging systems such as Debian's.  Apt is cool:-)

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From weil at sireconnect.de  Thu Jun 15 20:38:44 2000
From: weil at sireconnect.de (Helmut Weil)
Date: Fri, 16 Jun 2000 02:38:44 +0200
Subject: Problems with classes and __add__ and __mul__...
References: <8ibmn8$k5h$1@nnrp1.deja.com>
Message-ID: <3949779b@news2.sireco.de>

<abg at saturnsys.com> schrieb in im Newsbeitrag:
8ibmn8$k5h$1 at nnrp1.deja.com...
> Hi all,
>    I'm having a little problem with classes right now.  It's a class
> with a list as it 's main feature.  I've done the slices and such so that
> instance[:] and instance[x] work.  However, I can't seem to get
> __add__ and __mul__ to function the way that I want them to.
>

Just do this ...
def __add__(self, other_instance):
    result = spam()
    result.list =  self.list + other_instance.list
    return result

... and this will work as you wanted:
> z = spam()
> x = spam()
> z = z+x

Hope that helps.




From phil at geog.ubc.ca  Sun Jun 11 22:19:24 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 11 Jun 2000 19:19:24 -0700
Subject: Where are the math functions?
References: <slrn8k8dfv.tu.hzhu@mars.localdomain>
Message-ID: <m3bt17abf7.fsf@brant.geog.ubc.ca>

hzhu at mars.localdomain (Huaiyu Zhu) writes:

> I'm quite puzzled by not being able to find most of the common special
> functions, such as (incomplete) gamma, beta and bessel and erf.
> 

[snip]

Travis Oliphant (http://oliphant.netpedia.net/)
has wrapped the Cephes function library 
(http://www.netlib.org/cephes/index.html).  Here's the first few
paragraphs of the Cephes README:

          CEPHES MATHEMATICAL FUNCTION LIBRARY


     This computer software library is a collection of more than
400 high quality mathematical routines for scientific and
engineering applications.   All are written entirely in C
language.  Many of the functions are supplied in six different
arithmetic precisions: 32 bit single (24-bit significand), 64 bit
IEEE double (53-bit), 64 bit DEC (56-bit), 80 or 96 bit IEEE long
double (64-bit), and extended precision formats having 144-bit
and 336-bit significands.  The extended precision arithmetic is
included with the function library.

     The library treats about 180 different mathematical
functions.  In addition to the elementary arithmetic and
transcendental routines, the library includes a substantial
collection of probability integrals, Bessel functions, and higher
transcendental functions.

     There are complex variable routines covering complex
arithmetic, complex logarithm and exponential, and complex
trigonometric functions.

Phiil


From richard_chamberlain at ntlworld.com  Sun Jun  4 06:50:27 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 4 Jun 2000 11:50:27 +0100
Subject: Windows application with python - Tcl/Tk. How?
References: <393A2B0E.6879595F@tin.it>
Message-ID: <ooq_4.3901$JO2.85305@news2-win.server.ntlworld.com>

from Tkinter import *
root=Tk()
label=Label(root,width=12)
def clicker():
    label.configure(text="Hello Jilani")
button=Button(root,text="Click Me",command=clicker)
label.pack(side=TOP)
button.pack(side=TOP)
root.mainloop()

Try www.pythonware.com/library.htm

Also there is a good book by John Grayson (Python and Tkinter Programming)
and one on the way by Frederik Lundh published by O'Reilly.

Richard


Jilani Khaldi <jilanik at tin.it> wrote in message
news:393A2B0E.6879595F at tin.it...
> Hi All,
> could somebody send me a little example, or point me where can find it,
> about how to use python with Tcl/Tk to write GUI windows applications?
>
> Thank you.
>
> Jilani
> --
> <jk>
> ___________________________________________________
> http://space.tin.it/scuola/jkhaldi
> FreeBSD ~ The Power to Serve
>
>
>
>
>




From fstajano at uk.research.att.com  Thu Jun 29 11:25:02 2000
From: fstajano at uk.research.att.com (Frank Stajano)
Date: Thu, 29 Jun 2000 16:25:02 +0100
Subject: ANN: a better getopt.py
Message-ID: <4.3.2.7.2.20000629140130.00e2d5c0@localhost>

This is to announce an unofficial new version of the getopt.py module that 
helps parsing command line options. I wrote it because it scratched an 
itch; if it's useful to others too, so much the better.

http://www.cl.cam.ac.uk/~fms27/getopt/

This is a backwards compatible replacement that does not take away or 
change any existing functionality. It adds a new function getoptdict() 
which does the same things as getopt() but returns the result in a 
dictionary as opposed to a list of pairs. From my humble experience, the 
caller usually likes random (not sequential) access to the supplied 
options, so the dictionary appears to be the best match.

The only problem comes up with repeated options (e.g. "-v for verbose, -v 
-v for very verbose, -v -v -v for really very verbose indeed"). This of 
course doesn't quite fit in the dictionary model, since each key must be 
unique. The getoptdict() lets you choose what to do; you can REJECT 
repeated options (default behaviour) if you abhor them like I do; but you 
can also APPEND them to a list or KEEP_EARLIEST or KEEP_LATEST.

Another minor update concerns exceptions: the lonely GetoptError has been 
turned into a small class hierarchy. This allows the caller to distinguish 
the various errors if fine-grained control is desired. At the same time it 
is still possible to intercept all errors by simply calling the single 
GetoptError which is at the root of the hierarchy.

Note that the base file was the getopt.py that came with Python 1.6 alpha 
2. Therefore this module won't work with Python < 1.6, because the base 
file uses 1.6 features such as the index() method for a string.

This modification to getopt.py is released as open source under the same 
licence as the rest of Python. I, and my employer, disclaim all warranties 
just like Stichting Mathematisch Centrum and CNRI do. Feedback and bug 
reports to fstajano at uk.research.att.com.

   <P><A HREF="http://www.cl.cam.ac.uk/~fms27/getopt/">getopt.py</A> 1.3 - 
Compatible replacement for getopt that returns a dictionary instead of a 
list of pairs.  (29-Jun-2000)


  Frank (http://i.am/filologo.disneyano/)  http://www.uk.research.att.com/~fms/




From r2d2 at mao.acc.umu.se  Thu Jun 22 06:49:25 2000
From: r2d2 at mao.acc.umu.se (Niklas Frykholm)
Date: 22 Jun 2000 10:49:25 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <JEk45.262$Sn2.1571@news-server.bigpond.net.au>
Message-ID: <slrn8l3rpi.8ku.r2d2@mao.acc.umu.se>

>> However, on the NT system, the very same command takes 4.0 s.
>
>   One thing to try:
>      set path=
>      python -c "print 'hello'"

It doesn't do the trick. The running time is still 4 seconds.

// Niklas


From jar at mminternet.com  Tue Jun 13 00:10:35 2000
From: jar at mminternet.com (jim roush)
Date: Mon, 12 Jun 2000 21:10:35 -0700
Subject: Newbie questions
Message-ID: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>

I'm just learning Python.  The following code does not work.  It's
supposed to calculate the circumference after the user enters the
radius.  The program displays the prompt, but when I enter the radius
and the press enter, nothing happens.

FYI, I'm using Python 1.5.2 on WinNT.

pi = 3.141592654
radius = input("Enter is the radius: ")
circ = 2 * pi * radius
print circ


jar at mminternet.com
James A Roush


From jkraska1 at san.rr.com  Tue Jun 13 00:54:00 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 13 Jun 2000 04:54:00 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu> <8i3kbi$3a0$1@slb1.atl.mindspring.net> <3945AC6C.36BD396A@san.rr.com> <3945B8C6.8B4CF3A6@colconsulting.com>
Message-ID: <3945BF0E.5BAB3812@san.rr.com>

> > Although, to be fair, any seasoned C programmer has his or her
> > own preferred set of utility functions for doing hashtables,
> > stacks, lists, and so forth. I have a set of macro-expansion
> > tools for that that are far more efficient than the standard
> > template library could ever pray to be (written by a nut who
> > would meticulously process to assembly and analyze the number
> > and nature of assembly instructions issued for the equivalent
> > C... he was really freaky :).
> 
> But this is somethings we had to do.

This is doubtful, albeit you would have had to look somewhere to
find what you wanted... that I don't doubt. It wasn't so long ago
that STL was just a glimmer in the C++ standard's eye, and was
nothing more than a yet another template library from yet another
unix vendor. This isn't to argue the genuine usefulness of pythons
generic container classes, of course; obviously I'm sold. :)-





C/


From angband at blueyonder.co.uk  Sun Jun  4 07:29:50 2000
From: angband at blueyonder.co.uk (Bablos)
Date: Sun, 04 Jun 2000 11:29:50 GMT
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk> <393A3A9A.FB9F2E37@stud.ntnu.no>
Message-ID: <393A4C60.MD-1.4.4.angband@blueyonder.co.uk>

On Sun, 04 Jun 2000 13:16:42 +0200, Peter Schneider-Kamp wrote;
> Bablos wrote:
> > 
> > Is there a way that I can get Python to print this all on the same
> > line, so that it looks as if the line is being updated constantly, as
> > opposed to having a ridiculously long list of 'uploaded...' lines?
> 
> For unix you could just write:
> 
> print "\ruploaded",x,"bytes",

This is buffered on the Amiga, I assume.  I see very infrequent
updates, so I assume it saves up a large number of print statements
and then flushes them when it reaches a certain size.  Irmen?

But hey, at least '\v' works for me.  If it's all so very platform
dependant, surely it would be wise to standardise it somehow to
maintain some form of standard method?


Bablos.


From 12klat at sightreader.com  Thu Jun  1 19:50:04 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Thu, 01 Jun 2000 16:50:04 -0700
Subject: str(None) == '' ?
References: <8h5rpb$fl3$14$1@news.t-online.com>
Message-ID: <3936F6AC.3AB46D70@sightreader.com>

Eduard Hiti wrote:

> I would like to suggest that
>
>     str(None)
>
> should NOT return the string 'None', since most of the time, it complicates
> things.
> ...

In python (unlike C), the syntax and semantics are usually designed to simplify
the debugging process as the highest priority.  The None value generally
represents a special case.  It is usually a bad idea to -implicitly- handle
special cases invisibly; in fact, it's better if the special cases make lots of
noise until the programmer pays attention :-)  On the other hand, it is often a
good idea to -explicitly- hide special cases.

For the same reason, python throws exceptions in many situations where an error
return value such as None could be used instead.  This is another way python
makes the programmer pay attention to special cases.  Somebody once suggested
that open() should return None if the file is not found (that way you could say

(warning: hypothetical pseudo-python with a Bad Idea (TM))

if open('myfile')
   ...

Yuck!

- Ken Seehof

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000601/f00aac76/attachment.html>

From bjorn at roguewave.com  Fri Jun  9 16:38:20 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 09 Jun 2000 14:38:20 -0600
Subject: Problem with script and typing
References: <394123ad.89325968@nntp.interaccess.com> <8hr9jb$14g$1@sunnews.cern.ch> <394430d0.92688156@nntp.interaccess.com>
Message-ID: <394155BC.94F7E9E1@roguewave.com>


"Thaddeus L. Olczyk" wrote:
[snip]
> OK. That solves the specific problem, but what about the typing
> problem. Since projects only contains instances of Projects,
> FindProject returns a Project. The compiler seems to complain
> that it doesn't know that what FindProject returns is a project.
> So I can't call specific methods on the return value.

FindProject only returns a project if it finds one.  If it doesn't you
fall off the end of the for loop, and the default value of 'None' is
returned.  Perhaps you should throw an exception if you didn't find
anything?

-- bjorn



From dillera at isc.upenn.edu  Wed Jun 28 00:03:27 2000
From: dillera at isc.upenn.edu (Andrew Diller)
Date: Wed, 28 Jun 2000 00:03:27 -0400
Subject: XML-code examples - mini-review of Python XML
References: <394f7bc2.365514882@news.online.no> <00062201025104.08285@quadra.teleo.net>
Message-ID: <8jbteo$q3v$1@netnews.upenn.edu>

His book is out there! We got ours last week.

I like the book, but I also found it dissapointing. Most of it is an intro
into the Python language itself. What's left was mostly about his Pyxie and
PYX. I'm not sure why I'd want to learn PYX, when what I want is a solid
book on manipulating XML with python.

I didn't think I got that. I'm interested in learning about DOMs, SAX and
things that I'm reading and hearing about. PYX seems like a bizzare
off-shoot.

As for the into to Python part of the book (about 3/4 of it) that was a good
intro if you've never used python.

-andy diller


--
Andrew Diller       dillera at isc.upenn.edu
Information Systems & Computing
University of Pennsylvania
x3-7360
"Patrick Phalen" <python-list at teleo.net> wrote in message
news:00062201025104.08285 at quadra.teleo.net...
> [Thomas Weholt, on Wed, 14 Jun 2000]
> :: Is there more simple XML code out there? I mean the HowTos took about
> :: ten minutes, and it didn`t cover much.
> ::
> :: If somebody has some tips or code on how to traverse a DOM-tree,
> :: processing nodes based on tagname, and especially, based on relations
> :: with other nodes ( i.e. who`s your daddy, here`s your sister etc ),
> :: that would be great.
>
>
> While waiting for Sean McGrath's book to hit the shelves, you may find
> that David Mertz's nice Python XML introductory article gives you a leg
> up:
>
> http://www-4.ibm.com/software/developer/library/python1/
>




From kellyk at my-deja.com  Tue Jun  6 18:12:32 2000
From: kellyk at my-deja.com (kellyk at my-deja.com)
Date: Tue, 06 Jun 2000 22:12:32 GMT
Subject: win32Service.OpenService not working
References: <8hggdu$4fa$1@nnrp1.deja.com>
Message-ID: <8hjt03$nqb$1@nnrp1.deja.com>

In article <8hggdu$4fa$1 at nnrp1.deja.com>,
  tiddlerdeja at my-deja.com wrote:
> However, if I try to get a service with a space in the name, I get an
> error:
> >>> h = win32service.OpenService(hscm, "Windows installer",
> win32service.SERVICE_ALL_ACCESS)

As has been pointed out there is a descriptive name and an internal
name. To find out both names (and other info; see docs) for all services
use:

>>> import win32service
>>> SC= win32service.OpenSCManager("", None,
win32service.SC_MANAGER_ALL_ACCESS)
>>> services= win32service.EnumServicesStatus(SC)
>>> print services



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


From case at ironwater.com  Thu Jun 22 02:51:32 2000
From: case at ironwater.com (Case Van Horsen)
Date: Wed, 21 Jun 2000 23:51:32 -0700
Subject: snmp support in Python
References: <Pine.LNX.4.10.10006211416200.7273-100000@worf.mcnc.org>
Message-ID: <3951B774.24645582@ironwater.com>

I have been using pysnmp. It is available at
http://www.glas.net/~ilya/software/pysnmp.html

If you have any questions, please let me know.

Case Van Horsen

oliver wrote:
> 
> Hi,
> 
> Can someone update me the current status of SNMP support in Python? I
> remember a while ago, there were a few messages saying an development
> website setup on sourceforge, when I checked it, nothing was there yet.
> 
> Are there any workable solution? how mature is it?
> 
> Thanks for help
> 
> oliver


From rdudfield at my-deja.com  Mon Jun 26 08:22:25 2000
From: rdudfield at my-deja.com (rdudfield at my-deja.com)
Date: Mon, 26 Jun 2000 12:22:25 GMT
Subject: Can anyone solve this?
References: <8j737m$t5a$1@nnrp1.deja.com> <39570E2A.BBF53AB1@pik-potsdam.de>
Message-ID: <8j7htl$6m1$1@nnrp1.deja.com>

In article <39570E2A.BBF53AB1 at pik-potsdam.de>,
  Cezar Ionescu <ionescu at pik-potsdam.de> wrote:
> [Rene]
>
> [...snip]
> > I need the code to give every possible combination of the lists in
> > inputlist.
> [...snip again]
> >
> > Anyone able to solve this?
>
> Uhhh, I *think* so:
>
> def list_comb(inputlist):
>     outputlist = [[]]
>     for list in inputlist:
>         newoutputlist = []
>         for incomplete in outputlist:
>             for element in list:
>                 newoutputlist.append(incomplete + [element])
>         outputlist = newoutputlist
>     return outputlist
>
> if __name__ == "__main__":
>     a = [1, 2, 3]
>     b = ["c", 32]
>     c = [1, 2]
>     inputlist = [a, b, c]
>     print list_comb(inputlist)
>
> This prints the results you wanted. However, it probably won't do what
> you want for inputlist with one element, or with no elements, so
please
> take this as "a first approximation" only.
>
> Best,
> Cezar.
>

Thanks a lot for this, I tried for ages :)


Rene.



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


From jhe at webde-ag.de  Fri Jun 23 15:22:18 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 23 Jun 2000 21:22:18 +0200
Subject: adding mail attachments
References: <8ivdl4$90k$1@supernews.com>
Message-ID: <8j0daj$kch$1@pollux.ip-plus.net>

"geoff munday" <geoffm at nojunkmailqsiuk.com> schrieb im Newsbeitrag
news:8ivdl4$90k$1 at supernews.com...
> Can anyone advise where to look for instructions or examples??

Just did this today (you're lucky), you need a combination of cStringIO
(create file objects from your text, optional), MimeWriter (create the MIME
msg), mimetools (encode the parts, f.x. in base64), and smtplib (send the
mail).

Ciao, J?rgen

--
J?rgen Hermann (jhe<at>webde-ag.de)
WEB.DE AG, Amalienbadstr.41, D-76227 Karlsruhe
Tel.: 0721/94329-0, Fax: 0721/94329-22




From arta at NOSPAMx-stream.nl  Sat Jun 17 03:54:31 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Sat, 17 Jun 2000 09:54:31 +0200
Subject: locate
References: <8457258D741DD411BD3D0050DA62365907A21F@huina.oceanic.com>
Message-ID: <uSJ9zED2$GA.146@net003s>

> I'm assuming you mean character based user interface rather that
graphical.
> This is very platform dependent.  If on *nix there is the curses module.
On
> Win platforms there's some curses work alikes and other solutions.
>
> If you are on a unix or linux platform there are several resources I can
> point you towards that can get you started.  Let me know.

Yeah, I'm working on linux and windows, so if you could give me al the
information
I need.

Thnx in advance,

A[r]TA


--
-Life can kick your ass, but all you got to do is stand up and move on...




From jkraska1 at san.rr.com  Fri Jun  9 23:29:15 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 10 Jun 2000 03:29:15 GMT
Subject: windll
References: <3941ACC5.305B8830@videotron.ca>
Message-ID: <3941B6B4.264FA7A1@san.rr.com>

Jean-Yves Vinet wrote:
> 
> Is there someting like to "windll" to use
> dynamic librairy under Linux/Unix plateform.

What does windll do? "Using" dynamic libraries
under Unix is as easy as linking with them and
then making certain that they are in your
LD_LIBRARY_PATH. Unlike Windows, Unix distinguishes
between finding executibles (which must be in 
$PATH) and finding dynamic libraries (.so files,
which must be in $LD_LIBRARY_PATH).

If you want to know where a .so library is you
type "ldd" on the executable, ala:

% ldd myprog.exe

... and this will tell you which dynamic libraries
your program depends on, and whether or not it is
finding them, and where it is, when it is.

You can use nm to dump a libraries symbol table
if you want, by typing:

% nm mylib.so

... which will yield a complete dump of the symbol
table. Because this is huge, we often type:

% nm mylib.so | grep -i myfunction

... which will yield all those symbols of any case
matching "myfunction" in the library "mylib.so".

If it's building dynamic libraries that you want,
unlike in windows, you don't change the code at all.
Rather, you compile it differently and link it
differently. If you need to know those commands,
just say so.


C/


From kens at sightreader.com  Thu Jun  8 01:58:26 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 07 Jun 2000 22:58:26 -0700
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <7wc%4.7928$r7.1560935@news1.rdc2.tx.home.com> <393DDDF8.DE0401BB@sightreader.com> <W6v%4.8415$r7.1741613@news1.rdc2.tx.home.com>
Message-ID: <393F3602.CE771191@sightreader.com>

Rainer Deyke wrote:

> Ken Seehof <kens at sightreader.com> wrote in message
> news:393DDDF8.DE0401BB at sightreader.com...
> > Rainer Deyke wrote:
> > > Write an optimizing compiler that increase performance by a factor of at
> > > least 100 and decreases memory usage by a factor of at least 100.
> Python in
> > > its current incarnation is too inefficient to be useful as primary
> language
> > > for shrink-wrapped software.
> >
> > Because of it's dynamic nature, Python simply can't be optimized all that
> much
> > (read about all that type inference stuff and you'll see what I mean).
>
> A truly good optimizer might do wonders by analysing the entire system.  For
> example, it sees that a certain name can only reference objects of certain
> type, so it bypasses the generic access mechanism.  It sees that instances
> of class X only get attributes from a certain set, so the dictionary is
> replaced by a more efficient data structrure.  In general, if a human can
> read a python program and produce an equivalent efficient assembly program,
> a computer should be able to get close.

The type analysis is the tough part.  It is extremely difficult, and usually
impossible, to do meaningful and robust type analysis in a dynamic language such
as python.  Yes, it is true that perhaps 99% of python code doesn't actually
involve fancy abstract type hacking.  However, almost no python code can be
/verified not to involve fancy abstract type hacking/.  Almost every line of
code can invoke a function call.  A function call can have arbitrary side
effects on the local and global dictionaries.  In order for an optimizer to be
robust, it must never change the behaviour of a program, even if the program is
written badly.

I very much want somebody to prove me wrong.

These issues are so problematic that they have driven even Guido to come up with
some very strange ideas like having two kinds of python: the usual dynamic
python mode for most modules, and a non-pythonic python mode (invoked by a
special token at the beginning of the module) for modules that are to be
optimized.

> > Fortunately that just doesn't matter!
> >
> > 98% of code does not need to run fast.  You write the other 2% in C.
> > 98% of all objects (on a per-object basis) do not need to be packed tight.
> You
> > implement the other 2% in C.
> >
> > Conclusion: The primary language does not need to be efficient.
>
> But this involves complicated maneuvering between Python and C.  In many
> instances it is easier to just write the program in straight C.

> --
> Rainer Deyke (root at rainerdeyke.com)
> "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor

That's certainly true.  This kind of thing will get much easier as new
development tools are created.  For example, I've developed a code generator (a
Visual C++ wizard named PyExtWizard) some time ago to create python extension
modules and add new type and function boilerplates.  So for me, it's very easy
to get from python to C.  When some new (python based) build utilities come out
and get integrated into some well designed python IDEs, extension writing will
become easier still.

Getting from C to python is not as important as getting from python to C in this
context (C to python more important for adding scripting capabilities to
existing C/C++ applications).  I generally have a C layer at the bottom that
doesn't have any awareness of the higher python layer(s).  All high level coding
(including all of the user interface) is done in Python.

I'm picturing a world where python extension development will be as easy in a
Python IDE as pure C/C++ development in an IDE such as Visual C++.  This is
doable.

BTW, in this scenario, Python can be an excellent lab environment for testing
C/C++ code.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000607/8d2c6eb1/attachment.vcf>

From kens at sightreader.com  Thu Jun 15 23:51:00 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:51:00 -0700
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com> <3949A6BE.81B500E1@colconsulting.com>
Message-ID: <3949A423.13371179@sightreader.com>

"Frank V. Castellucci" wrote:

> Ken Seehof wrote:
> >
> > "Frank V. Castellucci" wrote:
> >
> > > Courageous wrote:
> > > >
> > > >
> > > > > As someone who uses and likes both C++ and Python, I think estimates
> > > > > like that are excessively simplistic.
> > > >
> > > > Me, too. And yet I will claim that I've become *more* productive
> > > > in Python after a brief 4-6 weeks of use than I am in C++ after
> > > > 15 years of use, and that's no exaggeration. If Python executed
> > > > as fast as other languages, I'd be tempted to *require* that our
> > > > staff use it. And please don't take this wrong; I'm no language
> > > > groupie -- I use C, C++, Java, Python, and Lisp -- all of them both
> > > > lately and frequently.
> > > >
> > > > > I use C++ for commercial programming, because it gives me the power and
> > > > > control I need.  Productivity isn't the best when measured in time to
> > > > > get the first prototype running, but I've found nothing better when
> > > > > measured in time to develop, deploy, maintain and support something a
> > > > > customer will pay for.
> > > >
> > > > Shrug. If it weren't for execution speed, your C++ environment would
> > > > be toast. Admit it. :)-
> > >
> > > While the last line seems rather toungue in cheek, there are other
> > > things that C++ brings to the table such as encapsulation, access
> > > control and type strictness.
> >
> > All three of these items were intentially left out of python because python
> > would not benefit from them, though C++ does.
> >
> > "Frank V. Castellucci" might have written (but actually didn't):
> > > A fish has certain definite advantages over a duck.  For example a fish has
> > gills.
>
> I didn't so it came from elsewhere.

I made it up.  That was a deliberate witty attempt to put words in your mouth :-)

> > Encapsulation, access control and type strictness are often erroneously
> > referred to as if they were features.  They aren't.  They are aspects of a
> > design philosophy.  To make a system less error prone, constrain the programmer
> > in order to prevent the programmer from writing bugs.  Of course, that's
> > impossible, since it is always possible to write bugs anyway.  Because C/C++
> > are extremely error-prone languages, they need stuff like encapsulation,
> > access, control and type strictness.  Because python is less error prone, it
> > doesn't need these "features".  Python is based on different philosophy: To
> > make a system less error prone, make it simple.
> >
> > You have to ask, "what is the actual value of these features?"  The answer is
> > that they prevent certain kinds of bugs from appearing.  Now ask, "what bugs
> > would a C++ programmer encounter if encapsulation, access, control and type
> > strictness were removed from C++."  Undoubtedly it would be a mess.  Get a
> > picture in your mind of these bugs.  Now ask a python programmer how often
> > these kinds of bugs come up.  Answer: "can't remember that ever happenning!"
> >
> > Look at the big picture.  Do encapsulation, access control and type strictness
> > really make C++ really less error prone than python?
>
> Somewhere in your conversation you went from the aspects of a particular
> language, while true they do reduce errors because the language is still
> low level enought to let you get in trouble, to implying that they were
> put in there to make a system less error prone. You then finish with
> justifying a position based on this posit.
>
> I disagree with the postulation.

My point is that encapsulation, access control and type strictness are relevant in a
comparison between C++ and python if they produce a specific advantage in C++ over
python that would make C++ preferable for some task.  My assumption is that
encapsulation, access control and type strictness produce robustness in C++, which
would be relevant if C++ were more robust than python as a result, which it isn't.

> Before these articles of disagreement, C had (and still does)
> structures. Systems built in C can be a mess, or not. That is based on
> the talent and experience writing the code. Python does not prevent you
> from writing "messy code", it just reduces the potential devastation
> caused by messy code. Python has the ability to hide data, you just use
> a different notation for the variable in the class. Python has the
> ability for type strictness, but you have to code it by hand as it is
> not a feature of the grammer.

Python doesn't /prevent/ messy code, but all other things being equal, it is easier
to write python code that is less messy because it's a simpler language. Typical ways
to be messy in C++ include: inconsistent choice of integer types and other basic
types, poor indentation practices, excessive abstraction layers, class inheritance
messiness, preprocessor messiness, etc.  Python has fewer basic types, no choice
about indentation, less need for abstraction layers, cleaner inheritance system and
less need for inheritance, and no need for a preprocessor.

Python is less messy /and/ messiness is less devastating in python.

> My original response was to the (lite hearted) statement that the only
> thing C++ brings to the table is execution speed. I will posit that
> languages for a solution space should be used that appropriatley address
> the design requirements. It ain't all C++, and it ain't all Python.

Agreed. So the question becomes: What are the relevant characteristics of the problem
spaces and corresponding characteristics of the candidate languages?  The best answer
I have so far is that python is a high level language and C++ is a low level
language, and that the two languages work well together.

> And the most original point of all these discussions was for
> qualification of productivity between the two languages. To which I
> still respond that if we measure productivity by the amount of time it
> takes to produce identical functionality (in something a bit more
> complex than "hello world") I would say Python.

Agreed.  Of course, python wins the "hello world" contest too. :-)

> I also believe that the larger the project, the more diminished the
> return on this particular measurement. Complexity takes it's toll on any
> language.

> --
> Frank V. Castellucci
> http://corelinux.sourceforge.net
> OOA/OOD/C++ Standards and Guidelines for Linux
> http://PythPat.sourceforge.net
> Pythons Pattern Package

In my experience, complexity takes less of a toll on python because it is a higher
level language.  I find that when I write a module in python, I tend to make it
behave as a component (i.e. having minimum dependency issues).  More often than not,
these components can be separately posted on Parnassus and used (hopefully) in
unrelated projects.  Component programming is extremely good for making robust large
scale projects.

In C++, which I've been using for much longer, I tend not to componentize quite as
well.  As a result, my code tends to hang together on a large scale in python quite a
bit better than in C++.  In addition, the small scale advantages hold true for large
scales projects as well.  So IMHO, based on experience, pythons productivity
advantages relative to C++ tend to increase as the project gets larger and more
complex.  The advantages of C++ tend to be more significant on a small scale.

Specific features, such as the requirement that one use semicolons, don't amount to
much unless they produce characteristics that differentiate between languages.  So
the real question is, what characteristics does C++ bring to the table, and for what
kind of jobs do these characteristics make C++ the better choice.  IMHO, C++ brings
speed and legacy to the table, so I use C++ for high speed components within a python
project, and for integration of legacy software.  Maybe it brings other things as
well, but I don't know what they would be.

The question of which language scales better is very important.  If C++ scales
better, python should perhaps be used primarily as a scripting language within C++
applications.  On the other hand if python scales better, C++ should primarily be
used for high speed components within python applications.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/39376e9d/attachment.vcf>

From sabren at manifestation.com  Sat Jun 17 18:00:02 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 17 Jun 2000 18:00:02 -0400 (EDT)
Subject: help compiling PIL
Message-ID: <Pine.BSF.4.10.10006171756050.6221-100000@chi.pair.com>


Hey all,

  I'm trying to compile PIL 1.1.. It seems to work well enough, but
when I actually run it, I get:

>>> import _imaging
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: Undefined symbol "_ImagingSavePPM" in python:./_imaging.so


   Can anyone tell me what I've done wrong?

Thanks,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------





From Dik.Winter at cwi.nl  Thu Jun  8 10:30:04 2000
From: Dik.Winter at cwi.nl (Dik T. Winter)
Date: Thu, 8 Jun 2000 14:30:04 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com>
Message-ID: <FvuAy4.51q@cwi.nl>

In article <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta at 4ax.com> urner at alumni.princeton.edu writes:
 > Recent number theory breakthrough:  you can get the
 > nth digit of PI without doing all the work needed 
 > to compute all the digits up to the nth digit (same
 > with some other numbers -- not sure if we know all
 > the relevant criteria as to which).

I checked the pages and have to make a correction here.  There is no such
claim.  The only claim is that it can be done without using much memory.
As one of the referred pages says:
  "we have not found a faster algorithm, nor have we proven that one does
  not exist."
Se essentially the amount of work is still the same, but the memory
requirents are smaller.
-- 
dik t. winter, cwi, kruislaan 413, 1098 sj  amsterdam, nederland, +31205924131
home: bovenover 215, 1025 jn  amsterdam, nederland; http://www.cwi.nl/~dik/


From DavidA at ActiveState.com  Tue Jun 13 13:31:32 2000
From: DavidA at ActiveState.com (David Ascher)
Date: Tue, 13 Jun 2000 10:31:32 -0700
Subject: O'Reilly Python Conference Early Registration Deadline Looming
Message-ID: <PLEJJNOHDIGGLDPOGPJJMEMMCHAA.DavidA@ActiveState.com>

It's that time of year again, the O'Reilly Open Source Convention, including
the Python track, is fast approaching.   You can save up to $280 if you
register by Friday July 16. The event will be held this July 17-20 in
Monterey, California.

Meet all the key Python players, meet fellow Python enthusiasts, learn about
new Python developments, find Python-related jobs, and have fun at this
yearly event in sunny Monterey, CA.

For details on the Python tutorials, see
http://www.oreillynet.com/pub/w/python_tutorials.html

For details on the Python talks, see
http://www.oreillynet.com/pub/w/python_presentations.html

For details on the Conference as a whole, see
http://conferences.oreilly.com/oscon2000/index.html

Looking forward to meeting you there,


-- David Ascher
   OSCON Python Track Program Chair
   Senior Developer, ActiveState Tool Corporation

Students and academics should remember to sign up for reduced fees:
http://conferences.oreilly.com/oscon2000/register/




From yawyi at gmx.de  Sat Jun  3 06:52:32 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Sat, 3 Jun 2000 12:52:32 +0200
Subject: String.join revisited (URGENT for 1.6)
References: <28052000.5@sanctum.jae.ddns.org> <8h15qr$rje$1@news.udel.edu> <mtzop76b2z.fsf@astron.berkeley.edu> <39346FAC.F2797EA5@cosc.canterbury.ac.nz> <8h320n$6fo$1@newshost.accu.uu.nl> <03062000.1@sanctum.jae.ddns.org>
Message-ID: <3938f8bb$0$25935@personalnews.de.uu.net>

Juergen A. Erhard <jae at ilk.de> wrote:
>It would seem (at least to me) to be the nicest (and cleanest) design
>to be able to do:
>
>   "some string".split().join()

Good point!

If split() splits a string into a list of strings, and split()
is a method of class String, and join() is the reverse operation
of split(), then join() should be a method of class List.

(Except that class List should really be ListOfStrings, which
isn't possible in Python  --> Back to the beginning of
the discussion about extended join() semantics.)

Johannes



From garry at sage.att.com  Wed Jun 14 11:32:47 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 14 Jun 2000 15:32:47 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net>
Message-ID: <3947A59F.5016D9F0@sage.att.com>

Aahz Maruch wrote:

> Like Grant, I've been using Unix about fifteen years, and I find Perl
> bizarre and difficult.

after reading several responses like this, i wonder if perl
is particularly irritating to old timers.  twenty years ago,
this kind of obscure but powerful notation (as embodies in awk
and such), was really cool.  but after a while the appeal kind
of wears thin, particularly when you find something that gives
you the same power but without the obscurity.

i find i'm more enamored of elegance these days.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From donn at u.washington.edu  Wed Jun 28 17:30:25 2000
From: donn at u.washington.edu (Donn Cave)
Date: 28 Jun 2000 21:30:25 GMT
Subject: Python design philosophy
References: <395A521F.4374B144@estinc.com> <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <8jdqph$7npc$1@nntp6.u.washington.edu>

Quoth Eric Lee Green <eric at estinc.com>:
| Steve Juranich wrote:
|> I am *brand* new to python (as of Sunday), and I just got to the section
|> in the tutorial about classes.  I was wondering why there really isn't
|> such an idea as a "private" member of classes? 
|
| Mostly because Python doesn't "really" have classes. What Python "really" has
| are hash tables whose entries are objects (either methods or other data
| objects), and some syntactic sugar for instantiating, calling methods within,
| and zapping these guys. Thus a class instance 'foo' with methods 'bar' and
| 'grr' and data values 'ugh' and 'umm' would actually be a hash table with four
| items in it, { 'bar': <some method> , 'grr': <some method>, 'ugh': 0, 'foo':
| 555 } and then 'foo.bar(x)' is syntactic sugar for foo['bar'](foo,x) . Though
| the interpreter of course will not accept the latter syntax (it enforces the
| class syntax). 
|
| This is one reason why inheritance is so efficient in Python, BTW. All
| inheritence does is add into the hash table for the overall class object those
| entries that were in the base class object. Then when you create a new
| instance, you actually are just creating a copy of the overall class object,
| without having to refer back to the base object(s). 

Hm, this doesn't really work for me.  Are you pulling my leg?

Does inheritance add to the hash table?  No, your derived class does
indeed refer to each of its parents' classes separately at run time,
you can even add functions at run time to a parent class and use them
from an old instance.

Does Python really have classes?  Yes!  Though we can explain some of
their implementation in terms of other Python features, the class is
absolutely integral to the language at a design level.  Though as with
any software there are accidents of implementation, there was also
an element of conscious language design.  Could a language otherwise
like Python be implemented with private functions like C++ (also, see
below!)?  Yes, and I think the price you would pay would be less in
the efficiency of the program than in the efficiency of programming.
It would slow you down as you try to write software, for little
practical benefit.

But in response to popular demand, there actually is a private class
member feature.  Define a member function with two leading underscores,
like "__private_function".  I have worked with code that does this
extensively, and it's kind of a pain in the neck.  It's a rare
programmer who will think to use it, and then really think before
using it.

	Donn Cave, donn at u.washington.edu


From bparsia at email.unc.edu  Thu Jun 22 13:22:51 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 22 Jun 2000 13:22:51 -0400
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu> <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <1ecjb1r.27wv4iz2rfkpN%bparsia@email.unc.edu>

Michael Hudson <mwh21 at cam.ac.uk> wrote:

> bparsia at email.unc.edu (Bijan Parsia) writes:
> 
> > William Dandreta <wjdandreta at worldnet.att.net> wrote:
> [snip]
> > > In addition, there are 3 function calls in each replace. Logic
> > > says eliminating one improve speed by 1/3.
> > 
> > Er.. no logic I know of :) Remember that "obvious" reasoning typically
> > leads you astray when thinking about optimization. 
> 
> Hear hear.

Too bad I don't always follow my own excellent advice.

> > The key, as I have written, is the *lookup* of the method/function.
> 
> Are you really sure about this?

No. :)

>  I was under the impression that it
> was the cruft at the top of eval_code2 in ceval.c that was the major
> overhead of a function call.  It certainly is in some circumstances,
> though I guess it may be possible that in others method lookup
> dominates.  Do you have numbers?

Nope. :( *Bad* Bijan! I was relying on the perhaps misinterpreted
reports of others. Which may also be a touch dated.

Ack, caught in a "common wisdom" fallacy in the very same *breath*
wherein I complain of another! Sheesh. ;)

[snip]
> > Try saving the function to a local var and using it. I'd be interested
> > to see what happens.
> 
> Does Python 1.2 have the local variable optimisation? 

What do you mean?

[snip]


-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From pinard at iro.umontreal.ca  Sat Jun 17 16:03:37 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Jun 2000 16:03:37 -0400
Subject: block comments
In-Reply-To: Thomas Wouters's message of "Sat, 17 Jun 2000 20:36:51 +0200"
References: <4mP25.63531$WS3.629201@typhoon.we.rr.com>
	<20000617203650.U26436@xs4all.nl>
Message-ID: <oqwvjoyt0m.fsf@titan.progiciels-bpi.ca>

[Thomas Wouters]

> Triple quotes may be what you want. Oh wait ;)

> I usually use normal comments: my editor makes it fairly easy to add '#' (I
> usually use ' ## ' to make clear that it's intended as a mass-comment-out)
> at the start of each line in a particular block.

Some people just use `if 0:' to introduce the block to comment, indented.
My preferred method is `# ' in the appropriate column.

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





From programmedintegration at home.com  Mon Jun 26 08:21:56 2000
From: programmedintegration at home.com (ColBoy)
Date: Mon, 26 Jun 2000 12:21:56 GMT
Subject: SMTP with Python 1.6
Message-ID: <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com>

I'm just learning the basics of Python having been a programmer for over 10
years.  I'm having  alittle difficulty with the following code.

#!c:\python\python.exe

import cgi, smtplib
s=smtplib.SMTP("localhost")
print "Content-Type: text\plain\n\n"
tolist=["colinmeeks at yahoo.com"]
The_Form=cgi.FieldStorage()
message=''
for name in The_Form.keys():
    message=message+"<B>Input:</B> "+name+" <B>Value:</B>
"+The_Form[name].value+"<BR>"
print message;
s.sendmail("colboy at programmedintegration.com", tolist, message)
print "Finished!"
s.quit;

If I comment out the SMTP specific things it works fine, printing out names
and values from a submitted webpage.  If I uncomment the SMTP stuff, nothing
appears and no email gets sent, also no error is shown.  Could someone give
me a pointer to what I'm doing wrong.

Colin




From Wolfgang.Grafen at marconicomms.com  Wed Jun 21 05:21:51 2000
From: Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen)
Date: Wed, 21 Jun 2000 11:21:51 +0200
Subject: List without duplicates?
References: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <3950892E.62BF3B@marconicomms.com>


> Is there a built in Python function that eliminates duplicates in a list?
>
> I have been sorting the list and compairing adjacent elements and if equal
> deleting one:
> -------------------------------
> mylist.sort()
>     j = len(mylist)-1
>     while j > 0:
>       if mylist[j] == mylist[j-1]:
>         del mylist[j]
>       j = j - 1
> ------------------------------------
> I recently tried using a dictionary. The list is the keys and I put and
> empty string for the entry
> --------------------
> mydict = {}
> for e in mylist:
>   if mydict.has_key(e): continue
>   else: mydict[e] = ''
> ------------------------------------
> Is ther a better way?
>

Try this ...

old_list = [1, 2, 3, 2, 1, 4, 3, 5, 2]


unique_list = []

for member in old_list:
   if not member in unique_list:
unique_list.append(member)

print unique_list
[1, 2, 3, 4, 5]


... or that ...

unique_list = []
for member in old_list:
    member not in unique_list and unique_list.append(member)

0
0
0
0
print unique_list
[1, 2, 3, 4, 5]

... or even following ...

unique_list = []

map(lambda member,unique_list=unique_list:
    member not in unique_list and unique_list.append(member),
    old_list)
[None, None, None, 0, 0, None, 0, None, 0]

>>> print unique_list
[1, 2, 3, 4, 5]
>>>

Wolfgang



From scarblac-spamtrap at pino.selwerd.nl  Wed Jun 28 04:25:05 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 28 Jun 2000 08:25:05 GMT
Subject: more python evangelising...
References: <B57E791A.264E%eableson@mac.com>
Message-ID: <slrn8ljdd0.36m.scarblac-spamtrap@flits104-37.flits.rug.nl>

Erik Ableson wrote in comp.lang.python:
> Just to confuse things:
> 
> We are talking about MS - and they're more interested in making money in
> training packages, ms certification programs, etc.  It stands to reason that
> they have no interest in an open source product that they don't have the
> ability to change arbitrarily to ensure that everyone in the corporate world
> is sent on MS Certified retraining courses.

Actually, they *can* do just that, because of its open sourceness. Nothing is
stopping them from taking Python, maybe improving its Windows support (they
can probably do a little better than Pythonwin since they built Windows,
although their effort would probably be more irritating...), and release it
as "MS Python"[1]. I just hope that *if* they do that, they won't break
compatibility and cross platform support, but they usually do...



[1] Or even Visual Python...
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From gmcm at hypernet.com  Fri Jun 30 20:23:57 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 01 Jul 2000 00:23:57 GMT
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8F635367Cgmcmhypernetcom@199.171.54.154> <962379073.1230449645@news.libertysurf.fr>
Message-ID: <8F63C1ABDgmcmhypernetcom@199.171.54.154>

Jerome Quelin wrote:

>gmcm at hypernet.com (Gordon McMillan) wrote:
>>>But how to pass this dictionnary to the parent class constructor?
>>class F(Frame):
>>  def __init__(self, *args, **kwargs):
>>    apply(Frame.__init__, (self,)+args, kwargs)
>>Generally you have to muck with the kwargs dict before passing it to the 
>>base class __init__.

>That is exactly what I want to know: what this mucking consists of?

Depends on what args the base class expects. If the base class expects 
**kwargs, then you can have anything in there.

But if the base class were defined:
class Frame:
  def __init__(self, x=0, y=0):
    ....

And the kwargs you pass it has {'snoo':33,...}, it's going to barf.

But maybe you've been playing with the interpreter, and have discovered all 
this by now...

- Gordon


From krussell4 at videon.home.com  Tue Jun 13 23:22:29 2000
From: krussell4 at videon.home.com (Kevin Russell)
Date: Wed, 14 Jun 2000 03:22:29 GMT
Subject: unsettling tkinter bug -- or is it just me?
References: <39458AFC.8AD5B850@tricity.wsu.edu> <3946b62a.507360500@news1.on.sympatico.ca> <3946BFD7.2DAE1899@uniserve.com>
Message-ID: <3946FA88.DF776F85@videon.home.com>

Bob van der Poel wrote:

> oppp, this looks like the same thing I was running with when I posted an
> earlier, unrelated post. In the meantime I've figured out that making
> the 'img' variable global solves the problem.... I guess this all makes
> sense, but it certainly is unsettling for newbies converting stuff from
> tcl/tk to python. Seems to me that their should be a better way.
> Thinking aloud (I may have to try later), if one were to change the
> variable containing the photoimage at a later time would the image in
> the button/label change?

No.  It simply vanishes.

-- Kevin




From pduffin at hursley.ibm.com  Fri Jun 30 06:18:17 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 30 Jun 2000 11:18:17 +0100
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <yYM65.2213$iN5.491623@ptah.visi.com>
Message-ID: <395C73E9.839ABEA6@hursley.ibm.com>

Grant Edwards wrote:
> 
> In article <39588741.FB7EB5FE at hursley.ibm.com>, Paul Duffin wrote:
> >Donn Cave wrote:
> >>
> >> Quoth ge at nowhere.none (Grant Edwards):
> >> | Or you can roll your own using the popen2 module.
> >
> >I think that you will find that Expect offers a lot more than a simple
> >pipe to the executable.
> 
> Yes, I know. That's where the "roll-your-own" bit comes in. If
> all you _want_ is a simple pipe to the executable, then expect
> isn't really the right choice.
> 

If all you _want_ is a simple pipe then you may be disappointed if it turns
out that what you _need_ is proper tty support. You are correct that a 
simple pipe is often enough, I was trying to explain that you cannot
roll your own version of Expect using a simple pipe.


From PClaerhout at CREO.BE  Thu Jun 22 08:27:53 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 14:27:53 +0200
Subject: win32print.GetPrinter()
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301216@msgeuro1.creo.be>

> Hi,
> 
> can anyone tell me which information the win32print.GetPrinter() function
> returns exactly? Any direction on where I can find this. It's not in the docs
> for the win32 extensions.
> 
> Kind regards,
> 
> 
> Pieter



From m.faassen at vet.uu.nl  Wed Jun 14 10:21:30 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 14 Jun 2000 14:21:30 GMT
Subject: Pointers in PYTHON
References: <20000614133324.10894.qmail@web1502.mail.yahoo.com>
Message-ID: <8i84da$j7c$1@newshost.accu.uu.nl>

Robin Porter <robinporter72 at yahoo.ca> wrote:

> I have been reading everything I can possibly get my hands on regarding
> PYTHON.  I have not yet come accross anything regarding the use of pointers
> for linked and double linked lists.

Why do you want a linked list? Anyway, linked lists aren't hard:

class Node:
    def __init__(self, next):
        self.next = next

linked_list = Node(Node(Node()))

Every name in Python's a reference, so no pointers are needed. Just think
of every name in Python as a pointer (to an object), if you like.

Doubly linked lists are along the same pattern, but have the trouble that
they introduce circular references, which is bad for Python's reference
counting based garbage collection scheme. You have to break one of the
references yourself for it to work:

class Node:
    def __init__(self, prev, next):
        self.prev = prev
        self.next = next

node1 = Node(None, None)
node2 = Node(None, None)
node1.next = node2
node2.prev = node1

# and now to clean up so that refcounting works:
node2.prev = None

>  I am just new to this language so please forgive my ignorance if it has
> been mentioned here before.

No problem.

> In addition to what ever information you can give me on pointers in PYTHON,
> could you please provide some info on combining C, Assembler, and Python
> for a project.

Take a look here:

Extending and Embedding the Python Interpreter

http://www.python.org/doc/current/ext/ext.html

And here:

Python/C API

http://www.python.org/doc/current/api/api.html

These are part of the standard Python documentation. You'll have to do
the assembler calls yourself, from C.

Regards,

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


From thomas at cintra.no  Tue Jun 20 07:35:23 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 20 Jun 2000 11:35:23 GMT
Subject: How to sell Python ( and OpenSoruce )
Message-ID: <39565119.358692532@news.online.no>

Hi,

I want to convince alot of my co-workers that OpenSource software,
especially Python, is ready for big business. ( is it ?! )

How do I sell an OpenSource solution like Python to hardheaded
closedsource-fanatics? I know they care about money, so that would be
my best angle. If anybody has any good examples of companies using
Python, or similar technology, with profit, that would be great. 

To be more exact :

My mission is to swap expensive, oversized databases solutions like MS
SQL 7.0 with PostgreSQL 7.0, crappy inconsistent script-languages like
VBScript with Python and MS SiteServer / IIS with Zope, Linux with
Windows  2000/ NT 4.0. I argue that the load on the system will never
be to great for PostgreSQL to handle as supposed to MS SQL-server,
VBScript, it sucks and they agree to that, Linux instead of Windows
will save lots of cash, so will the "reduced" cost of PostgreSQL
instead of MS SQL 7.0. 

The scenario above is normal in my experience, and I`d appreciate any
feedback from people with similar ideas. Related Case-studies would
also help alot.

The thing is that these people don`t thrust OpenSource. They laugh at
script-languages in general, and the idea of sharing code is even more
stupid. 

The growth of OpenSource, Python, Linux etc. depends on how well it
can be a suitable substitution for ClosedSource-alternatives. A more
agressive form of propaganda, showing off successful solutions, WITH
benchmarks, results of stress-tests etc. is vital. Right now there are
alot of talk about how great Python are, but few actual test of how
well Python performes under pressure, and not just Python alone, but
in a more complex environmet, like in Zope or using Python as
interface to a RDBMS.

I need selling points for OpenSource, the idea and philosophy,
examples of successful technology and eh .. moral support. I feel like
I`m in the lions den right now. ( " Think of a happy place. Calm blue
ocean, calm blue ocean ..." )

I truly love Python and would like to see it get the credit it
deserves.

Best regards,
Thomas Weholt


From cmedcoff at my-deja.com  Fri Jun 30 09:51:23 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Fri, 30 Jun 2000 13:51:23 GMT
Subject: Python COM Server Type Library
References: <8jfvvc$iql$1@nnrp1.deja.com> <qb5pls0uvkrbjg6hrkua0flurop7p1mpsb@4ax.com>
Message-ID: <8ji8kj$7n4$1@nnrp1.deja.com>

> As Mark said it cant be done automatically, but you can always create
> the type library by hand.
>
> Toby Dickenson
> tdickenson at geminidataloggers.com

Perhaps I've overlooked the obvious.  I guess you are suggesting that I
just type up the IDL and compile the type library.  That seems like it
should work.  I've done with for components I've written in C++/ATL
enough.

That gives me an idea.  Wouldn't it be nice to have a utility that you
could point at a type library and would generate a python script to
implement those components defined in the the library (with stubbed
methods of course).  I wonder if the pyhton win32 extensions expose
enought of the win32 api to do this?


--
Charles Medcoff
Senior Systems Engineer
Pyramid Solutions
1850 Research Drive, Suite 300
Troy, MI 48083


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


From hughett at mercur.uphs.upenn.edu  Fri Jun  2 09:59:09 2000
From: hughett at mercur.uphs.upenn.edu (Paul Hughett)
Date: 2 Jun 2000 13:59:09 GMT
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h7blm$j1b$1@slb2.atl.mindspring.net> <slrn8jeqfv.n4g.morpheus@teleute.rpglink.com>
Message-ID: <8h8ejd$3vv$1@netnews.upenn.edu>

Steve Lamb <morpheus at despair.rpglink.com> wrote:
: On Thu, 1 Jun 2000 22:04:22 -0600, Andrew Dalke <dalke at acm.org> wrote:
:>The GPL, which you are probably thinking of, does not allow similarly
:>unfettered use of their code.  Packages compiled with GPL code
:>restrict their use to only GPL-compatible licenses.

:     That's a new one to me.  I've never heard of the GPL requiring things 
: that
: I compile with a GPLed compiler to be distributed also under that license.
: I've only seen it where the distributed /code/, if modified and recompiled
: must also be distributed under the GPL.  

That's not what was meant.  If you compile and link a program that consists
of your own code plus *any* GPLed code, the whole thing must be distributed
under the GPL.  I much prefer a less restrictive license, such as Python's.

Paul Hughett


From thomas at cintra.no  Tue Jun 27 09:53:12 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 27 Jun 2000 13:53:12 GMT
Subject: Books
References: <8jarl9$9mi1@news.qualitynet.net>
Message-ID: <395cb143.112328700@news.online.no>

I think, allthough some disagree, that "Learning Python" is a good
start. But try the stuff online first. There are alot of links at
www.python.org, in several languages. 

A similar question was asked a few days ago. Read that posting and the
threads too. Might give a few more pointers.

Thomas

On Tue, 27 Jun 2000 15:30:28 +0300, "Helmit"
<helmit_world at hotmail.com> wrote:

>Can anyone sujest some good python books for beginners?
>



From timr at probo.com  Sat Jun 24 01:22:03 2000
From: timr at probo.com (Tim Roberts)
Date: Sat, 24 Jun 2000 05:22:03 GMT
Subject: Carrage Returns and Tabs in Windows Applications
References: <8ir53n$l5n$1@nnrp1.deja.com> <paf45.4867$qz4.36531@news-server.bigpond.net.au> <slrn8l36j8.b3d.quinn@pyro.ugcs.caltech.edu> <8it6p6$3b6$1@nnrp1.deja.com>
Message-ID: <J9H8LS8NJENMRP0CA8JBJTCRB6L5N2ARIC@4AX.COM>

jelizondo at my-deja.com wrote:
>
>Actually I may have worded my original message a little bit wrong.
>Basicly what I'm wondering is if there is a way to automate a series of
>commands in a windows application using python. For example.
>
>os.system(<program> <file>)
>
>then using python to automate a series of commands, sort of like batch
>and/or script file.

If you are trying to automate Windows applications, you have a hard row to
hoe.  However, it can be done.  There is enough of the Windows API in
win32all that you could use SendMessage to send WM_CHAR messages to an
application.  It's a pain, but it's possible.

Alternatively, if it is a popular commercial application, it may even have
a COM interface.  COM interfaces can be controlled from Python without too
much trouble.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From __kent__ at excite.com  Wed Jun  7 02:34:21 2000
From: __kent__ at excite.com (kent snyder)
Date: Tue, 6 Jun 2000 23:34:21 -0700
Subject: Real basic question about Entry box's data
Message-ID: <8hkqde$9cv$1@leopard.it.wsu.edu>

I'm trying to capture an Entry box's value and then print it out.  Actually
I'd like to do some math with it and return the answer to another Entry box.
I have the Grayson book and I can't find any SIMPLE, DIRECT guidance on
this.  Can anyone offer direction or a link to some sample code that's not
all cluttered up?

I'd like to NOT use PMW if at all possible.

Thanks a LOT in advance,
Kent




From paulb at infercor.no  Fri Jun  2 06:02:50 2000
From: paulb at infercor.no (Paul Boddie)
Date: Fri, 02 Jun 2000 12:02:50 +0200
Subject: SQL: don't use bind variables (was Re: Using strings with ' in them 
 in SQL-queries)
References: <39342055.1301760@news.c2i.net> <393510AB.D62AE9A5@infercor.no> <8h38ak$s1f$1@slb6.atl.mindspring.net>
Message-ID: <3937864A.DB218804@infercor.no>

Aahz Maruch wrote:
> 
> In article <393510AB.D62AE9A5 at infercor.no>,
> Paul Boddie  <paulb at infercor.no> wrote:
> >
> >As the ideal solution, though, I would recommend using bind variables
> >(placeholders) if your database module supports it:
> >
> >  db.query("insert into test (id, name) value (1, ?)", (var,))

[...]

> I've been keeping silent for a while, but I'm now going to start
> speaking forcefully against this style.  It's a real PITA to maintain
> when you have complex queries, because you can't use dicts and named
> parameters.  I find that it works much better to do something like this:
> 
> dict = {....}
> q = """
>     SELECT %(field)s FROM %(table)s
>     WHERE %(key)s IN (SELECT %(key)s FROM %(jointable)s)
>     """ % dict

This doesn't address the problem with ' characters in values. In fact, you're
configuring the query, which is quite distinct from typical applications of bind
variables. As far as I am aware, many database systems (if not all mainstream
database systems) do not allow you to use bind variables to specify which
columns and tables you're using.

Anyway, some database systems do support named parameters, and the DB-API
specification (version 2) says:

  Parameters may be provided as sequence or mapping and will be bound to
  variables in the operation.

The key word here is "mapping". I don't know how many database modules support
such functionality, though. It would look something like this with Oracle, if I
remember correctly:

  db.execute("insert into test (id, name) value (1, :var)", {"var" : var})

(Now I look again, I don't know where the 'query' method came from, but I assume
that the thread's initiator meant 'execute'.)

> (This is a stupid example, but it demonstrates the problem that
> frequently occurs when you have to use the same variable more than
> once.)  Remember that the Python Way is to write code as clearly as
> possible until such time as speed is proven to be an issue.

But speed isn't the only issue. As the initiator of the thread had found, there
are encoding issues which are conveniently solved by using the database module's
functionality. One reason why I don't do what you're doing is to reduce the
number of potential pitfalls connected with "trivial" implementation details -
if the database system provides API functions to pass string values directly to
the system itself, then why mess around with encoding string literals? (And I'm
not even considering inconsistencies between the documentation and how the
system might actually interpret encoded string literals.)

why-spend-all-that-time-making-a-nice-round-rolling-thing-when-someone-is-only
-going-to-tell-you-that-you're-reinventing-the-wheel-ly y'rs - Paul


From Sara.Fowler at usa.alcatel.com  Thu Jun  8 11:34:56 2000
From: Sara.Fowler at usa.alcatel.com (Sara Fowler)
Date: Thu, 08 Jun 2000 10:34:56 -0500
Subject: PLEASE HELP ON SOLARIS!!!
Message-ID: <393FBD20.E4664BD4@usa.alcatel.com>

Is ANYBODY out there trying to run Python on Solaris?  If so, how did
you get it to build?  

Does anybody know why the binary files are no longer supplied for Unix?

Sara
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Sara.Fowler.vcf
Type: text/x-vcard
Size: 330 bytes
Desc: Card for Sara Fowler
URL: <http://mail.python.org/pipermail/python-list/attachments/20000608/cc6be3d2/attachment.vcf>

From mwh21 at cam.ac.uk  Fri Jun 30 20:36:07 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 01 Jul 2000 01:36:07 +0100
Subject: Python MP3 Player
References: <395CE50A.3BA16777@bioeng.ucsd.edu>
Message-ID: <m3ituq7klk.fsf@atrus.jesus.cam.ac.uk>

Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:

> Does anyone know of a Python MP3 Player?

I've cobbled together some Python that you can use to control xmms if
that's what you want.  Anyone interested?  I can stick it on my
starship page if anyone is...

Cheers,
M.

-- 
  On the other hand,  the following areas are subject to boycott
  in reaction to the rampant impurity of design or execution, as
  determined after a period of study, in no particular order: 
    ...                              http://www.naggum.no/profile.html


From robin.escalation at ACM.org  Wed Jun 28 08:43:11 2000
From: robin.escalation at ACM.org (robin.escalation at ACM.org)
Date: Wed, 28 Jun 2000 08:43:11 -0400
Subject: postgreSQL successes
Message-ID: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com>

Does anyone have any success stories (or gotchas) related to
implementing a PostgreSQL database with Python that they'd like to
share? I am hoping to use it as a back-end to a b2b web site.

Technical info or just evangelism fine!

E-mail if you don't wish to share with all here.

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

"We told it about general relativity and gravity, the kind of people
that overcome hardships." 


From m.faassen at vet.uu.nl  Wed Jun  7 19:38:53 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 23:38:53 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net>
Message-ID: <8hmmed$ets$1@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <m33dmpj1b9.fsf at chinon.cnrs-orleans.fr>,
> Konrad Hinsen  <hinsen at cnrs-orleans.fr> wrote:
>>m.faassen at vet.uu.nl (Martijn Faassen) writes:
>>>
>>> Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
>>> nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?

That's fine, but it's really quite weird that so many people in this
newsgroup turn out to speak Dutch. How come?

>>*Blijken* te spreken. Misschien zijn we allemal bots die van een
>>enkele (nederlands) programmeur komen. Of misschien is er maar een
>>enkele mens in deze nieuwsgroep, die honderde adressen gebruikt om
>>de indruk te wekken dat Python heel populair is. Of misschien is
>>er een vertalingsprogramma dat alle teksten gewoon naar het Engels
>>vertaald, maar af en toe naar het Nederlands.

*Turn out* to speak. Maybe we are all bots created by a single
(Dutch) programmer. Or perhaps there's only a single human in this
newsgroup, who uses hundreds of addresses to give the impression
that Python is really popular [editor's note: Guido??]. Or maybe there
is a translation program that usually translates to English, but
once every while to Dutch.

>>Er zijn dus merdere rationale verklaringen, dus geen behoefte om
>>aan kosmische krachten te geloven!

Therefore there are multiple rational explanations, so no need to
believe in cosmic forces!

> Okay, okay, a joke's a joke, but if people are actually going to write
> full paragraphs in Dutch, it'd be nice if someone wrote translations for
> the small minority of us who don't know Dutch.

Ok, ok, een grapje is een grapje, maar als mensen echt hele alineas in het
Nederlands gaan schrijven, zou het leuk zijn als iemand vertalingen zou
schrijven voor de kleine minderheid onder ons die geen Nederlands kunnen.
[this for the majority]

Regards,

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


From kern at caltech.edu  Fri Jun  2 16:27:29 2000
From: kern at caltech.edu (Robert Kern)
Date: 2 Jun 2000 20:27:29 GMT
Subject: Python code editable/viewable by user
References: <KbRZ4.939$gr4.39365@news1.rdc1.il.home.com>
Message-ID: <8h95bh$euo@gap.cco.caltech.edu>

In article <KbRZ4.939$gr4.39365 at news1.rdc1.il.home.com>,
	"dsavitsk" <dsavitsk at e-coli.net> writes:
> I am looking for advice from someone who may have tried this before...or who
> thinks at least that it is not such a bad idea.
> i am working on a program where a user is presented a Tkinter interface.
> They push a button, some stuff happens.  The tricky part (i think) is that i
> want some of the even handler code to be in a text box where the user can
> edit it.

This is precisely what PTUI does (only in a more developed sense).  PTUI is
(was?) a pre-IDLE Python/Tkinter IDE that was written for a very similar
purpose to yours.  The author needed to be able to edit the program he was
running as he was running it.

More information and the last release are on the Starship:

http://starship.python.net/crew/zack/ptui

[snip]

> doug

-- 
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From darrell at dorb.com  Tue Jun 13 00:12:49 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 12 Jun 2000 21:12:49 -0700
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu>
Message-ID: <04c801bfd4ed$a398eff0$6401a8c0@home>

From: "Roy Katz" <katz at Glue.umd.edu>
>  Whatever, I now have three implementations of a virtual CPU to play 
> around with :)
> 

Now it's time to refactor. How easy will it be to add features to each?

--Darrell 





From ray at commsecure.com.au  Wed Jun 21 18:17:35 2000
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Wed, 21 Jun 2000 22:17:35 GMT
Subject: Frozen socket.
References: <Pine.GHP.4.21.0006161522100.3877-100000@don.et.byu.edu> <uln05qioe.fsf@ctwd0143.fitlinxx.com> <Pine.GHP.4.21.0006211035450.5957-100000@don.et.byu.edu>
Message-ID: <3950E734.E9206615@commsecure.com.au>

"C. Porter Bassett" wrote:
> 
> OK, I have been able to duplicate the problem that I was having, but I
> still don't know what to do about it.  First things first, I suppose.
> 
> I made two small programs, a server and a client, and I got the client to
> "freeze" in the exact same way.  My test server accepts three messages,
> but after that, it freezes.  For some reason that I don't understand, the
> client keeps sending messages (about 1000 of them, just like before), even
> though the server is not receiving any.  After about 1000 messages, it
> freezes, just like the program I am debugging.
> 
> Why is the client able to keep sending messages if the server is not
> receiving them?
> 
> In short, what is really happening, and how do I safeguard against it?
> 
> Here is the source for the two test programs I am talking about.
> sample client:
> from socket import *
> import time
> sock = socket(AF_INET, SOCK_STREAM)
> sock.connect("127.0.0.1",2778)
> for i in range(5000):
>    sock.send("This is a string of approximately the correct
> length.          "
>    time.sleep(0.001)
>    print i
> 
> sample server:
> from socket import *
> s = socket(AF_INET, SOCK_STREAM)
> s.bind(""<2778)
> s.listen(5)
> conn, addr = s.accept()
> counter = 0
> while 1:
>    counter = counter + 1
>    data = conn.recv(1024)
>    if couter > 3:
>       while 1:
>          pass
>    if data:
>       print data

Your server sample has a typo "couter", looping on "pass" is just
going to use the CPU, is this really what your real server is doing
when it stops processing input? The s.bind(""<2778)
syntax is another typo, you should have cut and pasted the real
client/server test progs.

With TCP/IP the client is able to stream data ahead of the server's
ability to process data, that is what you are seeing.
In this case the client continues to send until the socket
message buffers are exhausted, at which point it is put to
sleep waiting for more buffers to become available, presumably
because the server has processed some. If you want to limit this
to a smaller amount of data, see getsockopt and play with your
output low water mark. Then use select to ensure that you
are able to write prior to issuing the send operation.


From johngrayson at home.com  Thu Jun 15 06:32:22 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 15 Jun 2000 10:32:22 GMT
Subject: tkinter: wait_variable
References: <39482E96.350D115E@uniserve.com>
Message-ID: <8iabbm$i8o$1@nnrp1.deja.com>

In article <39482E96.350D115E at uniserve.com>,
  Bob van der Poel <bvdpoel at uniserve.com> wrote:
>
> How does wait_variable() work? I've got the following in a class:
>
> 	__marked = []
>
> 	....
>
> 	def mark(self):
> 		a callback to mark certain items
> 		self.__marked.append(x)
>
> 	def wait(self, root):
> 		root.wait_variable(self.__marked)
> 		return self.__marked
>
> and I call it, after setting up,etc:
>
> 	print myclass.wait(root)
>
> However, it never seems to return. With test print statements I have
> determined that my __marked variable is being updated...but
> wait_variable() doesn't seem to sense it.
>
> --
>    __
>   /  )      /         Bob van der Poel
>  /--<  ____/__        bvdpoel at uniserve.com
> /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>
>

You've got to wait for a Tkinter (Tk) variable. There is no ListVar
class, so you need to wait on an IntVar or BooleanVar, then in your
mark method, set self.__marked *and* your Tkinter variable...

    John


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


From andres at corrada.com  Thu Jun  8 10:42:54 2000
From: andres at corrada.com (andres at corrada.com)
Date: Thu, 8 Jun 2000 10:42:54 -0400
Subject: How to tell if I can open a file or not
In-Reply-To: <8ho4to0m8n@enews2.newsguy.com>; from Skip Hollowell on Thu, Jun 08, 2000 at 08:51:36AM -0400
References: <8ho4to0m8n@enews2.newsguy.com>
Message-ID: <20000608104254.B18466@corrada.com>

On Thu, Jun 08, 2000 at 08:51:36AM -0400, Skip Hollowell wrote:
> I have a indexer that I am writing that goest through a user-specified
> directory and all subdirectories and indexes all the .htm* file contained
> therein.  This works great until I get to a file that someone has chmod-ed
> so that I can't read it.  Then I bomb after the open().
> 
> Is there a way I can check the attrbutes of the file before I open it?  This
> is easy enough in c or perl, but have yet to find an equivalent in Python.
> 

Since python is bombing that means you have raised an IOError exception with
your open() statement. Try the "try" statement!

try:
    f = open('/path/to/file')
    .
    .
    .
except IOError:
    # We probably can't open this file for reading
    continue
    
Of course, there are many reasons why open may fail so this does not
guarantee that you are trapping only the files you cannot read.

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From elengvenis at uswest.net  Thu Jun 15 08:45:00 2000
From: elengvenis at uswest.net (Eric Lengvenis)
Date: Thu, 15 Jun 2000 07:45:00 -0500
Subject: PythonWin Debugging tools???
Message-ID: <Zc425.166$M85.9384@news.uswest.net>

I am very new to Python.
I just picked up the book Python Programming on Win32.
Early in the book, the authors describe the debugging features available.
None of them are available in my version. I noticed that the date on my
interactive window is April 13th 1999 and  the date on the figures in the
book is June 14th 1999.

Were the debugging features not included until this version? If so, how do I
get it. I went all over python.org and it definitely installed the April
13th version. (I have installed a couple of times from different downloads.)

Please advise. Where do I get the debugging tools???

Eric





From frankc at colconsulting.com  Tue Jun 13 00:29:58 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Tue, 13 Jun 2000 00:29:58 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu> <8i3kbi$3a0$1@slb1.atl.mindspring.net> <3945AC6C.36BD396A@san.rr.com>
Message-ID: <3945B8C6.8B4CF3A6@colconsulting.com>

Courageous wrote:
> 
> > >Then I move to C.  The process took about twice as long as the C++
> > >attempt; not counting the time it took for me to write miscellaneous
> > >libraries (stack, memory allocation), I noticed that I was spending far
> > >more time hunting pointer bugs and syntax errors than writing new
> > >code. oThis was the longest port; it took me about four times longer than
> > >it did in Python.
> 
> Although, to be fair, any seasoned C programmer has his or her
> own preferred set of utility functions for doing hashtables,
> stacks, lists, and so forth. I have a set of macro-expansion
> tools for that that are far more efficient than the standard
> template library could ever pray to be (written by a nut who
> would meticulously process to assembly and analyze the number
> and nature of assembly instructions issued for the equivalent
> C... he was really freaky :).

But this is somethings we had to do. I know you've heard the old "When I
was your age I walked to school", which the new cry of us greying and
balding is "When I was your age I had 64k on a mainframe". Now you look
on the sideboard of some pc product$ with requirements of 128 Mb just to
USE the damn things.

-- 
Frank


From nascheme at enme.ucalgary.ca  Fri Jun  9 17:22:36 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Fri, 09 Jun 2000 21:22:36 GMT
Subject: OT: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim>
Message-ID: <slrn8k2o30.3r4.nascheme@cranky.arctrix.com>

Tim Peters <tim_one at email.msn.com> wrote:
>I strongly recommend one deviation from the defaults, though:
>when it asks for the Python installation directory, force it to
>"c:\python15" (where "c:\" is the root of his boot drive, almost
>certainly exactly "c:\"). The default is some monster with
>embedded spaces in its name, and that just makes life harder.

This is something that pissed me off about Windows 95.  Microsoft
just had to show off their new feature allowing spaces in
filenames by putting a space in one of the most important
directory names on the system.  Unix has allowed weird characters
in filenames for ages.  Just because you can do it doesn't mean
its a good idea.  I wonder how much time has been wasted
worldwide because Microsoft did not call "Program Files"
"Programs".

    Neil

-- 
"Reliability means never having to say you're sorry." -- D. J. Bernstein


From Vladimir.Marangozov at inrialpes.fr  Wed Jun 14 20:12:22 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Thu, 15 Jun 2000 02:12:22 +0200
Subject: Static python on AIX
References: <3C1077AC64755A57.0FB8B24AA3155646.CE4465C98A055096@lp.airnews.net> <8gvvtt$s2n$1@slb2.atl.mindspring.net> <39476A09.E9D5E89C@soleil.serma.cea.fr>
Message-ID: <39481F66.E2950531@inrialpes.fr>

Franck Goubioud - SERMA/LENR wrote:
> 
> I'm trying to use Python on AIX.
> I'm trying to build a static executable with my modules and with the
> Numerical extension.
> The makefile builds it, but with an error message wich is:
> 
> The file access permissions do not allow the specified action.
> /usr/local/lib/python1.5/config/makexp_aix[41]:
> /usr/local/lib/python1.5/config/python.exp: 0403-005 Cannot create the
> specified file.
> 
> And the executable fails on the "import Numeric" instruction.
> On other platforms, there is no problems.

>From the above description, it's hard to tell what the problem is,
except that there is an access rights problem. When building the
interpreter, there are some temporary files that need to be written
to disk. You're probably trying to rebuild the interpreter once it is
installed in a place where you don't have privileges to write files.

Are you building from source? Some more info on the building context
might help us to give you a hint.

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


From hzhu at rocket.knowledgetrack.com  Wed Jun  7 21:47:14 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 08 Jun 2000 01:47:14 GMT
Subject: My wishlist for Python3K
Message-ID: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>

My wishlist for Python3K (This is not to start a flame war):

1. Everything is an object of a class that can be subclassed.

2. Augmentation operators: +=, *=, etc.  Dot operators .*, ./, etc (for
   numerical computation to distinguish matrix and elementwise operations).

3. The := operator for deepcopy as oposed to reference assignment.  Therefore
   - the assignment (a = b) guarantees (a is b)
   - the assignment (a := b) guarantees (a == b and a is not b).
   
4. Combinations of the above operators. 

These are not just syntactic sugars. They give big speed gains. And they are
very powerful when overloaded.

6. Have a clean mechanism to work with other programming languages other
than C.  (I have matlab/octave in mind.)

7. NOT case insensitive.  (No followup on this please.  Enough said already.)

When you see how much computation is done by the matlab/octave crowd you can
appreciate why convenience to scientific programmers (this term should be
clear) matters.  Want CP4E?  See http://www.mathworks.com/products/demos/

Huaiyu

PS. (Shameless advertising):  Want a Matlab clone in Python?  See
https://sourceforge.net/project/?group_id=6490


From andres at corrada.com  Mon Jun 26 22:05:47 2000
From: andres at corrada.com (andres at corrada.com)
Date: Mon, 26 Jun 2000 22:05:47 -0400
Subject: zope, python vs apache, perl
In-Reply-To: <8ish5s$kd4$1@nnrp1.deja.com>; from amitr@w-o-i.com on Thu, Jun 22, 2000 at 08:02:10AM +0000
References: <8ish5s$kd4$1@nnrp1.deja.com>
Message-ID: <20000626220547.B21453@corrada.com>

On Thu, Jun 22, 2000 at 08:02:10AM +0000, amitr at w-o-i.com wrote:
> I want to evaluate zope, python vs apache, perl.
> 
> which one can be used where!
> strong points and pitfalls.
> 

I would not put Apache in opposition to Zope, and Python. I use Apache to
serve Zope pages instead of Zope's built-in ZServer.

My experience with Zope was that it had a very steep initial learning curve.
It utilizes DTML tags that may trick Python users into thinking that they
can use straight Python but that is not the case. However, I now use Zope
exclusively to build web sites and find that I have become accustomed to the
Zopisms that made life at the beginning quite difficult.

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From rayleyva at my-deja.com  Fri Jun  2 20:44:42 2000
From: rayleyva at my-deja.com (rayleyva at my-deja.com)
Date: Sat, 03 Jun 2000 00:44:42 GMT
Subject: Found it with ( find /usr | grep config/Makefile )
References: <8h74jt$d9r$1@news2.isis.unc.edu> <m3ya4nzpjj.fsf@h0050047c4794.ne.mediaone.net>
Message-ID: <8h9kdo$is4$1@nnrp1.deja.com>

unfortunately the make command for PIL still doesn't work :

make -f Makefile.pre.in boot

Still looks for the file under :

/usr/lib/python/config/Makefile

instead of going to the appropriate directory

/usr/local/lib/python1.5/config/Makefile

I've been playing around with the Makefile.pre.in ... I'm a total novice
when it comes to shell-scripts ( wish I had payed more attention to them
now! ).

Any ideas would be really, really welcome.

Thanks
Ray


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


From anders.eriksson at morateknikutveckling.se  Tue Jun 13 05:28:06 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 13 Jun 2000 11:28:06 +0200
Subject: GUI editor
References: <8457258D741DD411BD3D0050DA62365907A1F9@huina.oceanic.com>
Message-ID: <mf74ksses09m77sdlerjjnr9fde6ljl9ps@4ax.com>

On Fri, 9 Jun 2000 09:21:02 -1000 , Doug Stanfield <DOUGS at oceanic.com>
wrote:

>Apparently Scriptics is having an identity crisis.  They are now Ajuba
>Solutions, but the SpecTCL stuff is still at
>http://dev.scriptics.com/software/spectcl/
>
>Trid to find SpecPy but couldn't find references to it at the above site,
>the Vaults, or www.python.org
>

You need to follow the SpecPerl link! Then you'll find a link to a
couple of SpecPython

It worked really nice! Maybe I will stay with this!

// Anders



From boud at rempt.xs4all.nl  Sat Jun  3 06:35:21 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 3 Jun 2000 10:35:21 GMT
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee> <m3g0qv91f2.fsf@wilma.george.org> <393855E8.DF87D1DB@home.com>
Message-ID: <8han19$4uc$1@news1.xs4all.nl>

Ivan Van Laningham <ivanlan at home.com> wrote:

> My brain is hurting from trying to understand XML in two days; please
> post *anything* you can to help out!


I was lucky when I was in your position - Lars Marius Garshol was just
asking for reviewers for his draft of his Python/XML book, and I got
to read it. It was very, very helpful ;-). Hope it gets published
soon. 

However, you're free to download my code - I use XML a lot in:

dbobj - the repository is XML based
kura  - webserver that takes XML templates and spews out HTML, and an
        XML data import utility.
kxml  - a small PyKDE xml file browser (unfinished!)

You can get them all at: http://www.valdyas.org/python/index.html
-- 

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


From greg at cosc.canterbury.ac.nz  Sun Jun 25 22:21:20 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Mon, 26 Jun 2000 14:21:20 +1200
Subject: Removing items from a list in a loop
References: <3952fdf6.950540640@nntp.interaccess.com> <8ipcln$bmd$1@news-int.gatech.edu>
Message-ID: <3956BE20.1E53E138@cosc.canterbury.ac.nz>

Andrew Henshaw wrote:
> 
> >Is there a safe way to iterate through a list and remove some items?

I sometimes do it this way:

  i = 0
  while i < len(l):
    if f(l[i]):
      del l[i]
    else:
      i = i + 1

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


From intmktg at Gloria.CAM.ORG  Mon Jun  5 16:25:26 2000
From: intmktg at Gloria.CAM.ORG (Marc Tardif)
Date: Mon, 5 Jun 2000 16:25:26 -0400
Subject: threading with time limit
Message-ID: <Pine.LNX.4.10.10006051515580.30679-100000@Gloria.CAM.ORG>

How can x threads be started at once and stopped y seconds later,
processing only the successfully returned threads?

For example, consider a situation where all the links from multiple URL's
must be retrieved at the same time. By setting a time limit, slow sites
don't slow down the whole process and appropriate error messages are
returned stating which sites wheren't completed in y seconds.

My current implementation (below), defines a list of links to parse and
creates a thread for each link. If the threads don't finish within 2
seconds, processing starts even the URL isn't retrieved.

Problems are:
- no error messages are returned for active threads killed after 2 seconds
- processing occurs on every thread, even if they were killed
- should the time limit occur in main() or in Page.run()?
- the time limit is applied to URL retrieval AND processing, whereas it
should probably apply to the former only

Any suggestions to solve any of the above problems would be much
appreciated.

--- getlinks.py ---
#!/usr/local/bin/python

links = ['http://www.python.org', 'http://www.cam.org']

import time
import string
import threading

import urllib
import urlparse
import sgmllib

def main():

    threadlist = []
    for link in links:
        thread = Page(link)
        threadlist.append(thread)

    for thread in threadlist:
        thread.start()

    for i in range(5):
        if threading.activeCount() == 1:
            break;
        time.sleep(0.5)

    linklist = []
    for thread in threadlist:
        tlinks = thread.getlinks()
        for tlink in tlinks:
            linklist.append(tlink)

    print linklist

class Page(threading.Thread):

    def __init__(self, url):
        self.url = url
        self.links = []
        threading.Thread.__init__(self)

    def run(self):
        text = self.readhtml()
        if text:
            self.links = self.readlinks(text)

    def readhtml(self):
        text = None
        f = urllib.urlopen(self.url)
        if f:
            text = f.read()
            f.close()
        return text

    def readlinks(self, text):
        parser = MyHTMLParser()
        parser.feed(text)
        parser.close()
        rawlinks = parser.getlinks()
        base = urlparse.urljoin(self.url, parser.getbase() or "")
        links = []
        for rawlink in rawlinks:
            t = urlparse.urlparse(rawlink)
            t = t[:-1] + ('',)
            rawlink = urlparse.urlunparse(t)
            link = urlparse.urljoin(base, rawlink)
            links.append(link)
        return links

    def getlinks(self):
        return self.links

class MyHTMLParser(sgmllib.SGMLParser):

    def __init__(self):
        self.base = None
        self.links = {}
        sgmllib.SGMLParser.__init__(self)

    def start_a(self, attributes):
        self.link_attr(attributes, 'href')

    def end_a(self): pass

    def do_area(self, attributes):
        self.link_attr(attributes, 'href')

    def do_img(self, attributes):
        self.link_attr(attributes, 'src', 'lowsrc')

    def do_frame(self, attributes):
        self.link_attr(attributes, 'src')

    def link_attr(self, attributes, *args):
        for name, value in attributes:
            if name in args:
                if value: value = string.strip(value)
                if value: self.links[value] = None

    def do_base(self, attributes):
        for name, value in attributes:
            if name == 'href':
                if value: value = string.strip(value)
                if value: self.base = value

    def getlinks(self):
        return self.links.keys()

    def getbase(self):
        return self.base


if __name__ == '__main__':
    main()



From mhammond at skippinet.com.au  Wed Jun 14 15:12:25 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:12:25 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com> <8i6u39$re3$1@nnrp2.deja.com>
Message-ID: <8i8le9$b6u$1@nnrp1.deja.com>

In article <8i6u39$re3$1 at nnrp2.deja.com>,
  millja at my-deja.com wrote:

> would like to integrate.  Is there some way to
> call this function from within the VBScript?
>  Or, it would work if I can get numbers,
> generated by the VBScript, into the Python
> function and the results of this function back to
> the VBScript.

VB should be able to call any Python function defined on the page,
simply by calling it by name.

Python should be able to call any VB function defined on the page with
some magic syntax that Im not sure of :-)

Under IE, you do it via "Window.VBFunc()".  I assume that under ASP, it
will be via the server or document object - most likely the document -
eg "Document.VBFunc()"

Feel free to post a sample once you get it working :-)

Mark.


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


From alex at magenta.com  Mon Jun 12 04:48:49 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 12 Jun 2000 10:48:49 +0200
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i1uv0$hki$0@216.39.141.21> <1d09kssdshm96qgilpm7rabmibt2blmv5j@4ax.com>
Message-ID: <8i28aa01nsu@news2.newsguy.com>

<mkx at excite.com> wrote in message
news:1d09kssdshm96qgilpm7rabmibt2blmv5j at 4ax.com...
> On 12 Jun 2000 06:11:44 GMT, dana_booth <dana at oz.net> wrote:
>
> >mec: So my question is this: Is Python a more "Windows-friendly"
language?
> >
> >Okay, I couldn't leave this alone. :)
> >
> >I think "Windows-friendly" is relative. I know Windows only programmers
who
> >can out-Perl just about anyone. (FWIW, I'm in the UNIX crowd. :)
>
> I should have replaced "Windows-friendly" with "less painful to
> non-Unix users"...

I think it's "less painful", period.  I come from a long and deep
Unix programming background, though with some mainframe and VMS at
the start and mostly-Windows/NT today, and I used Perl for many, many
years -- because I didn't know of anything better; when I finally
discovered Python, I soon started junking all of my Perl stuff, and
rewriting it in Python.  Today, I don't think I have any Perl left,
and the relief is still palpable.  Maintaining/tweaking something I
wrote 6 months ago thinking it was a use-once-then-throw-away (the
common case) is a light, enjoyable task rather than a nightmare.


> So, to a degree, some of the attraction to Python is that can me "more
> fun." And after a while, that can be important too.

Do get Hammond and Robinson's "Python programming on Win32", O'Reilly.

It's an excellent book, and will give you lots of hints about Windows
stuff that's good knowledge/background to know, *as well as* indications
and examples on how to work it with Python.  The contrast with the
equivalent Perl book, full of hacks and underuse/misunderstanding of
both Perl and Windows potential, couldn't be more striking IMHO.  It
bespeaks a different cultural mindset, "just hack together something
that may roughly work, with a couple of mysterious half-understood
idioms thrown in for good measure" vs "do it RIGHT"...


Alex





From sdhyok at email.unc.edu  Wed Jun 14 16:32:17 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 16:32:17 -0400
Subject: Numeric python
References: <Pine.LNX.4.10.10006141123440.26946-100000@intel3.wag.caltech.edu>
Message-ID: <8i8q4i$g8s$2@news2.isis.unc.edu>

Try www.sourceforge.com.

"Michael Todd Feldmann" <feldmann at wag.caltech.edu> wrote in message
news:Pine.LNX.4.10.10006141123440.26946-100000 at intel3.wag.caltech.edu...
> Could anyone point me in the right direction to get Numeric python?  I
> have a linux box and tried
>
> http://www.python.org/download/download_linux.html
>
> and went to
>
> http://andrich.net/python/
>
> which states that "This site is closed." since things may be outdated.  I
> am guessing there is somewhere else to find this and other RPM files.
>
> Thanks a bunch!
>
> Mike
>
> --------------------------------------
> Michael Todd Feldmann
> Beckman Institute (139-74)
> California Institute of Technology
> Pasadena, California 91125 USA
> Phone:(626)395-2722, FAX:(626)585-0918
> email: feldmann at wag.caltech.edu
> --------------------------------------
>
>
>






From MSteed at altiris.com  Tue Jun 20 12:21:30 2000
From: MSteed at altiris.com (Mike Steed)
Date: Tue, 20 Jun 2000 10:21:30 -0600
Subject: Why whitespace denotation of blocks doesn't work.
Message-ID: <65118AEEFF5AD3118E8300508B1248774CB34E@ALTNET>

> From: ted.drain at jpl.nasa.gov [mailto:ted.drain at jpl.nasa.gov]
> Sent: Tuesday, June 20, 2000 9:33 AM
> To: python-list at python.org
> Subject: Re: Why whitespace denotation of blocks doesn't work.
> 
> 
> The one thing that I really hate about the white space blocking is when
> you have to reformat because of a code change.  This is something that has
> hit me quite a few times.  
> 
> I'll write some code, and as I start debugging/prototyping/etc etc, I find
> that I have to enclose some existing code in a control block (if,etc). 
> With C/C++, I just put the code in brace, tell xemacs to re-indent the
> block, and everything's good.  With Python, I have to hand edit everything
> because the editor can't figure out how to do the indentation after I've
> inserted the control block.  Once you get a few nested indentation blocks,
> this gets very confusing and leads to bugs that are hard to find (code
> that did work suddenly doesn't).

In Vim it is easy to select a block of code and shift it in or out one
'shiftwidth'.  I don't know emacs, but I would be surprised if it did not
provide a similar facility.

> I love Python, but I'd really prefer to have open/close 
> brackets for the blocks.
> 
> Ted

--
M.



From jvickroy at sec.noaa.gov  Thu Jun 29 10:26:55 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Thu, 29 Jun 2000 08:26:55 -0600
Subject: exiting threads
References: <8jfcf9$2rp$1@nnrp1.deja.com>
Message-ID: <395B5CAF.5ED59A59@sec.noaa.gov>

The thread automatically terminates when the function being executed
returns.

Dale Burnett wrote:

> I am using thread.start_new_thread(self.foo,(x,y))
>
> Do I have to explicitly exit the thread or does it automatically exit
> when foo returns?
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From mwh21 at cam.ac.uk  Sun Jun  4 12:22:42 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 04 Jun 2000 17:22:42 +0100
Subject: Um? *boggle* Silly question about listobject.c
References: <000e01bfcddf$d2e09d40$1d2d153f@tim> <3939E84B.8AE86E66@san.rr.com>
Message-ID: <m3ln0lxvot.fsf@atrus.jesus.cam.ac.uk>

Courageous <jkraska1 at san.rr.com> writes:

> After running through a quick bit-shifting test, I discovered
> to my chagrine that I was thinking for some dumb reason that
> int was 16 bits. 

It can be, eg. on m68k Macs with appropriate compiler settings.
Isn't, on any commodity hardware newer than that, afaik.

> This is what I get for working on programming problems all day long
> during my leisure time. I heard a rumor that programming destroys
> brain cells more quickly than liquor.  Is it true? :)-

Well, my finals finish this week, so I'll let you know.

> *bonk self on head with rubber mallet*

Can't hurt :-)

Cheers,
M.

-- 
  This is an off-the-top-of-the-head-and-not-quite-sober suggestion,
  so is probably technically laughable.  I'll see how embarassed I
  feel tomorrow morning.            -- Patrick Gosling, ucam.comp.misc


From 12klat at sightreader.com  Thu Jun  1 19:04:34 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Thu, 01 Jun 2000 16:04:34 -0700
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no>
Message-ID: <3936EC02.60768A66@sightreader.com>

>>> x = 34.9
>>> s = "$%.2f" % x
>>> print s
$34.90

Read more in "More String Operations" in the "Python Library Reference"

Joseph Santaniello wrote:

> Hi,
>
> I have an SQL query which I can format to give me a nice value like
> 23,342.40 which Python squashes into 23342.4 when I stick it into a
> list element. Is there a nice way to format floats so they look like
> (US) currency?
>
> Thanks,
>
> Joe



From rupole at compaq.net  Mon Jun 12 17:35:55 2000
From: rupole at compaq.net (Roger Upole)
Date: Mon, 12 Jun 2000 21:35:55 GMT
Subject: Fun with sys.modules
References: <8i37m7$t96$1@pegasus.csx.cam.ac.uk>
Message-ID: <%Ic15.2292$c7.292284@pouncer.easynews.com>

How about b=a ?
         Roger Upole

"Duncan Grisby" <dgrisby at uk.research.att.com> wrote in message
news:8i37m7$t96$1 at pegasus.csx.cam.ac.uk...
> I'm attempting to make a module appear under two different names, and
> I've come up with the following. It does what I want, but I'm a bit
> worried by one of the side effects...
>
> # a.py
> print "a started"
> answer = 42
>
> -------
>
> # b.py
> print "b started"
> answer = 1234
>
> # Now the evil bit
> import sys, a
> sys.modules["b"] = sys.modules["a"]
>
> # This is what worries me
> print "sys is", sys
> print "answer is", answer
>
> print "b finished"
>
> -------
>
> $ python
> Python 1.5.2 (#1, Feb  1 2000, 16:32:16)
>  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import b
> b started
> a started
> sys is None
> answer is None
> b finished
>
>
> What happened to the globals in the running module!?  Presumably the
> entry in sys.modules was the only reference to module b, so it was
> deleted while it was still running. Is this safe?  My guess is that
> it's OK since it's basically the same situation as destructors which
> run after a module has been deleted.
>
> Any comments?  Am I totally crazy to be considering this?  If it is
> safe at the moment, what are the chances that a future Python release
> will behave differently?
>
> Cheers,
>
> Duncan.
>
> --
>  -- Duncan Grisby  \  Research Engineer  --
>   -- AT&T Laboratories Cambridge          --
>    -- http://www.uk.research.att.com/~dpg1 --




From romberg at smaug.fsl.noaa.gov  Tue Jun 27 13:35:16 2000
From: romberg at smaug.fsl.noaa.gov (Mike Romberg)
Date: 27 Jun 2000 11:35:16 -0600
Subject: Py_CompileString("lambda x : x == 2", ...) aborts
Message-ID: <ul3d7l39gdn.fsf@smaug.fsl.noaa.gov>

  I am attempting to convert a C++ string which holds the code for a
python lambda function into a PyObject so that I may use it in a later
call.  The code I'm using looks something like this:

PyObject *po = Py_CompileString("lambda x : x == 2", "<string>", 0);

This results in the process exiting with an abort():

Assertion failed
Abort (core dumped)

The stack trace looks like this:

#0  0x4011a111 in __kill ()
#1  0x40119d66 in raise (sig=6) at ../sysdeps/posix/raise.c:27
#2  0x4011b447 in abort () at ../sysdeps/generic/abort.c:88
#3  0x8090e09 in PyGrammar_FindDFA (g=0x819b420, type=0) at grammar1.c:50
#4  0x80bb807 in PyParser_New (g=0x819b420, start=0) at parser.c:129
#5  0x809116d in parsetok (tok=0x822f530, g=0x819b420, start=0, 
    err_ret=0xbffff42c) at parsetok.c:137
#6  0x80910a4 in PyParser_ParseString (s=0x81f7fa0 "lambda x : x == 2", 
    g=0x819b420, start=0, err_ret=0xbffff42c) at parsetok.c:78
#7  0x808d53a in PyParser_SimpleParseString (
    str=0x81f7fa0 "lambda x : x == 2", start=0) at pythonrun.c:966
#8  0x808d4b2 in Py_CompileString (str=0x81f7fa0 "lambda x : x == 2", 
    filename=0x8162481 "<string>", start=0) at pythonrun.c:932

  I'm wondering if someone may see what I'm doing wrong.  Or just be
able to tell me that I can't use lambdas like this.

Thanks,

Mike Romberg (romberg at fsl.noaa.gov)


From david_ullrich at my-deja.com  Fri Jun 16 13:09:09 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 16 Jun 2000 17:09:09 GMT
Subject: Problems with classes and __add__ and __mul__...
References: <8ibmn8$k5h$1@nnrp1.deja.com> <20000616011643.L26436@xs4all.nl>
Message-ID: <8idmv4$vrc$1@nnrp1.deja.com>

In article <20000616011643.L26436 at xs4all.nl>,
  Thomas Wouters <thomas at xs4all.net> wrote:
> On Thu, Jun 15, 2000 at 10:52:31PM +0000, abg at saturnsys.com wrote:
>
> > I would like to do this...
> > def __add__(self, other_instance):
> >     return self.list + other_instance.list
>
> > Which returns a list.  I would like to return an object of my class
type, but
> > it's not right obvious to me as to how to do that.
>
> Well, the result of 'x + y' is exactly what you return in x's __add__
> method, or if x doesn't have one, what y's __radd__ returns. So, if
you want
> 'x + y' to return an instance, make __add__ return an instance:
>
> class spam:
> 	def __init__(self, l=None):
> 		self.list = l or []
>
> 	def __add__(self, other_instance):
> 		return spam(self.list + other_instance.list)
>
> If you're worried about 'spam' not being accessible to '__add__',
because
> you are defining the class outside of the global namespace, you can do
this
> as well:
>
> 	def __add__(self, other_instance):
> 		return self.__class__(self.list + other_instance.list)
>
> A bit more clutter, but possibly more obvious.

   Another advantage to doing it this way is it does the
right thing if self is an instance of a subclass of spam.
(I started returning self.__class__(whatever) from
__xxx__ methods a while ago, wasn't sure whether it
would be regarded as evil to do it that way. I guess
not; that's a relief.)

DU


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


From cjc26 at nospam.cornell.edu  Sat Jun  3 10:59:31 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sat, 03 Jun 2000 14:59:31 GMT
Subject: Marking hyperlinks in a Text widget
References: <8h04jo$lgh$1@zingo.tninet.se> <oq7lcc41zr.fsf@titan.progiciels-bpi.ca> <8h0kn5$6pv$1@cubacola.tninet.se> <slrn8j94er.ebv.cjc26@localhost.mindriot.net> <8h4d2e$ap4$1@cubacola.tninet.se> <slrn8jblag.jua.cjc26@localhost.mindriot.net> <8h5hvi$djt$1@cubacola.tninet.se>
Message-ID: <slrn8ji7dm.5b1.cjc26@localhost.mindriot.net>

* Andr? Dahlqvist <andre at beta.telenordia.se> menulis:
| > That's actually a command-line argument of Unix versions of Netscape.  I
| > don't know if it would work on Windows.  If you're using Unix, try
| > typing the following at a shell prompt (with Netscape already running):
| > 
| > % netscape -remote 'openURL(www.python.org)'
| 
| Yeah, I'm on UNIX. I haven't really messed that much with the URL
| opening yet, but this is how I have thought it should work:
| 
| common_browsers = ['mozilla', 'netscape', 'lynx', 'w3m']
| 
| for browser in common_browsers:
| 	if not os.system(browser + URL)
| 		# found browser
| 		break

Running netscape this way will fail if the user is already running it;
to get around this you can check if it is running (using the output of
os.popen("ps ax")) and use the openURL command if it is.


| But the thing is that I haven't used your URL inserting, because
| it won't work in my application. I did however use the same code for
| tagging the URLs.
| 
| So for me doing this will only return the name of the tag for me:
| 
| URL = textwidget.tag_names("@%d, %d" % (event.x, event.y)
| 
| URL will be 'url' after this call. The last braces [1] that you use doesn't
| apply to my code since I haven't used that insert code you did, right?

Right, you'll have to read the text directly from the widget itself.
You won't even need to use tag_names in this case.


| Now this is what is driving me nuts, it should be so simple but I just
| can't get it to work! If you want I can send you my code so you can
| look at it.

You could even post it so that everyone could look at it :)


| > The callback function takes an argument (event), and you can get 
| > the x and y positions where the mouse was clicked using event.x and
| > event.y. So if you want to use tag_ranges, you could just look for the
| > range which contains (event.x, event.y).
| 
| Aha, that I can do.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From android at micron.net  Fri Jun  9 20:36:31 2000
From: android at micron.net (Peter Immarco)
Date: Fri, 09 Jun 2000 20:36:31 -0400
Subject: Probably a dumb question
References: <Pine.GSO.4.10.10006091955420.27416-100000@sundial>
Message-ID: <39418D8E.34D59FC4@micron.net>

Moshe,

How close to Python is JPython, that is, will I spend significant amounts of
time tracking the subtle differences between the two like Sun Java and MS
Java, or will I be blissfully ignorant of the minor difference? :)

Peter.

Moshe Zadka wrote:

> On Fri, 9 Jun 2000, Peter Immarco wrote:
>
> > Hi,  still waiting for my two books on Python from Amazon.  I know
> > Python is a full programming language so it can be used on web servers
> > for server side processing, but what about browser side processing?  Can
> > I also use it to create an 'applet' that runs in a compatible browser?
> > I'm guessing not likely because I don't think IE4+ or Netscape
> > Communicator support Python but I'm hoping I'm wrong.
>
> Well, the Time Machine at work once again: they all support Java, and
> Python can run inside JVMs: see http://www.jpython.org
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> http://www.oreilly.com/news/prescod_0300.html
> http://www.linux.org.il -- we put the penguin in .com



From matt at virtualspectator.com  Mon Jun 12 07:20:34 2000
From: matt at virtualspectator.com (matt)
Date: Mon, 12 Jun 2000 23:20:34 +1200
Subject: lambda and if
In-Reply-To: <07cd01bfd45e$9b0d07c0$1906a8c0@fc.fenx.com>
References: <8i2c9l$b0r$1@news1.wire.net.au> <0006122253431T.01005@localhost.localdomain> <07cd01bfd45e$9b0d07c0$1906a8c0@fc.fenx.com>
Message-ID: <0006122328441W.01005@localhost.localdomain>

sneaky ... it's nice.
might be nice to extend that with
["","big"][x[1] in ("big",)]  so we can extend the 'yes' set

but extending it to using the actual match is a little more difficult,
i.e.
 x[1] in ("big","little","wide",)  where the matched item would be appended if
it existed or else a blank


thanks
matt





On Mon, 12 Jun 2000, Emile van Sebille wrote:
> You could use this:
> 
> >>> filenames = map(lambda x:x[2] + "_" + x[0] + "_" +
> ["","big"][x[1]=="big"] + ".txt" ,items)
> 
> Emile van Sebille
> emile at fenx.com
> -------------------
> 
> 
> ----- Original Message -----
> From: matt <matt at virtualspectator.com>
> To: <python-list at python.org>
> Sent: Monday, June 12, 2000 3:46 AM
> Subject: lambda and if
> 
> 
> > say you had the following
> >
> > filenames = map(lambda x:x[2] + "_" + x[0] + "_" + x[1] +
> ".txt" ,items)
> >
> > but really wanted
> > filenames = map(lambda x:x[2] + "_" + x[0] + "_" + (x[1]
> in ("big",) ? "big" :
> > "") + ".txt" ,items)
> >
> > where (x[1] in ('big',) ? "big" : "") is a pseudo
> c++_pyhton expression that I
> > want to implement .....
> >
> > i.e. the result is to append "big" to the filename being
> created if it is the
> > 2nd element of item, otherwise don't append anything.
> >
> > regards
> > Matt
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From thomas at cintra.no  Tue Jun 13 03:05:54 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 13 Jun 2000 07:05:54 GMT
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au>
Message-ID: <3946dc6b.193651816@news.online.no>

On Tue, 13 Jun 2000 16:39:23 +1000, "Jason" <it at onestopent.com.au>
wrote:

>> import string
>> file = open ("file")
>> lines = file.readlines()
>> for line in lines:
>>     pair = string.split(line,":")
>>     print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
>> file.close()

The line :

file = open("file")

change "file" to the complete path of your text-file, like if your
file resides in c:\temp\music.txt, change it to :

file = open("c:\\temp\\music.txt")

And YES, the double backslashes are needed, or the \t in \temp is
interpreted as TAB. Just change that and run the script from PythonWin
or IDLE. They both come with Python, or eh ... at least IDLE does.

Thomas


From frankc at colconsulting.com  Mon Jun 19 00:06:56 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Mon, 19 Jun 2000 00:06:56 -0400
Subject: Circular import
References: <394D202B.1DF60864@colconsulting.com> <05a701bfd978$4fdb7ab0$6401a8c0@home>
Message-ID: <394D9C60.DEE41574@colconsulting.com>

Darrell Gallion wrote:
> 
> From: "Frank V. Castellucci" <frankc at colconsulting.com>
> >
> > With the assumption that class A and class B reference each other, how
> > do you avoid the Python error with:
> >
> > ---------- Composite.py -----------
> >
> > from Leaf import Leaf
> >
> > class Composite:
> > def __init__( self ):
> > self.__theLeaf = Leaf(self)
> >
> > -------- Leaf.py --------------
> >
> > from Composite import Composite
> >
> > class Leaf:
> > def __init__( self, aComposite ):
> > self.__theComposite = aComposit
> >
> >
> 
> I'm guessing you come from the land of C++, where the compiler wants to know
> every detail.
> Don't bother with this:
> > from Composite import Composite
> 
> Leaf can accept any kind of object as the composite.
> What ever methods the Composite has will be accessible.
> It's a beautiful thing once you get the hang of it.
> 
> --Darrell Gallion

Thanks!

-- 
Frank V. Castellucci


From mcalla at home.com  Mon Jun 26 19:04:34 2000
From: mcalla at home.com (Mike Callahan)
Date: Mon, 26 Jun 2000 23:04:34 GMT
Subject: Government Python application on the internet
Message-ID: <6kR55.17989$Yr4.303345@news1.rdc1.tn.home.com>

Fellow Pythoneers,

Check out www.crh.noaa.gov/lmk/asos/ksdf.htm if you have any interest in
aviation weather. This is a result of a Python app I wrote which plots
airport weather data using Mike Steed's gdchart 0.3. We eventually hope to
add more airports in the future. The data can be very interesting if a
thunderstorm goes over the airport.

Mike Callahan





From jasonic at nomadicsltd.com  Sun Jun  4 12:34:52 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Sun, 4 Jun 2000 12:34:52 -0400
Subject: [Human] Lanaguage translation
Message-ID: <sjl1243u5ri150@corp.supernews.com>

Hello

Does anyone have any ideas for language translation using Python?

There are on-line tools such as Babelfish and off-line software such as
'Unversal Language Translator 2000'[win32 -it's quite nice] etc.

I am looking for someway to script language engines like these off-line and
on-line, ideally some linux or cross-platform solution, but Win32 is ok for
now and near-term.

I have only just started doing research on this, so I may be overlooking
some obvious tools, but it appears that moset are stand-alone manual
applications which do not permit invisible automated remote-control via
scripts [Python modules] etc.

Also is there is any Python-based engines I could use directly?
 I would have thought that Python dictionaries, persistent objects like
ZOZDB could be idea for this job.

My immediate target is for European Languages. For both web and CD-ROM or
kiosk sintallations.
In developing for E-commerce in Mediterranean we are finding typically young
people (<30> often understand/read Egnlsih very well, but may or may not
speak it or write it. Richer kids tend to, because of more chance to travel
and spend time in UK etc. People over the age of 30, if they are
well-educated, may also understand and quite often speak or write well well,
but there are many business people and non-white collar jobs where the
English level is quite poor.  Thus we suspect a major obstacle is perhaps
lack of sufficient diversity of language. In email with European colleagues
the lack of response suggests same problems.. I note that many people
globally have a high-level of technical english reading, but may be lost in
conversation. This is very impressive to me

I personally note that looking at web-sites in a foreign language can be
extremely disturbing. Many web sites are so chock full of flashing heads and
subheads in columns and abreviated links, that one's mind-eye cannot rapidly
tell what is important - it is hard to 'scan the page and set mental
priorities ast the speed the page is designed for. In one's own naticve
lanaguage [English for me] cluttered or busy pages are manageable. Good
design is good design in any lanaguage.

I recently did an adhoc test visiting french and italian travel sites...
many tourist sites do have multilanguage buttons though not ubiquitous
pages. It seems to me that body of text to read is one not really the
issue - I am talking about labels, headings, buttons, site structure - often
single word noun/verb/functions and group identifiers.. These are the core
tools for web site navigation nda s recent surveys indicate many human
web-reading eyes may be first drawn to text and not graphics. Hard for me
say as I am a very visually oriented person and respond well to symbol,
images etc. But when it comes to web sites, In a foreign lanaguage I am soon
lost.... I speak French very fluently and amcomfortable reading books,
newspapers, magazines etc, but find websites hard. Yet french email or
newsgroup threads are fine.
I am asking myself why?

I welcome any comments from others about this topic and ways to acheive
dynamic translation with Python.
Thanks

- Jason

________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From jkraska1 at san.rr.com  Wed Jun 14 21:36:17 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 14 Jun 2000 21:36:17 -0400
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296B76@mail.sauder.com>

Huaiyu Zhu wrote:
> 
> >"Warren Postma" <embed at NOSPAM.geocities.com> writes:
> >
> >> Having something
> >> look correct on the screen, but not work because of misuse of tabs versus
> >> spaces, or incorrect indentation, is something that still continues to bug
> >> me.
> 
> If you use emacs:
> 
> C-x h M-x tabify


In vi/vim, you can globally replace all tabs with:

<esc>:1,$s/<tab>/    /g


Where the spaces are appropriate for the indentation
level you are looking at.


...................If I were running a python
project, the very first thing I'd do would be to forbid
tabs in python files. Both vim and emacs have modes
where hitting the tab key inserts enough spaces to
take you to the next tabstop.



C/



From thomas at xs4all.net  Wed Jun 21 14:19:20 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 21 Jun 2000 20:19:20 +0200
Subject: Why is Python so slow ?- revisited.
In-Reply-To: <8iq9tk$n8q$1@newshost.accu.uu.nl>; from m.faassen@vet.uu.nl on Wed, Jun 21, 2000 at 11:45:56AM +0000
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl> <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu> <20000620143255.F26436@xs4all.nl> <8iq9tk$n8q$1@newshost.accu.uu.nl>
Message-ID: <20000621201920.J26436@xs4all.nl>

On Wed, Jun 21, 2000 at 11:45:56AM +0000, Martijn Faassen wrote:

> > So, we have to check if range is still the same function. Which means we
> > have to look it up and compare identity -- and we've already lost any
> > advantage the cache could give us. Or we could place a marker on the
> > __builtin__ module, and invalidate the cached entry of 'range' whenver
> > 'range' is changed. But then we also have to add a marker to the global
> > namespace, because something can *add* a range there.

> You could do some analysis of local and global namespace, and see 
> that it is possible that 'range' is altered. If such a possibility
> exists at all (assignment to the name, or by from import), then range
> is simply never cached at all.

But how do you see, at compiletime, that *a user* of the code we are busy
compiling, will or wont change the global dictionary ? For instance:

import spam
def myrange(x):
    # insert funcky new range() for the purpose of debugging or such here
spam.range = myrange

You'd have to check at each compile which namespaces are assigned to, and
make sure, at runtime, those namespaces are not somehow optimized.

> If 'from module import *' is used, the user deserves what they
> asked for and *nothing* in that module is cached. I'm not sure
> how 'exec' can be used to screw things up, but let's say we can apply
> the same rule to that. Though that is possibly more tricky..

Take a look at the current translation from (LOAD|DELETE|STORE)_NAME
bytecodes to *_FAST and *_GLOBAL bytecodes. If the current namespace isn't
being changed dynamically (which you can determine, at compile time, for
*local* namespaces) the compiler can determine which name-lookups are going
to be local, and which are going to be global (if they do not fail).

However, there are two ways to alter a local namespace (one that isn't
global in some way, in any case): by using 'from <module> import *', and
by using 'exec'. Imagine 'exec "def range(...): ... "'. This adds 'range' to
the local namespace. 

It isn't tricky, though, the code for that is already there :)

> Something along the lines of this approach could work, I think? The
> approach would be to forbid the caching of certain names if it's
> detected certain things could happen to them that would make them
> uncachable.

These trick would work, yes, the question is: will they work enough? Will
they actually be more efficient than just trying the code ? Or will we be
special-casing so much circumstances that we'll end up optimizing a very
small percentage of code, at the cost of a much higher compile time and a
much more complicated compiler ?

> > Also, we need a way to invalidate the cached entry of egg.spam if any of egg or
> > any of its parent up to the one defining 'spam', have something done to
> > their 'spam' attribute.

> If the possibility exist that a method is changed, don't cache that
> method. Of course setattr() will be a major problem; very hard to determine
> whether setattr() won't affect some class at some point. And you won't
> know what class, or what name. Hmm.

Exactly :) Also, don't forget the fact that another module can change stuff
on an object, or an *exec* in another module (!) can change stuff on an
object, as long as that module has in some way access to the object, however
indirectly.

> > And that means than whenever you change an attribute
> > on a class or instance, you need to traverse the entire tree of subclasses
> > and instances (up to one that 'masks' the change) looking for cached entries
> > and invalidating them.

> Far cheaper would be simply to invalidate the entire cache in such
> a case. That is, if class attributes are changed. I'm not sure why
> you'd need to invalidate things when you change something on an instance,
> though? (except the thing that's changed itself)

That's what I meant. I said 'instances' because a single object can have a
lot of subclasses & instances, and *each of those* can have 'cached
entries'. We're not talking about optimizing a single instance, we're
talking about the whole can of worms.

> > And of course, if you do something like:
> 
> >   egg.ham.viking.spam.plate.drop()
> 
> > you have to consider *each lookup* cached ;-P I don't think lookups are
> > *that* expensive. I'm not sure what the penalty for a failed lookup is, I
> > seem to recall something about a successful lookup being relatively cheap,
> > compared to a failed one, and a 2nd lookup is not likely to fail ;)
> 
> I don't seem to understand this bit. What does it mean, considering
> each lookup cached? Could you explain?

The lookup of 'egg' in the local or global namespace, the lookup of 'ham' in
the namespace 'egg', the lookup of 'viking' in the namespace 'ham',
etcetera.

> I think you need the ideas of invalidation. I think you also need the
> idea of "don't ever cache this" (or at least "don't cache this until I
> tell you to"). I think that approach can help with some of the tricky bits.

Oh definately. I'm just not very sure there'll be enough non-tricky bits
left to actually matter ;)

Perhaps-we-should-discuss-this-over-a-beer-somewhere-in-Amsterdam-Martijn'ly
yr's, Thomas

(Though not during Euro2000, please)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From gmcm at hypernet.com  Mon Jun 12 15:47:15 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 12 Jun 2000 19:47:15 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au>
Message-ID: <8F51A5DF3gmcmhypernetcom@199.171.54.194>

Neurocrat wrote:

>gmcm at hypernet.com (Gordon McMillan) writes:
>
>[snip]
>
>> One point worth reiterating is that where the app (in a C++ design)
>> calls for lots of related classes, you will probably only have a
>> handful in Python. Of course, this means discarding classical "OOP"
>> thinking, and desinging more in the "component" style. But this can
>> yield an order of magnitude all by itself.
>
>If it's worth reiterating, would you mind elaborating on this
>somewhat?

Simply that in Python, inheritance doesn't necessarily have anything to do 
with runtime behavior. (You can, of course, introduce that dependency, but 
you'd be well advised not to.)

The classic example would be 2 base classes (say Event and View), and you 
want view->process(event) to get to both the right overload and the right 
override. In C++ (or Java) that takes spelling out the whole num(View 
derived classes) x num(Event derived classes) matrix, and bouncing the 
message back and forth (first to get the overload and then to get the 
override). In Python, it takes view.process(event).

That is, it's the interface that matters (not the derivation), and you 
don't even have to get formal about that.

Not that you should throw out everything you know about pre-condions and 
post-conditions and substitutability; but you don't have to spend all week 
convincing the compiler you know what you're doing.

- Gordon


From thomas at xs4all.nl  Mon Jun 26 03:24:41 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 07:24:41 GMT
Subject: smtplib
References: <CC59868A0A76D311B1F50090277B321C1F33E8@VRTMAIL>
Message-ID: <slrn8le19c.2oh.thomas@centurion.xs4all.nl>

On Sun, 25 Jun 2000 19:48:30 -0400, Mike Fletcher <mfletch at tpresence.com>
wrote:
>Same here, the Python newsgroup has posts showing up fine, but the list is
>only getting messages that are from other subscribed peoples.  Guessing that
>when they fixed the repeating messages they wound up shutting down the
>gateway :( .  Or possibly the gateway got spam-listed for flooding?

The news->list gateway was closed, by the list maintainer, in order to fix
some bugs in Mailman.

All in all an excellent time to try and get XEmacs + gnus to work properly
(and failing ;-)

Thomas.


From jkraska1 at san.rr.com  Wed Jun  7 01:37:00 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 07 Jun 2000 05:37:00 GMT
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com>
Message-ID: <393DE010.B268CC8D@san.rr.com>

> Can't quite figure out where the error occurs.

You do realize that Guido's copiously taking detailed notes,
in order to extend his list of reasons why he won't be including
lambdas in python 3k, right? :)


C/


From redaelli at inc.it  Sun Jun  4 06:16:00 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 12:16:00 +0200
Subject: float does not parse '0.000000E+00'
References: <39397EFF.E1B76E6E@inc.it> <8hbvv0$40h$1@news.wrc.xerox.com> <393A21C9.E955293F@inc.it> <slrn8jk8sk.hgn.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <393A2C60.C0C1AD3@inc.it>

Remco Gerlich wrote:
> 
> Paolo Redaelli wrote in comp.lang.python:
> > Mark Jackson wrote:
> He means that one of your 'from mod import *' commands may have overwritten
> the name 'float' in the current namespace. Rebound it, in other words. Or
> maybe some other command did that?
Thanks for the explanation. I wasn't aware of such dangers... BTW I
solved the problem
-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/


From dfan at harmonixmusic.com  Mon Jun 12 20:30:09 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 12 Jun 2000 20:30:09 -0400
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no>
Message-ID: <wkg0qiwhgu.fsf@turangalila.harmonixmusic.com>

Peter Schneider-Kamp <petersc at stud.ntnu.no> writes:

| I have to create a tuple (of two integer values) in a
| C extension. At the moment I use the following (rather
| ugly in my eyes) construction:
| 
| t = PyTuple_New(2);
| PyTuple_SET_ITEM(t,0,PyInt_FromLong(a));
| PyTuple_SET_ITEM(t,1,PyInt_FromLong(b));
| PyErr_SetObject(PyExc_IndexError,t);
| 
| where
| PyObject* t;
| long a,b;
| 
| Is there "A Better Way(TM)"?

I just started using CXX and wholeheartedly recommend it, if you're
coding in C++.  This example basically becomes:

  Tuple t(2);
  t[0] = Int(a);
  t[1] = Int(b);

It deals with the reference counts automatically too.

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


From tismer at tismer.com  Sat Jun  3 19:19:33 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 04 Jun 2000 01:19:33 +0200
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <8hbo11$bo2$1@gaia.cdg.acriter.nl>
Message-ID: <39399285.BBD4C2B4@tismer.com>


Cees de Groot wrote:
> 
> Jeff Collins  <Jeffery.D.Collins at aero.org> said:
> >I'm currently porting a 1.5.2+ (soon to be 1.6) to the Palm.
> >Hopefully, I'll be making an announcement soon.
> >
> Drool...

Hee hee :-)
Guess which 1.5.2 this will be :)

going-to-release-SLP-1.2-tomorrow-if I-understand-CVS-WIN-well-
-ly y'rs - (stack|clue)less-chris

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



From aahz at netcom.com  Thu Jun  8 14:25:36 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jun 2000 18:25:36 GMT
Subject: wishlist Python3K: subclass ints
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <393EFF38.C023F10F@san.rr.com> <393FD0C7.D70479A2@visionart.com>
Message-ID: <8hoof0$r9c$1@slb1.atl.mindspring.net>

In article <393FD0C7.D70479A2 at visionart.com>,
Pete Shinners  <pete at visionart.com> wrote:
>
> - Also gone are all the nit-picking type checking routines.
>   instead of first checking if a variable is a "class" type,
>   then checking if the object is a certain class, all variables
>   hold objects of the same "rank". the best move would be to
>   have "int" and "float" be subclassed from a base class like
>   "realnumber".

Close, but no cigar.  Start with the base class "number".  Then
everything else follows: int, float, rational, complex.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From petersc at stud.ntnu.no  Fri Jun  2 15:10:11 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 02 Jun 2000 21:10:11 +0200
Subject: And now for something completely different. Was: String.join 
 revisited    (URGENT for 1.6)
References: <28052000.5@sanctum.jae.ddns.org> <sj3ejda85pj169@corp.supernews.com> <hjoY4.6071$wYl.223614464@newsb.telia.net> <39328edc$0$18384@personalnews.de.uu.net> <_pyY4.10102$Za1.160118@newsc.telia.net> <3932e628$0$6681@personalnews.de.uu.net> <39348990.F7E82EC0@Lugoj.Com>
Message-ID: <39380693.BA37C303@stud.ntnu.no>

James Logajan wrote:
> 
> var = 2["abc"];
> 
> Points awarded for determining the type of "var" and the value assigned to
> it. Extra credit points for explaining why this is valid C/C++. (Hint: this
> has been valid C for as long as the language has been around.)

Okay, it is a char. The corresponding Python expression is
"abc"[2], so the value has to be 'c' (or as it is C 99).

I like it, but I did not know it was legal C. Why?

hoping-there-is-no-slice-operator-too-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From kgw-zamboni-news at stiscan.com  Mon Jun 26 21:49:59 2000
From: kgw-zamboni-news at stiscan.com (Ken Walter)
Date: Tue, 27 Jun 2000 01:49:59 GMT
Subject: stopping threads
References: <3957C1DC.E7C77E2E@sec.noaa.gov> <8j8id2$v3h$1@nntp9.atl.mindspring.net> <etdg0q0hqvt.fsf@w20-575-109.mit.edu>
Message-ID: <eDcSHHgdulIC-pn2-JdOkYStddguU@zamboni.stiscan.com>

On Wed, 26 Jun 2000 21:08:54, Alex <cut_me_out at hotmail.com> wrote:

>
>> I'm currently reading up on Java threads, and while such a function
>> exists, the authors recommend very strongly against doing so.
>> Instead, your thread should watch for some condition and return/exit
>> from its main function/method when that condition becomes true.
>
>Do they say why?
>
Cleaning up resources and closing files properly.
It is always safer to "send a message" to the thread/process
and let it terminate itself.  Only misbehaving ones should be 
terminated from
the outside and then fixed.

>Alex.

Ken Walter

Remove -zamboni to reply
All the above is hearsay and the opinion of no one in particular





From thomas at xs4all.nl  Sat Jun 24 11:28:42 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 24 Jun 2000 15:28:42 GMT
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu> <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk> <1ecjb1r.27wv4iz2rfkpN%bparsia@email.unc.edu> <m34s6l4flf.fsf@atrus.jesus.cam.ac.uk> <8iumph$887$1@nnrp1.deja.com>
Message-ID: <slrn8l9ksv.elg.thomas@centurion.xs4all.nl>

On Fri, 23 Jun 2000 03:50:13 GMT, Andy Freeman <anamax at earthlink.net> wrote:
>In article <m34s6l4flf.fsf at atrus.jesus.cam.ac.uk>,
>  Michael Hudson <mwh21 at cam.ac.uk> wrote:
>> Certainly at some point there was no LOAD_FAST opcode; every name
>> lookup would first look into a local dict, then the global one, then
>> the builtins.  Now references to locals are just indexes into (C)
>> vectors; this is much quicker.

>Is there a standard way to turn off this optimization on a def
>by def basis?

Not really, and it shouldn't be necessary, either. A function's locals are
fairly unreachable, so they cannot be modified by something other than the
function itself(*), and the function itself is analyzed at compile time, to
see if it does any self-modifying ('from module import *', and 'exec', to be
precise.) So, you can turn the optimization 'off' by using something like:

exec "1"

somewhere in your function (though not in an 'unreachable' branch, like in a
'if 0:' -- that'll get optimized away.)

But because a function's locals are so unreachable, there should really be
no need to turn off the optimization. (Which is also why it's default on,
instead of dependant on the '-O' or '-OO' switch, or the like.

(*) It might be possible to change a functions' locals using the
bytecodehacks package, but I've never looked at that package before, so I
wouldn't know.

Newsgroup-instead-of-list-for-the-first-time-ly yr's,
Thomas.


From hzhu at knowledgetrack.com  Wed Jun 14 20:30:26 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 15 Jun 2000 00:30:26 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com>
Message-ID: <slrn8kg937.328.hzhu@rocket.knowledgetrack.com>

>"Warren Postma" <embed at NOSPAM.geocities.com> writes:
>
>> Having something
>> look correct on the screen, but not work because of misuse of tabs versus
>> spaces, or incorrect indentation, is something that still continues to bug
>> me.

If you use emacs:

C-x h M-x tabify

will tabify the whole buffer.  I always do this when doing copy and paste
from news groups, and do an untabify the other way round.  This is because
the default tab space is 4 in emacs (which is sane) but 8 everywhere else.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From pinard at iro.umontreal.ca  Tue Jun  6 19:40:28 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 19:40:28 -0400
Subject: __getattr__ and __setattr__ troubles
In-Reply-To: Michael Hudson's message of "06 Jun 2000 21:11:28 +0100"
References: <8gehq1$kq6$1@news1.xs4all.nl> <8ggo6m$2v6h$1@news.aros.net>
	<oq1z2ad4wi.fsf@titan.progiciels-bpi.ca>
	<m3u2f6wowf.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <oqem6abcpf.fsf@titan.progiciels-bpi.ca>

[Reissued, after a reject from python.org.  Mailman apparently failed because
 of a disk full situation.]

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Have you caused Python to crap out by infinite looping with __getattr__
> yet?

Not once!  Maybe I've been lucky.  In fact, it all goes rather well now.

> > ---------------------------------------------------------------------->
> > class See:
> >     def __getattr__(self, key):
> >         print '**', key
> >         return None
> > print See()
> > ----------------------------------------------------------------------<

> > will yield:

> > ---------------------------------------------------------------------->
> > pinard at titan:~ $ python see.py
> >  ** __str__
> > Traceback (innermost last):
> >   File "see.py", line 7, in ?
> >     print See()
> > TypeError: call of non-function (type None)
> > ----------------------------------------------------------------------<

> > * By the way, and a bit unrelated, why the space before the two asterisks?

> Probably to do with softspace.  Not sure.

It should go between printed element, not before the first one, should it not?

> > * The diagnostic would be rather cryptic, if the program was not so short.
> >   Usability studies could be driven to show that the diagnostic should
> >   rather be changed to something more informative :-).

> Your wish is my command [...]  Not sure it's a sufficiently common error
> to justify this though.  I'll send it to patches at python.org and see what
> they think of it.

I hope they will accept them.  Each and every diagnostic should be as
clear as possible, and _especially_ those diagnostics which are uncommon!

> Overriding __getattr__ is good for "oo, ain't that cool" value, but it
> is probably best to try to avoid it - if you need it, fine, but consider
> alternatives first.

Of course, a "need" is a pretty subjective matter.  We can do a great
deal with very little, and there are not many so-called needs that would
really stand scrutiny.  In my case, I have a working program already,
and implementing __getattr__ is part of an intermediary cleanup, before
attempting to move further, so to simplify code all over and gain better
maintainability.  So in a sense, I could say that I considered at least one
alternative, which is already implemented and working.  I think that the
increase in legibility is worth the slowdown, which the project can afford.

P.S. - I hope that Guido will do his best so future Python stays legible.

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







From jhe at webde-ag.de  Fri Jun 23 15:33:58 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 23 Jun 2000 19:33:58 GMT
Subject: XML-code examples
References: <394f7bc2.365514882@news.online.no> <00062201025104.08285@quadra.teleo.net>
Message-ID: <8J0E0G$KLC$1@POLLUX.IP-PLUS.NET>

> [Thomas Weholt, on Wed, 14 Jun 2000]
> :: Is there more simple XML code out there? I mean the HowTos took about
> :: ten minutes, and it didn`t cover much.

For a real-world example, you can check out

http://center.oftheinter.net/~jh/xml/bookmarks/checkurls.py

and the other files in that directory. It parses a XBEL file, and checks the
status of the bookmarks in that file via the httplib module.

My very first non-trivial Python program, BTW. :)




From ssthapa at midway.uchicago.edu  Tue Jun 20 13:17:45 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Tue, 20 Jun 2000 17:17:45 GMT
Subject: A few python questions
Message-ID: <slrn8ktmmj.2md.ssthapa@ntcs-ip48.uchicago.edu>

    I've been trying to learn how to program in python and as
an exercise been trying to write a simple database app.  However,
I've that a few questions that don't seem to be answered in the 
library reference, tutorial or reference.  
    The first problem I've encountered is trying to determine 
the size of a file.  The file type doesn't seem to have any
function that returns the length of a file.  I've been seeking
to the end of the file and then using tell.  However this seems
kind of inelegant, is there a way to get the file size of a file
without doing this or using the stat function from the os module?
    Also is it possible to use PyKDE or PyQT without installing SIP?
I already have SWIG and don't really want to install another tool
that has basically the same functionality. 
    Finally, is there some documentation for the python interface 
to postgresql?  The rpm package that comes with redhat 6.2 doesn't
have any documentation. 

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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From m.faassen at vet.uu.nl  Thu Jun  1 18:54:28 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 1 Jun 2000 22:54:28 GMT
Subject: "non-essential difficulty" - syntax highlighting?
References: <D15A35708943D3119BA70060979C3C663828D5@PETRIFIED> <8ge7ka$ouo$1@slb6.atl.mindspring.net> <39351648.ED1F1288@uab.edu> <8h37ve$djt$1@slb7.atl.mindspring.net> <39352F52.B5A5B462@uab.edu> <8h4egl$pkc$4@newshost.accu.uu.nl> <393690B2.ED517A8C@uab.edu>
Message-ID: <8h6pj4$ecq$3@newshost.accu.uu.nl>

Shae Erisson <shapr at uab.edu> wrote:
[snip ideas about incremental parser]

> I was thinking of doing it after the Python Refactoring Browser is
> finished, but if someone wants to join me in coding this up, I can
> change priorities. I would guess it would take a month or two.

Any references to this browser? I did a few searches on the web but
couldn't find anything, but interesting. :)

> Any suggestions, comments or offers of help?

Do you know about the compiler-SIG? One of the reasons it got started
was exactly to create the infrastructure to create intelligent 
editors (such as IDLE); incremental parsers and all.

Regards,

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


From rcc at nospamthanks_jennic.com  Mon Jun 19 12:48:19 2000
From: rcc at nospamthanks_jennic.com (Robert Cragie)
Date: Mon, 19 Jun 2000 16:48:19 GMT
Subject: simple thread problem
References: <394E2F9C.6BCD96B5@iue.tuwien.ac.at> <8ilbfp$6a8$1@nntp9.atl.mindspring.net>
Message-ID: <newscache$2quewf$nzd$1@jenpc07.jennic.com>

The mysterious message you get is when 'threading' encounters threads
created in 'thread', so it sounds to me that you are mixing 'threading' and
'thread' objects - are you using 'threading.Semaphore' or such like? As Aahz
says, try to do it all using 'threading' objects.

BTW, it's not actually a problem as such - it will probably still work.
'threading' calls these 'Dummy threads' - look at threading.py for more
info.

Robert Cragie

Aahz Maruch <aahz at netcom.com> wrote in message
news:8ilbfp$6a8$1 at nntp9.atl.mindspring.net...
| In article <394E2F9C.6BCD96B5 at iue.tuwien.ac.at>,
| Markus Gritsch  <gritsch at iue.tuwien.ac.at> wrote:
| >
| >I started a very simple program to get familiar with Python threads.
| >It works fine, except for some misterious messages which I get every
| >time a thread exits:
| >
| >currentThread(): no current thread for xxxx
|
| I'm not sure how to tell you to fix the problem in your current context;
| instead, let me push you strongly to subclass threading.Thread() instead
| of using the thread module.  I guarantee this will make your life a lot
| simpler.  If you still have trouble, come back here and I'll show you
| how to do your example using Thread().
| --
|                       --- Aahz (Copyright 2000 by aahz at netcom.com)
|
| Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
| Hugs and backrubs -- I break Rule 6
|
| "The only problem with Microsoft is they just have no taste." --Steve Jobs
| (From _Triumph of the Nerds_ PBS special)




From johnvert at my-deja.com  Thu Jun  8 14:35:24 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Thu, 08 Jun 2000 18:35:24 GMT
Subject: getting a URL
Message-ID: <8hop1b$c1b$1@nnrp1.deja.com>

Hello,

As an exercise I'm trying to write a simple cgi script that will get the
source code of an html file and echo it to the screen.  Here's what I
have:

import cgi, urllib

(snip)

		url = urllib.urlopen("http://www.server.org/foo.html")
		print "<pre>"
		print cgi.escape(url.read()),
		print "</pre>"

But nothing is printed.  The problematic line seems to be:

url = urllib.urlopen("http://www.server.org/foo.html")

But the URL I gave it in the actual program is correct.  I tried a few.
If I run this on the shell (it's meant to be a cgi) it raises IOError
sayin 'network unreachable'.  What could be wrong?  Also, why does
execution stop at:

      print "<pre>"

Why isn't the rest (print "</pre>" among others) printed?

Thanks,
 -- John


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


From digitig at cix.co.uk  Sat Jun  3 19:49:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 4 Jun 2000 00:49 +0100 (BST)
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <Pine.GSO.4.10.10006031335390.17165-100000@sundial>
Message-ID: <memo.20000604004939.14417C@digitig.compulink.co.uk>

In article <Pine.GSO.4.10.10006031335390.17165-100000 at sundial>, 
moshez at math.huji.ac.il (Moshe Zadka) wrote:

> On 2 Jun 2000, Florian Weimer wrote:
> 
> > There's the convention to use "Foo" for classes and "foo" for
> > instances.  I don't think people want to give up that.
> 
> OK, I've heard this one once too many times. 
> 
> What's wrong with:
> 
> >>> class Foo:
> ...     pass
> ... 
> >>> aFoo = Foo()
> 
> Certainly much clearer?

Not, IMHO, when you get up to things like:

>>> class Foo:
...    pass
...
>>> def bar(foo = None,
...    # /lots/ of other optional parameters
...    ):
...    # some implementation
...
>>> baz = Foo()
>>> bar(foo = baz)

I consider it a /very/ useful idiom.


From owen at astroNOJNK.washington.edu.invalid  Mon Jun 12 12:55:55 2000
From: owen at astroNOJNK.washington.edu.invalid (Russell E. Owen)
Date: Mon, 12 Jun 2000 09:55:55 -0700
Subject: Threads -- play well with others? If not, what instead?
Message-ID: <8i34mr$9hti$1@nntp6.u.washington.edu>

David Beazley's excellent "Python Essential Reference" says in the 
section on threads: "In addition, many of Python's most popular 
extensions such as Tkinter may not work properly in a threaded 
environment."

I assume it's true, but it was quite a bombshell. I was hoping to write 
a networked GUI client, hence:
- read data from a socket and fill in a GUI display
- accept input from the user and write data to the socket
I assumed I'd use two threads, one for input, one for output. Now I have 
no idea what to do. Any suggestions?

-- Russell


From richard_chamberlain at ntlworld.com  Sat Jun 17 17:58:55 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 17 Jun 2000 22:58:55 +0100
Subject: Problem Installing/Using Windows Extensions
References: <2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1@lp.airnews.net>
Message-ID: <BnS25.1127$fw6.20615@news6-win.server.ntlworld.com>

Hi David,

I had the same problem with a PC at work running W95, but not on any of the
others. I finally narrowed it down to be the only one without Internet
Explorer 5 installed. And from that I presume that DCOM needs to be
installed.

Hope that helps,

Richard


David Steffen <steffen at biomedcomp.com> wrote in message
news:2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1 at lp.airnews.net...
> Howdy:
>
>   I'm trying to use Pythonwin with Python 1.52 under Windows 95 and am
> consistently running into a problem that I don't understand at all,
> probably because I know relatively little about Windows.
>
>   When I selected "COM Makepy Utility" from the Tools menu of
> Pythonwin, I got the following error message:
>
> """
> >>> Failed to execute command:
> from win32com.client import makepy;makepy.main()
> Traceback (innermost last):
>   File "C:\Program Files\Python\Pythonwin\pywin\framework\toolmenu.py"\
> , line 103, in HandleToolCommand
>     exec "%s\n" % pyCmd  File "<string>", line 1, in ?
>   File "C:\Program Files\Python\win32com\client\makepy.py", line 46, \
> in ?
>     import genpy, string, sys, os, types, pythoncom
>   File "C:\Program Files\Python\win32com\client\genpy.py", line 16, in ?
>     import pythoncom
>     Import Error: DLL load failed: A device attached to the system is \
> not functioning.
> """
>
>   I noted that what I had was version 1.25, and that version 1.32 was
> up on Mark Hammond's site on starship python, so I installed the newer
> version.  The installed completed, but with three error messages near
> the end:
>
> """
> Registering COM Server AXScript Engine
>
> Registration of the AXScript Engine COM server failed.
>
> Installation will continue, but this server will require manual
> registration before it will function.
>
> exceptions.ImportError: DLL load failed: A device attached to the
> system is not functioning.
> """
>
> ...with the other two messages being similar ones about the Python
> Interpreter COM server and the Dictionary COM server.
>
> I'd be delighted to RTFM, if I could find the relevant M (I've looked
> in the included Python docs and on the python.org and starship python
> websites).
>
> I'd be happy to register the servers manually, if that would solve the
> problem and if I had a clue as to what than meant.
>
> If there is a device not functioning on my laptop, I'd love to know
> that, but a disk scan indicates no problems and I haven't noticed any
> other problems with my laptop.  I'm thinking I may be missing a DLL,
> driver, or other software component; if anyone could suggest which one,
> I'd be pleased as punch.
>
> Thanks in advance!
>
> -David Steffen-




From akuchlin at mems-exchange.org  Thu Jun  8 16:46:12 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Jun 2000 16:46:12 -0400
Subject: ANNOUNCE: early alpha Quixote Web development tool
References: <200006080135.VAA04781@mira.erols.com> <393FD07E.20EAB1B2@easics.be>
Message-ID: <3du2f3yk8b.fsf@amarok.cnri.reston.va.us>

Jan Decaluwe <jand at easics.be> writes:
> - Didn't find the doc subdirectory mentioned in the README file.

Oops!  I've made a new tarball that fixes this, and also copied the
three files in that directory (INSTALL, ptl.txt, ZPL.txt) 
into http://www.mems-exchange.org/exchange/software/files/quixote/ .

> - It seems the top-level package should be placed in a 'quixote'
>   directory, but this is not explicitly documented. Very good
>   that all the code is put into a package (unlike the Zope components).

Running setup.py should handle this for you, though, and install the
modules correctly in a 'quixote' directory.
 
> One thing I'm using from Zope is Persistent objects - objects
> that are automatically written and restored from disk in response
> to requests. (However, I'm completely confused about the continuing

We're planning to use the ZODB, and eventually ZEO on top of the ZODB.
ZODB's interface is quite natural and transparent to a Python
programmer.  No need to reinvent *that* wheel, at least for our
project's needs.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
  "You? What are you?"
  "Me? Lady, I'm your worst nightmare -- a pumpkin with a gun."
  -- The Furies and Mervyn, in SANDMAN #66: "The Kindly Ones:10"






From tomsantos at email.com  Sun Jun 18 16:37:32 2000
From: tomsantos at email.com (Tom Santos)
Date: Sun, 18 Jun 2000 13:37:32 -0700
Subject: Getting Password from User
References: <8ij9kr$qa5$1@bob.news.rcn.net> <87ln027nof.fsf@ifm.uni-kiel.de>
Message-ID: <8ijcfq$gef$1@bob.news.rcn.net>

Just what I was looking for. :)

Thanks,

-Tom-

Janko Hauser <jhauser at ifm.uni-kiel.de> wrote in message
news:87ln027nof.fsf at ifm.uni-kiel.de...
> Look for the module getpass in the library reference. It also supports
> windows according to the document string.
>
> HTH,
> __Janko
> --
>   Institut fuer Meereskunde             phone: 49-431-597 3989
>   Dept. Theoretical Oceanography        fax  : 49-431-565876
>   Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
>   24105 Kiel, Germany




From thomas at xs4all.net  Wed Jun 14 04:18:37 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 14 Jun 2000 10:18:37 +0200
Subject: TCP packet size?
In-Reply-To: <39487D5B.FC96BE3F@rpgarchive.com>; from chris@rpgarchive.com on Thu, Jun 15, 2000 at 01:53:15AM -0500
References: <39487D5B.FC96BE3F@rpgarchive.com>
Message-ID: <20000614101837.Z20700@xs4all.nl>

On Thu, Jun 15, 2000 at 01:53:15AM -0500, chris wrote:

> Is there a way to ensure or predict packet size when using TCP?  I'm
> developing a Multi-user application that functions much like a
> muli-player game.  The applications are sharing xml data.  The xml data
> can get very big 3000-6000 bytes.  Unfortunately, when the size of the
> network message gets over about 2000, its to get broken up into smaller
> packets.  So, when I send a 3000 byte message it gets cut up into 2-3
> packets.  Anther problem I get is when I send two small messages back to
> back, they are sometimes received together.  Is there something I'm
> missing here?

> I tried using headers and breaking up messages into smaller chunks, but
> the size of the receiving packets doesn't always match what I sent.  I'd
> appreciate any help or suggestions on how I can tackle this problem.
> I'd be happy to send someone my network code.

Are you sure you are using TCP ? A TCP connection is supposed to createa a
'virtual circuit' that relieves you of all the troubles of determining
packet size, handling out of order arrival, losing packets, etc. What you
describe sounds more like UDP than TCP, and if you are working with UDP, you
better be damned certain you want UDP ;-)

IP packet sizes are limited by the 'Maximum Transmission Unit', which should
be between 68 and 1500 bytes. Most modern TCP/IP stacks do path MTU
discovery to lower the MTU when it is necessary (the other side, or a router
on the link to the other side, apparently can't or won't handle a large MTU,
so it is lowered.) so it's not really safe to rely on that size.

This restriction goes for both TCP and UDP, though TCP automatically breaks
up any data you send through the connection into packets of the right size,
and reassembles them on the other size, in order. TCP packets are also
guaranteed to arrive in-order, and there is error-control so that lost
packets are re-sent.

UDP does not have all these control functions (that's why it's called the
Unreliable Datagram Protocol ;-) and that means you have to do all those
things yourself: keep a packet counter, notify the other end when packets
are lost, wait for them to arrive in-order, and make sure your packets
aren't too large(*). For large datatransfers it is likely not worth it. If
you really need the unreliable properties of UDP (like not stalling when you
are experiencing packetloss), make two connections, TCP for data and UDP for
signaling the other side.

You should also keep in mind that using UDP will probably fail when the
connection has to pass through a firewall or NAT (Network Address
Translation) situation.

*) I haven't done enough UDP things in Python to know how to do these things
easily, so I can't help you there, sorry ;-P

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From kens at sightreader.com  Fri Jun 16 01:39:44 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 22:39:44 -0700
Subject: Python class browser
References: <XKQ15.1$oL1.1552@paloalto-snr1.gtei.net>
Message-ID: <3949BDA0.DF21A426@sightreader.com>

Brian wrote:

> Anyone know of a class browser for Python?  Preferrably written using
> wxPython?
> Thanks!

There's a good one that's part of PythonWin.  I haven't looked at the
code though.  PythonWin and wxPython are very similar, so it shouldn't be
too difficult to port.

Let me know if you find a solution.  I'm gonna need it too in a couple
months.

Thanks,

Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/7703f821/attachment.vcf>

From JamesL at Lugoj.Com  Sat Jun 10 00:12:03 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Fri, 09 Jun 2000 21:12:03 -0700
Subject: Reloading on the fly: was: Re: Python Productivity over C++
References: <20000609185450.91797.qmail@hotmail.com> <8hrjg3$dn2$2@news1.xs4all.nl>
Message-ID: <3941C013.F1F6DE85@Lugoj.Com>

Boudewijn Rempt wrote:
> Could you post an example of designing the main program to reload
> submodules? I'd like to get that working, too, and coming from
> languages where that's impossible, I've a bit of trouble coming up
> with a solution...

Several ways are possible. Here is one way. In the main program you might
have:


# main.py
# Run-time modify of a module.
import sys
import os
import m1
m1.F(1) # Try first version of m1.py
del sys.modules['m1']  # The magic bit.
print "Please edit m1.py; press return when ready."
sys.stdin.readline()
import m1
m1.F(1) # Try second version of m1.py

And the module to be modified looks like:

# m1.py
def F(score):
    print "Banana"
    #print "Mango"

Now a typical run with commenting out the first pritn and uncommenting the
second print in m1.py should lead to an output like:

Banana
Please edit m1.py; press return when ready.

Mango


Hope this helps!

P.S. If you don't even know the name of the module ahead of time, you can
get creative with the "exec" statement!


From smst at bigfoot.com  Thu Jun 15 03:20:32 2000
From: smst at bigfoot.com (Steve Tregidgo)
Date: Thu, 15 Jun 2000 08:20:32 +0100
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com> <ug0qflns9.fsf@ctwd0143.fitlinxx.com>
Message-ID: <394883C0.AF4822BB@bigfoot.com>

David Bolen wrote:
> 
[much of `verify_class' snipped]
>
>                 except AttributeError, TypeError:
>                     print "Method %s not supported" % curmethod

Note that this code won't pick up the TypeError -- to do that, make a
slight change:

    except (AttributeError, TypeError):

The second `argument' in an except clause is used to get the error that
was raised -- so the code you posted will assign the details of the
AttributeError in the name TypeError!  To catch multiple types of error,
you must list them all as the first argument in the except statement,
using a tuple as above.  (Note also that you can use multiple except
clauses to do different things depending on what error is raised).

Feeling-pedantic-this-morning-ly y'rs,
Steve 

-- 
Steve Tregidgo
Software Developer
http://www.businesscollaborator.com



From mskott at image.dk  Mon Jun  5 14:47:04 2000
From: mskott at image.dk (Martin Skøtt)
Date: Mon, 05 Jun 2000 18:47:04 GMT
Subject: Recompiling Python with threads enabled
Message-ID: <n1l03rcl.fsf@jumbo.skott.dk>

Hi

Sorry if you don't think this belongs in a Python group, but I think
its more relevant here than in a Linux group. Sorry for the
inconvenience :-)

And now on to the question:
I wish to recompile Python 1.5.1 on my Red Hat 5.2 system to enable
the threads module, but how do I do? Do I need to install the src RPM
and which files should I change to enable threads??

I hope anyone will help me
-- 
Martin Sk?tt
mskott at image.dk

'In a certain sense, all decent programming languages are equally powerfull'
Dr. Alan Turing


From hellan at acm.org  Mon Jun  5 16:09:33 2000
From: hellan at acm.org (Jon K Hellan)
Date: 05 Jun 2000 22:09:33 +0200
Subject: MimeWriter and Excel file
References: <20000605.17343286@jcoruna2.bio.umd.es>
Message-ID: <8766rnj3eq.fsf@parus.parus.no>

juan Carlos Coru?a <jcoruna at umd.es> writes:

> Hello all,
> I have the following problem:
> I send an email with python using the MimeWriter module and an excel 
> file attached. When I send the to myself inside our intranet the 
> attchmente arrives well, but if I send the email to anyone outside our 
> intranet and open the attachment with MS Excel I see the base64 
> encoded characters. Any ideas.

I would be very surprised if this turned out to be a Python
problem. The MIME apparently doesn't arrive intact at the recipient.

As to why, I can only guess. It could be an anti-virus measure at the
receiving site. 

Regards

Jon K?re


From hansen at fsl.noaa.gov  Wed Jun 21 12:59:57 2000
From: hansen at fsl.noaa.gov (Tracy Lee Hansen)
Date: Wed, 21 Jun 2000 10:59:57 -0600
Subject: Color Wheel Editor in Python?
Message-ID: <3950F48D.580D1499@fsl.noaa.gov>

Does anyone have a "standard" Color Wheel Editor implemented in Python? 
The ones we've found off of the Python website are the slider-bar rather
than color wheel variety...

Thanks!
Tracy


From just at letterror.com  Sun Jun 11 04:43:35 2000
From: just at letterror.com (Just van Rossum)
Date: Sun, 11 Jun 2000 09:43:35 +0100
Subject: tuple creation in C extensions
In-Reply-To: <3942AFEE.5155481@stud.ntnu.no>
Message-ID: <l03102801b569016fbd23@[193.78.237.174]>

At 11:15 PM +0200 10-06-2000, Peter Schneider-Kamp wrote:
>I have to create a tuple (of two integer values) in a
>C extension. At the moment I use the following (rather
>ugly in my eyes) construction:
>
>t = PyTuple_New(2);
>PyTuple_SET_ITEM(t,0,PyInt_FromLong(a));
>PyTuple_SET_ITEM(t,1,PyInt_FromLong(b));
>PyErr_SetObject(PyExc_IndexError,t);
>
>where
>PyObject* t;
>long a,b;
>
>Is there "A Better Way(TM)"?

t = Py_BuildValue("(ii)", a, b);

Just





From alex_c at MIT.EDU  Wed Jun 14 13:05:51 2000
From: alex_c at MIT.EDU (Alexander S Coventry)
Date: Wed, 14 Jun 2000 13:05:51 -0400 (EDT)
Subject: tuple creation in C extensions
In-Reply-To: <20000614184508.B26436@xs4all.nl> (message from Thomas Wouters on
	Wed, 14 Jun 2000 18:45:08 +0200)
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu> <20000614184508.B26436@xs4all.nl>
Message-ID: <200006141705.NAA14660@w20-575-109.mit.edu>

> The docs already have that:
> 
>    PyObject* PyList_GetItem (PyObject *list, int index)
> 
>    Return value: Borrowed reference.
> 

Oh, right.  I saw that before, but I didn't realize that 'Borrowed
reference' means the reference count to the object returned isn't
incremented.

Alex.



From moshez at math.huji.ac.il  Tue Jun  6 01:59:06 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 6 Jun 2000 08:59:06 +0300 (IDT)
Subject: indirect references
In-Reply-To: <20000605160915.G11994@activestate.com>
Message-ID: <Pine.GSO.4.10.10006060857010.4711-100000@sundial>

On Mon, 5 Jun 2000, Trent Mick wrote:

> Strings are immutable in
> Python and 

Right

> passed/assigned by value (not by reference, which is what you are
> asking for here).

Wrong, they are passed by reference. But what Shaleh is asking is for
mutating the reference, and *that's* impossible.

> Arrays and dictionaries are passed by reference.

Just like everything else in Python. But arrays and dictionaries are
mutable.

just-picking-some-random-nits-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From jkraska1 at san.rr.com  Fri Jun  9 23:33:08 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 10 Jun 2000 03:33:08 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com>
Message-ID: <3941B79D.AAE9ECAF@san.rr.com>

> I find it easier and quicker to lay out high level design and in
> experimenting with finding the best class/data schema. And I can then run
> it. The syntactical constructs of C++ tend to distract from the design
> thought processes.

I strongly agree. In python, I get to think entirely about the problem,
without worrying about the language/programming environment one bit,
really. Furthermore, and this is a MAJOR win, python programs are *small*,
which makes them easier to maintain, easier to read, and faster to code
in general. Python is also very syntactically lightweight, which has to
me my favorite thing about it.

Right now, I strictly use it as a rapid prototyping environment. It's
great for testing out concepts.


C/


From tanzer at swing.co.at  Wed Jun 14 01:27:41 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 14 Jun 2000 07:27:41 +0200
Subject: SV: Python Productivity over C++ 
In-Reply-To: Your message of "Tue, 13 Jun 2000 17:01:45 +0200."
             <39464CD8.38FB435B@muc.das-werk.de> 
Message-ID: <m1325iH-000wcDC@swing.co.at>

Thomas Thiele <thiele at muc.das-werk.de> wrote:

> The only difference I see between developing algorythms in C++ and
> Python is that python needs no time for compiling. But the advantage
> of C++ is that I get the most error messages during compilation (->
> earlier!)

If this is so, then you aren't using Python's full potential.

And as other people pointed out in this stream, C++'s type checking is
overrated. (If you talked about Ada or Eiffel, the situation would be
different but Python would still have strong advantages).
 
> And I think the developing overhead (algorithms, class design) is
> larger than translation into a special language.

This is definitely not true -- if you are using Python in a pythonic
way.

If you use Python, design and implementation can go hand in hand. In
fact, using a notation like UML just wastes time. This has a lot of
consequences:

- less work to do (instead of drawing diagrams you get working code)

- immediate feedback

- it is much easier to reconsider a specific design decision and
  decide on different strategy

Fast feedback and less legacy design result in a better designed
system.

Proceeding in a similar way with C++ would be suicidal -- except for
the tiniest problems. When using C++, you constantly have to consider
lots of issues totally unrelated to the problem to be solved (what
Brooks called accidental complexity). Because that interferes with
the goals of a good design, you first have to make a
language-independent design.

To translate that design into C++, you need a second low-level design
step. With experience, you can do that while implementing -- though I
would still contend that this step needs more effort than
designing/implementing in Python.

The real difference between a Python and a C++ program appears only
after you finished it. Changing a well designed C++ program is much
harder than changing an eqivalent Python program. 

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From invalid.address at 127.0.0.1  Fri Jun  9 13:45:10 2000
From: invalid.address at 127.0.0.1 (David)
Date: Fri, 09 Jun 2000 17:45:10 GMT
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl>
Message-ID: <39442c45.54923673@news.telus.net>

On Fri, 9 Jun 2000 16:39:38 +0200, Thomas Wouters <thomas at xs4all.net>
wrote:

>As promised, here is my patch to add augmented assignment (+=, -= and
>family) to Python. It's still Work In Progress, and probably requires a
>rewrite of some magnitude before Guido will accept it.


Why would the Python community want this?  It obfuscates code, in as much
that newbies aren't likely to intuite what the symbols mean.
 
We just had a thread suggesting that = and == are frequently confused by
newbies.   How much more confused would they be with a plethora of
alternative symbols?
 
Python seems to stick to the "one right way" model of programming: most
people's code tends to look the same.  It increases maintainability and
readability.
 
I think we'll end up with a confusion of styles if augmented assignments
are allowed.
 
And it's not like it adds new functionality nor improves code readibility.
It just saves the lazy programmer a whopping one or two keystrokes.
 
Bleargh, I say.  IMO, YMMV, IANALL, etc.


From scarblac-spamtrap at pino.selwerd.nl  Wed Jun 21 08:40:46 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 21 Jun 2000 12:40:46 GMT
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee> <394e8973.14164217@news.iol.ie>
Message-ID: <slrn8l1dpg.n23.scarblac-spamtrap@flits104-37.flits.rug.nl>

Sean Mc Grath wrote in comp.lang.python:
> On Fri, 02 Jun 2000 18:01:12 +0300, Kaimar Karu <kaimar at ut.ee> wrote:
> >Where can I find some kind of helping material (tutorials etc) about how
> >to use xml in python?
> >I need some examples, just pure specification is not enough.
> 
> My book "XML Processing with Python", ISBN 0 13 021119 2,
> has just been published by Prentice Hall. This should
> help.

That sounds good. Is there an independent review of it somewhere on the web
already?

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  2:38pm  up 107 days,  2:51,  8 users,  load average: 0.17, 0.37, 0.27


From eric at estinc.com  Thu Jun 29 17:15:33 2000
From: eric at estinc.com (Eric Lee Green)
Date: Thu, 29 Jun 2000 21:15:33 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com> <MdK65.2186$iN5.485797@ptah.visi.com>
Message-ID: <395BBC97.F8939100@estinc.com>

Grant Edwards wrote:
> 
> In article <395B6960.26AB52C1 at estinc.com>, Eric Lee Green wrote:
> 
> >> >Mostly because Python doesn't "really" have classes. What Python
> >> >"really" has are hash tables whose entries are objects (either methods
> >> >or other data objects), and some syntactic sugar [...]
> >>
> >> By this logic, all C++ has is pointers and some syntactic sugar.
> >
> >Exactly :-).
> >
> >C++ is syntactic sugar for "C". As should be obvious from the fact that most
> >early C++ compilers generated "C" as their output :-).
> 
> And C is just semantic sugar for assembly...

You mean you've never noticed the resemblence between "C" and PDP-11 assembly
language? :-).

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From hzhu at knowledgetrack.com  Wed Jun 21 15:35:13 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 21 Jun 2000 19:35:13 GMT
Subject: Computer Algebra System
References: <8inpa4$8iv$1@vishnu.jussieu.fr> <m366r41nqr.fsf@chinon.cnrs-orleans.fr>
Message-ID: <slrn8l26d4.q2.hzhu@rocket.knowledgetrack.com>

On 20 Jun 2000 15:33:00 +0200, Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
>Rodolphe Pollet <pollet at liliput.lct.jussieu.fr> writes:
>
>> Is anyone aware of a Computer Algebra System (CAS) ---
>> i.e a mathematical software that is able to derivate, integrate,
>> simplify, find limits, ... analytically --- in python ?
>> It should compete with Mathematica or Maple.
>
>Pythonica isn't there yet, but it's an interesting start. More at:
>
>       http://www.strout.net/python/pythonica.html

That looks like an emulator of Mathematica, but I think there could be a much
better symbolic package given python's oo capabilities. 

One should be able to define classes of mathematical objects with various
properties. Python allows a level of abstraction much closer to true
mathematics than Mathematica can.

For example, if one defines polynomial on a ring and someone defines a ring
of algebraic numbers then one immediately get polynomials of algebraic
numbers.

Python's simple interface mechanism that is not tied up to inheritance makes
it extremely easy to define mathamatical concepts with any given property.

I have some ideas on how this might be done, although my time is tied up for
numerical computations.  If anyone is interested I can take some time to
write a prototype to get it started.  An open source symbolic algebra
package base on python stands a chance of greatly outperforming Mathematica
or Maple, especially in its ease of user-defined extensions.  It would also
be a big boost to python itself.

Huaiyu

PS. And a lot of mathematicians just love to use indentation!

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From kpmurphy at my-deja.com  Fri Jun 30 16:21:33 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Fri, 30 Jun 2000 20:21:33 GMT
Subject: backslash in regular expression problem
References: <394A55BA.69269793@wolfson.co.uk> <02de01bfd826$cf05bd80$6401a8c0@home>
Message-ID: <8jivfr$qc2$1@nnrp1.deja.com>

what if instead of having '\\stuff' you have a str = '\\stuff'  ?

i know that you can print r'\\stuff' but you can't print rstr...

how do you use the raw with a variable?

Thanks,

	-->keith

In article <02de01bfd826$cf05bd80$6401a8c0 at home>,
  "Darrell Gallion" <darrell at dorb.com> wrote:
> Try to avoid assigning to "string" that will cause problems when you
import
> string
> Especially when your using "\" try to use raw strings. Note the
r"xxxx"
>
> >>> s=r"\END\\"
> >>> print re.sub('[\\\]','',s)
> END
> >>> print re.sub(r'[\\]','',s)
> END
>
> --Darrell Gallion
>
> ----- Original Message -----
> From: "Gordon McLeod" <Gordon.McLeod at wolfson.co.uk>
> > import re
> > string="\END\\"
> > x=re.sub('[\]','',string)
> > print x
> >
> > and expect x to be 'END' but I get:
> >
> > Traceback (innermost last):
>
>


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


From paul at prescod.net  Thu Jun 22 14:56:31 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 22 Jun 2000 13:56:31 -0500
Subject: Any Python project similar to Jserv and Cocoon?
References: <6D8A17398E28D3119F860090274DD7DBF13C74@pces.cadlab.it>
Message-ID: <3952615F.F4FE49EF@prescod.net>

Alessandro Bottoni wrote:
> 
> Hi Snake-tamers,
> lately, I have been playing with XML, Apache, Jserv and Cocoon (see
> http://www.apache.org , http://java.apache.org and http:/xml.apache.org ).
> 
> All of these tools are written in Java (and this makes almost impossible to
> make a "distribution" CD-ROM for them, because of the strict licence policy
> of Sun...).

Actually, Apache neither Apache nor XML are written in Java. :)

Anyhow, the Python equivalent to JServe and Cocoon is called "Zope". I
think its XML support is not yet as strong as Cocoon's but it has other
features which are well beyond Cocoon.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Floggings will continue until morale improves.





From robin at jessikat.co.uk  Wed Jun 28 04:55:31 2000
From: robin at jessikat.co.uk (Robin Becker)
Date: Wed, 28 Jun 2000 09:55:31 +0100
Subject: Controlling Acrobat Reader with COM
References: <200006262159.AA3991339306@mail.mamey.com>
 <s7khls0in7rrfbqnqdo4cha6chh9qqhstr@4ax.com>
 <Q38VarAdKTW5Ewcx@jessikat.demon.co.uk>
 <kkb65.30$Tb7.278@news-server.bigpond.net.au>
Message-ID: <EYdKnFAD2bW5Ewdz@jessikat.demon.co.uk>

In article <kkb65.30$Tb7.278 at news-server.bigpond.net.au>, Mark Hammond
<MarkH at ActiveState.com> writes
>"Robin Becker" <robin at jessikat.demon.co.uk> wrote in message
>news:Q38VarAdKTW5Ewcx at jessikat.demon.co.uk...
>> ...
>
>> I tried already with correctly escaped filenames and always got the same
>> error ie 'c:\\Python\\Installs\\test.pdf' got the same error as
>> 'gnarg.pdf' (and I don't have any docs called gnarg.pdf :)).
>
>Its likely that Acrobat needs a full-blown control container.  You may
>have to run it like the OCX demos in Pythonwin are run - check out
>pywin\demos\ocx.
>
>Mark.
...
Yep I reckon that's the problem. For those interested wxPython now has a
activex dem pdf doc in a window, but it fails for me for some reason.
-- 
Robin Becker


From bjorn at roguewave.com  Fri Jun  2 17:28:34 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 02 Jun 2000 15:28:34 -0600
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no> <3936EC02.60768A66@sightreader.com> <Pine.LNX.4.21.0006020157430.25547-100000@lade.src.no> <8h7ar6$nb8$1@slb6.atl.mindspring.net>
Message-ID: <39382702.83F4E9CA@roguewave.com>

Andrew Dalke wrote:
> 
> Joseph Santaniello wrote:
> >Ah, yes. But I'm still missing my commas [for 23,342.40]... A nice
> > clean way to do that?
> 
> I think the following is only somewhat tricky, but perhaps the
> cleanest.  (Okay, there's also supposed to be a way to do this with
> locale, but I don't know how.)

import locale
locale.setlocale(locale.LC_ALL,"")  # set to default locale
locale.format('%.2f', 5555.555, 3)  # (format, value, grouping)
# outputs 5,555.56 on my system

--bjorn



From avv at quasar.ipa.nw.ru  Sat Jun 24 13:41:52 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Sat, 24 Jun 2000 10:41:52 -0700
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net> <3gh3lsgg9dtkfgrsq1ml2i9jcfkknrb6ei@4ax.com> <Qfq45.15292$C44.879449@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <3954F2E0.6768A127@quasar.ipa.nw.ru>

Hi

William Dandreta wrote:
> Eventually we plan to migrate to Linux. So I need something platform
> independent.
> 
> The DB has 100,000+ entries and it grows at about 1000 to 2000 per month.
> 
> We have it indexed on 6 fields, and fast access is important. Users are used
> to typing in the first couple of letters and holding the down arrow key
> until they find what they are looking for.

If fast access is really important, you may take a look at an
open-source descendant of the famous db_Vista. Which for sure runs under
DOS and Linux. I used it that way for years, with my own wrappings for
C++ and Python. With this database the best strategy would be to develop
enough higher-level domain specific persistent C/C++ entities,
implemented via the database, then wrap them for Python. For my
experience, it rules. This implies that the database evolves slower than
your Python scripts.

Regards

Alexander



From jim at interet.com  Tue Jun 20 09:16:02 2000
From: jim at interet.com (James C. Ahlstrom)
Date: Tue, 20 Jun 2000 09:16:02 -0400
Subject: Kjbuckets-Module, Python 1.6 and NT
References: <394E4B02.5F4A0AB4@szs.ira.uka.de>
Message-ID: <394F6E92.42AD72B9@interet.com>

Thomas Freytag wrote:
> 
> Dear all,
> 
> we have severe trouble getting the kjbuckets-module running with Python
> 1.6 and Windows NT. We have tried the following so far:
> 
> 1)Compile the kjbucketsmodule.c as a dll. Fails and produces some
> "Initialiser is not a constant" error, whatever...

Use the compiler switch to compile as a C++ module.  Be sure to
declare initkjbuckets() as an exported symbol with declspec().

> 2)Re-compile the Python interpreter using Visula C++ 6.0 and adding
> kjbucketsmodule.c to the files to be considered in the project. Plus we
> edited the "setup.in" in the Modules directory.  C++ claims to copile
> "kjbucketsmodule.c", however, when we try  "import kjbuckets" with the
> freshly compiled interpreter the answer is "No module....".

This only works on Unix.  You must edit ./PC/config.c by hand to
add the module.  This is the easiest solution.

Jim Ahlstrom


From akuchlin at mems-exchange.org  Fri Jun  9 10:29:11 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 09 Jun 2000 10:29:11 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <m3snunuj1s.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3dwvjzlyh4.fsf@amarok.cnri.reston.va.us>

Michael Hudson <mwh21 at cam.ac.uk> quoted Moshe Zadka:
> I  was glad I did it in Python, because it only cost me one afternoon
>   to discover the idea sucks.            

Indeed.  This is a point missed by much of the flamage on the
linux-kernel list about the CML2 kernel configuration system.  If CML2
turns out to be a bad idea, or can't be made to work correctly, I'm
sure that ESR would much rather throw out 1000 lines of Python code
than 10,000 of C code that had to be carefully debugged to get it to
run enough to discover that it's a bad idea.  If CML2 does turn out to
be a good idea, it can be recoded into C if the Python dependency
really is unacceptable.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
I'm willing to accept that someone might just have all the answers. However,
I'm pretty sure he won't be posting them on a GeoCities page.
  -- From mathew's October 1999 web.scan






From paul at fizzylab.com  Sat Jun 17 17:27:22 2000
From: paul at fizzylab.com (paul beard)
Date: Sat, 17 Jun 2000 14:27:22 -0700 (PDT)
Subject: string.join question (was Re: urllib question)
In-Reply-To: <Pine.LNX.4.21.0006171015510.25296-100000@fep132.fep.ru>
Message-ID: <Pine.LNX.4.10.10006171424430.19831-100000@mail.fizzylab.com>

The length isn't the issue. I'm not able to do the string.join for some
reason. I need to feed many URLs from a common host. I have the URLs (from
an access log) and want to just append the URL to the host. 



On Sat, 17 Jun 2000, Oleg Broytmann wrote:

> On Fri, 16 Jun 2000 paul at fizzylab.com wrote:
> > u = urllib.urlopen(string.join(["http://some.host.com/",
> > "montstrousQueryString?withParameters"]))
>    ^^^^ How long is it? HTTP allows only 1024 characters here. For longer
> queries use POST method instead of GET.
> 
> Oleg.            (All opinions are mine and not of my employer)
> ---- 
>     Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
>            Programmers don't die, they just GOSUB without RETURN.
> 





From dieter at handshake.de  Sat Jun 10 16:49:59 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 10 Jun 2000 22:49:59 +0200
Subject: Tkinter: treeview widget
References: <393E3782.FF7A5DCA@proceryon.at> <Lpm05.3258$jX6.60871@news2-win.server.ntlworld.com>
Message-ID: <x7hfb15ki0.fsf@lindm.dm>

"richard_chamberlain" <richard_chamberlain at ntlworld.com> writes on Sat, 10 Jun 2000 08:57:40 +0100:
> A couple more:
> 
> You could also look at:
> 
> http://www.magicnet.net/~gcash/tree.html
> or
> http://www2.zyvex.com/OpenChem/index.htm
or
  URL:http://www.handshake.de/~dieter/pyprojects/treewidget.html


Dieter


From kpmurphy at my-deja.com  Thu Jun 29 16:14:24 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Thu, 29 Jun 2000 20:14:24 GMT
Subject: writing in gif format
Message-ID: <8jgamf$rli$1@nnrp1.deja.com>

does anyone know how to write in gif format.  I am using the PhotoImage
class, not the Python Imaging Library.  Python tells me "TclError: image
file format "gif" has no file writing capability"

Thanks,

-->keith


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


From bzimmer at ziclix.com  Tue Jun 27 11:40:21 2000
From: bzimmer at ziclix.com (brian zimmer)
Date: Tue, 27 Jun 2000 15:40:21 GMT
Subject: [ANNOUNCE] zxJDBC 1.2.1 - Database Connectivity in JPython
Message-ID: <8jaht0$esp$1@nnrp1.deja.com>

The zxJDBC package provides a nearly 100% Python DB API 2.0 compliant
interface for database connectivity in JPython. It is implemented
entirely in Java and makes use of the JDBC API. This means any database
capable of being accessed through JDBC, either directly or using the
ODBC-JDBC bridge, can be manipulated using zxJDBC.

In addition to the standard API, it adds the same extension methods as
mxODBC to provide seamless database access regardless of implementation
language.

You can get the latest information at:

http://www.ziclix.com/zxjdbc

thanks,

brian


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


From jeremylynch at my-deja.com  Tue Jun 27 08:02:11 2000
From: jeremylynch at my-deja.com (jeremylynch at my-deja.com)
Date: Tue, 27 Jun 2000 12:02:11 GMT
Subject: learning-python question
References: <8j8ep3$tkn$1@nnrp1.deja.com> <20000626170753.C2492@activestate.com> <20000626194537.A3557@activestate.com>
Message-ID: <8ja53r$4gr$1@nnrp1.deja.com>

Thanks for all the help, I think i've got the hang of it now


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


From michael at stroeder.com  Mon Jun 19 14:34:00 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 19 Jun 2000 20:34:00 +0200
Subject: Python vs Perl
References: <8ikai5$v3d$1@nnrp1.deja.com> <394ebe95.255648282@news.online.no> <slrn8krpbf.gc.andyl@azaal.dircon.co.uk>
Message-ID: <394E6798.44A819B0@stroeder.com>

Andy Leighton wrote:
> 
> On Mon, 19 Jun 2000 06:41:39 GMT, Thomas Weholt <thomas at cintra.no> wrote:
>
> >* Perl is harder to read than Python. This makes debugging and re-use
> >of other peoples code harder.
> 
> Of course if you only know Perl, I would think Perl would be the easiest
> to read.  Although it would be fair to say that Python is more transparent
> than Perl.

I learned Python for CGI-BIN programming because it was more
readable to me. Although every document / book about web programming
had Perl examples. I knew neither Python nor Perl before.

Ciao, Michael.


From jeremy at beopen.com  Thu Jun  1 00:03:55 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 01 Jun 2000 00:03:55 -0400
Subject: Scoping: Is Python and does it matter?
References: <392C3782.3675@europa.com> <392d1d14$0$23937@wodc7nh0.news.uu.net> <39306b73$1_1@news.nwlink.com>
Message-ID: <16DA8EBEBD6C00D1.D306BA21A3057D9E.A35767208E34AFEB@lp.airnews.net>

nanotech at pacifier.com (noone) writes:

> What is Python's philosophy? Guido has a whitepaper on Style; has
> he or someone done one on Scoping?
> 
> Here is another example that concerns my coworkers (though I have not had
> this issue, do not see the issue, and would not code this way--what can
> you do?!!):
> 
> Python 1.5.1 (#37, Apr 27 1998, 13:36:17)  [CW CFM68K w/GUSI w/MSL]
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> x=1
> >>> def print_X():
> ...   print x
> ... 
> >>> print_X()
> 1
> >>> def print_X():
> ...   print x
> ...   x=2
> ... 
> >>> print_X()
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 2, in print_X
> NameError: x
> >>> 
> 
> In Perl, the second 'print_X' would succeed because 'x' would be found in
> the outer scope. Python scopes the var to the function due to the
> assignment (it seems). One could use 'global', but then the assignment
> would affect the module var (where in Perl, the assignment might be
> my($x)=2, in which case the "global" x would not be modified).
> 
> Again, I am looking for Python's philosophy (reasoning, excuses,
> whatever! :) so that my coworkers will switch and I will not have to look
> and ugly/bad code because it was fast hacked!
> 

There's a joke that Barry is fond of:  A patient comes in to see a
doctor.  He says, "Doc, it hurts when I move my arm this way."  The
doctor says, "Then don't more your arm that way."

It's hard to give you good advice about the example code you present,
because it's not at all clear what it is supposed to do.  In general,
it is confusing for the same variable name to be re-used for different
purposes within the body of a single function.  It is also confusing
to use the same name for a global and a local when you want to access
both.  Python's rules make it harder for you to write that confusing
code.  Instead, you should come up with different names.

Jeremy


From smoriano at sicon.net  Thu Jun 29 05:07:12 2000
From: smoriano at sicon.net (sergio)
Date: Thu, 29 Jun 2000 11:07:12 +0200
Subject: NEWBIE QUESTION
Message-ID: <395B11C0.4C619958@sicon.net>

 Hi all:
    I`m working with oython a few days, and i want to now if i can work
directly with a .dll file, or i must run makepy with the .dll,  or i
must make a .pyd file, if it's like this, Can you tell me how can i make
a pyd file




From david_ullrich at my-deja.com  Thu Jun  8 13:19:31 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 08 Jun 2000 17:19:31 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news> <8hlrkn$5t2$1@nnrp1.deja.com> <8hlssa$r3$1@nntp9.atl.mindspring.net>
Message-ID: <8hokig$85g$1@nnrp1.deja.com>

In article <8hlssa$r3$1 at nntp9.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <8hlrkn$5t2$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com>
wrote:
> >
> >    Not if they never hear about it. I thought he went out of his
> >way to be polite, to make it clear that he wasn't saying Python
> >bad Ruby good. Sounds to me like something that might be of
> >interest to a lot of Python people.
> >    Maybe it doesn't matter to people who are somehow instantaneously
> >informed of everything, but the rest of us appreciate a little bit
> >of marginally on-topic information. (I first heard about Delphi when
> >I was lurking in a VB group trying to figure out whether it was more
> >awful or less awful than the BASIC I was using at the time...)
>
> Well, Ruby comes up often enough naturally.

     I didn't know that.

> I think the objection
here
> comes from the explicit advertisement combined with the repetition
(this
> isn't the first time).

     Or that - does put a different flavor on it.

     I do think that I first heard about Python on a Perl newsgroup.
(Just a random fact of no direct relevance...)

DU

>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "You could make Eskimos emigrate to the Sahara by vigorously arguing
--
> at hundreds of screens' length -- for the wonder, beauty, and utility
of
> snow."  -- PNH to rb in r.a.sf.f
>


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


From Norman_Shelley-RRDN60 at email.sps.mot.com  Thu Jun 29 12:26:19 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Thu, 29 Jun 2000 09:26:19 -0700
Subject: postgreSQL successes (Interbase?)
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com>
Message-ID: <395B78AB.69DDC4F4@email.sps.mot.com>

Have you looked at Interbase (newly made OpenSource)?  We used it years ago and
it was a very nice and easy to use DB. And because it was commercial I've got to
believe they worked very hard on speed as well as consistencty/safety.


Eric Lee Green wrote:

> robin.escalation at ACM.org wrote:
> > Does anyone have any success stories (or gotchas) related to
> > implementing a PostgreSQL database with Python that they'd like to
> > share? I am hoping to use it as a back-end to a b2b web site.
>
> Yes. Can't tell you exactly at the moment though (NDA).
>
> Basic plusses of PostGreSQL: transactions, cursors, stored procedures, joins
> with subselects, all mostly work as you'd expect of any SQL database. The same
> cannot be said to be true of most other embedded databases out there (e.g.
> RAIMA, C-Tree, MySQL, ...). Read performance is good, and the DBMS engine
> supports all the usual kinds of locking that you'd expect from an RDBMS.
>
> Minuses: The page length restriction on size of a record can be the pits, if
> you're not careful about your design. If you need BLOBs, you're out of luck
> (the Python module supposedly supports the PostGreSQL BLOB type, but it's not
> reliable). Insert speed is horrible -- inserting data takes forever and a
> half, though PostGreSQL 7.0 now has a bulk import facility that may make it
> faster (I haven't had a chance to benchmark it yet). And, finally, the
> documentation is cryptic, incomplete, and scattered all over the place... I
> had to reference two different manuals in order to figure out how to create
> (and destroy) a PL/SQL stored procedure, for example. Lack of explicit support
> for replication is a pain too, though it's no worse than any other embedded
> database (i.e., a 'modified' field and an explicit procedure can be used to do
> it).
>
> The biggest minus for the back-end of a b2b web site, though, would be the
> fact that it is bog-slow at opening a connection. PostGreSQL forks a new
> process for each connection, and the new process must do all sorts of things
> to establish authentication etc... PHP3's PostGres support gets around this by
> creating "persistent" connections that hang around with the Apache process,
> but I have no idea if Python's Apache module support is robust enough to
> handle this situation. Not to mention that Python's Apache module support is
> almost entirely undocumented... it is quite unclear, for example, how to run
> your former Python CGI scripts via the PyApache module.
>
> My basic conclusion is that PostGreSQL makes a fine alternative to current
> embedded databases. It has better functionality than the xBase types, and
> unless you need massive bulk inserts, the performance is excellent.  On the
> other hand, I personally would not use Python with PostGreSQL and Apache for
> implementing a b2b web site. Python with ZOPE/Zserver and PostGreSQL, however,
> might be reasonable (I have not investigated that yet).
>
> --
> Eric Lee Green                         eric at estinc.com
> Software Engineer                      Visit our Web page:
> Enhanced Software Technologies, Inc.   http://www.estinc.com/
> (602) 470-1115 voice                   (602) 470-1116 fax



From aaron.ginn at motorola.com  Fri Jun  2 13:34:48 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 02 Jun 2000 10:34:48 -0700
Subject: Sorting a list
Message-ID: <snk8g856mf.fsf@motorola.com>

I'm a python newbie, so please be kind! :)

I'm trying to sort a list.  Fairly simple, but I'm not getting
at all what I expect.  When I attempt to sort the list, the
return value is None, and I can't understand why at all.

Here's my code:

def get_files_in_directory( dir ):
    # Returns a list of files in a directory.

    print os.listdir(dir).sort()
    print os.listdir(dir)

    return os.listdir(dir).sort()

The directory 'dir' does exist.  From the docs, os.listdir() should
return a list of files in a directory, and sort() should sort
them, but here's the output from the two print statements:

None
['tsmc18_5lm.03.drc', 'tsmc18_5lm.04.drc', 'tsmc18_5lm.05.drc',
'tsmc18_5lm.06.drc', 'tsmc18_5lm.07.drc', 'tsmc18_5lm.08.drc',
'tsmc18_5lm.10.drc', 'tsmc18_5lm.01.drc', 'tsmc18_5lm.02.drc',
'tsmc18_5lm.09.drc']

Why is sort returning an empty list?


Thanks,
Aaron

-- 
Aaron J. Ginn

Opinions mine, not Motorola's


From neilh at scintilla.org  Mon Jun 12 06:33:10 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 12 Jun 2000 10:33:10 GMT
Subject: How to display UNICODE text in idle?
References: <8i255k$1r0j$1@news.cz.js.cn>
Message-ID: <G%215.2903$cQ.24621@news-server.bigpond.net.au>

> I want to write a python program in IDLE, but idle cannot display
> other language text. How to do it?

   It works for me with the version of IDLE which came with the alpha Python
1.6a2. Hope the characters survive the news handlers:

Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
IDLE 0.6 -- press F1 for help
>>> print "?a??"
?a??
>>>

   Neil




From jfr at genyinfo.com  Sun Jun 25 14:01:11 2000
From: jfr at genyinfo.com (Jean-Francois Richard)
Date: Sun, 25 Jun 2000 14:01:11 -0400
Subject: smtplib
References: <000c01bfdec7$30f84b60$bdcb45c6@timaratz>
Message-ID: <uOr55.1895$LX4.17861@weber.videotron.net>

You just have to append a "Subject: Your Subject" to your message body and
it should work.

"Peter Timaratz" <timaratz at lisco.com> wrote in message
news:000c01bfdec7$30f84b60$bdcb45c6 at timaratz...
> I'm using smtplib with a CGI script. I'm puzzled as to why this module
> doesn't support sending a subject for the e-mail.
>
> And the e-mail I receive doesn't show the to address in my e-mail client.
>
>




From shapr at uab.edu  Sat Jun 17 08:16:08 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sat, 17 Jun 2000 12:16:08 GMT
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>
Message-ID: <394B6C86.C3CD4BCF@uab.edu>

John Schmitt wrote:
> 
> Hi.
> 
> I'd love to take the diversion of writing Make using Python.  The things
> that are keeping me from doing this are, in order:

http://www.software-carpentry.com is/was sponsoring a cash reward for
the best make in python. Several proposals were submitted, you may get
some design ideas from them.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
 squirrels furnished by Paul E Black, the world is safe.


From bparsia at email.unc.edu  Tue Jun 13 18:00:50 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Tue, 13 Jun 2000 18:00:50 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au> <394484C0.A5E412D9@san.rr.com> <slrn8ka8kq.1r0.jajs@Pascual.es>
Message-ID: <642A954DD517D411B20C00508BCF23B001296979@mail.sauder.com>

Juanjo ?lvarez <jajs at NOSPAMretemail.es> wrote:

> Courageous dijo sobre Re: Python Productivity over C++,
> en fecha: Mon, 12 Jun 2000 06:33:05 GMT
> 
> |Python is a simply wonderful prototyping environment, IMO.
> |Now only if it had an optimizing compiler...
> 
> I think every Python fan dream with that... 
> 
> It's really so difficult?

1) Look at Vyper, a Python clone+extensions. I haven't heard much about
it in a while. It looked very promising to me. It intended to do a lot
with whole module/program analysis, espeically wrt to type inferencing.
For compiling to machine code executables, this seems like a very good
approach. Stalin (a scheme compiler) takes this path, plus giving up
some runtime dynamaticity (well, you kinda have to if you're going to
analyize the *whole* program) and it achieves quite amazing code.
Outpaced C for certain numeric code, where reasonably idiomatic C and
Scheme were used, and even when someone skanky C was thrown in, IIRC.

2) Common Lisp style compiler. Big grab bag of tricks there, including
optional type declarations. Probably not quite the right match for
Python.

3) Smalltalk/Self style VMs. Improving method/function dispatch alone
can do a hell of a lot, as simple hand optimization shows (e.g., binding
a function to a local variable in an inner loop). Self style is more
ambitious, and probably not worth the implementation effort.

My understanding is the CPython VM is designed for simplicity,
portability, embedability, and easy of using C for extendability. Not
speed. My guess is that one of these has to be reduced for speed to go
up. Acutally, more than one, since simplicity will always be compromised
as you at optimization, and if you add trying to preserve all the
others, you get more complicated quickly.

Vyper shows, in a preliminary way (as does JPython), that there may be
serious gains made by moving to a different, higher level implementation
language. There are indeed some whispers of YA implementation...

...but if I told you about it, you'd tell me back, and then you'd have
to kill me, and my life insurance isn't paid up.

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.



From nobody at nowhere.nohow  Sat Jun 17 12:02:09 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 17 Jun 2000 16:02:09 GMT
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com>
Message-ID: <slrn8kn881.sdl.nobody@isis.visi.com>

In article <394AEFE7.6F3B718C at uniserve.com>, Bob van der Poel wrote:

>> I can't figure out how to create a Tk bitmap and then use it in
>> a button.  According to the docs I can find, something like this
>> ought to work:
>> 
>>     myBitmap = BitmapImage(name='myBitmap', data='[...]')
>> 
>>     Button(root,bitmap='myBitmap')
>> 
>> But, Tk complains that there is no bitmap named 'myBitmap'.  If
>> I leave out the "name=" when I create the bitmap and then use
>> the name retrieved with str(myBitmap), I get the same error.

>The data is an exact copy of the file. For example:
>
>	bobpic = BitmapImage (data = """
>	#define bob_width 107 
>        #define bob_height 154
>	static char bob_bits[] = {
>
>0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,  
>          0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff
>< blessedly snipped >
>	0xff,0xff,0xff}; """ )  

Yes, I know.  The call to BitmapImage() works fine, the error occurs when I
try to use the bitmap in a button like:

   Button(root, bitmap='queen')

This is where the error occurs -- it doesn't matter if I use the name I
specified in the call to BitmapImage or let Tk generate one and retrieve it
with str().  What I can't figure out is how to use bobpic as the bitmap for
a Button.

-- 
Grant Edwards                   grante             Yow!  An air of FRENCH
                                  at               FRIES permeates my
                               visi.com            nostrils!!


From d at d.d  Fri Jun 16 14:41:10 2000
From: d at d.d (Dave Pettypiece)
Date: Fri, 16 Jun 2000 14:41:10 -0400
Subject: newbie: dynamic code loading
Message-ID: <394A74C6.C95C5646@somanetworks.com>

hi everyone,

i am extremely new to the python world, so any suggestions would be
helpful.

what i would like to do is be able to dynamically load python classes
either remotely from a network connection or locally, it doesn't really
matter to me. the catch is that the program which will be loading these
classes will have no prior knowledge of what the classes it will load
are named, only that they implement a well known interface.

essentially my question boils down to: does anyone know a nice way to
implement a "plug-in" architecture in python?

as a side note, is it possible to get around the security restriction
placed on the rexec module? (eg i want to be able to use sockets)

any help is greatly appreciated!

dave



From db3l at fitlinxx.com  Tue Jun 27 19:55:31 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Jun 2000 19:55:31 -0400
Subject: compilation
References: <8j84cf$6bj$1@bmerhc5e.ca.nortel.com> <slrn8lf95c.fu7.thomas@centurion.xs4all.nl> <8jahlv$rui$4@bmerhc5e.ca.nortel.com>
Message-ID: <uya3q65n0.fsf@ctwd0143.fitlinxx.com>

msoulier at nortelnetworks.com (Michael P. Soulier) writes:

> I actually hate finding errors at runtime. It's my biggest complaint about
> Perl, in that I can make a function call that doesn't exist, and I won't find
> out about it until I hit that piece of code. I suppose python has the same
> flaw? Dynamic binding is fine, but a check for typos like that would be good. 
Ah, but the rub is how do you perform that check without actually executing
the script, since its the execution of the script that may actually yield
the availability of the function or variable :-)

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


From v2ng at my-deja.com  Tue Jun 27 15:01:28 2000
From: v2ng at my-deja.com (v2ng at my-deja.com)
Date: Tue, 27 Jun 2000 19:01:28 GMT
Subject: Excel Currency problems
Message-ID: <8jatls$p0i$1@nnrp1.deja.com>

I'm having problems reading currencies formatted fields in Excel from
Python through COM.

If I type in a value (say : 941221.6) and format it in Excel to :

Currency, 2 decimal places, symbol = $, Negative numbers = -$1,234.10

In Python, I'll use the win32com.client modules and do:

        return sheet.Cells(row, col).Value

where sheet is the worksheet I'm using.

Python will return the following value:

(2, 822281408)

If I reformat the cell to General, everything seems ok.

Any ideas?
Vic


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


From shapr at uab.edu  Thu Jun 22 10:58:38 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 22 Jun 2000 14:58:38 GMT
Subject: Python interface to Oracle? - yes.
References: <3951C636.2E985CE@ust.hk> <87og4t26zz.fsf@ns.f7.ph4.lan>
Message-ID: <39522A9C.57D500F4@uab.edu>

Andreas Rottmann wrote:
> 
> David YEUNG <dyeung at ust.hk> writes:
> 
> > Does anyone know if any python module has been developed for
> > interfacing with Oracle8i database on Linux?
> >
> That would be interesting for me, too. I'm likely to develop one, if
> it doesn't exist, since I have the infrastructure ready - see
> GQL.sourceforge.net.

http://www.zope.org/Products/DCOracle
Digital Creations appears to be the main source of experts on Python and
Oracle interaction.

The changelog for DCOracle mentions Oracle 8 specifically in several
places, therefore I would assume they have support for it.

It supports Python 1.4 or higher, and Oracle 7.3 and higher.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From m.faassen at vet.uu.nl  Thu Jun 15 10:52:51 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 14:52:51 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <8iagu3$ici$1@newshost.accu.uu.nl> <8iak0c$aau$1@slb3.atl.mindspring.net>
Message-ID: <8iaqk3$dru$1@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <8iagu3$ici$1 at newshost.accu.uu.nl>,
> Martijn Faassen <m.faassen at vet.uu.nl> wrote:
[snip]
>>I'm definitely not an old timer and it's frustrating to me too.
>>(Gordon McMillan makes comments like "Even Martijn knows it, and he's
>>young" :)

> If you recall the age-of-FORTRAN-use discussion, I'm on the youngish
> side for comp.lang.python, myself, though I expect I'm slowly creeping
> up on the average.  ;-)

Me too! My birthday was last friday! :)

> Say, whatever happened to Gerrit, anyway?

Don't know; haven't seen him around for a while. Last thing I saw of him
recently was a book review on a Python book on a Dutch site; don't know
when that got posted, but if it was anytime recent he's still around,
just quiet.

Regards,

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


From tolot at utolot.toppoint.de  Sun Jun  4 15:03:05 2000
From: tolot at utolot.toppoint.de (Marc Christiansen)
Date: Sun, 4 Jun 00 20:03:05 +0100
Subject: Um? *boggle* Silly question about listobject.c
References: <000e01bfcddf$d2e09d40$1d2d153f@tim>
Message-ID: <tolot.09oq@utolot.toppoint.de>

Tim <tim_one at email.msn.com> wrote:

> of type size_t, but that won't happen in time for 1.6.  In the meantime, if
> you really need lists with more than 2 billion elements, your hardware
> budget is so large you can afford to pay someone to patch you up a custom
> Python <wink>.

What makes you sure, that sizeof(int)==4?

-- spam.c --
#include <limits.h>
#include <stdio.h>

int main(void)
{
  printf("%d %d %d\n",sizeof(short),sizeof(int),INT_MAX);
  return 0;
}
-- EOF --

(0;0)13.SRam:> spam
2 2 32767

(Ok, I admit using a switch to get short==int ;-)

There-are-sizeof(int)!=4-systems-ly y'rs
 Marc


From DOUGS at oceanic.com  Fri Jun  9 15:21:02 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Fri, 9 Jun 2000 09:21:02 -1000 
Subject: GUI editor
Message-ID: <8457258D741DD411BD3D0050DA62365907A1F9@huina.oceanic.com>

[Blu Dragoon wrote:]
> > Is there a gui editor written in python for tkinter?

[and Alan Gauld replied:] 
> SpecTCL with its Python bolt-on SpecPy produces 
> reasonable Python Tkinter code. Personally I find 
> Tkinter easy enough to use that a GUI designer is
> hardly necessary but some people like 'em and 
> SpecTCL works with Python/Tkinter... with Perl/Tk, 
> Java/AWT, HTML and of course Tcl/Tk too.
> 
> Find it on www.scriptics.com or do a search via 
> your favouraite engine.

Apparently Scriptics is having an identity crisis.  They are now Ajuba
Solutions, but the SpecTCL stuff is still at
http://dev.scriptics.com/software/spectcl/

Trid to find SpecPy but couldn't find references to it at the above site,
the Vaults, or www.python.org

Got any leads? It isn't bundled with the SpecTCL installer I grabbed.

-Doug-



From jkraska1 at san.rr.com  Wed Jun 21 20:21:38 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 22 Jun 2000 00:21:38 GMT
Subject: Python complements C++ for productivity
References: <NEBBIOGBDMDPEANLFKGHKEGJCEAA.jschmitt@vmlabs.com> <395128A4.3D70A4E4@san.rr.com> <PFc45.211883$MB.3849912@news6.giganews.com>
Message-ID: <39515C40.DEA9FAED@san.rr.com>

> >I haven't seen any of these phrase books, but Python is just
> >soooo easy. There's really not much to it.
> 
> The appalling thing is that I can't really see all that much that you
> gain for the added complexity.

You don't really.

In the early days of the STL, when the compilers wouldn't support
debugging through it, I refused to use it categorically. I used a
set of void* containers instead. They always worked like a charm
and did everything I wanted them to. Like python, it was almost
always hash/vector. I seldom had an application for anything else.

These days, having ventured into the simulation business, I do
make use of deques and priority queues, but still. That's not
a lot of containers.




C/


From redaelli at inc.it  Sun Jun  4 06:11:57 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sun, 04 Jun 2000 12:11:57 +0200
Subject: Found the problem! Was Re: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home>
		<39397FD4.67C7FD73@inc.it> <867lc64a47.fsf@g.local>
Message-ID: <393A2B6D.4B33A5C8@inc.it>

I found the problem. It's related to lists; it seems to me that python
does not allow smoothly things like:
numbers = [n,m] =  map(int, numbers)

Gareth McCaughan wrote:
> I tried running this. It didn't work, but not for reasons
[...]
> there were no variables called xx,yy. I replaced them with x and y.
oops.. I tried catching the exceptions in that version. It wasn't the
one that fails on float.

> The output you quoted doesn't look complete. It has a
> traceback but doesn't say what kind of exception was
> raised. Could you post the whole message?

> I wonder whether your file might contain some funny
> characters (though I'd have thought they'd show up if
> so). Try adding, after that diagnostic "print", another
> line that says
>     print map(ord, x)
> and see if there are any values in the list it prints
> outside the range 32..126. Or change the print line
> to
>   print "x(%s)=%s y(%s)=%s" % (type(x),`x`,type(y),`y`)
> for more readable output that should give the same
> information: look for unexpected escapes in the output.
Thanks for the tip!!!!

> Does the following program produce a similar error for you?
> 
> for line in open("LCASE.001").readlines():
>   for item in string.split(line):
>     print "About to try parsing:", item
>     x = float(item)
>     print "Result is:", x
> 
No. it runs smoothly. 

> Is the LCASE.001 file you posted the only LCASE* file
> in the relevant directory, or might it actually have been
> during the parsing of a *different* file that it fell over?
LCASE.001 is the only file parsed.

> Something else I happened to spot: you have "lenght"
> instead of "length" in a few places. Shouldn't make
> any difference to the running of the program, though.
...well... that's becuase I'm not a native english speaker and often I
misspell words... 8-)

-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/
-------------- next part --------------
 1
 4 3
      1  0.000000E+00  0.000000E+00
      2  0.000000E+00  0.300000E+03
      3  0.500000E+03  0.300000E+03
      4  0.500000E+03  0.000000E+00
      1     1     2
      2     2     3
      3     3     4
      1  0.300000E+03  0.000000E+00  0.100000E+01
  0.000000E+00 0.000000E+00 0.000000E+00 0.250000E+04 -.146922E+03 0.652654E+04
  0.150000E+02 0.478073E-01 -.395235E-04 0.250000E+04 -.146922E+03 0.668798E+04
  0.300000E+02 0.956146E-01 -.174419E-03 0.250000E+04 -.146922E+03 0.720229E+04
  0.450000E+02 0.143422E+00 -.429172E-03 0.250000E+04 -.146922E+03 0.806948E+04
  0.600000E+02 0.191229E+00 -.828271E-03 0.250000E+04 -.146922E+03 0.928954E+04
  0.750000E+02 0.239036E+00 -.139620E-02 0.250000E+04 -.146922E+03 0.108625E+05
  0.900000E+02 0.286844E+00 -.215745E-02 0.250000E+04 -.146922E+03 0.127883E+05
  0.105000E+03 0.334651E+00 -.313651E-02 0.250000E+04 -.146922E+03 0.150670E+05
  0.120000E+03 0.382458E+00 -.435786E-02 0.250000E+04 -.146922E+03 0.176985E+05
  0.135000E+03 0.430266E+00 -.584599E-02 0.250000E+04 -.146922E+03 0.206830E+05
  0.150000E+03 0.478073E+00 -.762539E-02 0.250000E+04 -.146922E+03 0.240203E+05
  0.165000E+03 0.525880E+00 -.972055E-02 0.250000E+04 -.146922E+03 0.277105E+05
  0.180000E+03 0.573687E+00 -.121559E-01 0.250000E+04 -.146922E+03 0.317535E+05
  0.195000E+03 0.621495E+00 -.149561E-01 0.250000E+04 -.146922E+03 0.361495E+05
  0.210000E+03 0.669302E+00 -.181454E-01 0.250000E+04 -.146922E+03 0.408983E+05
  0.225000E+03 0.717109E+00 -.217484E-01 0.250000E+04 -.146922E+03 0.460000E+05
  0.240000E+03 0.764916E+00 -.257897E-01 0.250000E+04 -.146922E+03 0.514546E+05
  0.255000E+03 0.812724E+00 -.302936E-01 0.250000E+04 -.146922E+03 0.572620E+05
  0.270000E+03 0.860531E+00 -.352846E-01 0.250000E+04 -.146922E+03 0.634223E+05
  0.285000E+03 0.908338E+00 -.407874E-01 0.250000E+04 -.146922E+03 0.699355E+05
  0.300000E+03 0.956146E+00 -.468262E-01 0.250000E+04 -.146922E+03 0.768016E+05
      2  0.500000E+03  0.100000E+01  0.000000E+00
  0.000000E+00 -.956146E+00 0.105359E+06 0.000000E+00 -.810241E-12 0.118531E+15
  0.250000E+02 -.956146E+00 0.354593E+09 0.000000E+00 -.250000E+03 0.118531E+15
  0.500000E+02 -.956146E+00 0.553234E+09 0.000000E+00 -.500000E+03 0.118531E+15
  0.750000E+02 -.956146E+00 0.615784E+09 0.000000E+00 -.750000E+03 0.118531E+15
  0.100000E+03 -.956146E+00 0.561998E+09 0.000000E+00 -.100000E+04 0.118531E+15
  0.125000E+03 -.956146E+00 0.411631E+09 0.000000E+00 -.125000E+04 0.118531E+15
  0.150000E+03 -.956146E+00 0.184438E+09 0.000000E+00 -.150000E+04 0.118531E+15
  0.175000E+03 -.956146E+00 -.998256E+08 0.000000E+00 -.175000E+04 0.118531E+15
  0.200000E+03 -.956146E+00 -.421405E+09 0.000000E+00 -.200000E+04 0.118531E+15
  0.225000E+03 -.956146E+00 -.760544E+09 0.000000E+00 -.225000E+04 0.118531E+15
  0.250000E+03 -.956146E+00 -.109749E+10 0.000000E+00 -.250000E+04 0.118531E+15
  0.275000E+03 -.956146E+00 -.141248E+10 0.000000E+00 -.275000E+04 0.118531E+15
  0.300000E+03 -.956146E+00 -.168578E+10 0.000000E+00 -.300000E+04 0.118531E+15
  0.325000E+03 -.956146E+00 -.189761E+10 0.000000E+00 -.325000E+04 0.118531E+15
  0.350000E+03 -.956146E+00 -.202822E+10 0.000000E+00 -.350000E+04 0.118531E+15
  0.375000E+03 -.956146E+00 -.205787E+10 0.000000E+00 -.375000E+04 0.118531E+15
  0.400000E+03 -.956146E+00 -.196678E+10 0.000000E+00 -.400000E+04 0.118531E+15
  0.425000E+03 -.956146E+00 -.173522E+10 0.000000E+00 -.425000E+04 0.118531E+15
  0.450000E+03 -.956146E+00 -.134343E+10 0.000000E+00 -.450000E+04 0.118531E+15
  0.475000E+03 -.956146E+00 -.771639E+09 0.000000E+00 -.475000E+04 0.118531E+15
  0.500000E+03 -.956146E+00 -.105359E+06 0.000000E+00 -.500000E+04 0.118531E+15
      3  0.300000E+03  0.000000E+00  -.100000E+01
  0.000000E+00 0.000000E+00 -.129038E+07 0.000000E+00 0.000000E+00 0.000000E+00
  0.150000E+02 0.000000E+00 -.156394E+10 0.000000E+00 0.000000E+00 0.161438E+03
  0.300000E+02 0.000000E+00 -.243946E+10 0.000000E+00 0.000000E+00 0.675750E+03
  0.450000E+02 0.000000E+00 -.271495E+10 0.000000E+00 0.000000E+00 0.154294E+04
  0.600000E+02 0.000000E+00 -.247753E+10 0.000000E+00 0.000000E+00 0.276300E+04
  0.750000E+02 0.000000E+00 -.181429E+10 0.000000E+00 0.000000E+00 0.433594E+04
  0.900000E+02 0.000000E+00 -.812342E+09 0.000000E+00 0.000000E+00 0.626175E+04
  0.105000E+03 0.000000E+00 0.441215E+09 0.000000E+00 0.000000E+00 0.854044E+04
  0.120000E+03 0.000000E+00 0.185928E+10 0.000000E+00 0.000000E+00 0.111720E+05
  0.135000E+03 0.000000E+00 0.335473E+10 0.000000E+00 0.000000E+00 0.141564E+05
  0.150000E+03 0.000000E+00 0.484049E+10 0.000000E+00 0.000000E+00 0.174938E+05
  0.165000E+03 0.000000E+00 0.622943E+10 0.000000E+00 0.000000E+00 0.211839E+05
  0.180000E+03 0.000000E+00 0.743446E+10 0.000000E+00 0.000000E+00 0.252270E+05
  0.195000E+03 0.000000E+00 0.836847E+10 0.000000E+00 0.000000E+00 0.296229E+05
  0.210000E+03 0.000000E+00 0.894436E+10 0.000000E+00 0.000000E+00 0.343718E+05
  0.225000E+03 0.000000E+00 0.907502E+10 0.000000E+00 0.000000E+00 0.394735E+05
  0.240000E+03 0.000000E+00 0.867335E+10 0.000000E+00 0.000000E+00 0.449280E+05
  0.255000E+03 0.000000E+00 0.765225E+10 0.000000E+00 0.000000E+00 0.507355E+05
  0.270000E+03 0.000000E+00 0.592461E+10 0.000000E+00 0.000000E+00 0.568958E+05
  0.285000E+03 0.000000E+00 0.340332E+10 0.000000E+00 0.000000E+00 0.634090E+05
  0.300000E+03 0.000000E+00 0.129038E+07 0.000000E+00 0.000000E+00 0.702750E+05
-------------- next part --------------
#!/usr/bin/env python

import glob # to look for all LCASE.* files
from string import * ## to process informations
from math import * ## we need almost everything from math....

class Telaio:
    def __init__(self):
        ## Loading all the load cases 
        ## Looking for all LCASE.xxx files
        self.filelist = glob.glob("LCASE*")
        self.load_cases = [None] * len(self.filelist) 
        
        for filename in self.filelist:
            print "Processing %s" % [filename]
            file = open(filename)
            case_n = int( split(filename,".")[1] )
            print "Case",case_n
            ## Reading the trailing 1
            file.readline()
            ## read nodes and rods numbers
            number_list = split( file.readline() )
            numbers = map(int, number_list)
            nodes_n = numbers[0]
            rods_n  = numbers[1]
            print numbers,"numbers type", map(type,numbers)
            ##print  "nodes: %i, rods: %i" % [numbers]
            for node_i in range(0,nodes_n):
                [node, x, y] = split( file.readline() )
                node = int(node);
                print "Trying to parsa '%s' '%s' '%s'" % (node,x,y)
                #print "Looking for strange characters:",map(ord,x)
                x = float(x)
                y = float(y) 
                print "Parsed to '%s' '%s' '%s'" % (node,x,y)
                ## Note: node_i is NOT used!
        ## End reading files.
        
if __name__ == '__main__':
    t = Telaio()



From olipt at mayo.edu  Sat Jun 17 10:48:41 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Sat, 17 Jun 2000 09:48:41 -0500
Subject: Python vs. Matlab
In-Reply-To: <724s6sa0k5.fsf@rogn.ii.uib.no>
References: <724s6sa0k5.fsf@rogn.ii.uib.no>
Message-ID: <Pine.LNX.4.10.10006170944090.6103-100000@us2.mayo.edu>

> 
> I have become almost addicted to Matlab after using it extensively for
> some years, however, the licencing feels very strict after getting
> used to free and very good quality software like Emacs, Linux, etc.

I was in this same position a couple of years ago.

> 
> My main use of Matlab is reading in large binary files into
> arrays(output from simulations), do some processing with various
> matrix routines, then extract sections of the 3D arrays for plotting.
> 
> Spotted Python some months ago, and liked it!  Can Python, Numpy and
> the many plotting extensions be used as a replacement for Matlab, is
> Python slower/faster etc.?  
> 


Python can be used for all the things you describe.  I have been using
Python exclusively for about two years and my only wish is that there were
more toolboxes available.  For the operations you describe, there is
already more than enough capability.

It is easy to add your own C or Fortran extensions as well so you can fill
in the holes in the toolbox.

It can be tough to get started with NumPy since the commands and ideas are
a little different than Matlab.  But, once you are over the small learning
hump, you can never go back to Matlab's restrictive environment.

Incidentally, I've found that Python is faster than Matlab.  Matlab has
wonderful speed for raw matrix operations, but if you have any program
flow, Python seems faster.

Good luck,

-Travis




From stephen at cerebralmaelstrom.com  Sun Jun 18 13:01:00 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sun, 18 Jun 2000 17:01:00 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com> <39520785.690843718@nntp.interaccess.com>
Message-ID: <gf735.63946$WS3.647412@typhoon.we.rr.com>

    I *completely* don't get this section of your argument. What about
whitespace makes Python more 'difficult' to extend?
Right-click Cut, Right-Click Paste, Right-Click->Source Code->Indent Region.
That's no more difficult then having to arrange your curley braces.

    Incidentely, I don't think i've ever had more then a handful (ever!
Like, five?) errors pop-up due to indendation. And this is being generous to
your side. It is easy to simply LOOK at a Python Source and see, usually,
where any such errors pop up. I don't generally make a habit of sitting on
my keyboard -- do you? -- so don't have the problem of characters
mysteriousely just popping into my editor.

--S

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
news:39520785.690843718 at nntp.interaccess.com...
> On Sat, 17 Jun 2000 08:12:03 -0500, "Bob Hays, Computer Geek"
(snip)
> >>3) Finally. Python is a language where it is almost easy to write
> >>extensible programs. To illistrate what I mean consider this example
> >>of a python program I recently wrote:
> >
> >Here's were we will agree to disagree, I think.  I think its way
> >faster and easier to extend Python stuff than C++ (and Java) due to
> >the lack of compile/link cycle and due to inheritance inside of
> >Python.  Yep, you have to be careful how you indent, but if you have a
> >good editor it will cut/copy/paste as-is without reindenting.  I see
> >more problems due to bad cut/copy/paste reuse in all programming
> >languages than I do from mis-indenting so far.
> >
> I think you misunderstood what I said. I think that Python is not now
> easily extensible do to the  problems with whitespaces. If that were
> fixed then Python would be easily extensible.
(snip)




From famvandenberg at home.com  Mon Jun 12 13:10:04 2000
From: famvandenberg at home.com (bo Vandenberg)
Date: Mon, 12 Jun 2000 17:10:04 GMT
Subject: Jpython
References: <3944F806.36BE6A29@student.liu.se>
Message-ID: <3945194E.A4EB024D@home.com>

Good luck with your project.

You might get more useful help if you send a detailed idea of what you want
to do.

If you are in the formative stages you may find browsing the
Python/Jpython tutorials most useful as they give you considered early
steps.

I don't know your background but some links to start with are:

The first stop for Python (IMHO)
http://www.pythonlabs.com/

JPython
http://sourceforge.net/project/?group_id=5473

General Python Portal -- with easy links to important pages
phttp://www.oreillynet.com/pub/a/python/2000/04/14/python-intro.html


SOURCE MODULES ETC...
http://www.vex.net/parnassus/apyllo.py?a=l
http://tor.dhs.org/~zephyrfalcon/snippets/index.html



I hope these help in your search. If you meet with some success share your
success as well as you questions :)

bo



From neurocrat at one.net.au  Mon Jun 12 00:37:33 2000
From: neurocrat at one.net.au (Neurocrat)
Date: 12 Jun 2000 14:37:33 +1000
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155>
Message-ID: <85snuj346q.fsf@one.net.au>

gmcm at hypernet.com (Gordon McMillan) writes:

[snip]

> One point worth reiterating is that where the app (in a C++ design) calls 
> for lots of related classes, you will probably only have a handful in 
> Python. Of course, this means discarding classical "OOP" thinking, and 
> desinging more in the "component" style. But this can yield an order of 
> magnitude all by itself.

If it's worth reiterating, would you mind elaborating on this
somewhat?

I've been using C++ and Eiffel for a few years, but I've recently
started experimenting with Lisp and Python. (Lisp has been the biggest
eye-opener of all, but that's another story). In Python, I've found an
immediate productivity increase, but I'm still too new to the language
to know exactly why.

My first impression is that its dynamic typing, uncluttered syntax and
ample built-in types encourages more "natural" designs. It makes
coding easier and more fun (which is important to me). Even as a
newcomer, I feel that I'm using the language with more pleasure and
confidence rather than carefully paying attention to all its
idiosyncracies (C++) or forcing my design to fit the paradigm
("classical OO" / Eiffel). I find that Python translates thought into
code without too much fuss, and virtually no linguistic cruft. But is
there more to it than that?

What is it about Python that encourages the "component" style you
described?


From dgoodger at bigfoot.com  Sat Jun  3 22:08:22 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Sat, 03 Jun 2000 22:08:22 -0400
Subject: exit(); halt; in pyhton
In-Reply-To: <3939B4C1.3FACC11D@tin.it>
References: <3939B4C1.3FACC11D@tin.it>
Message-ID: <B55F3255.458E%dgoodger@bigfoot.com>

on 2000-06-03 21:45, Jilani Khaldi (jilanik at tin.it) wrote:
> How can I write this Pascal code in python?
> var
> x: double;
> begin
> if x = 0 then halt
> else
> do others things
> end;
> 
> Freeing everything.

Does this do the job?

    if x == 0 then:
        import sys  # only if it's not imported by the module itself
        sys.exit()

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From fw at deneb.cygnus.argh.org  Fri Jun  2 16:51:50 2000
From: fw at deneb.cygnus.argh.org (Florian Weimer)
Date: 02 Jun 2000 22:51:50 +0200
Subject: mixed solution: unicase (unique allowed case) (was: Re: Case-sensitivity: )
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <20000522183552.A29224@mudge> <8gcoij$60n$1@news-int.gatech.edu> <8h92ko$5e1@news.Informatik.Uni-Oldenburg.DE>
Message-ID: <87n1l3u7q1.fsf@deneb.cygnus.argh.org>

Johannes Nix <Johannes.Nix at mail.uni-oldenburg.de> writes:

> why don't turn it this way to get advantages of both alternatives:
> 
> - the language is case sensitive.
> 
> - an assignment to the same name with different case in the same
> module is a Case Error exception.

This idea is not new.  GNAT (GNU Ada Translator), for example,
implements it.  Although Ada is case-insensitive, there's a compiler
switch which enforces casing which matches the style used in the
declaration of an entity.  Code which compiles even though this switch
is turned on certainly looks cleaner.

> But you would _not_ allow any programmer to distinguish names only by
> case (which in my eyes is not the best style, and most probably a
> mistake which leads to a name error anyway). A small percentage of
> code would have to be fixed up. (And this could be done
> automatically).

There's the convention to use "Foo" for classes and "foo" for
instances.  I don't think people want to give up that.


From dillera at isc.upenn.edu  Mon Jun 26 20:59:34 2000
From: dillera at isc.upenn.edu (Andrew Diller)
Date: Mon, 26 Jun 2000 20:59:34 -0400
Subject: Python for Palm ?
References: <8j56ir$7fr$1@ctb-nnrp2.saix.net> <8j5n9o$grt$1@netnews.upenn.edu> <8j6nne$4bv$1@slb6.atl.mindspring.net>
Message-ID: <8j933o$eps$1@netnews.upenn.edu>

"Corvus" <Software at TheMysticEye.com> wrote in message
news:8j6nne$4bv$1 at slb6.atl.mindspring.net...
> It's available for WindowsCE and it works pretty good...
>

yeah, but when your done, its still just windows ce.




From m.faassen at vet.uu.nl  Wed Jun 14 07:41:48 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 14 Jun 2000 11:41:48 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de>
Message-ID: <8i7r1s$l96$1@newshost.accu.uu.nl>

Thomas Thiele <thiele at muc.das-werk.de> wrote:

> How should a REAL python program looks like.
> I design classes like in C++, thats true. But I also use typical python features
> like lists, dicts, map etc. I'm not trying to write everything by myself.

> How have I to think as a pythonprogrammer?

There are differences in approach and philosophy between Python and C++ (or 
Java or one of the other statically typed languages):

Some of my observations on Python 'style':

  * Pythoneers care less about encapsulation. That is, they like 
    encapsulation but they don't go out of their way doing getters/setters
    and the like; the idea is to encapsulate later, when it becomes
    necessary. Python allows this as prototyping can go really quickly.
    The dynamic typing aspect also allows a lot of easy loose-coupling 
    straight away.

  * Pythoneers care less about inheritance hierarchies. When a class
    supports an interface it supports an interface; inheritance often
    isn't needed. 

  * Pythoneers care less about static type safety (obviously). They don't worry
    a lot inside any method or function what type the passed
    arguments are (they may sometimes query if it supports an interface,
    though). They'll worry about that when passing the arguments
    in.

  * Pythoneers care more about interfaces, but worry less about them.
    Lots of interfaces are implicit, but that's fine, as Python's
    dynamicity allows them to make the code do the right thing 
    anyway. Interfaces aren't really types, so there's less worry
    about breaking interfaces.

  * Pythoneers care more about rapid prototyping. The language allows it,
    so they often write one to throw it away. Or at least overhaul it
    drastically. Or not, as it may turn out.

  * Pythoneers care more about (unit) tests. Again, tests are easier and quicker
    to write because the language allows rapid prototyping.

  * Pythoneers worry less about reuse. While reuse with Python is very
    easy to accomplish, they don't have to *worry* about it; it just
    happens when it does. When it doesn't, it's because it's too easy
    to whip something up quickly yourself.

  * Pythoneers use Python's introspection abilities such as hasattr() and
    isinstance(). Not all the time, not even frequently, but often enough
    it's very convenient glue to make things work the way they want, 
    without having to complicate (or even touch) everything else. Use it
    when the design would've painted you in a corner in a statically
    typed language.

Of course, any of these may be wrong. :)

Regards,

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


From cut_me_out at hotmail.com  Mon Jun 19 18:08:52 2000
From: cut_me_out at hotmail.com (Alex)
Date: 19 Jun 2000 18:08:52 -0400
Subject: How does __setitem__() work?
References: <slrn8kt5iu.2cn.bob@slim.instantappointment.com>
Message-ID: <etdaegh5nnv.fsf@w20-575-109.mit.edu>

If you do this:

class Test:
    
    def __setattr__(self, attribute, value):
        print 'setting', attribute, 'to', value
        self.__dict__[attribute] = value
    
    def __setitem__(self, index, value):
        print 'setting index', index, 'to', value
        ## THE_BLANK

t = Test(); t.list = [1,2,3]
t[0] = 7

The result is 

setting list to [1, 2, 3]
setting index 0 to 7

You have to assign to an 'item' of the instance itself for the
__setitem__ method to be called.  Assigning to 'items' of members won't
involve the instance itself.

Alex.


From mhammond at skippinet.com.au  Fri Jun 16 23:05:43 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 17 Jun 2000 03:05:43 GMT
Subject: Pythonwin: editor nits
References: <39439B04.76F15F39@columbus.rr.com>
Message-ID: <bWB25.700$qz4.6064@news-server.bigpond.net.au>

Check out the documentation in the Pythonwin help file about the tabs (in
the overviews section).  If you can't find it, then update your build!

Basically, "Smart Tabs" is designed to work the way you are seeing it
work - it uses the first indent in the source file to set the options for
the file.  If you want your options to be in effect _always_, turn
smart-tabs off.

This is identical to the way IDLE works - we share the same auto-indent
code (courtesy of the timbot)

Mark.

"Dan Underwood" <danunder at columbus.rr.com> wrote in message
news:39439B04.76F15F39 at columbus.rr.com...
> I'm using build 129 on Win98SE. I find that the editor does not
> (usually) remember my tab and indent settings. Nor does it remember that
> I set "Use Tabs" on and "Show Whitespace" on. Actually, the options
> dialog shows these settings as remembered but the editor seems to ignore
> them for existing files. I do not want to deal with questions about why
> I change tab and indent ;-) but I would like to know if I am just not
> setting something right or are these just some bugs to be eventually
> fixed. If a later build has already fixed them, "never mind".
> Thanks.
> Dan.




From pkop at brb.midat.de  Wed Jun 21 07:12:13 2000
From: pkop at brb.midat.de (Peter Koppatz)
Date: Wed, 21 Jun 2000 13:12:13 +0200
Subject: 123.4+2=125.40000000000001
Message-ID: <3950A30D.DEE4CF4D@brb.midat.de>

>>> 123.4+2
125.40000000000001   # is this OK?
>>> print 123.4+2
125.4
>>>

On my calculator: 123.4+2 = 125.4 !?
I am using Python 1.6a2

Peter




From matz at netlab.co.jp  Sun Jun 11 23:44:00 2000
From: matz at netlab.co.jp (Yukihiro Matsumoto)
Date: 12 Jun 2000 12:44:00 +0900
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available  now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news>
 <8hlrkn$5t2$1@nnrp1.deja.com>
 <1ebvwau.af3wl21301xdcN%bparsia@email.unc.edu>
 <87em67fpp6.fsf@ev.netlab.co.jp> <394123E9.F3293D77@roguewave.com>
Message-ID: <87ya4bsgvz.fsf@ev.netlab.co.jp>

Hi,

Bjorn Pettersen <bjorn at roguewave.com> writes:

|The reason I haven't done any programming in it yet, however, is that
|the English language documentation is not yet good enough.  I.e. I get a
|feeling that it is not complete (vs. the Japanese docs), and I'm afraid
|that if I got stuck, gramatical issues would prevent me from solving my
|own problems...  I'll contine to come back to it though.

The book will be out from Addison Wesley in autumn, if things go as
planned.  It will change the situation.

							matz.


From pmueller at decrc.abb.de  Tue Jun  6 10:57:25 2000
From: pmueller at decrc.abb.de (Peter Mueller)
Date: Tue, 06 Jun 2000 16:57:25 +0200
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <393D1155.D067A821@decrc.abb.de>

Hi,

it looks like that I have the same problem. I use the pickler. When I import
gtk/pygnome modules, the pickler can't read my pickled object (containing
floats) anymore.
This behaviour is new since redhat 6.2. On redhat 6.1 the same code with the
same pickled object works without any problem. The python versions on both
redhat 6.1/6.2 are the same. There must be a difference one of the modules.

Peter

Remco Gerlich wrote:

> Paolo Redaelli wrote in comp.lang.python:
> > Look at the this short program:
> >
> > =======
> > #!/usr/bin/env python
> >
> > s = "0.000000E+00"
> > print "Converting",s," into: ", float(s)
> > print "Importing gtk"
> >
> > import GDK
> > from gtk import *
> > from gnome.ui import *
> > import GdkImlib
> > print "Converting",s," into: ", float(s)
> > =======
> > Here's its output:
> > =======
> > [paolo at aldebaran paolo]$ python float.py
> > Converting 0.000000E+00  into:  0.0
> > Importing gtk
> > Converting 0.000000E+00  into:
> > Traceback (innermost last):
> >   File "float.py", line 11, in ?
> >     print "Converting",s," into: ", float(s)
> > ValueError: invalid literal for float(): 0.000000E+00
> > =======
> >
> > The only lesson I can extract from it is that GTK breaks float!
>
> It still looks to me like gtk or gnome.ui is overwriting float, because of
> your from module import * statements. What does "print repr(float)" say?
>
> (I don't have gtk, can't test this myself).
>
> --
> Remco Gerlich,  scarblac at pino.selwerd.nl
> This space intentionally left blank.



From MarkH at ActiveState.com  Fri Jun 30 01:04:41 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 30 Jun 2000 05:04:41 GMT
Subject: Newbie making noises (play xxxHz tone on Win32?)
References: <_RJ65.1649$7%3.135352@news.flash.net>
Message-ID: <JTV65.1836$Tb7.13207@news-server.bigpond.net.au>

The standard "winsound" module in the next version of Python has a Beep
function.

For Python 1.5, you can use the same function, but from the win32api
module; hence you will need the win32 extensions.

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

"Tim Hamza" <thamza.nospam at geocities.com> wrote in message
news:_RJ65.1649$7%3.135352 at news.flash.net...
> First Post
> Newbie
>
> I have a need to play a tone of arbitrary frequency on a Win9x system
> through the soundcard.
>
> Something like play_tone(frequency, duration)
>
> Also I would like to be able to play multiple tones at once.
>
> ie play_chord(freq1, freq2, ..., duration)
>
> I look around on the Vaults of Parnassus and through the standard
libraries,
> etc and found nothing like this.
>
> Anybody: Been there done that?  Ideas?
>
> Thanks,
>                 Tim
>
>




From phd at phd.russ.ru  Mon Jun 19 13:31:04 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 19 Jun 2000 17:31:04 +0000 (GMT)
Subject: Bookmarks database and Internet robot version 3.1.2
Message-ID: <Pine.LNX.4.21.0006191729180.22931-100000@fep132.fep.ru>

                    BOOKMARKS database and internet robot

WHAT IS IT
   Here is a set of classes, libraries, programs and plugins I use to
manipulate my bookmarks.html. I like Netscape Navigator, but I need more
features, so I write and maintain these programs for my needs. I need to
extend Navigator's "What's new" feature (Navigator 4 named it "Update
bookmarks").


WHAT'S NEW in version 3.0
   Complete rewrite from scratch. Created mechanism for pluggable storage
managers, writers and robots.


WHAT'S NEW in version 3.1.2
   Minor speed improvements in parser (just removed debug output if not in
debug mode).
   New program check_dups.py prints a list of duplcat URLs in the DB.


WHERE TO GET
   Master site: http://sun.med.ru/~phd/Software/Python/#bookmarks_db

   Faster mirrors: http://skyscraper.fortunecity.com/unix/797/Software/Python/#bookmarks_db
   http://members.xoom.com/_XMCM/phd2.1/Software/Python/index.html#bookmarks_db


AUTHOR
   Oleg Broytmann <phd2 at earthling.net>

COPYRIGHT
   Copyright (C) 1997-2000 PhiloSoft Design

LICENSE
   GPL

STATUS
   Storage managers: pickle, FLAD (Flat ASCII Database).
   Writers: HTML, text, FLAD.
   Robots (URL checker): simple, forking.

TODO
   More storage managers: shelve, SQL, MetaKit.
   More robots: threading, async.

Oleg.
---- 
   Oleg Broytmann  http://skyscraper.fortunecity.com/unix/797/  phd2 at email.com
           Programmers don't die, they just GOSUB without RETURN.




From emile at fenx.com  Thu Jun 15 08:46:16 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 15 Jun 2000 05:46:16 -0700
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com>
Message-ID: <11c701bfd6c7$b8ef7500$1906a8c0@fc.fenx.com>

I think I've got this working so that sub-classed
instances can be compared to subclassed references.

----- Original Message -----
From: Huaiyu Zhu <hzhu at knowledgetrack.com>
<snip>
>
> Cool!  Things always look easier in python. :)
>
> Now only if somebody could enhance that to check inherited
methods too (the
> "more general case").  That is, if a method is not found,
go recursively
> into the superclasses.  The following example is not
correct at the moment:
>
<snip>
> --
> Huaiyu Zhu
hzhu at users.sourceforge.net
> Matrix for Python Project
http://MatPy.sourceforge.net
> --
> http://www.python.org/mailman/listinfo/python-list
>


import types

def meths(items, meths=None):
  if meths == None:
    meths = {}
  for obj in items:
    for k, v in obj.__dict__.items():
      if type(v) == types.FunctionType:
        meths[k]=None
  return meths

def bases(obj, items=None):
  if items == None:
    items = []
  for b in getattr(obj, '__bases__', []):
    bases(b, items)
    items.append(b)
  if type(obj) == types.ClassType:
    items.append(obj)
  return items


def verify_class (Reference, Instance):
  r_meths = meths(bases(Reference))
  i_meths = meths(bases(Instance))

  for meth in r_meths.keys():
    if not i_meths.has_key(meth):
      print "Method %s not supported" % meth


Portions/ideas taken from both Kevin Jacobs
and from David Bolen.

Emile van Sebille
emile at fenx.com
-------------------




From jerome.quelin at insalien.org  Fri Jun 30 08:14:41 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 14:14:41 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8jhum101vku@news2.newsguy.com>
Message-ID: <962367281.1198008296@news.libertysurf.fr>

"Alex Martelli" <alex at magenta.com> wrote:
>> I was wondering how to pass variable-length keyworded arguments list
>> to my base-class constructor? Look at the example code below:
>Doesn't apply work for __init__ just as for any other function?
Sure it works for __init__ too, I've already tried this approch, but it doesn't
work either. In fact, I'm asking the question for __init__ because I run into
this problem when calling constructors, but the question is the same for every
inherited and overloaded method.

When using the *args syntax, I get an error when calling my constructor because
I'm using keywords arguments.

When using the **args syntax, python fetches all keyworded arguments in _one_
dictionnary in my constructor, and I want to keep the arguments in multiple
key/value pairs (and not flattened in one dictionnary) since the parent
constructor is waiting for keyworded arguments too.
If I use the **args syntax and then call the base constructor with:
Parent.__init__(self, args)
I got an error since it doesn't wait for a dictionnary as an argument (and args
is a dictionnary in this case) but for key=value arguments, with key and value
taken from the args dictionnary.

Any idea?
Jerome
--
jerome.quelin at insalien.org


From mhammond at skippinet.com.au  Sun Jun 18 00:28:02 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 18 Jun 2000 04:28:02 GMT
Subject: Problem Installing/Using Windows Extensions
References: <2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1@lp.airnews.net>
Message-ID: <mdY25.1544$qz4.13082@news-server.bigpond.net.au>

"David Steffen" <steffen at biomedcomp.com> wrote in message
news:2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1 at lp.airnews.net...

> I'd be delighted to RTFM, if I could find the relevant M (I've looked
> in the included Python docs and on the python.org and starship python
> websites).

Sheesh - On the "downloads" page (which is where you _must_ have got 132
from), just under the explanation about 1.5 and 1.6 living together, it
says:


Should you have any problems installing these packages, please refer to
the <a ..>Installation Problems Guide</a>.

Ive now made it BOLD, and also copied the line to the info about each
specific build.

Please check it out, and let me know how I can possibly make it any
clearer.  Your exact problem is listed there...

Mark.





From dan at cgsoftware.com  Fri Jun  9 20:21:44 2000
From: dan at cgsoftware.com (Daniel Berlin)
Date: Fri, 9 Jun 2000 17:21:44 -0700 (PDT)
Subject: WIDESPREAD INCOMPETENCE AT BELL ATLANTIC
In-Reply-To: <394188BF.B39F1255@san.rr.com>
References: <B60FEC7EB.pduegowihyk@ooobpqvldjd.com> <394188BF.B39F1255@san.rr.com>
Message-ID: <Pine.LNX.4.10.10006091721010.17120-100000@propylaea.anduin.com>

That would be unfortunate, for while this isn't the approriate forum,
everything that was in that message, is true.



On Sat, 10 Jun 2000, Courageous wrote:

> 
> > Perhaps the US Government should take a look at Bell Atlantic in
> > much the same way it's currently looking at Microsoft, with a similar
> > remedy in mind.
> 
> Bell Atlantic just Acquired GTE and doubled in size. It appears
> the FTC disagrees....
> 
> 
> C/
> -- 




From claird at starbase.neosoft.com  Fri Jun 16 12:54:45 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 16 Jun 2000 11:54:45 -0500
Subject: Conversion from proprietary satellite control script language
References: <20000616084700.29919.qmail@web2006.mail.yahoo.com>
Message-ID: <C5BE370AAB754BF0.2ECE2983386628A0.03ABBB0029C64DFE@lp.airnews.net>

In article <20000616084700.29919.qmail at web2006.mail.yahoo.com>,
TJ Judson  <judson at iname.com> wrote:
			.
			.
			.
>I'm afraid my only other answer is to use a language that supports
>"goto", like Tcl.
			.
			.
			.
How does Tcl support goto?
-- 

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


From neurocrat at one.net.au  Tue Jun 13 18:55:44 2000
From: neurocrat at one.net.au (Neurocrat)
Date: 14 Jun 2000 08:55:44 +1000
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <ur9a1mmdk.fsf@ctwd0143.fitlinxx.com>
Message-ID: <85zoopfaxb.fsf@one.net.au>

David Bolen <db3l at fitlinxx.com> writes:

[ compile time errors]

> This does raise an interesting question.  Of the total set of errors
> (which "most" you receive during compilation), how many of them are
> related to syntactic or language overhead versus algorithmic issues.
> 
> It seems to me that the former does not necessarily represent a
> benefit to C++ over Python, because in my experience C++ has more
> syntactic "sugar" that can increase errors in that respect, so the
> fact that you get more of them during compilation isn't something that
> translates over to Python anyway.
> 
> Some amount of such errors (say errors in type declarations or other
> issues related to the strong, static typing of C++) can certainly
> point to algorithm issues during compilation.  But again, many of
> those issues arise from the requirement of strong typing an algorithm
> in the first place, so need to be considered as errors whose existence
> are largely a by-product of the development environment in use.

My (admittedly very limited) experience with Python bears this
out. Many of my compile time errors caught by C++ are errors that
would not have been present in a language with cleaner syntax. 

I know this is the exceptional case, but sometimes Python even has the
advantage here. Every C++ programmer has done this at some point:

for (int i = 0; i < 10; ++i);
        cout << "Hello" << endl;

This, of course, prints "Hello" only once, not 10 times as we
intended. In a toy example it's easy to spot the offending ; at the
end of the loop construct, but is there a C or C++ programmer alive
who hasn't spent hours trying to track down insidious problems caused
by that single legal semicolon? I know I have.

If I made the corresponding mistake in Python:
        for x in range(10)      # forgot the :
                print "hello"
I'd know it straight away.

> But true, there's the area of name errors during runtime, which my
> first inclination was to consider a real problem with Python.  Even
> with some of the 'lint'-like tools around, it didn't seem to be
> considered a major problem by the community.  But the more code I
> write the less concerned I become as I find that surprisingly it just
> isn't showing up as a major problem, 

I too find it surprising, but my experience is the same.

> ... and it's easy enough to program
> defensively against anyway.

I think the use of assertions to define the intended behaviour of a
routine is a better way of both preventing and catching bugs than any
compile time mechanisms. That's just my personal opinion. (Of course,
that much is possible with C++ too).


From anthony at interlink.com.au  Tue Jun 27 10:06:57 2000
From: anthony at interlink.com.au (Anthony Baxter)
Date: Wed, 28 Jun 2000 00:06:57 +1000
Subject: more python evangelising...
Message-ID: <200006271406.AAA06218@mbuna.arbhome.com.au>

in a mainstream-ish news article, no less...

http://www.theregister.co.uk/content/1/11606.html

when talking about C#, and it's relationship to Java...


[snippity]

And an even more cunning route was at hand, as the redoubtable maintainer
of the Java site Cafe Au Lait, Rusty Harold Elliot points out. Elliot,
a long time critic of Sun's licensing tactics, notes: "If Microsoft
wanted to really challenge Java, they should have gone with Python. I
just don't believe it's possible for any major advances in language
design to be made while restricting oneself to the mistakes Kernighan
and Ritchie made 30 years ago."

Well, quite. The open source language Python has not only proven itself
macho enough in heavy duty production environment, it's simple enough
to be taught as a first language to kids. And it's interfaced with COM
interface for several years now. Maybe someone in Redmond thought it
was time to unleash some "innovation" at the world, and cometh the hour,
cometh the new programming language.

[snippity]

--
Anthony Baxter     <anthony at interlink.com.au>
It's never too late to have a happy childhood.



From BgPorter at NOacmSPAM.org  Fri Jun 30 12:21:22 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Fri, 30 Jun 2000 16:21:22 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com>     <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com>     <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com>     <geT65.2262$iN5.511399@ptah.visi.com> <m2lmznzk2t.fsf@mycroft.actrix.gen.nz> <slrn8lp673.u8u.thomas@centurion.xs4all.nl> <8jidt1$2k5s$1@noao.edu> <u9375.2324$iN5.533105@ptah.visi.com>
Message-ID: <6O375.34996$lU5.251428@news1.rdc1.nj.home.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:u9375.2324$iN5.533105 at ptah.visi.com...
> In article <8jidt1$2k5s$1 at noao.edu>, Richard Wolff wrote:
> >thomas at xs4all.nl (Thomas Wouters) writes:
> >
> >>On 30 Jun 2000 19:47:38 +1200, Paul Foley <see at below> wrote:
> >>>On Fri, 30 Jun 2000 02:03:56 GMT, Grant Edwards wrote:
> >
> >>>> C predates the PDP-11 so we'll chalk it up to convergent
> >>>> evolution.
> >
> >>>Oh no it doesn't.
> >
> >>Oh yes it does ;) The first bits of C, including, for instance, the
> >>auto-increment operators, were designed and used on the PDP-7, not the
> >>PDP-11.
> >
> >>> C was written for the Unix port to PDP-11.
> >
> >>More like 'during' than 'for'. UNIX and C both, together, evolved on for
> >>quite a number of years (witness the new C & UNIX 'standards')
> >
> >"C was originally written for the PDP-11 under UNIX ... "
> >  The C Programming Language, by D. M. Ritchie, S. C. Johnson, M. E.
Lesk,
> >  and B. W. Kernighan, Bell System Technical Journal, Vol 57, #6, 1978,
Pg 1991
>
You can read Dennis Ritchie's own history of the development of C at
http://cm.bell-labs.com/cm/cs/who/dmr/chist.html
He describes C's origins on the PDP-7, and from my reading of the paper, how
the language went from its embryonic PDP-7 state into something more like
classic K&R C while working on the PDP-11. I really think that it's a floor
wax /and/ a dessert topping..





From timr at probo.com  Thu Jun  8 00:40:33 2000
From: timr at probo.com (Tim Roberts)
Date: Thr, 8 Jun 2000 04:40:33 GMT
Subject: win python and graphics
References: <393692FB.8610EBCA@dallas.net> <393D7BFD.F50F9932@sightreader.com> <Bah%4.22817$Rq.2324142@pouncer.easynews.com> <0006071724260A.00959@localhost.localdomain>
Message-ID: <GK8UJSKOS0IC0J1QC27N95LSAGVOTEN7V6@4AX.COM>

matt <matt at virtualspectator.com> wrote:

>Hi, I pretty much only use Linux, but have to do some things on windows.  In
>particular I need to take screen shots of the desktop or particular windows.  I
>would like do this using python rather than in borland C++.  Can anyone help out
>here ?

Not in straight python, even with the Win32 extensions.  You can locate
windows and even get a DC, but the necessary GDI calls aren't available.

If you're willing to include a manual step, it might be possible.  The
PrntScrn key on PC keyboards takes a snapshot of the current desktop to the
clipboard, from which many graphics apps can process it.  Alt-PrntScrn
snapshots the topmost window to the clipboard.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From moshez at math.huji.ac.il  Thu Jun  8 04:06:20 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 8 Jun 2000 11:06:20 +0300 (IDT)
Subject: Why not a Python Package Manager
In-Reply-To: <393EDA6D.3427CBE1@visionart.com>
Message-ID: <Pine.GSO.4.10.10006081101490.3146-100000@sundial>

On Wed, 7 Jun 2000, Pete Shinners wrote:

> > I agree. I agree with 'something like Debian' so much I think dpkg and
> > friends is pretty much the way to go. Why not completely use the debian
> > package manager?
> 
> don't leave us windows users out in the cold!
> i'm not sure of the window'y'ness of the debian packages,
> but perhaps dpkg can be applied to many platforms?

Debian is a "Universal Operating System". Debian GNU/Hurd is a development
project (but then again, so is Hurd), Debian GNU/FreeBSD is an idea that's
definitely floating around, and Debian GNU/Solaris and GNU/NT have been
thrown out in the open. There's no reason dpkg *won't* work on Solaris,
and it would actually be cool to have a good, upgradable OS on the Solaris
kernel (yes, some of us actually have to work with that kernel). The NT
case is a bit harder, since NT doesn't have a concept like "/", but as
long as you map "/" to "wherever windows was installed:\", you can just
use the normal Debian setup. The big question here is the compiler used --
it wouldn't be that easy. 

debian-GNU/Red Hat Linux-anyone?-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From ssthapa at midway.uchicago.edu  Sat Jun 24 18:54:50 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 24 Jun 2000 22:54:50 GMT
Subject: need insight...
References: <3954671F.51E53AE6@mindspring.com> <Zh_45.85985$WS3.752637@typhoon.we.rr.com>
Message-ID: <slrn8la4f4.769.ssthapa@ntcs-ip48.uchicago.edu>

Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:
>    The problem is that you really do not want to change the sequence you're
>using in 'for' -- the documentation, if memory serves, shows an example of
>precisely why. You'll want to make a copy of it to iterate through.. like;
>
>for item in cwd[:]:
>    ...
>    cwd.remove(item)
>
>Like that..see? The [:] slice copies the whole sequence.
>
    Actually I don't think code even needs to make a copy or have the
cwd.remove(item) call since it deletes cwd right after the for loop.
That should save a little space and speed up the code a bit.

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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From olipt at mayo.edu  Fri Jun  2 16:46:52 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Fri, 2 Jun 2000 15:46:52 -0500
Subject: Help, Image Filters with numpy
In-Reply-To: <393803A2.9AE23E9C@visionart.com>
References: <393803A2.9AE23E9C@visionart.com>
Message-ID: <Pine.LNX.4.10.10006021542070.4508-100000@us2.mayo.edu>

> i have some image data, greyscale, and i'd like to do
> some simple filter work on it. i'm trying to think of
> a "most optimized" way to get this done. i'm not sure
> about all the tools that numerical python gives me, so
> i thought i'd ask to see if parts of this are already
> taken care of.
> 
> say i define a quick soften filter...
> 
> soften = array([[1,2,1], [2,4,2], [1,2,1]])
> 
> how can i quickly/easily apply this to my image
> which is stored in another array? i've been hunting
> examples that would do this for me, but no luck


Charles Waldman already told you this is called convolution and that it
can be efficiently performed using FFT's if the kernel sizes are large. 

He did not tell you that I have a general convolution routine that will do
what you want in the package signaltools.  I use it constantly for just
this kind of operation.  signaltools is available at
http://oliphant.netpedial.net

The routine is quite general and will do N-dimensional convolution with
arbitrary kernels for arbitrary data types (even object types).  For your
purposes you would use it like this:

output = signaltools.convolveND(someimage, soften, 'same')

The code is written in C and is quite fast.  The 'same' is to return an
image with the same size.

Good luck,

Travis





From porter at et.byu.edu  Fri Jun 16 13:45:37 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Fri, 16 Jun 2000 11:45:37 -0600 (MDT)
Subject: Frozen socket.
Message-ID: <Pine.GHP.4.21.0006161113130.3877-100000@don.et.byu.edu>

I am very new at this sort of this, so I am hoping that there is a simple
and obvious (if you know it) solution to my problem.  I looked in the FAQ
and tutorials, but to no avail.

I have written a simple python program on linux that uses sockets.  My
program connects to an existing server.  There is some data exchanged for
handshaking purposes, and then I am supposed to start sending a long list
(not a python list) of strings to the server.  I don't get any responses
from the server - I keep sending out these strings until I am done, at
which time I disconnect the socket.

At least, that's what is supposed to happen.  Somewhere between string #
1000 and 1020 (it varies somewhat), my program freezes. It does this
whether I send the strings out as fast as I can, or if I space send out
the strings to one per second.  
I do a ctrl-C to escape, and the last line of the exception printout is:
    self.sock.send(messages.packMM(request))
This seems to me that the program is getting stuck in the
sock.send() funciton. 

I have looked around, but I haven't been able to figure out why a program
would choke on a send() command.  I don't even know if it is because of
something I am doing on my end or something happening on the server
end.  If it is my fault, how do I fix it?  If it is because of something
happening on the server side, how do I compensate?

Thank you for your help.  






From speed at ?.com  Fri Jun 30 15:00:59 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 30 Jun 2000 19:00:59 GMT
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <962367319.379353174@news.libertysurf.fr>
Message-ID: <slprjb10jev179@corp.supernews.com>

jerome.quelin at insalien.org (Jerome Quelin) wrote in
<962367319.379353174 at news.libertysurf.fr>: 

>Then, is it better/cleaner to access private members with accessors or
>not? Is it a matter of style? Or are there hidden caveheats using (or
>not using) accessors? It _seems_ cleaner to use accessors, but man, it's
>quite awkward. 

The usual trick is to write __getattr__ / __setattr__ methods.

Here's a class that'll try to call 'get_var' for
any undefined field 'var', and 'set_var' whenever
set_var exists:

class Object:
    def __getattr__(self, name):
        try:
                return getattr(self.__class__, "get_" + name)(self, name)
        except AttributeError:
                raise AttributeError, name
    def __setattr__(self, name, value):
        try: 
            getattr(self.__class__, "set_" + name)(self, name, value)
        except AttributeError:
            self.__dict__[name] = value

    def get_x(self, name):
        return self.__x
    def set_x(self, name, value):
        self.__x= value

obj = Object()
obj.x = 2
print "obj.x =", obj.x
print "obj.__dict__ =", obj.__dict__;

and voila, whenever your program calls 'obj.x = val', your
class calls 'set_x', which acesses its private field __x --
and does whatever else it needs to do.  

I think JPython uses JavaBeans this way, so you can write
'window.Background = Color.blue', and it calls 
'window.setBackground(Color.blue)' which might generate a
redraw event or something.

-Speed!


From jkraska1 at san.rr.com  Thu Jun 15 02:03:03 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 15 Jun 2000 06:03:03 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net>
Message-ID: <39487243.B90DBD85@san.rr.com>


Re: python vs c++

> However my C++ friends are badgering the crap out of me...

Most of your C++ friends would be picking their tongues
up off the floor if they saw the LOC of python programs
in comparison to their C++ counterparts.

Python may be too slow for some tasks, but in the area
of programmer productivity, very little beats it, and
arguably the environments that do (e.g., lisp) have too
many other negatives associated with them.



C/


From pj at sgi.com  Thu Jun 22 15:41:22 2000
From: pj at sgi.com (Paul Jackson)
Date: 22 Jun 2000 19:41:22 GMT
Subject: boolean true and false values.
References: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk> <0b0801bfdc5f$b3cca3c0$1906a8c0@fc.fenx.com>
Message-ID: <8itq52$1m6l8$1@fido.engr.sgi.com>

Emile van Sebille" <emile at fenx.com> wrote:
|> It's so much easier to recognize 122 octal as having alternate
|> bits set than ...

Huh - how is it that 122 (base 8) == 1010010 (base 2) has alternate
bits set?
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From feldmann at wag.caltech.edu  Wed Jun 14 14:29:20 2000
From: feldmann at wag.caltech.edu (Michael Todd Feldmann)
Date: Wed, 14 Jun 2000 11:29:20 -0700 (PDT)
Subject: Numeric python
Message-ID: <Pine.LNX.4.10.10006141123440.26946-100000@intel3.wag.caltech.edu>

Could anyone point me in the right direction to get Numeric python?  I
have a linux box and tried 

http://www.python.org/download/download_linux.html 

and went to

http://andrich.net/python/

which states that "This site is closed." since things may be outdated.  I
am guessing there is somewhere else to find this and other RPM files.  

Thanks a bunch!

Mike 

--------------------------------------
Michael Todd Feldmann
Beckman Institute (139-74)
California Institute of Technology
Pasadena, California 91125 USA
Phone:(626)395-2722, FAX:(626)585-0918
email: feldmann at wag.caltech.edu
--------------------------------------





From bvdpoel at uniserve.com  Thu Jun 29 20:38:59 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Thu, 29 Jun 2000 17:38:59 -0700
Subject: tracing a variable
Message-ID: <395BEC23.9A3097F0@uniserve.com>

Is is possible, as in tcl/tk, to put a watch on a variable? I've got a
tkinter script and I'd like to use this to enable/disable menu options
depending on the state of a buffer, etc.


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



From JoeSmith at bogusaddress.com  Sun Jun 18 00:17:59 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Sun, 18 Jun 2000 04:17:59 GMT
Subject: Python 3000 -> No, but there's 'Ruby'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com>
Message-ID: <394C4E3D.73DBB25C@bogusaddress.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000618/ecad364b/attachment.html>

From moshez at math.huji.ac.il  Wed Jun 28 01:41:34 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 28 Jun 2000 08:41:34 +0300 (IDT)
Subject: Speed with Python
In-Reply-To: <oq7lbbmhoc.fsf@titan.progiciels-bpi.ca>
Message-ID: <Pine.GSO.4.10.10006280840140.27161-100000@sundial>

On 27 Jun 2000, [ISO-8859-1] Fran?ois Pinard wrote:

> [Ken Seehof]
> 
> > I do all GUI work in python (wxPython) and I have not noticed a difference
> > in speed as compared to C++, probably because most time is spent in
> > library calls anyway.
> 
> I experimented some GUI writing in Python, and initially expected that things
> would be slow.  Not at all: I was very pleasantly surprised by the speed.

Indeed: in most GUIs, most time is spent in the user's head: IOW, the
computer is too fast, and most time it is just sitting quietly, waiting
for the user to make up his mind. Python's idle loop is almost as good
as C++'s <wink>

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From gtalvola at NameConnector.com  Wed Jun 14 14:56:46 2000
From: gtalvola at NameConnector.com (Geoff Talvola)
Date: Wed, 14 Jun 2000 14:56:46 -0400
Subject: tkinter/windows/starting html
References: <39476C94.DA0AA2AD@stud.ntnu.no> <028398e6.cd93152e@usw-ex0108-063.remarq.com>
Message-ID: <3947D56E.4FCD93CA@NameConnector.com>

This works for me (assuming you have the win32 extensions installed):

import win32api
win32api.ShellExecute(0, 'open', r'c:\junk.html', None, None, 0)

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com





From richard_chamberlain at ntlworld.com  Tue Jun 20 02:14:45 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 20 Jun 2000 07:14:45 +0100
Subject: {Q} OOP Animal class.
References: <8iluf0$le$1@nnrp1.deja.com> <FwF6Iy.G0I@world.std.com>
Message-ID: <aQD35.4118$fw6.82978@news6-win.server.ntlworld.com>

I believe Zargroves get pretty upset if someone mentions Blarndoogle's
splointers. So we shouldn't say anything more.

Richard ( ;-)

Will Ware <wware at world.std.com> wrote in message
news:FwF6Iy.G0I at world.std.com...
> johnvert at my-deja.com wrote:
> > 1. I call the method reply()
> > 2. The interperter sees reply() does not exist in class Hacker
> > 3. It goes one class up the hierarchy to class Primate
> > 4. reply() doesn't exist in class Primate
> > 5. It goes one class up the hierarchy to class Mammal
> > 6. reply() doesn't exist in class Mammal either
> > 7. It goes one class up the hierarchy to class Animal
>
> Good so far. As you suspected, this is where things start to
> get a bit dicey. My understanding (which perhaps others will correct
> if necessary) is that each method search is a separate thing, and in
> each case we are prepared to climb the inheritance tree. So when
> reply() calls speak(), we begin the whole search process anew, starting
> at Hacker. Hacker doesn't have a speak() method, but Primate does,
> and that's why we use Primate.speak().
>
> > I really want to have a solid understanding
> > of this because it's important to understand OOP.
>
> OOP is primarily a flexible way to let you make say things like this:
>
> A Blarndoogle is an object with this set of behaviors.
> A Flugestroid is almost exactly like a Blarndoogle, except
> that it splointers the way a Zargrove does, and it has its
> own unique way of flithering, different from either a
> Blarndoogle or a Zargrove.
>
> The idea would be that the behaviors for Blarndoogles and Zargroves
> would probably already be thoroughly debugged when you were ready to
> define a Flugestroid, so the only really new thing you need to do is
> define how it flithers. That is, all your effort is focussed entirely
> on what's new about this situation, not on re-implementing and
> re-debugging existing behaviors. At least that's the theory.
>
> Some languages (Python generally not among them) stress a "privacy"
> aspect of objects, that there should be information available only to
> the object and not to anybody else, and the object can therefore define
> how others get to view or modify that information. Python mostly
> doesn't worry about object privacy much; you can create private
> attributes by starting attribute names with two underscores, but you
> don't have the elaborate machinations of public/protected/private you
> get with C++ and Java.
> --
>  - - - - - - - - - - - - - - - - - - - - - - - -
> Resistance is futile. Capacitance is efficacious.
> Will Ware email:    wware @ world.std.com




From aahz at netcom.com  Mon Jun 26 17:31:28 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jun 2000 21:31:28 GMT
Subject: Final garbage collection?
References: <962028703.1248416950@news.libertysurf.fr> <FwrnM8.MMA@world.std.com> <962045446.1993274611@news.libertysurf.fr>
Message-ID: <8j8i3g$ht2$1@slb7.atl.mindspring.net>

In article <962045446.1993274611 at news.libertysurf.fr>,
Jerome Quelin  <jerome.quelin at insalien.org> wrote:
>
>But, (and the following is taken from perl documentation -man perlobj):
>       When an interpreter thread finally shuts down (usually
>       when your program exits), then a rather costly but
>       complete mark-and-sweep style of garbage collection is
>       performed, and everything allocated by that thread gets
>       destroyed.  This is essential to support Perl as an
>       embedded or a multithreadable language.
>
>I was wondering why Python does not implement such a mechanism, that
>seems to me a Good Thing (tm), because _in final_, every object _is_
>destroyed, either if their reference count is nul or not, since the
>memory is reclaimed by the system.

My memory may be faulty, but IIRC one of the primary reasons for not
doing this is the problems that result from memory allocations done by
extensions, and the possibility of having weird holes in memory.

That said, there *is* work being done on garbage collection, but it most
likely will need to be explicitly called.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From embed at NOSPAM.geocities.com  Fri Jun 30 15:22:21 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Fri, 30 Jun 2000 15:22:21 -0400
Subject: Reference Counting Woes in Python 1.5.2! Help!
Message-ID: <wr675.13560$227.259336@nnrp1.uunet.ca>

I am embedding Python and until recently had no problems with Reference
Counting.  Now I have one, and it has crept in during about a 6 day stint of
intensive coding, involving converting my application from a single threaded
python application to correctly using the python global interpreter lock and
creating thread state information for each thread.

Now i have a subtle reference count underflow bug, which seems to be causing
objects to be freed while still referenced.

I am using Visual C++ 6. It has debug libraries that are catching this
problem printing an error message and quitting. When I go to release version
of code, I get access violations and then the code quits. Alas, what I'd
like is to see the access violation location in source code, but the darn
debug library gets in the way!

Has anyone else debugged pesky reference count problems in Visual C++ with
Python 1.5.x and come out on the winning end? Your help or tips or
suggestions on how to debug this sort of thing would be most welcome!

Warren Postma




From nascheme at enme.ucalgary.ca  Tue Jun  6 14:24:03 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Tue, 06 Jun 2000 18:24:03 GMT
Subject: Memory leak: assigning function object to instance variable
References: <8hh0pf$i7$1@phys-ma.sol.co.uk> <slrn8joise.ik.nascheme@cranky.arctrix.com> <3%5%4.54$vh.18462@news>
Message-ID: <slrn8jqgfo.314.nascheme@cranky.arctrix.com>

OPUS <rhicks at nospam.rma.edu> wrote:
>Where is it??? Does it work with 1.5.2?

    http://www.enme.ucalgary.ca/~nascheme/python/gc.html

Support for method objects has not been backported to 1.5.2 yet.

    Neil

-- 
"If you're a great programmer, you make all the routines depend on each
other, so little mistakes can really hurt you." -- Bill Gates, ca. 1985.


From bwarsaw at beopen.com  Thu Jun 29 22:33:07 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Thu, 29 Jun 2000 22:33:07 -0400 (EDT)
Subject: Aspect-oriented Python?
References: <8jg821$vll$1@gaia.cdg.acriter.nl>
Message-ID: <14684.1763.210191.842107@anthem.concentric.net>

>>>>> "CdG" == Cees de Groot <cg at gaia.cdg.acriter.nl> writes:

    CdG> I've been reading on aspect-oriented programming a bit, and
    CdG> it sounds really interesting. A Java implementation, AspectJ
    CdG> (www.aspectj.org) looks promising, but has the drawback that
    CdG> it only supports static weaving (wouldn't know how to do
    CdG> dynamic weaving with Java, so it's not a big surprise).

I don't know of anybody doing AspectPy, but there is a Python
connection here -- Jim Hugunin invented JPython and left CNRI to work
on the AspectJ compiler.

-Barry



From Michael.Scharf at gmx.de  Mon Jun 19 17:55:30 2000
From: Michael.Scharf at gmx.de (Michael Scharf)
Date: Mon, 19 Jun 2000 23:55:30 +0200
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news and 
 links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net>
Message-ID: <394E96D2.5787F2A3@gmx.de>

"Andrew M. Kuchling" wrote:
>     In the same thread, David Bolen and Huaiyu Zhu write a script to
>     verify that a class implements all the methods of a template class:
>         http://x53.deja.com/getdoc.xp?AN=635109618
> 

Trying to extract the code or even read the code in the daja.com
representation seems impossible. Either I am an old-fashioned ascii
formatting fetishist or just to stupid to figure out, how to display
a deja article in simple plain ascii. Looking at the source of
this article: the entire script is put into one line.....

Maybe that's the price we have to pay for all the "non-computer-people"
in the internet, but I think with each change deja gets less and less
useful (at least for me). 

Does anybody know another news database with a more (white space sensitive)
programing language friendly interface?


Michael
-- 
     ''''\     Michael Scharf
    ` c-@@     TakeFive Software GmbH, a Wind River Company
    `    >     http://www.WindRiver.com
     \_ V      mailto:Michael.Scharf at gmx.de


From akuchlin at mems-exchange.org  Mon Jun  5 10:11:53 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 5 Jun 2000 09:11:53 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun  5)
Message-ID: <BA7ED618087F752B.EE3239CC079D7980.5169D95D2EE3FD08@lp.airnews.net>

    It's finally public; Guido, Fred Drake, Jeremy Hylton, and Barry
    Warsaw are now working for BeOpen.
        http://www.pythonlabs.com/letter.html

    BeOpen ran an interview with Guido:
        http://www.beopen.com/features/interviews/rossum_index.html

    PythonWare announced the long-awaited release of PythonWorks 1.0.
    "PythonWorks is a rapid application development tool, designed for
    efficient development in Python.  PythonWorks includes editors,
    a debugger, a user-interface design tool, project browser, deploy-
    ment tools, integrated on-line help, and a scriptable environment."
        http://www.pythonware.com/products/works/main1.htm      

    Tim Peters showed how to implement the Dining Philosophers in Python:
        http://x62.deja.com/getdoc.xp?AN=628752464.1

    PyGreSQL v3.0, an interface to PostgreSQL, has been released.  The
    new version implements the DB-SIG API.
        ftp://ftp.druid.net/pub/distrib/PyGreSQL.tgz

    Eric S. Raymond has been working on CML2, a replacement for the Linux
    kernel configuration system that's written in Python.
        http://www.tuxedo.org/~esr/kbuild/

    Also, Raymond's article in the Linux Journal's Python supplement (the
    one with the controversial cover) is now available online:
        http://www2.linuxjournal.com/lj-issues/issue73/3882.html
    Unfortunately, most of the other articles aren't available:
        http://www2.linuxjournal.com/lj-issues/issue73/lstoc.html

    Other recent articles include an intro to PyQt:
        http://www.sunworld.com/sunworldonline/swol-05-2000/swol-05-qt.html

    And two introductory articles:
        http://www.oreillynet.com/pub/a/network/2000/06/02/magazine/python_first_language.html
        http://www.geeks404.com/programming/articles/prog.052600.intro_python.php


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

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From noone at do.not.use  Mon Jun  5 12:11:25 2000
From: noone at do.not.use (Russell Turpin)
Date: Mon, 05 Jun 2000 11:11:25 -0500
Subject: Cool!  (was: "dlist" -- beta is ready)
References: <393B4144.4C1DC04C@san.rr.com>
Message-ID: <393BD12D.BFCA6672@do.not.use>

Courageous wrote:
> There are three items which I will finish soon
> (perhaps sooner if someone asks for them):
> 
> 2. reversing.

Since this is a more symmetric implementation than
the standard list, I wonder if it makes sense just
to reverse the sense, ie, subtract indices from 
the top, rather than add from the bottom. If this
works, then reversing becomes O(1) instead of O(n).
(OTOH, this adds a test to the index operator, and
there likely will not be much need for reverse, 
since dlist supports operations at either end. 
With standard lists, reverse provides a way to 
work at the other end efficiently, as long as you 
can justify the cost of the reverse. So this likely
is a BAD idea.)

OTOH, here's another possible optimization that
symmetry makes possible: If you support
indexing from either end, it is possible to support
O(1) deletes, as long as all other operations are
in the first or last segment of the list. Reallocation 
would occur only when there is an operation that has
deleted spots between it and both ends of the list.
This would change linear scan with selected delete
from an O(n^2) operation to an O(n) operation. On
the other hand, this would also add some complexity
to your current implementation, since it requires
maintenance of a delete list, with first and last
indices. It also adds one more comparison to the 
index operator.

Excited-about-dlist-ly yrs,
Russell


From cg at gaia.cdg.acriter.nl  Wed Jun 28 17:06:41 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 28 Jun 2000 23:06:41 +0200
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <8jdpd1$ogt$1@gaia.cdg.acriter.nl>

Steve Juranich  <sjuranic at condor.ee.washington.edu> said:
> Also, what benefit is gained by allowing a user to
>add/delete members from existing classes?  Wouldn't it be more desirable
>to create his/her own derived class?
>
I've been reading up on aspect-oriented programming (quickstart link
would be www.aspectj.org, although that's a bit Java specific). It makes
a lot of sense to fiddle around with someone else's class code at
run-time. 

Also, don't forget that with dynamically typed languages like Python,
you often aren't interested in types at all, just in the kind of methods
an object offers you. So mixing in methods, even at run-time, can be
a very useful operation in order to make the object accessible to
more other objects (who expect these methods to be present). 

Besides, it's fun.


-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From nascheme at enme.ucalgary.ca  Mon Jun  5 20:52:44 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Tue, 06 Jun 2000 00:52:44 GMT
Subject: Memory leak: assigning function object to instance variable
References: <8hh0pf$i7$1@phys-ma.sol.co.uk>
Message-ID: <slrn8joise.ik.nascheme@cranky.arctrix.com>

John Parkey <jparkey at nospamrapidbs.com> wrote:
>I have registered this bug, with code fragment below, but wondered if
>anybody else was falling foul of memory leaks of this type.

The latest version of my GC patch will prevent this leak.
Hopefully it will be an optional feature in 1.6.

    Neil

-- 
The internet: Learn what you know.  Share what you don't.


From paul at prescod.net  Tue Jun  6 13:49:55 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 06 Jun 2000 12:49:55 -0500
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <m3wvk2wy1u.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <393D39C3.11486684@prescod.net>

Michael Hudson wrote:
> 
>...
> 
> Tangentially, does anybody know of any implementations of JavaScript
> (not necessarily connected to a browser) that let you type interactive
> commands into a listener and genereally muck about with the language?

You can write it in about fifteen lines of HTML and Javascript.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Alabama's constitution is 100 years old, 300 pages long and has more
than
600 amendments. Highlights include "Amendment 393: Amendment of
Amendment
No.  351", "Validation of Laws Regulating Court Costs in Randolph
County",
"Miscegenation laws", "Bingo Games in Russell County", "Suppression
of dueling".  - http://www.legislature.state.al.us/ALISHome.html



From stevena at permanent.cc  Sat Jun  3 20:00:33 2000
From: stevena at permanent.cc (Steven D. Arnold)
Date: Sat, 03 Jun 2000 20:00:33 -0400
Subject: auto-debugging features?
Message-ID: <4.3.1.2.20000603191943.01621258@phear.dementian.com>

Hey folks,

Throwing this out for consideration and criticism.

I'd like a generic system-wide method of logging the activities of my 
programs for debugging purposes.  If I write and distribute a program, I'd 
like to be able to allow the user to set the debug level to a variety of 
different places, perhaps through an interface I provide, resulting in a 
log of the program's activity that can be parsed by any program that can 
use the debugging info protocol.  The log would probably be in 
human-readable text and would simply be parsed by a more sophisticated 
program intended to help with reviewing the log information.

I'd like to get all this without cluttering up my code with print 
statements, or worse yet, `if' statements, a la:

	if (g_debuglevel > 5):
		print "[NOTICE/reports.cell.print_cell] Called function \"print_cell\" 
with params:"

Debugging info would go to stderr.  The user could of course replace stderr 
with an object of their choosing, resulting in an app that can 
automatically transmit debugging info to any place on the network, for example.

Something that would be cool in such a debugging info utility:

	- Notification on entry to a function.
	- On entry to a function, list the parameters and values for the params.
	- When data is changed, note the before and after values.
	- When an if statement is executed, list the expressions that were 
compared and the values of each.

It 'feels' to me that the most general way to solve this problem would be 
to introduce a series of meta-magic functions in python that would trigger 
when important events occured; but this would only happen if some kind of 
flag were turned on.  So, for example, a __func__ meta-magic method could 
be triggered whenever any method in a class were called.  Paramters to 
__func__ would be the method name and the entire parameter list.  The coder 
of __func__ could determine exactly how the method would react to any 
function being called -- anything from no action, to writing a line to a 
log file, to sending a message to a network host.

Similarly, we might have a function for __data_to_change__, which would be 
passed a stack trace, the value of the variable about to change and the 
value to which it would be changed.

True, __getattr__ would handle some aspects of this, but I would prefer to 
keep __getattr__ clean and oriented around a single purpose -- implementing 
the class of which it is a part.  This suite of meta-magic functions would 
hopefully cover every meta-action a program could take and would therefore 
allow for very sophisticated debugging applications.  If we had a 
__startup__ meta-magic function, we could write a program that would 
attach, when the program ran, to a remote debugger.  Since each action the 
program could take would be represented by a meta-magic function, we could 
then remotely step through the program.  We could support exactly the level 
of debugging we desired and would not be locked into having either too much 
or too little information.  This would be a miracle for CGI debugging, for 
example.

Thoughts and/or suggestions for improvement are welcome!


steve

--
Steven D. Arnold            Que quiero sera         stevena at permanent.cc
"Fear: The best way out is through."                     -- Helen Keller




From fiona at sitegnome.com  Wed Jun 28 06:27:12 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 28 Jun 2000 10:27:12 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 28th, 2000
Message-ID: <20000628102712.23862.qmail@synop.com>

Hi All,

Only a couple of entries entered into http://python.faqts.com today.

cheers,

Fiona


## New Entries #################################################


-------------------------------------------------------------
How can I access a method with variable amount of parameters?
http://www.faqts.com/knowledge-base/view.phtml/aid/4048
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich

You use apply(). apply(f, args) calls the function f with the arguments
given in the tuple 'args'.

To make a function that takes multiple arguments, use

def my_apply(func, *args):
   return apply(func, args)

(args is a tuple containing the values passed in).

To handle keyword arguments as well, use

def my_apply(func, *args, **kw_args):
   return apply(func, args, kw_args)
   
In this case, kw_args is a dictionary with keyword: value pairs. Just 
play around in the interpreter a bit, it's quite simple but looks 
complicated the first time you see it...


-------------------------------------------------------------
What is the best way to make multiple replacements of text strings in a file?
http://www.faqts.com/knowledge-base/view.phtml/aid/4049
-------------------------------------------------------------
Fiona Czuczman
Steve Holden, Steve Nordby,David Goodger, Fredrik Lundh

You could clean the source up a bit with:

   foobar = ( ('foo1', 'bar1'),
              ('foo2', 'bar2'),
              ('fooN', 'barN') )
   source = open(source_file,'r')
   contents = source.read()
   source.close()
 
   for foo, bar in foobar:
        contents = replace(contents, foo, bar)
 
   dest = open(dest_file, 'w')
   dest.write(contents10)
   dest.close()

For speedup, you could write the whole loop as a single statement,
but it will get horrible quickly:

   contents = replace(
                 replace(
                 replace(contents,
                 'fooN', 'barN'),
                 'foo2', 'bar2'),
                 'foo1', 'bar1)

and, of course, ths code is much less easy to maintain.

How about using a dictionary and a loop:

replaceme = {'foo1': 'bar1', 'foo2': 'bar2', 'foo3': 'bar3'}
for key in replaceme.keys():
    srch = key
    rplc = replaceme[key]
    contents = string.replace(contents, srch, rplc)

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

Perhaps a bit complex for a newbie, but the most definitive answer I've 
seen so far has been Fredrik Lundh's 2000-04-12 reply to the "Python 
idiom: Multiple search-and-replace" thread:

> Is there a Python feature or standard library API that will get me 
> less Python code spinning inside this loop?   re.multisub or 
> equivalent? 

haven't benchmarked it, but I suspect that this approach is more 
efficient:

...

# based on re-example-5.py

import re
import string

symbol_map = { "foo": "FOO", "bar": "BAR" }

def symbol_replace(match, get=symbol_map.get):
    return get(match.group(1), "")

symbol_pattern = re.compile(
    "(" + string.join(map(re.escape, symbol_map.keys()), "|") + ")"
    )

print symbol_pattern.sub(symbol_replace, "foobarfiebarfoo")

...






From gmcm at hypernet.com  Tue Jun 13 21:35:40 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Jun 2000 01:35:40 GMT
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu> <8F51D9DB3gmcmhypernetcom@199.171.54.154> <jl7lbtm60s.fsf@endor.bbb.caltech.edu>
Message-ID: <8F52DC18Agmcmhypernetcom@199.171.54.194>

[posted and mailed]

Michael Vanier wrote: 

>gmcm at hypernet.com (Gordon McMillan) writes:
>
>> Michael Vanier: 

[snip]

>> >To fix this, I changed the flag to be a list with one integer
>> >element. Now, when I change the list element in the parent, the child
>> >sees it too.  
>> 
>> 
>> Let me guess. You did something like make "flag" a module level
>> global, and then did "from ... import *" in the other.
>
>Even dumber than this; it was all in the same file.  I set the global in
>the parent thread, and expected that the new thread would see the new
>value when it was changed.

And did the routine that set it say "global flag"? If not, it set a local 
(within the routine).

>> 
>> If you had referenced "module.flag", all would have worked. But that 
>> stinks, too. Your thread should have a flag member, and the parent
>> thread should set it (through a setter, if you really want to be
>> virtuous <wink>). That way the flag becomes part of the thread's API,
>> instead of a bass- ackwards dependency.
>> 
>
>Well, the thread wasn't an object; I just called a function and expected
>that it would see the same data (see below).

Ah. I assumed that you were using threading, not thread. The latter is the 
low level primitives. The former is the one to use.

>So what you're saying is that a thread's namespace is not the same as
>the global namespace of the parent.  This is puzzling, since that's what
>I thought a thread was: effectively a new process which shares the
>parent's namespace.  Is this in a FAQ somewhere?  Or am I just not
>getting it? 

No you're not. This has NOTHING to do with threads.

>To further complicate things, here is a stripped-down version of the
>code that actually works!

If you'd post the code that *doesn't* work, then we could quit this 
guessing game!


5-to-1-on-the-missing-global-ly y'rs

- Gordon


From avv at quasar.ipa.nw.ru  Wed Jun 28 23:45:58 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Wed, 28 Jun 2000 20:45:58 -0700
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com>
Message-ID: <395AC676.F748BE39@quasar.ipa.nw.ru>

Hi

Eric Lee Green wrote:
> Basic plusses of PostGreSQL: transactions, cursors, stored procedures, joins
> with subselects, all mostly work as you'd expect of any SQL database. The same
> cannot be said to be true of most other embedded databases out there (e.g.
> RAIMA, C-Tree, MySQL, ...). 

Hmm. It's hardly to say about Raima exactly the same as you may say
about traditional relational databases, because Raima is a different
animal. You can go relational with Raima Data Manager, but you gain its
real power if you use its navigational network model. It's more flexible
than just 'cursors'. It supports transactions too. Besides this, Raima's
(Centura's) client/server database Velocis supports all this plus normal
SQL access: your application may have Velocis engine embedded into it
and gain speed of using native library calls to access the database
(sitting on a local filesystem, with the app). In the same time this
application can serve as a full fledged SQL-server for the rest of the
world.

I have some alpha-stage Python wrappings for Velocis, which everybody
can test with Centura's demo version of the database. If anybody is
_really_ interested please let me know. I didn't plan to release these
wrappings now, but if you ask...

As far the Python interface to its native network model API is concerned
I still believe than it's better to developed higher-level application
objects with these API and then wrap them for Python. YMMV.

Regards

Alexander

PS. I do not work for Centura :-)



From cjensen at bioeng.ucsd.edu  Mon Jun 19 19:42:41 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Mon, 19 Jun 2000 16:42:41 -0700
Subject: spaces in filenames
References: <000201bfd350$0d818660$f6a2143f@tim>
Message-ID: <394EAFF1.CC812E26@bioeng.ucsd.edu>

You can also use Null characters (ASCII code 255).  Win 95/98 could see
it, but couldn't access it.  I think Win2000 can handle it ok.

-- 
Curtis Jensen


From frankc at colconsulting.com  Mon Jun 19 00:06:56 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Mon, 19 Jun 2000 04:06:56 GMT
Subject: Circular import
References: <394D202B.1DF60864@colconsulting.com> <05a701bfd978$4fdb7ab0$6401a8c0@home>
Message-ID: <394D9C60.DEE41574@COLCONSULTING.COM>

Darrell Gallion wrote:
>
> From: "Frank V. Castellucci" <frankc at colconsulting.com>
> >
> > With the assumption that class A and class B reference each other, how
> > do you avoid the Python error with:
> >
> > ---------- Composite.py -----------
> >
> > from Leaf import Leaf
> >
> > class Composite:
> > def __init__( self ):
> > self.__theLeaf = Leaf(self)
> >
> > -------- Leaf.py --------------
> >
> > from Composite import Composite
> >
> > class Leaf:
> > def __init__( self, aComposite ):
> > self.__theComposite = aComposit
> >
> >
>
> I'm guessing you come from the land of C++, where the compiler wants to know
> every detail.
> Don't bother with this:
> > from Composite import Composite
>
> Leaf can accept any kind of object as the composite.
> What ever methods the Composite has will be accessible.
> It's a beautiful thing once you get the hang of it.
>
> --Darrell Gallion

Thanks!

--
Frank V. Castellucci


From colin at kaizensoft.com  Wed Jun 14 22:08:41 2000
From: colin at kaizensoft.com (colin)
Date: Wed, 14 Jun 2000 19:08:41 -0700
Subject: 2 embedding in C++ questions
Message-ID: <39483AA9.471D8AE1@kaizensoft.com>

1. Is there a way I can, from C++, figure out whether there exists in a
loaded python module any class which is derived from a particular other
class that I know the name of, and if so it's name?  Then further, to
instantiate that class from C++?  Except for parsing the module's python
source file myself...

2. Is there a way to call a method of a python object from C++ except
for "PyObject_CallMethod"?  I tried doing a getAttr with a name, saving
that attr, and calling PyObject_CallFunction, and PyObject_CallObject
with no success.  Mebbe I'm just doing it wrong?

thanks,
Colin



From android at micron.net  Fri Jun  9 11:02:17 2000
From: android at micron.net (Peter Immarco)
Date: Fri, 09 Jun 2000 11:02:17 -0400
Subject: Probably a dumb question
Message-ID: <394106F9.A58F87CE@micron.net>

Hi,  still waiting for my two books on Python from Amazon.  I know
Python is a full programming language so it can be used on web servers
for server side processing, but what about browser side processing?  Can
I also use it to create an 'applet' that runs in a compatible browser?
I'm guessing not likely because I don't think IE4+ or Netscape
Communicator support Python but I'm hoping I'm wrong.

Thanks,
Peter.



From claird at starbase.neosoft.com  Tue Jun  6 14:30:23 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 6 Jun 2000 13:30:23 -0500
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <393D3948.821E1DB6@prescod.net>
Message-ID: <881B6C214E393C4E.E01D545A5E6A6673.A4C267911C483151@lp.airnews.net>

In article <393D3948.821E1DB6 at prescod.net>,
Paul Prescod  <paul at prescod.net> wrote:
			.
			.
			.
>> I'm aware of CGI scripting with Python. It's great. But JavaScript
>> pages load faster than CGI pages, and are easier to program, to
>> boot. Plus, JavaScript can directly call Java Applet methods, enabling
>> it to do functions that are simply impossible with CGI scripts.
>
>Actually, Java applets were last year's programming model.  In fact,
>Java applets are something of an endangered species.
>
>Java is almost entirely used on the server side these days and Python is
>just as good there as Java is. In fact, with Zope, Python is better.
>
>Client-side Javascript is quite popular but in terms of volume of code
>it is a drop in the bucket compared to what is going on on the server
>side.
			.
			.
			.
My colleague Paul is known for inciting (virtual) riots
over Python.  I wonder what that makes me, because I find
his words here far too mild.

The corporate developers I see have, if anything, taken to
UNDERappreciating JavaScript.  They simply abandon it,
rather than deal with browser incompatibilities.  Client-
side is all toys, they agree (I don't, but it's not worth
fighting about).

The server is where the action is--ESPECIALLY for conventional
notions of performance.  The way organizations juice up their
perceived performance is to scale servers.  They are *much*
better at spending money on bigger boxes than on replacing
a jillion client desktops, or improving client-side code.

And what software has the biggest buzz on the server side?
Python-based Zope.  Not Beans, not even AOLServer or mod_perl,
and certainly not JWS.

Were these "set-up" questions?  The other thing I get from
big organizations is that the Web isn't good enough for them,
and they need to deploy stuff that works both over the Web
and in better versions as client-server.  Scripting languages
are *great* for that kind of development.  The Java vision
and marketing are good, but ... well, I'll just say that
Python has nothing for which it should apologize (apart from
a few technical quibbles on which I'm working).
-- 

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


From darrell at dorb.com  Sat Jun 17 00:52:57 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 16 Jun 2000 21:52:57 -0700
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com>
Message-ID: <028001bfd817$e9dc9ca0$6401a8c0@home>

I'm a strong believer in expanding tabs into four spaces.
Spaces reliably render even in email.

Not to mention the pain to edit a file with the tabs screwed up.
I use a script to strip the evil tabs and my life has been better for it :-)

--Darrell Gallion

"Huaiyu Zhu" <hzhu at knowledgetrack.com> wrote in message >
> OK, so I agree that the (setq-default tab-width 4) in my .emacs is evil
(it
> sits there since way back when I copied my first .emacs from somebody
else),
> I'm wondering what's the easiest way to revert this on multiple files,
> either within or without emacs.  If I'm going to change my default emacs
> behavior I want to do this on all the files, and safely.
>





From siva1311 at my-deja.com  Tue Jun 27 08:47:22 2000
From: siva1311 at my-deja.com (siva1311 at my-deja.com)
Date: Tue, 27 Jun 2000 12:47:22 GMT
Subject: multiple replacements
Message-ID: <8ja7ok$6be$1@nnrp1.deja.com>

Hi,
I am a newbie Python programmer and was wondering the best way to make
multiple replacements of text strings in a file. I have 10-15 strings to
replace and right now I am doing this like:

  source = open(source_file,'r')
  contents = source.read()
  source.close()

  contents1 = string.replace(contents, 'foo1', 'bar1')
  contents2 = string.replace(contents1, 'foo2', 'bar2')
  ...
  contents10 = string.replace(contents9, 'foo10', 'bar10')

  dest = open(dest_file, 'w')
  dest.write(contents10)
  dest.close()

I know this must be a terrible kludge.

Using sed I can use a single statement to carry out the replacements:
    sed "s/foo1/bar1/g
         s/foo2/bar2/g
         ...
         s/foo10/bar10/g" source_file > dest_file

so I assume there must be a much better method than I am currently using
in my Python version.

Any suggestions for improvement to the Python code are appreciated.

-Brad


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


From jbauer at rubic.com  Mon Jun 26 17:53:00 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 26 Jun 2000 16:53:00 -0500
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <03bf34ab.667a04d2@usw-ex0105-040.remarq.com>
Message-ID: <3957D0BC.64D36612@rubic.com>

RickZ wrote:
> I'd like to get my employer to add Python to its "approved" list
> of development tools, too. But which Python? It's an extra
> handicap to say, "Well, the language now is changing constantly
> in ways that will potentially break what we develop now.

I've been writing Python code since 1995 and haven't had
to change anything related to the language through 4 major
releases.  Most of my code changes have been related to 
environments or interfaces (e.g. PostgreSQL -> MySQL).

> I've personally become sold on OpenSource and its 
> implications. But my employer is a Microsoft Business 
> Partner, and they think they're doing the Right Thing.

Talk about coding standards in a constant state of flux! 
Going from one version to another in Microsoft (VBasic, 
VC++, Foxpro, J++, CE SDK) is like stepping through a 
minefield. In comparison, Python is conservative to 
the point of slothfulness.

don't-bother-to-wake-me-when-we-have-a-new-release-ly y'rs,

Jeff Bauer
Rubicon Research



From m.faassen at vet.uu.nl  Tue Jun 20 11:28:47 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jun 2000 15:28:47 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <8inq3u$p0i$1@newshost.accu.uu.nl> <39586f3d.366407626@news.online.no> <395978c6.368848746@news.online.no>
Message-ID: <8io2jf$jcg$2@newshost.accu.uu.nl>

Thomas Weholt <thomas at cintra.no> wrote:
[snip switching from Linux to Windows confusion]
> Eh ... uhm ... what did I just say ?? Let me clarify; I`m not a
> schizo. I meant use Linux instead of Windows. Ok?

I knew you meant that from the start! It was just funny the way you
responded to my admittedly confusing correction. :) I hope no harm was
done though!

Regards,

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


From Lance.Ellinghaus at eConnections.com  Tue Jun 27 14:43:59 2000
From: Lance.Ellinghaus at eConnections.com (Lance.Ellinghaus at eConnections.com)
Date: Tue, 27 Jun 2000 11:43:59 -0700
Subject: more python evangelising...
Message-ID: <OFD3784693.E2406D84-ON8825690B.0066AB04@econnections.com>

I disagree.

If you would like to look at some large scale applications that run under 
Python, just take a look at Yahoo!, eGroups, GeoCities. Not only are these 
sites handling a large traffic load, but I would guess that these are not 
small applications.

lance





"Tom" <tom-main at REMOVEME.home.com>
Sent by: python-list-admin at python.org
06/27/2000 08:57 AM

 
        To:     python-list at python.org
        cc: 
        Subject:        Re: more python evangelising...

But you're falling into the old trap of thinking that one language is best
for everything.

Python is excellent for what it was designed for, but this does not 
include
large scale app development.

On the other hand, maybe it will be possible to add 'optional' features to
Python 3000 that will make it suitable for that kind of development 
without
interfering with its simplicity and (development) speed.

Tom.

"Anthony Baxter" <anthony at interlink.com.au> wrote in message
news:200006271406.AAA06218 at mbuna.arbhome.com.au...
> in a mainstream-ish news article, no less...
>
> http://www.theregister.co.uk/content/1/11606.html
>
> when talking about C#, and it's relationship to Java...
>
>
> [snippity]
>
> And an even more cunning route was at hand, as the redoubtable 
maintainer
> of the Java site Cafe Au Lait, Rusty Harold Elliot points out. Elliot,
> a long time critic of Sun's licensing tactics, notes: "If Microsoft
> wanted to really challenge Java, they should have gone with Python. I
> just don't believe it's possible for any major advances in language
> design to be made while restricting oneself to the mistakes Kernighan
> and Ritchie made 30 years ago."
>
> Well, quite. The open source language Python has not only proven itself
> macho enough in heavy duty production environment, it's simple enough
> to be taught as a first language to kids. And it's interfaced with COM
> interface for several years now. Maybe someone in Redmond thought it
> was time to unleash some "innovation" at the world, and cometh the hour,
> cometh the new programming language.
>
> [snippity]
>
> --
> Anthony Baxter     <anthony at interlink.com.au>
> It's never too late to have a happy childhood.
>


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


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000627/53949830/attachment.html>

From aahz at netcom.com  Tue Jun 13 10:19:10 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 13 Jun 2000 14:19:10 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de>
Message-ID: <8i5fsu$t7$1@nntp9.atl.mindspring.net>

In article <394625A3.4489C696 at muc.das-werk.de>,
Thomas Thiele  <thiele at muc.das-werk.de> wrote:
>
>I think the more complex the program is the more C++ is the better
>choice.
>
>It is very easy and fast to write short effective programs with python.
>But in larger programs a lot of time is spend for developing and
>testing the algorithm.

Yup.  And that's precisely why Python looks better and better as your
programs get larger.  Python allows you to concentrate on your problem
rather than the language.  Python literally allows you to develop the
code for your algorithm in realtime, rather than designing your
algorithm and *then* writing the code to match your algorithm.

See Eric Raymond's article in the Linux Journal for more details:
http://www2.linuxjournal.com/lj-issues/issue73/3882.html
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From g_will at cyberus.ca  Mon Jun 19 10:05:36 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: 19 Jun 2000 14:05:36 GMT
Subject: Pythonwin-wxPython
References: <394D3024.7E2B1A11@hotmail.com> <8F57DAB28gmcmhypernetcom@199.171.54.154>
Message-ID: <01bfd9f7$8061b620$8b54c3d1@Cyberus>

Yes this is a real drag, but the easy way to get around it is to include a
try and except and then it works with pywin and you get your traceback.

    class myFrame(wxFrame):
        def __init__(self, parent=NULL):
            wxFrame.__init__(self, parent, -1, "wx_plot_lines",
                wxDefaultPosition,wxSize(600,520))          
            try:
                plotCanvas(self)  #your panel
            except:
                self.Destroy() # causes main loop to exit
                traceback.print_exc() # and you still get your trace
         
As well you need to import traceback.

Regards,

Gordon Williams


Gordon McMillan <gmcm at hypernet.com> wrote in article
<8F57DAB28gmcmhypernetcom at 199.171.54.154>...
> kupisch at hotmail.com (ratze) wrote:
> 
> >My  problem is that when I am developing a wxPython application
> >everytime the the start of my app fails due to an exception I cannot
> >start it again and get 'Only 1 wxApp per process!'. How do I get around
> >this??
> 
> Run the wxPython from the command line. You can get around this if you 
> really want to, but it's not worth the struggle.
> 
> - Gordon
> 


From nobody at nowhere.nohow  Thu Jun 29 22:03:56 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Fri, 30 Jun 2000 02:03:56 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com> <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com>
Message-ID: <geT65.2262$iN5.511399@ptah.visi.com>

On Thu, 29 Jun 2000 21:15:33 GMT, Eric Lee Green <eric at estinc.com> wrote:
>Grant Edwards wrote:

>> >C++ is syntactic sugar for "C". As should be obvious from the fact that most
>> >early C++ compilers generated "C" as their output :-).
>> 
>> And C is just semantic sugar for assembly...
>
>You mean you've never noticed the resemblence between "C" and PDP-11 assembly
>language? :-).

Sure have.  Back when I was an undergrad, PDP-11 was what you
learned assembly on.  Later I was quite struck by how well C
operators mapped to PDP-11 addressing modes (never saw any C
until after I graduated).

C predates the PDP-11 so we'll chalk it up to convergent
evolution.  I don't remember much about the PDP-8 architecture
(which I think comes closer to being concurrent with the origin
of C), but I don't think it matched C or the PDP-11 very well.
Wasn't it the really odd stack-based thing, or was that an old
HP machine?  Or maybe the DEC-10...  time to get out the old
computer architecture textbooks.

-- 
Grant Edwards                   grante             Yow!  I want another
                                  at               RE-WRITE on my CEASAR
                               visi.com            SALAD!!


From Rod.Haper at sage.nelesautomation.com  Wed Jun 21 17:59:04 2000
From: Rod.Haper at sage.nelesautomation.com (Rod Haper)
Date: Wed, 21 Jun 2000 16:59:04 -0500
Subject: Visual Tabs in Vim (was: Why whitespace denotation of blocks doesn't 
 work.)
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au> <39505457.5223ADC5@san.rr.com> <20000621155338.C11308@fangorn.csse.monash.edu.au> <slrn8l22bv.ma2.nascheme@cranky.arctrix.com> <39512E52.CC80781A@sage.nelesautomation.com>
Message-ID: <39513AA8.4DF00511@sage.nelesautomation.com>

The following is an improvement to the autocmds block:

if !("autocommands_loaded")
    let autocommands_loaded = 1
    autocmd BufNewFile,BufRead,BufEnter *.py set list
    autocmd BufLeave *.py set nolist
endif

Now this will automatically turn off list mode when you exit a Python
filetype buffer.
I also forgot to mention that the mapping of the F8 function key below
is all one line.


Rod Haper wrote:
> 
> Add the lines below  to your .vimrc file to show <TAB> characters as "?"
> when vim list mode is on, automatically switch on list mode when
> entering a Python filetype buffer and allow taggling of list mode using
> the F8 function key:
> 
> set lcs=tab:?\
> set nolist
> if !("autocommands_loaded")
>     let autocommands_loaded = 1
>     autocmd BufNewFile,BufRead *.py set list
> endif
> map <F8> :if &list <Bar> set nolist <Bar> else <Bar> set list <Bar>
> endif <CR>
> 
> Neil Schemenauer wrote:
> >
> > Toby J Sargeant <tjs at cs.monash.edu.au> wrote:
> > >:set lcs=tab:?\
> > >:set list
> >
> > Unfortunately both those settings are global instead of per
> > buffer.  I guess you could hack some macro to change them when
> > you enter a .py buffer.

-- 
+----------------------------------------------------------------------+
| Rod Haper, Senior Systems Analyst      |  Neles Automation           |
| Phone: 713-346-0652 - direct           |  SCADA Solutions, Inc.      |
|        713-939-0393 - fax              |  7000 Hollister             |
| Rod.Haper at sage.nelesautomation.com     |  Houston, Texas 77040, USA  |
+----------------------------------------------------------------------+


From fiona at sitegnome.com  Sat Jun 24 03:21:24 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 24 Jun 2000 07:21:24 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 24th, 2000
Message-ID: <20000624072124.12950.qmail@synop.com>

Hello Python users,

Below are the entries I've entered into http://python.faqts.com today.

regards, Fiona


## New Entries #################################################


-------------------------------------------------------------
Has anyone created typemaps for SWIG for Microsoft __int64 to bind them to Python long ints?
http://www.faqts.com/knowledge-base/view.phtml/aid/3961
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond

Not specifically - but note that 1.5.2 and later has LONG_LONG support,
meaning you just pass the __int64 directly to Python - eg:

>From pywintypes:
PyObject *PyLong_FromI64(__int64 ival)
{
 return PyLong_FromLongLong(ival);
}
PyObject *PyLong_FromUI64(unsigned __int64 ival)
{
 return PyLong_FromUnsignedLongLong(ival);
}

BOOL PyLong_AsI64(PyObject *val, __int64 *lval)
{
 *lval = PyLong_AsLongLong(val);
 return *lval != -1 || !PyErr_Occurred();
}

BOOL PyLong_AsUI64(PyObject *val, unsigned __int64 *lval)
{
 *lval = PyLong_AsUnsignedLongLong(val);
 return *lval != (unsigned __int64)-1 || !PyErr_Occurred();
}

Should be pretty easy to wrap with SWIG...


-------------------------------------------------------------
Is it possible to configure emacs so that it automatically starts in python mode when opening a file *.py?
http://www.faqts.com/knowledge-base/view.phtml/aid/3963
-------------------------------------------------------------
Fiona Czuczman
Piet van Oostrum

You do this in the standard emacs way:

(setq auto-mode-alist
      (cons '("\\.pyw?$" . python-mode) auto-mode-alist))

or

(add-to-list 'auto-mode-alist '("\\.pyw?$" . python-mode))

It is also useful (at least on Unix) to have

(add-to-list 'interpreter-mode-alist '("python" . python-mode))


-------------------------------------------------------------
Is is possible to dynamically add methods to classes in Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3964
-------------------------------------------------------------
Fiona Czuczman
Ken Seehof, Remco Gerlich

class some_class:
   pass
   
instance = some_class()

def some_method(self):
   print "Hello!"
   
some_class.method = some_method

instance.method()

eg:

>>> class C:
...  pass
...
>>> x = C()
>>> def f(self, z):
...  print self, z
...
>>> C.f = f
>>> x.f(12)
<__main__.C instance at 190d450> 12
>>>


-------------------------------------------------------------
Can anyone tell me how I would go about disabling a pmw combobox?
http://www.faqts.com/knowledge-base/view.phtml/aid/3960
-------------------------------------------------------------
Fiona Czuczman
Richard Chamberlain

In the ToDo list for Pmw disabling is mentioned so hopefully we'll have 
a solution from the source before long. In the meantime...

This is a bit of hack (in the derogatory sense of the word) but hey!

So you basically want to import the myComboBox class and create an 
instance just as you would with Pmw. I've added a couple of methods - 
disable and enable suprisingly enough.

Included below is some test code.

It works in the same way as you mentioned, by unbinding the events, and 
rebinding them to enable it.

Richard


from Tkinter import *
import Pmw
root = Tk()
root.title('Disabling ComboBox Hack')
Pmw.initialise()

class myComboBox(Pmw.ComboBox):
    def disable(self):
        # Rebind things to my empty handler
        self.component('arrowbutton').bind('<1>',self.handler)
        self.component('arrowbutton').bind('<3>',self.handler)
        self.component('arrowbutton').bind('<Shift-
3>',self.handler)
        self.component('entryfield_entry').configure
(state='disabled',fg='grey')

    def enable(self):
        # bind the events back up to the original methods
        self.component('arrowbutton').bind('<1>',self._postList)
        self.component('arrowbutton').bind('<3>',self._next)
        self.component('arrowbutton').bind('<Shift-3>',
self._previous)
        self.component('arrowbutton').configure(takefocus=1)
        self.component('entryfield_entry').configure
(state='normal',fg='black')

    def handler(self,event):
        # so it doesn't propagate the event
        return('break')

# Now test it

comboentries = ("Pmw Should", "have a nicer way", "to
disable", "a combo box")
theEnabler = Button(root, text='Enabled',padx=20, pady=10)
theEnabler.pack(expand=1, fill=BOTH, padx=8, pady=8)

combobox = myComboBox(root, label_text='Disabling Combobox:',
labelpos='wn',
                        listbox_width=24, dropdown=1,
                        scrolledlist_items=comboentries)
combobox.pack(fill=BOTH, expand=1, padx=8, pady=8)

combobox.selectitem(comboentries[0])

def disable(event):
    if theEnabler['text']=='Enabled':
        combobox.disable()
        theEnabler.configure(text='Disabled')
    else:
        combobox.enable()
        theEnabler.configure(text='Enabled')
theEnabler.bind('<Button-1>',disable)
root.mainloop()


## Edited Entries ##############################################


-------------------------------------------------------------
I'm new to Python, where should I start?
Can you give me an overview of the Python Documentation?
Do I need books to learn Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/1356
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Tom Funk,Simon Brunning,Dparsavand, Phil Austin

The handbook is part of the on-line documentation available at the
Python web site and with the Python installation -- and it's free.

You may not need to go to a bookstore if you peruse the following page:

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

The tutorial is quite complete:

  http://www.python.org/doc/current/tut/tut.html

If you work through the tutorial, it should carry you pretty far along
in your quest.  I found it to be *quite* useful.

The Library Reference discusses the modules that ship with Python. 

  http://www.python.org/doc/current/lib/lib.html

The Language Reference is a bit more abstract, and *much* more dry.  
However, it does completely describe the core Python language
constructs, grammar and syntax.   It's often referred to as material for
"Language Lawyers."

  http://www.python.org/doc/current/ref/ref.html

I like the Module Index:

  http://www.python.org/doc/current/modindex.html

It allows you to jump straight to the module of your choice.

If you're using Win32 (as I do), then you may find the MS HTML Help 
version to be useful (it's my favorite).  If you use Win32, you might 
want to check out:

  http://www.orgmf.com.ar/condor/pytstuff.html

Best of all, these very complete works of non-fiction are FREE.... gotta 
love that!  

I own five Python books, but I still find myself referring back to the 
Python documentation regularly.

---

As one newbie to another, I can recommend
<http://www.idi.ntnu.no/~mlh/python/programming.html> for an
introduction to programming, and
<http://www.idi.ntnu.no/~mlh/python/instant.html> for a bit more on
Python. After that, you might want to look at
<http://starship.python.net/crew/amk/grimoire/html/> for a
'cookbook' of useful techniques. If none of these take your fancy,
there are other links at the python site -
<http://www.python.org/doc/Intros.html>.

If you have any questions look at the FAQs -
<http://www.python.org/doc/FAQ.html>

If you follow http://www.pythonlabs.com/ to
http://www.vex.net/parnassus/ and then choose the  Info/Books/Tutorials 
link then Tutorials, you will get quite a few alternatives.

Also, have a look at:

http://www.networkcomputing.com/unixworld/tutorial/005/005.html

http://yhslug.tux.org/obp/thinkCS/thinkCSpy

The current numpy tutorial (supercedes the one that's cited on 
Parassus):

http://numpy.sourceforge.net/

Richard P. Muller's python short course:

http://www.wag.caltech.edu/home/rpm/python_course/

Konrad Hinsen's Python for Science tutorial

http://starship.python.net/crew/hinsen/






From rajarshi at home.com  Mon Jun  5 15:55:08 2000
From: rajarshi at home.com (Rajarshi Ray)
Date: Mon, 05 Jun 2000 19:55:08 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com>
Message-ID: <393C053A.58DD8220@home.com>

Besides the gee-whiz aspect of such algorithms, do they have any value?
After all, what makes pi pi is ALL of its digits in a particular
sequence.

Kirby Urner wrote:
> 
> Here's another "math through programming" challenge:
> 
> Recent number theory breakthrough:  you can get the
> nth digit of PI without doing all the work needed
> to compute all the digits up to the nth digit (same
> with some other numbers -- not sure if we know all
> the relevant criteria as to which).
> 
> The original algorithm worked for PI expressed as
> a hexadecimal number, but I found a web page purporting
> to give the algorithm for a base 10 PI, but I couldn't
> figure out how to make it work in Python.[1][2]
> 
> Anyone?
> 
> Another question:  does the algorithm for getting
> the nth digit of PI actually do less work (in terms
> of flipflops) than would an algorithm that gets
> all digits up to n inclusive?  I assume so, as I
> read about someone getting a trillionth digit or
> something -- but I don't think we've got PI to that
> many places.
> 
> Kirby
> 
> [1] http://www.mathsoft.com/asolve/plouffe/plouffe.html
> [2] http://www.lacim.uqam.ca/plouffe/Simon/articlepi.html

-- 
Sir, I have found you an argument. I am not obliged to find you
an understanding.

                                               - Samuel Johnson


From buzzard at urubu.freeserve.co.uk  Mon Jun 19 20:47:39 2000
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Tue, 20 Jun 2000 01:47:39 +0100
Subject: Naive question
Message-ID: <8imf14$dff$1@newsg3.svr.pol.co.uk>

I run the following script (Test2.py), query f.id at the command-line prompt
and get NameError: f.


def start():
    f = Forest()


class Forest:
    def __init__(self):
        self.name = 'Forest'
        self.id = 1
        self.graphs = []
        self.count = 0
        self.d = DiGraph('initial', self)

class DiGraph:
    def __init__(self, label, pForest):
        self.owner = pForest
        self.name = label
        self.id = pForest.count
        pForest.graphs.append(self.id)
        pForest.count = pForest.count + 1


>>> from Test2 import *
>>> start()
>>> f.id
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
NameError: f
>>>


If I comment out the first two lines and create the Forest instance at the
command line, then I can query f.id etc.


#def start():
#    f = Forest()


class Forest:
    def __init__(self):
        self.name = 'Forest'
        self.id = 1
        self.graphs = []
        self.count = 0
        self.d = DiGraph('initial', self)

class DiGraph:
    def __init__(self, label, pForest):
        self.owner = pForest
        self.name = label
        self.id = pForest.count
        pForest.graphs.append(self.id)
        pForest.count = pForest.count + 1

>>> from Test2 import *
>>> f=Forest()
>>> f.id
1
>>> f.d.owner.id
1
>>>


How can I achieve this without creating the Forest instance at the command
line?  Cheers.  Any help appreciated.

Duncan




From jkraska1 at san.rr.com  Wed Jun 21 16:41:03 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 21 Jun 2000 20:41:03 GMT
Subject: Python complements C++ for productivity
References: <NEBBIOGBDMDPEANLFKGHKEGJCEAA.jschmitt@vmlabs.com>
Message-ID: <395128A4.3D70A4E4@san.rr.com>

John Schmitt wrote:
> 
> When I first opened Pythonwin, I used the Python/Perl phrasebook.  Very
> useful to me.  I don't know the STL very well (I don't even know what the
> SWIG is) so I couldn't write one, but how hard would be to write a Python
> <-> C++/STL phrasebook?

Well, while there is much more to the STL these days than
in its early years, in many ways the STL is about containers,
and iterating over them. While in C++, this is a complicated
subject matter, in Python, it's not.

There are really only "two and a half" container classes
in Python:

Dictionaries,
Lists (which are analagous to C++ vectors),
and Tuples (which are immutable lists).

I haven't seen any of these phrase books, but Python is just
soooo easy. There's really not much to it.



C/


From herzog at online.de  Wed Jun 28 19:56:19 2000
From: herzog at online.de (Bernhard Herzog)
Date: 29 Jun 2000 01:56:19 +0200
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3k8f9cqcc.fsf@greebo.nodomain.de>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> Given the following directory structure,
> 
> python/
>    AAA.py
>    BBB/
>       __init__.py
>       AAA.py
>       
> if sys.path contains 'python', then 
> 
> from AAA import *
> from BBB.AAA import * 
> 
> would do the right thing at most locations, except when pwd is python/ and
> when '.' is before 'python' in sys.path, in which case they both import from
> python/AAA.py.

If your cwd is python then the sys.path entry for python/ and '.' are
the same directory. Why would that lead to problems? The only thing I
can think of is that other directories in sys.path also have AAA
modules.

> This situation occurs quite often in practice.  For example, AAA could be a
> module for doing something.  BBB could be a wrapper package for doing things
> somewhat differently.  So within BBB/AAA.py one calls 'import AAA' to get
> the real job done, and oops, it imports itself!

That's the way packages are designed in python, regardless of your cwd.
'import xyzzy' in a module in a package will look for a module 'xyzzy'
in the same package first and if not found there will look in the
directories in sys.path.


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


From ktlehmaNOktSPAM at uswest.com.invalid  Sat Jun 10 11:05:34 2000
From: ktlehmaNOktSPAM at uswest.com.invalid (Kip Lehman)
Date: Sat, 10 Jun 2000 08:05:34 -0700
Subject: Programming Poll
References: <8hrim7$fjv$1@nnrp1.deja.com> <20000609210521.A6218@bigskytel.com> <3941DCEC.5E5A0C06@stud.ntnu.no>
Message-ID: <2e506186.ac63c595@usw-ex0102-084.remarq.com>

Take the ending comma off the URL, you'll be able to reach
the site.

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



From ame at swipnet.se  Wed Jun 21 16:13:39 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Wed, 21 Jun 2000 22:13:39 +0200
Subject: PIL and HPGL
Message-ID: <4682ls4ukdce8uipv7hcgfjkihv6teoppj@4ax.com>

Hello!

I need to create a lot of PDF files of some drawings (made in AutoCAD)
and I thought I try to do it with Python. The easiest way should be by
using PIL to read hpgl/2 and then create a PDF. One catch though.....
PIL can't read HPGL/2. 

Anyone that have created this 'plug-in' for PIL?
Anyone that knows where I can get the specification of HPGL?

// Anders



From sdm7g at virginia.edu  Fri Jun 16 16:11:07 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 16 Jun 2000 16:11:07 -0400 (EDT)
Subject: Lot's of one-time-pads available for $15 each! [was: Nth digit of PI]
In-Reply-To: <8idus1$enb$1@slb3.atl.mindspring.net>
Message-ID: <Pine.A32.3.90.1000616155834.16829E-100000@elvis.med.Virginia.EDU>

On 16 Jun 2000, Aahz Maruch wrote:


> Close, but no cigar.  What you're describing that's vulnerable is the
> use of a book as an encryption key rather than a one-time pad.  If it
> were literally a one-time pad, there'd be no particular physical marks.
> Of course, I don't think the concept of a one-time pad existed as such
> back then.

  Well -- the concept of one time pads goes back to the 1920's or earlier,
and the ones the German's used were written down on paper -- the "one
time" implied that they were destroyed after they were used. 
  It's a modern day computer notion that even better than a real physical
pad of paper that can be intercepted or stolen, is a shared algorithm that 
generates a random one time pad. 
  But your "Close, but no cigar" comment is justified on other grounds: 
the one time pad is supposed to be random, and using a non random pad 
makes it theoretically vulnerable. I just wonder how practically 
vulnerable it is, given a extremely large choice of non-random bit-streams
to choose from. 

 I believe the idea of using a book text as an easily shared but very
large key is a pretty ancient idea --certainly predating the one-time pad.
Does anyone have a reference ? I can't remember where I first saw it 
mentioned. 


-- Steve Majewski 
   Cut wood, carry water. 




From cjensen at bioeng.ucsd.edu  Fri Jun 23 20:48:46 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Fri, 23 Jun 2000 17:48:46 -0700
Subject: simple question...
References: <3953e13a.761314922@news1.on.sympatico.ca>
Message-ID: <3954056E.E3C18DF@bioeng.ucsd.edu>

chibaA at TinterlogD.Tcom wrote:
> 
> Hello,
> 
> I'm sure this is a simple question...
> 
> how would I call another python script from within a script.  ie. I
> want to basically stop the script that's running, and start running
> another.  I don't have any particular functions in the target script,
> and I don't want to continue running the script I'm currently in (ie.
> I don't want to use import xxxxx).  I would also like to avoid a
> system call if possible.
> 
> Any ideas?
> 
> TIA!

I'm not sure what you're asking.  However, anytime you want to run a
script, you have to import it.  If you run it from the command line,
it's implicitely imported.










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


From wlav at hpatl26.cern.ch  Wed Jun 14 04:30:53 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 14 Jun 2000 08:30:53 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB333@ALTNET> <8hr8fj$fu$1@sunnews.cern.ch> <B9e05.1145$vp3.81025@news1.rdc2.tx.home.com>
Message-ID: <8i7frt$o02$1@sunnews.cern.ch>

"Rainer Deyke" <root at rainerdeyke.com> writes:

>Wim Lavrijsen <wlav at atlas03.cern.ch> wrote in message
>news:8hr8fj$fu$1 at sunnews.cern.ch...
>> All that STL containers require is that your dtor and copy assignment
>> don't throw. The compiler generated members fullfill that requirement.

>This statement is both incorrect and misleading.

Bold words.

>Incorrect because there
>may be other requirements: for example, std::set (along with several others)
>requires that operator< and operator== are defined for the object.

Nonsense. You can specialize std::set with your own version of comparison
class instead of the standard less and you can make sure that your version
of it does not require said operators.

>Misleading because for many classes the compiler generated functions cause
>incorrect behavior.

As I have explained in another post, 'many' is only true for buggy code
that is not exception safe.

Of course, your statements only underline the original point, that is, that
the STL is to difficult to use for C++ novices. However, although it has
its flaws, the STL is well designed and all the simple things have been
taken care of real good.

Best regards,
         Wim Lavrijsen



From tom-main at REMOVEME.home.com  Tue Jun 27 11:40:58 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 27 Jun 2000 15:40:58 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <eW365.71228$7o1.1703374@news2.rdc1.on.home.com>

This should not be ignored (as suggested above).

Python's ability to seamlessly use COM is valuable to many using Python on
Windows, and is a potential inducement to VB programmers to switch to
Python.

Now MS claims that the environment that C# targets will be language neutral.
If this is true, and if Python can be used to develop for this environment,
then this too will be valuable to a reasonable number of Python (or
potential Python) developers.

BTW, MS claims that they have already submitted C# to the European standards
body that Sun screwed over.  I assume that the catch is that they are not
submitting the associated run-time environment.  Does anybody know?

Tom.

"Don Tuttle" <tuttledon at hotmail.com> wrote in message
news:qJ165.8938$nZ5.153196 at typhoon.southeast.rr.com...
> Consider this a heads up NOT a promotion!!
>
> In case you haven't heard Microsoft is coming out with a massive new Web
> stratagy that, in part, uses a new language called C# (That sharp, like in
> music).  Here's a part of what they have to say:
>
> "Because of its elegant object-oriented design, C# is a great choice for
> architecting a wide range of components-from high-level business objects
to
> system-level applications. Using simple C# language constructs, these
> components can be converted into Web services, allowing them to be invoked
> across the Internet, from any language running on any operating system."
>
> "More than anything else, C# is designed to bring rapid development to the
> C++ programmer without sacrificing the power and control that have been a
> hallmark of C and C++. Because of this heritage, C# has a high degree of
> fidelity with C and C++. Developers familiar with these languages can
> quickly become productive in C#."
> .......
>
> "The modern design of C# eliminates the most common C++ programming
errors.
> For example:
>
> Garbage collection relieves the programmer of the burden of manual memory
> management.
> Variables in C# are automatically initialized by the environment.
> Variables are type-safe. "
>
> Hmmm, sound familar?
> Here's the link
> http://www.msdn.microsoft.com/vstudio/nextgen/technology/csharpintro.asp
>
> I look forward to you comments.
>
> Don
>
>
> --
> Believe nothing no matter where you read it or who has said it, not even
if
> I have said it, unless it agrees with your own reason and your own common
> sense.
> --Buddha
>
>




From jjlucsy at concentric.net  Fri Jun 23 14:58:49 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Fri, 23 Jun 2000 14:58:49 -0400
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <JEk45.262$Sn2.1571@news-server.bigpond.net.au> <slrn8l3rpi.8ku.r2d2@mao.acc.umu.se> <m0o45.6859$5f7.130092@news-east.usenetserver.com> <slrn8l4cmn.e2m.r2d2@mao.acc.umu.se> <g4m6ls0orjo2djeevc3i6ia2vg089mul8l@4ax.com>
Message-ID: <6qO45.11009$Q7.326348@news-east.usenetserver.com>

Yes.

"Toby Dickenson" <mbel44 at dial.pipex.net> wrote in message
news:g4m6ls0orjo2djeevc3i6ia2vg089mul8l at 4ax.com...
> Does that version include the winsound module?

-
Joel Lucsy (jjlucsy at concentric.net)





From dgomezm at etheron.net  Mon Jun 12 22:44:01 2000
From: dgomezm at etheron.net (dgomezm at etheron.net)
Date: Tue, 13 Jun 2000 02:44:01 GMT
Subject: Wondering about Domingo's Rational Mean book"
References: <pn5djsceirmrdhdd2ncht9c26hv0q5gkdq@4ax.com> <ralyh3rlx4xa@forum.mathforum.com> <8hp1g8$47o$1@barcode.tesco.net>
Message-ID: <8i475e$48n$1@nnrp1.deja.com>


Very sorry, for this message appearing twice, my fault.


Subject:      Re: Wondering about Domingo's Rational Mean book"
Author:       Iain Davidson <Sttscitrans at tesco.net>
Organization: Tesco ISP
Date:         Sat, 10 Jun 2000 23:27:24 +0100


Domingo Gomez Morin wrote:
>> Given a set of two initial fractions:
>> [3/2, 4/3]
>> According to my definition and notation (see home page),
>> Rm[{3/2, 4/3},{2*3/2*2, 4/3}]=[7/5,10/7]
>> Rm[{7/5, 10/7},{2*7/5*2, 10/7}]=[17/12,24/17]
>> Rm[{17/12, 24/17},{2*17/12*2, 24/17}]=[41/29,58/41]
>> Rm[{41/29, 58/41},{2*41/29*2, 58/41}]=[99/70,140/99]
>> and so  on....
>> is the most simple example of a Rational Process for aproximating
>> the square root of 2 (yielding sets of two fractions whose product
>> is always 2).
>> Thus, as everyone can see, this very simple rational process
>> yields two column of values, the first one
>> (3/2, 7/5, 17/12, 41/29, 99/70,...) and
>> the second one
>> (4/3, 10/7, 24/17, 58/41, 140/99,...)


"Iain Davidson" <Sttscitrans at tesco.net> replied:

>I don't see why you choose [3/2, 4/3]
>as the two initial fractions and why you choose
>to multiply top and bottom by 2.
>What you seem to be doing is choosing two fractions
>a/b and c/d such that ac= 2l and bd = 1.

I always do prefer to bring numerical examples, this brings
more fun to many people.
The set of two initial fractions used in a rational process
for approximating the square root of any number P  is:
[a/b, P*b/a]

It is clear that  [3/2, 4/3] and consequently all the other new sets
satisfy the well known condition for generating reduced fractions  |ad-
bc| = 1.
This particular rational process is not only related to CFs
but to Daniel Bernoulli?s method for solving algebraic equations.
As a matter of fact Bernoulli?s method for  solving
second order algebraic equations is exactly the same procedure that
the CFs expressions of algebraic numbers of second order (Traditional
CFs).

I didn?t feel any need for explaining the above facts, specially when
considering that my favorite point is about the significance of the
lacking on any precedents on such trivial rational processes in the
whole history of mathematics,
worst when considering that by agency of the rational mean
(simple arithmetic) you can easily find Bernoulli?s,
Newton?s and Halley?s methods, among other infinite number
of new methods.
That is what I call "general and fundamental".
Unfortunately, it seems that you are not willing
to discuss such a very important point, very important indeed,
specially for all those readers-purchasers of math books  who
have never read anything about such trivial way (simple arithmetic)
for finding Bernoulli?s, Newton?s and Halley?s methods.

What do you think a humble reader-purchaser of math books
should say about that?

Just say: Oooops.....just two or three thousands years to reach that?

I think one should say:
Are all those math-books on warranty?
What about "rigorousness" in mathematics?

This a very important point for me and some others,
and hope no to be plunged into a one-way discussion.


>While the fractions get closer and closer to sqrt(2), there
>is no guarantee that best approximations will be found.


Well, let me try to explain all this in a more general way,
even though it is very difficult by using this media:

****(Notice that I won?t use the symbol _ as subscripts, I don?t
find it neccesary for the following explanation, so the
numerical value of the variables C1, C2, a1, a2, b1 means subscripts)

It is well known that for any  CF whose coefficients
are:
Sucesive numerators: b1, b2, b3, b4,...
Sucesive denominators: a1, a2, a3, a4,...

that is:  CF= b1 + b2/(a1+ b3/(a2+ b4/...

Being C1, C2, C3, ... the successive convergents of the CF
then each convergent can be computed in terms of the two
preceding convergents by using a lineal homogeneous
recurrence relation (LHRR) of the second order:

Thus being the first two convergents of the CF:
C1= b1/1,  		let?s denote this fraction: N1/D1
C2= (b1*a1 + b2)/ a1, 	let?s denote this fraction: N2/D2

then the third convergent is:

C3= (a2*(b1*a1 + b2)+ b3*b1)/ (a2*a1 + b3*1) =
N3/D3= (a2*N2+ b3* N1)/(a2*D2+b3*D1)

As you can see according to the general definition of
the rational mean stated in my page:
www.etheron.net/usuarios/dgomez/RmDef.htm
the convergent C3 is just the rational mean between the
two preceding convergents C1 and C2.
Of course, all this also applies for all convergents Ci.

Therefore,  all CFs are just a rational process ruled by the rational
mean, including of course the particular case of the
simple CFs (mediant, reduced fractions, b1=1, b2=1, b3=1,...).
All this is fully explained in my book and even though
I?m pleased to answer any question, it is hard to put all this
by using Newsgroup format. An english version of my book
is coming out soon and will be available via internet.



Any convergent, any reduced fraction,
any best approximation condition (|ad-bc| = 1), any new idea,
that is, anything  coming out from Cfs concept is fully embraced
by the Rational mean concept and the Rational Process.
The rational mean rules traditional CFs.
So if you have some doubts, something to say,  about the rational mean,
then your arguments are straightaway directed to the CFs, Bernoulli?s,
Newton
and Halley?s method, because all of them are ruled by the rational
mean as stated in my book and summarized in my home page:
www.etheron.net/usuarios/dgomez/
All of them are just rational processes ruled by the rational mean.

I find of so much importance to consider the striking fact that western
culture needed more than two or three thousands years to find
Bernoulli?s,
Newton?s and Halley?s methods by means of the infinitesimal calculus
and the cartesian system, also that all  their most outstanding
mathematicians never realized that such methods could have been
easily developed by using simple arithmetic.
That?s just striking, specially when considering that any
algebraic equation can be formulated in terms of a "parametric"
equation (may be the term "multiple proportion" would be more
adequate),
a very important point that allows us to find the roots
of any algebraic equation by agency of the rational mean, that is,
by means of simple arithmetic.
See for example, the most simple (trivial) example on this matter
at the end of my page:
www.etheron.net/usuarios/dgomez/arithmonic.htm
which deals with the golden mean (assuming x=p/q you can formulate
the equation x^2+x=1 in terms of a "parametric" equation (proportion),
then applying the rational process as shown in the web page, so any new
set of two approximations of x is always satisfying the golden
proportion.


Just striking, indeed, specially for a humble purchaser
of "rigurous" math books containing so much rigorous talk about
best approximations and roots solving.



>No, you are making the claim "Again: the rational process embraces
_all_
>CFs."
>There are CFs based on generalizations of Euclid"s algorithm that find
>simultaneous approximations to three quantities, say, Cubrt(4):Cubrt
(2): 1
>So how does the rational process embrace these ?



I think that by means of this very long message
(very sorry for its length) I have clearly stated
that the rational mean rules the CFs, so they are just part of a
rational process of second order.
So, again,  if you have some doubts, something to say,
about the rational mean, then your arguments are immediately
directed to the traditional CFs, Bernoulli?s, Newton and
Halley?s method, (among others) because all of them are
ruled by the rational. That?s it. I think we could save a lot of time
if you were could  read the contents of :
www.etheron.net/usuarios/dgomez/RmDef.htm
and
www.etheron.net/usuarios/dgomez/Roots.htm
which explain sufficiently the rational mean and rational process
definition

Now, considering that this not a one-way discussion I hope
you take some time to formulate here those CFs method you mentioned
for finding a sequence of best approximations for the cube root of
any number, I?m sure it will be amusing and very useful not only
for me but for the whole audience of this newsgroup.
Also I would like to get any response about the other topics
I have mentioned on the precedents and history of developing
Bernoulli?s, Newton?s, and specially Halley?s method
by means of simple arithmetic, if I?m wrong on this then
there must be, at least, two or three thousands years, sorry,
I meant to say "thousands of references" (books, articles, etc)
on such a very simple arithmetical matter.
:-)


Domingo




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


From wjdandreta at worldnet.att.net  Tue Jun 20 20:56:16 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 00:56:16 GMT
Subject: Dictionary vs if-elif?
References: <UlS35.15101$Xx5.781176@bgtnsc06-news.ops.worldnet.att.net> <rg10ls89rd67ufpp48s2pr8j78q657kbdm@4ax.com>
Message-ID: <QoU35.15351$Xx5.796413@bgtnsc06-news.ops.worldnet.att.net>

Hi Ralf,

That looks good , I had not thought of it.

Bill

>bogusCodes = ( '1.', '31', '90', 'AZ' ...... )
>...
>if item in bogusCodes:
>  do something useful





From cobrien at Radix.Net  Fri Jun 30 14:20:39 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 30 Jun 2000 14:20:39 -0400
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com> <ku66qtv945.fsf@lasipalatsi.fi> <395B6D35.DA83A7ED@estinc.com>
Message-ID: <8jiodn$jc8$1@saltmine.radix.net>

In article <395B6D35.DA83A7ED at estinc.com>,
Eric Lee Green  <eric at estinc.com> wrote:

[big snip]

>Last time I looked at ZOPE, it was not clear how to integrate new Python
>methods with it. A shame, it looks like a great product. Maybe I'll get a

Adding external python functions to zope is painless.  You put the .py file
in the Extensions directory and add a "ExternalMethod" from the web
management interface.  

You can also wrap python classes around queries, i.e. to calculate
results from the raw data in the database, and make it look like there
are extra columns.  They call these "Pluggable Brains" of all things.


>chance later on to really get into it and figure all that stuff out, it looks
>like it would solve some of the performance problems I've encountered using
>PostGreSQL and Python with Apache. 
>

Zope will provide a persistant connection to the database, and can
optionally cache results.  There are some newer PostgreSQL adapters
that are supposed to be thread safe, so you can one PostgreSQL
connection per thread, with a thread pool to handle requests.

Zope does take a while to get your head around.

-- cary


From tismer at tismer.com  Sun Jun  4 15:17:25 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 04 Jun 2000 21:17:25 +0200
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <393AA3E6.253C88D1@tismer.com>
Message-ID: <393AAB45.1C7256F4@tismer.com>

[Forth Python]
> One idea is to adopt the OTA standard which
> you can find at http://www.europay.com .

Sorry this is the right URL:
http://www.europay.com/smartcard/Smartcard_ota_solutions.html

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



From sdm7g at Virginia.EDU  Mon Jun 26 20:48:01 2000
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Mon, 26 Jun 2000 20:48:01 -0400 (EDT)
Subject: Darwin (Not Python, yet...)
In-Reply-To: <8j7skg$1f$1@bob.news.rcn.net>
References: <Pine.A32.3.90.1000624142526.18059A-100000@elvis.med.Virginia.EDU> <8j7skg$1f$1@bob.news.rcn.net>
Message-ID: <Pine.A32.3.90.1000626195403.20434A-100000@elvis.med.Virginia.EDU>

http://softrak.stepwise.com/ finds two python packages:

 A binary package for Mac OS X Server from Scott Anguish. 
 It doesn't seem to be a Debian package, which is what Darwin wants. 
 I would guess it used whatever packaging NeXT used. 
 Unpacking in manually and trying to run python I get: 

    dyld: ./python Undefined symbols:
    __dummy
    __objcInit 

 So the OSX Server version really needs OSX Server shared libraries. 

 The other version is an OS X Source distribution by Serge Beauchamp
 which says that it's the "Same as the standard unix distribution except
 that I changed the Make files so that it can successfully compile 
 and install on a HFS+ volume." 



     I assume this was the case insensitivity fix. 


     BTW:  <http://www.mit.edu/people/wsanchez/papers/USENIX_2000/> 
      is a paper on "The Challenges of Integrating the Unix and Mmac OS
      Environments"  which has some interesting info about some of the
      differences in HFS+/UFS filesystem semantics and MacOS/Unix 
      file semantics. 


 
  This version actually has the binaries included and they do run 
  under Darwin! Problem is, it is distributed as a .zip file, which
  (I think) unpacked under MacOS9 (/w Stuffit Expander) with the 
  wrong line endings, so that some of the imports don't seem to 
  work properly. ( I think this is what's wrong -- I've seem similar
  symptoms on Mac when unix .py files get transfered as binaries.) 

  This also seems to confuse make. It can't find the targets. 

  This zip file doesn't seem to be in gunzip format so I couldn't 
  unzip it from Darwin to get the right line endings. 

  When I run it, I get:
 
    'import exceptions' failed; use -v for traceback
    Warning! Falling back to string-based exceptions
    'import site' failed; use -v for traceback
    Python 1.5.2 ( ... 
    Copyright ... 
    >>> 
  
 
    >>> import posix 
    >>> dir(posix) 

    returns what I expect, but importing a text .py file like:

    >>> import os
    >>> dir(os) 

   returns only: 
    ['__builtins__', '__doc__', '__file__', '__name__' ]  


  I'll try to find another way to unpack the .zip file and try again. 
  Still ironing out a few kinks, but it's nice to know that the 
   binaries are compatible. 

  If there are typo's in the message it may be because the keyboard 
   under X seems a bit glitchy compared to the same keyboard under 
   either the Darwin console or Mac OS9. 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
                       CHOP WOOD, CARRY WATER.


From petersc at stud.ntnu.no  Tue Jun 20 01:10:55 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 20 Jun 2000 07:10:55 +0200
Subject: Naive question
References: <8imf14$dff$1@newsg3.svr.pol.co.uk>
Message-ID: <394EFCDF.7EDAE76F@stud.ntnu.no>

Duncan Smith wrote:
> 
> def start():
>     f = Forest()

"f" is variable local to the start() function. When
start exits (after executing its only command) there
is no reference left for f and it is cleaned up.
Even if it was not cleaned up, it would still be local
and thereby unaccessible to you.

> How can I achieve this without creating the Forest instance at the command
> line?  Cheers.  Any help appreciated.

The obvious solution would be:

def start():
    return Forest()

If for some reason you don't want to have the corresponding
f = start() call, you can always leave out these evil
"from x import *" statements and use a variable global to
your module:

f = None

def start():
    global f
    f = Forest()

then you cann access it by
mymodulesname.f after running mymodulesname.start()

Doing this with the "from x import *" statement does not
work as the f imported thereby will still reference None.

If the start method is only to be run once, than you can
put the call to it inside the module:

def start():
    return Forest()

class Forest:
    pass

f = start()

I'd prefer the obvious solution (just use return Forest())
though. Why do you want to do it like that anyway?

what-channel-is-the-timbot-on-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From parker at gol.com  Sun Jun 18 17:57:31 2000
From: parker at gol.com (Ian Parker)
Date: Sun, 18 Jun 2000 21:57:31 GMT
Subject: [FAQTS] Python Knowledge Base Update -- June 16th, 2000
References: <20000616111056.18561.qmail@synop.com>
Message-ID: <Nkh6gDAQVUT5EwmS@gol.com>

In article <20000616111056.18561.qmail at synop.com>, Fiona Czuczman
<fiona at sitegnome.com> writes
>
>Hi!
>
>My arrangement with FAQTS was to get 225 entries into 
>http://python.faqts.com by June 16.  I've done it! With a lot of help 
>from you guys.  Thanks everyone.
>
>This doesn't mean my daily entries will cease, I'm going to continue 
>working on the FAQTS Python Knowledge base for a while to come.  Its fun 
>and educational :-)
>
>Regards,
>
>Fiona Czuczman
>


Fiona

Congratulations but please don't stop yet.  This is good.

Regards

Ian



From rumjuggler at cryptarchy.org  Wed Jun  7 18:23:02 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Wed, 07 Jun 2000 22:23:02 GMT
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com> <p2l%4.40729$WS3.454680@typhoon.we.rr.com>
Message-ID: <hsitjs0135i7o2o1umcfqs070gf62n16fk@4ax.com>

On Wed, 07 Jun 2000 05:26:45 GMT, "Stephen Hansen"
<stephen at cerebralmaelstrom.com> wrote:

>Ack. _Why_? Ick, yick, sick!

Well, you know, any step to make Python less readable is bound to
result in its increasing adoption by Perl programmers.

I'm only trying to help!

-- 
Barnabas T. Rumjuggler

This is the Great Theatre of Life. Admission is free but the taxation
is mortal. You come when you can, and leave when you must. The show is
continuous. Good-night. 
 -- Robertson Davies, _The Cunning Man_


From npnigro at my-deja.com  Thu Jun 29 13:40:35 2000
From: npnigro at my-deja.com (npnigro at my-deja.com)
Date: Thu, 29 Jun 2000 17:40:35 GMT
Subject: ExitWindowsEx()
Message-ID: <8jg1mk$e1s$1@nnrp2.deja.com>

I'm having a problem executing ExitWindowsEx() to
force reboot my Win95/Win98 machine.  The
following is my code:

import win32api
win32api.ExitWindowsEx(4 Or 2,0)

The call does not reboot the machine as it is
supposed to.  Am missing something?  I'm using
build 132... Thanks in advance

Nick


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


From kain at dynup.net  Wed Jun  7 18:09:24 2000
From: kain at dynup.net (Scott Barron)
Date: Wed, 7 Jun 2000 22:09:24 GMT
Subject: SQL Query Question
References: <8hm9as$ho3$1@news-int.gatech.edu>
Message-ID: <Fvsyn3.Ms@boss.cs.ohiou.edu>

Perhaps you are looking for something like this:

query_string = "SELECT AAA FROM BBB WHERE cc = %s"

then you could do:

query_function (query_string % `pp`)  # if pp is an integer type

or
query_function (query_string % pp)   # if pp is a string type

or change %s to %d or whatever,  depending on what pp is.

-Scott

In article <8hm9as$ho3$1 at news-int.gatech.edu>, liang at cad.gatech.edu
(Shengquan Liang) wrote:
> 
> 
> if i need a variable to be used as the creteria for selection, how do i
> construct the query?
> 
> say i have a variable pp=1000  in the program and i'd like to  use the
> following query:
> 	SELECT AAA FROM BBB where cc = 1000
> 
> if i want to replace that 1000 with the variable 'pp' in the Python
> program, how do i modify the query statement?
> 
> thanks.




From Wolfgang.Grafen at marconicomms.com  Wed Jun 21 08:44:06 2000
From: Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen)
Date: Wed, 21 Jun 2000 14:44:06 +0200
Subject: Dictionary vs if-elif?
References: <UlS35.15101$Xx5.781176@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <3950B896.18C0E757@marconicomms.com>

William Dandreta wrote:

> Here is a code snippet
> -------------------------------------------------------------------------
> bogusCodes = { '1.':'', '31':'', '90':'', 'AZ':'', 'BH':'', 'C1':'',
>                'CB':'', 'FR':'', 'HD':'', 'JF':'', 'KL':'', 'KU':'',
>                'LU':'', 'ON':'', 'PF':'', 'Q3':'', 'UP':'', 'W4':'',
>                'WO':''}
>
> item  = data.readline()
> while item:
>   if (item[1:2] in digits) and (item[0:1] == 'D'):
>     goodVend.write(item)
>   elif vendors.has_key(item[0:2]):
>     goodVend.write(item)
>   elif bogusCodes.has_key(item[0:2]):
>     pass
>   else:
>     missingVend.write('Vendor code %s is missing from file VendCode.txt\n' %
> item[0:2])
>   item  = data.readline()
> ---------------------------------------------------------------
> originally I had it written like so:
>
> if ...:
>   pass
> elif...:
>   pass
> etc.
> -------------------------
> another possibility is and if statement with a very complex condition:
> if a or b or c ...:
>   pass
> else :
> etc
> ---------------------
> Is there a better way to do it?
> Which of these three would be the preferred method?
>
> Bill

Hi Bill,

try a very fast version:

bogusCodes = ['1.', '31', '90', 'AZ', 'BH', 'C1',
              'CB', 'FR', 'HD', 'JF', 'KL', 'KU',
              'LU', 'ON', 'PF', 'Q3', 'UP', 'W4',
              'WO']

DDIGITS    = map(lambda x:'D%s'%x,range(10)

vendorCodes = ['2.','41']  # as an example

def PASS(key):
    pass

dict = {}

for key in bogusCodes:
    dict[key] = PASS

for key in vendorCodes + DDIGITS:
    dict[key] = goodVend.write

while 1:
    item = data.readline()
    try:
        key = item[:2]
        dict[key](item)
    except KeyError:
        missingVend.write(
          'Vendor code %s is missing from file VendCode.txt\n'%item[0:2])

Wolfgang



From loredo at spacenet.tn.cornell.edu  Thu Jun  1 15:09:49 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 01 Jun 2000 15:09:49 -0400
Subject: Stand-alones on the Mac?
References: <393645EF.17424CDA@kfunigraz.ac.at>
Message-ID: <3936B4FD.ECA88284@spacenet.tn.cornell.edu>

Siegfried-

Mac-specific Python info is here:

http://www.cwi.nl/~jack/macpython.html

Post your message to the PythonMac SIG (Special Interesty Group):

http://www.python.org/sigs/pythonmac-sig/

Lots of Python-on-Mac experts there!

> And I isntalled also the Mac version of Python and
> also there I could not found anything about Python and stan-alones for
> the Mac.

Look a little harder; there is an HTML file somewhere in the
distribution that explains this in some detail.  I'm not on
a Mac at the moment so I can't tell you where it is exactly....

Yes, you can make stand-alone Python applications on the Mac
that do not require the user to install Python.  MacPython
comes with the Numerical module pre-installed, so you needn't
worry about that.  It also has Mac-specific GUI modules to
build the user interface.  You can write graphical output
using the PIL library, also included.  To do x-y plotting is
more problematical; you might want to look at the PIDDLE module.

Check the PythonMac SIG for more expert advice.

Good luck!
Tom Loredo


From stephen at cerebralmaelstrom.com  Thu Jun 22 05:24:51 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 22 Jun 2000 09:24:51 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <DXk45.79957$WS3.716413@typhoon.we.rr.com>

I have absolutely *no* idea if it will run with Python 1.2, or under DOS,
but when I was looking for database support awhile back, someone pointed me
to MetaKit, and I absolutely *love* it...

http://www.equi4.com/metakit/

Creating a database is as easy as...

>>> import Mk4py
>>> mk = Mk4py # Done for ease-of-typing!

>>> db = mk.Storage("filename.mk",1)
>>> table =
db.getas("mytable[StringField1:S,IntField1:I,StringField2:S,BLOBField:B]")
>>> table.append(StringField1='This is a string',IntField1 = 2,StringField2
= 'Lalala',BLOBField = 'Binary data stored here :P')
>>> table[0].StringField1
'This is a string'

You use the 'getas' to dynamically create, or alter, the structure of the
database/table/view... later, if you don't want to change the structure, you
can just use "db.view('name')".

--S



William Dandreta <wjdandreta at worldnet.att.net> wrote in message
news:Uw945.16309$Xx5.881927 at bgtnsc06-news.ops.worldnet.att.net...
> Hi Anders,
>
> >Do you REALLY need to use dBASE? Are you interacting with some other
> >programs that uses the same database?
>
>
> I don't really need to use Dbase but I need something that works with DOS
> and is platform independent.
>
> I found XBase which is an open source C++ database library that uses dbase
> data structures and is platform independent. If Python does not already
have
> Dbase access, I guess I will have to figure out how to put a Python
wrapper
> around Xbase.
>
> >Maybe you could use SWIG to create a Python module for using it.
> >If you're not in a hurry I could maybe do it, but not until late
> >august.
>
> What is SWIG?
>
> >// Anders
> >
>
>




From thomasm at legend.co.uk  Tue Jun 13 08:38:00 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Tue, 13 Jun 2000 13:38:00 +0100
Subject: No Boolean type in "types"
References: <Pine.GSO.4.10.10006131512230.20303-100000@sundial>
Message-ID: <39462acb.0@katana.legend.co.uk>

> > Can someone explain me why the module "types" have no boolean type
defined ?

I knew this, sorry I wasn't clear enough,
Considering a "structural" point of view, having true and defined in a
standard way
and being able to "indicate" that the type is Boolean is a nice programming
practice.
In C one will use TRUE and FALSE. Those define are no more than integer too
but help
code readability. That is my point.

If types don't define this every programmer "have the charge" to do it.
Python credo isn't "one way to do it" ??

Thomas




From stianh at ifi.uio.no  Tue Jun 13 04:08:47 2000
From: stianh at ifi.uio.no (Stian Husemoen)
Date: 13 Jun 2000 08:08:47 GMT
Subject: pipes and popen2
Message-ID: <slrn8kbqug.pd0.stianh@localhost.localdomain>

I was a bit bored this weekend, and passed time coding python. In one
snippet I wanted to make a UI to ispell. The idea was to create a
bidirectional pipe, keep that pipe open, and feed ispell with words to
spell. Like in this pseudo-code:

	stdout, stdin = popen2.popen2('ispell -a')
	
	for word in wordlist:
		# spell this
		stdin.write('pyhton')

		# result
		print stdin.readline()

Problem is popen2 wont give me any output until I close stdin. Is there any
way to both read and write to a program while keeping the pipe and
process open? I can't seem to find any postings or documentation on this
subject.


Cut-down-a-tree-with-a-herring?-It-cant-be-done-ly y'rs
-- 

Stian Husemoen <stianh>


From mvanier at endor.bbb.caltech.edu  Thu Jun  8 20:43:57 2000
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 08 Jun 2000 17:43:57 -0700
Subject: tkinter -- why doesn't this work?
Message-ID: <jl66rj3cqa.fsf@endor.bbb.caltech.edu>

Hi,

Here's a trivial tkinter script meant to catch and print key events in a
frame.  It doesn't work on my system (RH Linux 6.1, pentium pro, python 1.6a2
or python 1.5.2), but if I modify it to catch (say) button presses or
anything else but key bindings it does work.  Any ideas?

TIA,

Mike

--------------------------------------------------------------------------
#! /usr/bin/env python

from Tkinter import *

root = Tk()

def callback(event):
    print dir(event)
    print "key: %s" % event.char 
    
frame = Frame(root, width=200, height=200)
frame.bind("<Key>", callback)
frame.pack()
root.mainloop()
--------------------------------------------------------------------------


--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.


From cut_me_out at hotmail.com  Tue Jun  6 23:54:28 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 23:54:28 -0400
Subject: Example of Python code (C code to Python code)
References: <Ych%4.758$0l2.6362@news2.tor.primus.ca>
Message-ID: <etdpupu9mdn.fsf@w20-575-109.mit.edu>

def get_float (prompt):
    while 1:
        number = raw_input (prompt)
        try:
            number = float (number)
        except ValueError:
            print 'Please enter a numerical value (just ' \
                  'digits and maybe a period.)'
            continue
        break
    return prompt

hours = get_float ('Please enter the hours worked per week? ')
rate  = get_float ('Please enter the rate of pay? ')

regular_hours = 40

if hours > regular_hours:
    overtime = hours - regular_hours
    overtime = overtime * rate * 1.5
else:
    overtime = 0

gross = (hours - overtime) * rate + overtime

min_salary = 300

if gross > min_salary:
    tax = (gross - min_salary) * .23
else:
    tax = 0

net = gross - tax

print 'Gross: %f, Net: %f' % (gross, net)




From pete at visionart.com  Wed Jun  7 13:50:54 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 07 Jun 2000 10:50:54 -0700
Subject: Why not a Python Package Manager
References: <393f42e4.77383471@news.online.no>
Message-ID: <393E8B7E.12751F79@visionart.com>

maybe a step towards what Debian has with its DEB packages?

some sort of python central library with all packages
in a standard environment. there would be precompiled
packages for maintained architectures and/or source.
each package would list its dependencies

it would then be trivial to download and choose which
python packages you wanted since they could automatically
take care of their dependencies.

after downloading source and creating the module
(with distutils in 1.6) you could have the option
to submit that module back to the repository as a 
precompiled version for your architecture?

heh, anyways just a wild idea that struck me.
likely a lot of work, but then again definitely 
worth the investment it seems.

now i need someone to get inspired and make it happen :]


From embed at NOSPAM.geocities.com  Tue Jun 27 09:47:16 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 27 Jun 2000 09:47:16 -0400
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com>
Message-ID: <4f265.9121$227.218658@nnrp1.uunet.ca>

I find callbacks from C to python to be no problem. I use a simple
publish/subscribe event model implemented as a C extension that has a
dictionary with the event names as the keys (strings), and the callbacks as
a list of Python functions.  The C code is both portable and elegant, IMHO.

Adding COM just to get callbacks would be ugly for two reasons:

1. You already have TWO sets of  primitive types (C builtin variable types,
and Python's native types), why mix in a third?

2. COM on Linux is a fudge, and not a very stable one at that. There is a
COM implementation for Linux but it's not open source, so if you have a
problem, you can't fix it, you're stuck.


Warren

<scoota at my-deja.com> wrote in message news:8ja7gv$687$1 at nnrp1.deja.com...
> Hi,
>
> I'm writing an automation interface for a large system embedding Python
> as the script language.  To keep the scripts as easy to write as
> possible (for the end-user) I'm providing a set of extension objects
> written in C++.  What I would ideally like is for these objects to fire
> events which are then handled in the Python script.  So far I can see
> no platform independent way of doing this except using explicit
> callbacks (which is messier than I would like).
>
> Has anyone any experience in this area?
>
> Scoota
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From 12klat at sightreader.com  Thu Jun  1 01:24:30 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Wed, 31 May 2000 22:24:30 -0700
Subject: Calling Function in DLL on Windows
References: <dThZ4.101952$fF5.3223396@news1.rdc1.il.home.com>
Message-ID: <3935F38E.FF0061DC@sightreader.com>

dsavitsk wrote:

> i need to call a DLL from within a python program. in VB i would declare the
> function like...
>
> Declare Function LoadSFW Lib "Sfwlnk32" (ByVal strPath As String, ByVal
> strConfig As String) As Long
>
> then use it later like...
>
> intErrorCode = LoadSFW("C:\SFW", 6)
>
> i don't know how to do this in Python, so any quick explanation would be
> helpful.
>
> doug

I don't know of any easy way to do it.  The hard way to do it would be to
write an extension module that wraps your DLL function(s).  If you don't
know how to do this, read about "Extending Python".  If you don't know
C or C++, I'm afraid you are probably out of luck until someone implements
a generic solution....

Here's an interesting project for Windows programmers: Write a general
purpose DLL caller.  Functions should be declared in python (like in VB)
and the DLL should be accessible from python without requiring the user
to write any C code.  This is doable (VB does it).  The hard part will be
converting the DLL function arguments.

 - Ken Seehof




From fgbrown at ionet.net  Wed Jun 14 21:59:23 2000
From: fgbrown at ionet.net (F. G. Brown)
Date: Wed, 14 Jun 2000 20:59:23 -0500
Subject: Math statements and Freeze problem
Message-ID: <8i9d2e$t45$1@ionews.ionet.net>

Greetings
I haven't exactly hit a wall on this one, it's more like a thorn in my
side.I have two lines that figure conversion from Fahrenheit to Celsius and
vice versa.  Before I started writing this I was almost sure that I would
need only one line and the two conversions would both reference the same
line. But that hasn't been the case, as you can see below.  Was it just my
imagination or is there a more "elegant" way to do this? Also I tried to run
the Freeze program on this (first time) and something is not working right.
Does this have to be done at a DOS prompt or is it done in the python
shell,IDLE or PythonWin?
I think there is some kind of problem with my path statement, what does
yours look like?
Regards,
Fred


????????????????????????

# temp.py
         # description of program
print "The following program converts the values"
print "of Fahrenheit and Celsius. Select the known"
print "value you have below."
decide = 'y'
while decide == 'y':
         # selects known value to determine if or else statement
         print
         print "If the known temperature is in"
         print "degrees Fahrenheit enter (1) if"
         print "Celsius enter (2)"
         degrees = raw_input("Fahrenheit (1) or Celsius (2)?   ")
         if degrees == "1":
              fahrenheit = input (" What is the temperature in degrees
Fahrenheit?   ")
              # is there a better way to do the math?
              celsius = (fahrenheit - 32)*.556
              print "That equals %0.1f degrees Celsius" %celsius
         else:
              celsius = input (" What is the temperature in degrees
Celsius?")
              # ditto
              fahrenheit = (celsius *1.8)+32
              print "That equals %0.1f degrees Fahrenheit" %fahrenheit
              # gives user choice of converting another
         decide = raw_input ("Another? (y)es or (n)o")
         if decide == "y":
              print 'OK, no problem'
else:
              print 'See you later!'






From DOUGS at oceanic.com  Sat Jun 10 17:42:45 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Sat, 10 Jun 2000 11:42:45 -1000
Subject: PATCH: Augmented assignment 
Message-ID: <8457258D741DD411BD3D0050DA62365907A202@huina.oceanic.com>

[Christian Tanzer wrote:] 
> > It doesn't obfuscate code looking like:
> > 
> >    nested_structure [some_index] [another_index] = 
> nested_structure [some_index] [another_index] + 1
> > 
> > And it saves more than a few keystrokes.

[Moshe Zadka replied:] 
> Anyone doing *that* isn't  familiar with the pydiom:
> 
> s, i = nested_strucre[some_index], another_index
> s[i] = s[i] + 1

Wow, I like that.  Its great!!!

Could you pass on the link to that storehouse of great pydioms.
I'd like to see more  ;-Q

but-I-guess-the-point-was-there's-usually-already-a-clearer-way-to-write-it'
ly yr's
-Doug-




From jkraska1 at san.rr.com  Tue Jun 13 02:06:50 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 13 Jun 2000 06:06:50 GMT
Subject: can python do this?
References: <3945c87a$0$505$7f31c96c@news01.syd.optusnet.com.au>
Message-ID: <3945D022.63951249@san.rr.com>


Q: "Can python do this?"
A: "Python was born for it, baybee!"

import string
file = open ("file")
lines = file.readlines()
for line in lines:
    pair = string.split(line,":")
    print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
file.close()

--

>>> import process
FORM: ['Popular Recording']
CMTYPESWITCH: ['Popular']
RELOCORIGINAL: ['55864A2603AC94ACCA2568F200224388']
CMCOUNTRY_ORIGIN: ['']
CMGENRE_CODE: ['']
**CPTITLE: ['INFINITE POSSIBILITIES']
**CPARTIST: ['AMEL LARRIEUX']
CPOARTISTS: ['']
**CMCATNUMBER: ['4948792']
**CMAPN: ['9399700067507']
**CMARIAPRICECODE: ['C9921']
**CMARIAMEDIACODE: ['21']
**CMARIADISTRIBUTORCODE: ['SMA']
CMARIAPACKAGECODE: ['']
**DMRELEASE: ['19/06/2000']
cMDateFormat: ['19/06/2000 12']
DMDELETE: ['']
cMDeleteStatus: ['No']
**CMLISTPTRACKS: ['GET UP', 'I N I', 'SWEET MISERY', "SEARCHIN' FOR MY SOUL", 'EVEN IF', 'INFINITE
POSSIBILITES', 'SHINE', 'DOWN', 'WEATHER', 'MAKE ME WHOLE', 'GET UP (THREAD HAD FUN MAIN MI', 'GET
UP (MV MIG MIX (RADIO))']
CMLISTPARTISTS: ['', '', '', '', '', '', '', '', '', '', '', '']
RELOCMEDIA: ['F73C80183A01A8D7CA2568F2002A5517']
**CURMPRICE: ['15.91']
**CMTYPE: ['Compact Disc']
CMRECORDCOMPANY: ['SONY MUSIC']
CMPACKAGE: ['']
CMARIADISTRIBUTORHOUSECODE: ['']
cDistributorHouse: ['']
CMLOCALE: ['Y']
$UpdatedBy: ['CN=PPT/OU=AEROS/O=JUKEBOX']
$Revisions: ['02/06/2000 05']
>>>


From michael at rcp.co.uk  Wed Jun 28 12:24:39 2000
From: michael at rcp.co.uk (Michael Abbott)
Date: Wed, 28 Jun 2000 17:24:39 +0100
Subject: Current date and time
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com> <20000620010356.A892@better.net> <395017fb@newsprime.tidalwave.net>
Message-ID: <8jd8s7$pik$1@panther.rmplc.co.uk>

Try
    time.strftime('%A %d %B %Y %H:%Mhr', time.gmtime(time.time()))

The library documentation is quite adequate to tell us this.

"Aeneas @hotmail.com>" <shan_mu<nospam> wrote in message
news:395017fb at newsprime.tidalwave.net...
> How do you format the result as--
> "Tuesday  20 June 2000  21:18hr" ?
> Thanks.





From moeller at dophnic.yi.org  Sun Jun 18 14:08:59 2000
From: moeller at dophnic.yi.org (Derek Moeller)
Date: Sun, 18 Jun 2000 13:08:59 -0500
Subject: Python 3000 -> No, but there's 'Ruby'
In-Reply-To: <394C4E3D.73DBB25C@bogusaddress.com>; from JoeSmith@bogusaddress.com on Sun, Jun 18, 2000 at 04:17:59AM +0000
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <394C4E3D.73DBB25C@bogusaddress.com>
Message-ID: <20000618130858.A13798@dophnic.yi.org>

On Sun, Jun 18, 2000 at 04:17:59AM +0000, Joe Smith wrote:
[html]

I would respond privately, alas, you don't seem to have a valid
email address for responses (@bogusaddress.com).

So, just try not to use HTML mail in the future. It's annoying,
has no added advantage over ASCII for mail, and makes it difficult
to read for people with mail readers that have chosen not to have
embedded web browsers (after all, that's what a web browser is for,
not a mail reader).


-- 
Derek Moeller

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \



From mwh21 at cam.ac.uk  Fri Jun  9 08:48:01 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 13:48:01 +0100
Subject: pyc -> py can i transform it?
References: <3940B9F5.639FD84A@muk.tu-darmstadt.de>
Message-ID: <m3pupruike.fsf@atrus.jesus.cam.ac.uk>

Frank Panse <panse at muk.tu-darmstadt.de> writes:

> Can I transform pyc files to py files?

There's decompyle at 

http://www.csr.uvic.ca/~aycock/python/content.html#de-about

(good google-able name, that one).

Writing a decompiler using my bytecodehacks package is one of my
perpetual projects, but I'd advise quite strongly against holding your
breath until I finish it...

It's not exactly hard to use `dis' and write equivalent code to any
given module, but it's not exactly fun either.

Cheers,
M.

-- 
  There are 'infinite' number of developed artifacts and one cannot
  develop appreciation  for them all.  It would be all right to not
  understand something, but it would be imbecilic to put judgements
  on things one don't understand.               -- Xah, comp.lang.lisp


From tiddlerdeja at my-deja.com  Mon Jun  5 11:35:26 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Mon, 05 Jun 2000 15:35:26 GMT
Subject: set Registry User Security
Message-ID: <8hghbt$5bc$1@nnrp1.deja.com>

In regedt32 you can assign user privilleges for a Key.

Can you do this from within Python, perhaps using the win32api?


Any help appreciated.


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


From igorr at ifi.uio.no  Mon Jun 19 17:33:11 2000
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 19 Jun 2000 23:33:11 +0200
Subject: split() with a separator set -> thank you
References: <XFMail.000619174023.mikael@isy.liu.se>
Message-ID: <xjv1z1ts6eg.fsf@gjallarhorn.ifi.uio.no>

* Mikael Olofsson

[snip]

Thanks a lot for the hints, everyone.



ivr, 're'-solution seemed to be the easiest :)
-- 
Simula-konsulent? Simsulent? "I eat Java for breakfast!".
                                           -- thorkild


From wjdandreta at worldnet.att.net  Thu Jun 15 18:39:59 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 15 Jun 2000 22:39:59 GMT
Subject: List of functions called in Python source files?
Message-ID: <3Xc25.3507$Uw3.210307@bgtnsc04-news.ops.worldnet.att.net>

I need to get a list of the external (ones not in the same source file)
functions called in each Python source file. Does anyone know an easy way to
do this?

Bill




From thomas at xs4all.net  Mon Jun 19 04:18:16 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 19 Jun 2000 10:18:16 +0200
Subject: Why is Python so slow ?- revisited.
In-Reply-To: <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net>; from wjdandreta@worldnet.att.net on Mon, Jun 19, 2000 at 03:22:06AM +0000
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com> <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <20000619101816.X26436@xs4all.nl>

On Mon, Jun 19, 2000 at 03:22:06AM +0000, William Dandreta wrote:

> Christian Tismer wrote in message
> <394D6579.8B2FF840 at appliedbiometrics.com>...

> >The time spent in comment lines is very small,and it is reasonably
> >smaller if you use Stackless Python. Anyway, if you are concerned
> >about multi-line comments, then try this out:
> >Put a """ line in front of a longer couple of comments,
> >and end it the same way.
> >I should wonder if execution time would change reasonably.

> The """ syntax would seem to be slower than putting # at the begining of
> everyline because with # it only needs to test the first character of the
> line, if it's # it can skip the rest of the line. But with """, it needs to
> check every character after until it finds the matching """.

The tokenizer has to search those ends together anyway, and besides, that's
a compiletime penalty, not runtime. It won't matter a hoot in function
calls, as all that happens string literals in a void context is that they
get added to the functions' constants tuple (function.func_code.co_consts.)
And that happens at compiletime (which can be at runtime, of course, if you
use exec() or import an uncompiled module, but if you do that often you
shouldn't whine about performance ;-)

However, all this is based on Python 1.5.2, *not* 1.2. I dont think there
are many people who can give useful tips about 1.2 performance, other than
'either upgrade or stop caring about it.'. I believe quite a lot has changed
since 1.2 ;-P

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From liang at cad.gatech.edu  Wed Jun  7 15:55:08 2000
From: liang at cad.gatech.edu (Shengquan Liang)
Date: 7 Jun 2000 19:55:08 GMT
Subject: SQL Query Question
Message-ID: <8hm9as$ho3$1@news-int.gatech.edu>


if i need a variable to be used as the creteria
for selection, how do i construct the query?

say i have a variable pp=1000 
in the program
and i'd like to 
use the following query:
	SELECT AAA FROM BBB
	where cc = 1000

if i want to replace that 1000 with the variable 'pp'
in the Python program, how do i modify the query statement?

thanks.


From schneiker at jump.net  Wed Jun  7 06:32:07 2000
From: schneiker at jump.net (Conrad Schneiker)
Date: Wed, 7 Jun 2000 05:32:07 -0500
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
Message-ID: <8hl866$bt5$1@news.jump.net>

Well, at least _some_ people regard Ruby as a better OO Perl than Perl or as
a means of getting anticipated Python 3000 features now. Maybe if you knew
more about Ruby, you might become one of them.

Obviously for a great many people, situations, and applications, Perl or
Python are clearly their best respective choices. (I still use Perl from
time to time and I still sometimes recommend that people check out Python.)

However I suspect there are still a great many other people who generally
like Perl or Python, but who for reasons of need and temperament are not
locked into them, and who also would like something that provides a more
powerful and more uniform and easier to use variation on the same sorts of
themes that made Perl or Python initially attractive and useful power tools.
The purpose of this note is simply to let such people know about Ruby. (I
discovered Ruby in a passing reference I chanced across when perusing
comp.lang.python last year, and I am trying to pass on that inadvertent
favor in this experiment.)

1 About Ruby.

1.1 What is Ruby?

    Ruby is a relatively new, very high level, fully OO language that
    integrated many good ideas from Perl, Python, Smalltalk, Eiffel,
    ADA, Clu, and Lisp. (Ruby is more fully OO than Python.) In
    addition to common OO features, Ruby has threads, singleton
    methods, mix-ins, fully integrated closures and iterators, plus
    proper meta-classes.  Ruby is a relatively easy to learn and easy
    to read language, yet it is very powerful and sophisticated. Ruby
    has a true mark-and-sweep garbage collector, which makes code more
    reliable and simplifies writing extensions. For many of us, Ruby
    is "better Perl than Perl" and a "better Python than Python". You
    could characterize the Ruby philosophy as "there's a better way to
    do it" (TABWTDI).

1.2 Where can I find out more about Ruby?

    Ruby's home web site:

        http://www.ruby-lang.org/en (Ruby English language home page.)

            Follow the links to documentation, downloads, the Ruby
            application archive, the Ruby mail list archives, and lots
            of other interesting information.

        http://www.ruby-lang.org/en/whats.html (Ruby feature list.)

    Ruby FAQ:

        http://www.rubycentral.com/faq/

    Ruby classes, modules, and methods reference:

        http://www.rubycentral.com/ref/

    English language Ruby books:

        Two forthcoming English-language Ruby are planned for
        publication later in Y2K:

            <<Don't know title yet.>> by Yukihiro Matsumoto

            "Pragmatic Ruby" by Dave Thomas and Andrew Hunt

    Ruby newsgroup:

        comp.lang.ruby




From jcm at bigskytel.com  Thu Jun  1 05:20:41 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 1 Jun 2000 03:20:41 -0600
Subject: Python sighting in the wild: Red Hat's Anaconda GUI installer
In-Reply-To: <3934AF57.706E14C7@softhome.net>; from etoffi@softhome.net on Wed, May 31, 2000 at 02:21:11AM -0400
References: <JUUY4.38612$HG1.681183@nnrp1.uunet.ca> <3934AF57.706E14C7@softhome.net>
Message-ID: <20000601032041.C1234@bigskytel.com>

* etoffi <etoffi at softhome.net>:
> 
> ps: i have a biased dislike towards lisp, and parentheses in general.
> 

That's just because you don't use GNU-Script (yet).

http://catalog.com/hopkins/text/gnuscript.html

  david.
  
-- 
Four red rare deer.  Four lean hounds crouched low and smiling the
cruel bugle sang before. -- e.e.cummings in a blender



From tim_one at email.msn.com  Tue Jun 13 12:39:24 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 13 Jun 2000 12:39:24 -0400
Subject: How many times inner functions are compiled ?
In-Reply-To: <960897680.1890969865@news.libertysurf.fr>
Message-ID: <000e01bfd555$ef3f4220$582d153f@tim>

[Jerome Quelin]
> def outer():
>     def inner():
>         print "spam"
>     inner()
> outer()
> outer()
>
> In the snippet code above, I call outer() two times. I would like to
> know how many times python will compile the inner function ?  One time
> when compiling the script/module, or two times, ie at every invocation
> of the outer function ?

Just van Rossum gave you the best answer, but since he was out-voted 2 to 1
I thought I'd jump in to even the odds <wink>.  "inner" is compiled exactly
once, independent of how often "outer" is called (even if you never call
outer, "inner" is compiled once).  However, the "def" stmt is executable in
Python, and at runtime, each time "outer" is called, upon hitting the "def
inner():" block the (pre)compiled code object for "inner" is wrapped up in a
function object and bound to "outer"'s local vrbl name "inner".  The
"function object" adds stuff like values for default arguments to the code
object (but you don't have any default arguments in this example, so you'll
have to use your imagination).

yet-another-reason-python-isn't-a-democracy<wink>-ly y'rs  - tim





From scherbi at bam.com  Thu Jun 22 12:55:55 2000
From: scherbi at bam.com (Bill Scherer)
Date: Thu, 22 Jun 2000 16:55:55 +0000
Subject: mailing list duplicates?
References: <C5A7720BDEB7D3118A5200E018C200A05D40EA@EXCHANGE>
Message-ID: <3952451B.E1CAE9D7@bam.com>

OK. This is getting serious now.  Barry, are you out there, what's the
scoop?  If this goes on much longer, I'm going to unsubscribe for week
before I check back in to see if it's been fixed.

If somebody *is* working on this problem, can we *please* have a
progress report of somekind?

Thanks.

    desparate-enough-to-use-a-timbot-sig-for-the-first-time-ly yrs,

    Bill Scherer

Gaetan Corneau wrote:

> Yes, a lot.
>
> ______________________________________________________
>    Gaetan Corneau
>    Software Developer
>    Copernic.com
>    http://www.copernic.com
>    E-mail: corg at copernic.com
> ______________________________________________________
> "Golf is a game whose sole aim is to hit a very small
> ball into an even smaller hole, with weapons
> singularly ill designed for that purpose." -- Winston Churchill
>
> > -----Original Message-----
> > From: python-list-admin at python.org
> > [mailto:python-list-admin at python.org]On Behalf Of Bill Scherer
> > Sent: Wednesday, June 21, 2000 8:34 AM
> > To: python-list at python.org
> > Subject: mailing list duplicates?
> >
> >
> > Is anyone else seeing a lot of duplicates on the list?
> >




From cartho at mordor.ch  Thu Jun  8 03:17:31 2000
From: cartho at mordor.ch (Cyrille Artho)
Date: Thu, 08 Jun 2000 09:17:31 +0200
Subject: TypeError: unsubscriptable object
References: <393ECFC3.879BC357@mordor.ch> <8hmp6i$ets$6@newshost.accu.uu.nl>
Message-ID: <393F488B.37F13682@mordor.ch>

> What's possible (but it's not possible to determine this from your snippets)
> is that you accidentally assign something else to W elsewhere.
self.W is equal to
[[0, 0, 0], [0, 0, 0], [0, 0, 0]]
just before the assignment.
self.W[i][j] = self.W[i][j]  + ...
I attached the entire script to this post (it's not too big).
--
Regards,
Cyrille Artho - http://artho.com/ - Tel. +41 - [0]1 - 313 08 92
Alcohol is the anesthesia by which we endure the operation of life.
                -- George Bernard Shaw
-------------- next part --------------
import copy
import random
import whrandom
import math

class NN:
    global n0, alpha, nInput, nNeuron
    n0 = 0.2
    alpha = 0.01
    nInput = 2
    nNeuron = 2
    
    def __init__(self):
        self.V = [0, 0, 0]
        self.W = [[0,0,0], [0,0,0], [0,0,0]]


    def playRound(self): # one round of rolling dice
        X1 = random.choice(range(5)) + 1
        X2 = random.choice(range(5)) + 1
        X3 = random.choice(range(5)) + 1
        I = [0, 0, 0]
        I[0] = 1
        I[1] = min(X1, X2)
        I[2] = abs(X1 - X2)
        S = copy.deepcopy(I)

        # calculate output
        P = copy.deepcopy(I)
        sum = 0
        for i in range(nNeuron+1):
            # input
            for j in range(nInput+1):
                sum = sum + self.W[i][j]*I[j]
            P[i] = 1 / (1 + math.exp(-sum))
            if (P[i] > whrandom.random()):
                S[i] = 1
            else:
                S[i] = 0
            # hidden layer
            sum = 0
            sum = sum + self.V[i]*S[i]
        P = 1 / (1 + math.exp(-sum))
        if (P > whrandom.random()):
            O = 1
        else:
            O = 0

        # analyze result
        YA = I[2]
        if (abs(X1 - 3.5) >= abs(X2 - 3.5)):
            Xmax = X1
        else:
            Xmax = X2
        YB = abs(Xmax - X3)

        if (O == 1):
            Y = YA
        else:
            Y = YB

        if (Y == max(YA, YB)):
            r = 1
        else:
            r = 0

        n = n0 * (max(YA, YB) - min(YA, YB))

        # change weights
        for i in range(nNeuron):
            for j in range(nInput+1):
                print self.W
                print self.W[i][j]
                self.W[i][j] = self.W[i][j]  + n * (
                    r(S[i] - P[i]) - alpha * (1 - r)
                    * (S[i] - (1 - P[i]))) * I[j]
            self.V[i] = self.V[i] + n * (
                r(O - P) - alpha * (1 - r) * (O - (1 - P))) * S[i]

if __name__ == '__main__':
    n = NN()
    n.playRound()

From petersc at stud.ntnu.no  Tue Jun  6 03:09:23 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 06 Jun 2000 09:09:23 +0200
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com>
Message-ID: <393CA3A3.C701EB54@stud.ntnu.no>

Kirby Urner wrote:
> 
> The original algorithm worked for PI expressed as
> a hexadecimal number, but I found a web page purporting
> to give the algorithm for a base 10 PI, but I couldn't
> figure out how to make it work in Python.[1][2]

Can you give the algorithm in some kind of pseudo code?
Pseudo python would be appreciated ...

If I understood it right the calculation has to be done
like this for every pair of primes in the denominator:

1) calculate k1 and k2 for 1/(a*b) = k1/a+k2/b
2) calculate m1, m2, m3 for (k1/a+k2/b)/c = m1/a+m2/b+m3/c

So that means one basically needs a continued fraction
algorithm and a factorization algorithm, right?

thinking-the-article-is-not-really-clear-about-it-ly
y'rs Peter

BTW: The algorithm for base 16 works ... very slowly:

def hexdcm(a,b,d,e):
  x = long(a % b)
  if d:
    for k in range(d):
      y = x << 4
      x = y % b
  else:
    y = a
  c = [y-x]
  for j in range(e):
    y = x << 4
    x = y % b
    c.append(y-x)
  sum = 0
  for j in range(e+1):
    sum = sum + (1L << ((e-j) << 2)) * (c[j] / b)
  return sum

def sum(d,e):
  sum = 0L
  for n in range(d+1):
    sum = sum + hexdcm(120L*n**2L+151L*n+47L,\
(8L*n+1L)*(2L*n+1L)*(8L*n+5L)*(4L*n+3L),d-n,e)
  return sum 

def pi(d,e):
  s = sum(d,e) % 16L**(e+1L)
  return int((s - s % 16L**e) / 16L**e)

def upto(n):
  for i in range(n+1):
    if not i % 50:
      if i:
        print s+"\n  ",
      else:
        print "3.",
      s = ""
    s = s + "%x" % pi(i+1,4)
  print "\n"
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From gtalvola at NameConnector.com  Fri Jun  2 16:42:54 2000
From: gtalvola at NameConnector.com (Geoff Talvola)
Date: Fri, 02 Jun 2000 16:42:54 -0400
Subject: Calling Function in DLL on Windows
References: <dThZ4.101952$fF5.3223396@news1.rdc1.il.home.com>
Message-ID: <39381C4E.434783F3@NameConnector.com>

dsavitsk wrote:

> i need to call a DLL from within a python program. in VB i would declare the
> function like...
>
> Declare Function LoadSFW Lib "Sfwlnk32" (ByVal strPath As String, ByVal
> strConfig As String) As Long
>
> then use it later like...
>
> intErrorCode = LoadSFW("C:\SFW", 6)
>
> i don't know how to do this in Python, so any quick explanation would be
> helpful.

I can think of 3 ways:

 - if it's a COM DLL, use the win32com package.
 - you could write your own extension module.
 - the easiest way is to use Sam Rushing's calldll and windll modules.

For simple stuff, the windll module makes it super-easy.  For example, this puts
up a message box using the Win32 MessageBox function which lives in user32.dll:

from windll import module, cstring
user32 = module('user32')
user32.MessageBox(0, cstring('This is a really easy way to call a DLL'),
                  cstring('Wowee'), 0x30)

Calldll lets you call more complicated functions.

You can download them from http://www.nightmare.com/software.html and you'll
need both the calldll package and the DynWin package, which contains the windll
module.

The "Python Programming on Win32" book contains information like this, and is
really useful if you're using Python in a serious way on Windows.

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com





From boud at rempt.xs4all.nl  Sat Jun  3 14:43:44 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 3 Jun 2000 18:43:44 GMT
Subject: Python GUI requires tcl?
References: <8hbeol$obo$1@samba.rahul.net>
Message-ID: <8hbjl0$611$1@news1.xs4all.nl>

Ed Ohsone <eosn at rahul.net> wrote:

> Is there another graphic package which is more 
> compact, I mean, which requires python + the graphic toolkit only
> without the extra burden of loading another language like tcl?

All other solutions require the GUI toolkit plus a Python wrapper
library. Wxpython even requires three parts: the GUI toolkit,
the Wxwindows C++ wrapper/extension, and the Python wrapper. 

> How about the stability of the python's tk-based graphic package?

It seems pretty stable. However, if you need more complicated widgets,
like treeviews, you've got to make them up yourself from simpler
tk widgets - and you might introduce bugs yourself ;-).

> As far as I know, tk is intricately connected to tcl in its C source.
> Therefore, the python's GUI may be buggier than the original tk based on tcl.
> Of course, it can be less buggier :).
> If you have experience in both, please let me know.

You can safely try all options and determine which one fits your
requirements best. I don't think there are any major problems with
either Tkinter, PyQt/KDE, Wxpython, pyGTK, PyGnome, PyFLTK, Foxpy or
Pythonwin, other than the absolute impossibility to remember the right
CaPiTaliSaTion for each one ;-). Oh, and installation might sometimes,
for some versions, for most toolkits, prove to be a bit of a poser.  

I prefer PyQt/PyKDE myself - it nice, clean, full-featured and stable,
but I think that having all these choices is great.

Go to http://www.vex.net/parnassus/apyllo.py/808292924.247038364 for
an overview of all your choices...

--

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


From nobody at nowhere.nohow  Sat Jun 17 13:07:33 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 17 Jun 2000 17:07:33 GMT
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <pfO25.1380$i06.309216@ptah.visi.com>

On Thu, 15 Jun 2000 10:44:03 +0800, root at sirius.ftsm.ukm.my <root at sirius.ftsm.ukm.my> wrote:

>has anyone wrote the 8 non-attacking queen problem and conway's
>game of life in python? i try but face a few newbie problems.

While sitting around waiting for a software test to fail (I
really hate the bugs that will only happen once every day or
two), I translated my 8-queens program from Scheme to Python:

    <ftp://ftp.visi.com/users/grante/stuff/queens.py>

[If anybody knows how to avoid having to write the bitmap data
to a temp file and the reading it back in with the Button() call,
please let me know!]
    
>2. 8 non-attacking queen.
>
>   use backtracking. the function call itself back. how to do recursive call
>   in python? 

Pretty much the same as most other procedural languages.

roughly, the C program I study look like this;
>   
>   queen_count = -1
>   
>   main()
>   {
>     addqueen();
>   }
>   
>   addqueen()
>   {
>   queen_count++
>   if queen_count != 8
>        addqueen()
>   else
>        print result
>   }
>	
>   how to pass a global variable (queen_count) to a function?

Declare it global in your function:

def addqueen():
    global queen_count
  
    queen_count = queen_count + 1
    
    if queen_count != 8:
        addqueen()
    else:
        print result

queen_count = -1

addqueen()

-- 
Grant Edwards                   grante             Yow!  The PILLSBURY
                                  at               DOUGHBOY is CRYING for
                               visi.com            an END to BURT REYNOLDS
                                                   movies!!


From mikael at isy.liu.se  Tue Jun 13 11:11:46 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 13 Jun 2000 17:11:46 +0200 (MET DST)
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 200
In-Reply-To: <bkhckscbhq9e9i18ijci51eh4r82f3qaqk@4ax.com>
Message-ID: <XFMail.000613171146.mikael@isy.liu.se>

On 13-Jun-00 Anders M Eriksson wrote:
 >  It's a pity that this packages doesn't work on Windows NT! 

The same goes for Win95.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    13-Jun-00
Time:    17:11:17

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


From yawyi at gmx.de  Mon Jun 26 09:29:54 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Mon, 26 Jun 2000 15:29:54 +0200
Subject: chroot'ed demon process
References: <395727E3.D4C1388A@stroeder.com>
Message-ID: <39575bc1$0$18726@personalnews.de.uu.net>

Michael Str?der <michael at stroeder.com> wrote:
>Is it possible to call a chroot() in a demon process running under
>Python after reading all configuration data?

If python would have a chroot(), you would either have to call it
after any modules have been imported or make the Python library
and any shared libraries and support files needed for them
available in the chroot "jail". You cannot access files outside of
the chroot environment (that's the whole point of chroot).

Swigging chroot() or adding it to module posix would be trivial.

Have you read the "Secure UNIX Programming FAQ" at
http://www.whitefang.com/sup/ ?

Also see the chroot utility:
man 1 chroot
info chroot

Johannes



From alex at magenta.com  Mon Jun 12 08:25:46 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 12 Jun 2000 14:25:46 +0200
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com> <8hlr8r$77a$1@newshost.accu.uu.nl>
Message-ID: <8i2l2f022ku@news2.newsguy.com>

Martijn Faassen <m.faassen at vet.uu.nl> wrote in message
news:8hlr8r$77a$1 at newshost.accu.uu.nl...
> Courageous <jkraska1 at san.rr.com> wrote:
> [snip]
> > Well, yeah, but [0:3] as "element zero through three, inclusive"
> > is about as intuitive as you can get. While I understand completely
> > how the current implementation works (reading the source code and
> > all, dontcha know), I still fail to see *why* this was chosen.
>
> a[0:i] + a[i:] == a
>
> a[0:len(a)] == a
>
> Inserting a list at index i:
>
> a[:i] + b + a[i:]
>
> In practice, no need to think about a *lot* of off-by-one errors.

Koenig's excellent book on C's traps and pitfalls is where I
first saw advocacy for the style of "lower bound included,
upper bound excluded" and lots of good reasons for it.  Today,
of course, the style is pervasive in C++'s standard library,
and in Python too.  It's truly a good idea.


Alex





From matt at null.net  Thu Jun 22 18:33:51 2000
From: matt at null.net (Matthew Schinckel)
Date: Thu, 22 Jun 2000 16:33:51 CST
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <20000619184000.639516@210.9.22.17> <8il6sm$g00$1@slb3.atl.mindspring.net>
Message-ID: <20000622163352.938487@210.9.21.161>

Matthew Schinckel <matt at null.net> wrote:
> I have come across one example where re was much faster than using
> string.find: I was stripping out html tags, and doing a one line
> re.sub('<*?>','') was much faster than using string.find() and slice
> notation.

aahz at netcom.com (Aahz Maruch) wrote:
> Sure! But then you're not really doing just a string.find(). The point
> is that for any operation where string.find() alone can be used, it
> will be faster than an re. Any time you add complexity, such as
> find/replace with a pattern rather than a fixed string, chances are
> good that the re will be faster. -- --- Aahz (Copyright 2000 by
> aahz at netcom.com)

Yeah, I would have been better off doing (in fact I did, once I had
figured it out), a re., but what I was doing was:

1. Find the next occurance of '<'

2. Find the next occurance of '>'

3. Keep everything before 1., and after 2.

4. Repeat until there are no more '<' or '>'.

Obviously, this was stupidly slow, and if there were unmatching signs,
caused a huge memory leak, in one case using up 100+Mb of RAM in several
seconds :-)

Matt.





From spamfranke at bigfoot.de  Fri Jun 16 13:50:20 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Fri, 16 Jun 2000 17:50:20 GMT
Subject: Function that knows the names of its actual parameters
Message-ID: <394a6395.7217524@news.btx.dtag.de>

I found myself in need for the following functionality a couple of times:

>>> a=1
>>> b="foo"
>>> c=b
>>> magic(c, b, a)
{'a': 1, 'b': 'foo', 'c': 'foo'}

that is, a function that goes one step further than **dict arguments and returns
a dictionary with names and values of its actual parameters.
It's useful every time one tries to write generators for Python or other code (right
now I'm writing some SQL code emitting functions).

I guess it would be possible to implement using the a similar trick than the one 
to find out the caller's name - generating an exception and examining the
traceback.
But maybe someone knows a more efficient solution or one that has chances 
to work with future Python version as well?

Stefan



From jhe at webde-ag.de  Wed Jun 28 04:48:09 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Wed, 28 Jun 2000 10:48:09 +0200
Subject: XML-code examples - mini-review of Python XML
References: <394f7bc2.365514882@news.online.no> <00062201025104.08285@quadra.teleo.net> <8jbteo$q3v$1@netnews.upenn.edu>
Message-ID: <8jce1g$kc0$1@pollux.ip-plus.net>

"Andrew Diller" <dillera at isc.upenn.edu> schrieb im Newsbeitrag
news:8jbteo$q3v$1 at netnews.upenn.edu...
> I like the book, but I also found it dissapointing. Most of it is an intro
> into the Python language itself.

I planned to purchase the book, but the "contains an introduction to Python"
stopped me when I read the amazon description. And after what I hear from
you, that was the right decision it seems.




From aa8vb at yahoo.com  Thu Jun  8 07:33:37 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 08 Jun 2000 07:33:37 -0400
Subject: CPAN for python?
In-Reply-To: <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us>; from
 akuchlin@mems-exchange.org on Wed, Jun 07, 2000 at 10:05:21PM -0400
References: <393fbb15.100053125@nntp.interaccess.com>
 <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us>
Message-ID: <20000608073337.A324967@vislab.epa.gov>

Andrew Kuchling:
 |Shae Erisson <shapr at uab.edu> writes:
 |> If you did it IETF[1] style, I think people would go with it.
 |> [1] IETF motto: 'rough consensus, running code'
 |
 |The distutils work seems to be freezing down nicely...the basic principle
 |of running setup.py seems stable.

Definitely!  I was amazed how easy it was to package a C extension last week.
Very impressive; the feedback I've gotten from early pkg testers mirrors that.

Open problems: We can install packages, but we can't clean them up (save
ad-hoc methods and the "Microsoft approach" [reinstall the OS/Python]).
Hopefully that's next.  Listing what module dists you have, and handling
module dist dependencies isn't covered yet either.


 |So, what would people want on top of distutils?  A Debian-style "do
 |apt-get, upgrade" to update currently installed packages?  A
 |CPAN-style "fetch and install this module for me, please"?  Simple
 |version checking?

Someday, hopefully:

         setup.py install

or some tool will fetch, build, and install, not just the module dist, but
all of its dependencies too.

I guess I've been spoiled by the FreeBSD ports system.  Fetch the small
port for a package and just kick off "make install".  Under the hood, it
fetches source dists from ftp/www sites, and patches, configures, builds,
and installs the package _and_ all its dependent packages, all
automagically.  Simple (particularly good for the newbies).

Wonder if Python could have something similar.  Folks could post URLs to a
"getit.py" (or setup.py) script, and users could literally cut-and-paste
this URL into a Python install tool that would just "do the install".  That
is, don't bother them with where to FTP or surf, how to get the files,
where to put the packages, what the name of the distdir in the tarball is,
etc.  Just install the package and do what it takes to make it happen.
That'd be cool.

-- 
Randall Hopper
aa8vb at yahoo.com



From ehagemann at home.com  Sat Jun 17 19:42:27 2000
From: ehagemann at home.com (Eric Hagemann)
Date: Sat, 17 Jun 2000 23:42:27 GMT
Subject: Python Grep (was: writing (Gnu)MAKE in Python)
References: <8457258D741DD411BD3D0050DA62365907A23A@huina.oceanic.com>
Message-ID: <D1U25.4636$_54.87028@news1.rdc1.md.home.com>

If your file can fit in memory I believe you will find that reading the
whole thing in (using readlines() ), then searching over the lines is a bit
faster.  I was doing something similar for files in the 5MB range and the
speed dropped from 5 sec to about 2 sec ( I was doing more than the grep but
.....)

Also you might have better luck with the regular expression stuff (rather
than the find command) and precompiling the string

Cheers


"Doug Stanfield" <DOUGS at oceanic.com> wrote in message
news:8457258D741DD411BD3D0050DA62365907A23A at huina.oceanic.com...
> [John said:]
> > > ...I would like to have Bash, sed, and grep as Python
> > > programs rather than compiled C.
> >
> [and Corageous asked:]
> > Why?
>
> Consider this an attempt at translation.
>
> There have been posts in the past about having a Python shell.  Perhaps
> thats what John wants.  Not that I'd like that.  It may be that John wants
> the functionality of sed and grep more easily accessible in Python.  That
is
> something I care about (thus the self serving attempt at manipulating this
> thread. ;-)
>
> Grep functions in particular are something that I wonder about.  The
> following is an experiment:
>
> #!/usr/bin/python
> #
> #       pygrep.py
> #
> #
> """     An attempt to compare the use of the grep
>         command with a 'Pythonic' method of finding
>         all lines in a file that contain a string.
> """
> import string, os
>
> def pygrep(the_string,the_file):
>     """ Search for and return all occurrances of lines
>         in the_file that contain the_string.
>         This is simplistic and unprotected but its
>         purpose is only to learn how to make it fast. """
>
>     find = string.find
>
>     holder = open(the_file,'r')
>     while 1:
>         line = holder.readline()
>         if not line:
>             break
>         if find(line,the_string) <> -1:
>             print line
>
> def mygrep(the_string,the_file):
>     """ This is usually what I do when I need this. """
>
>     command = 'grep %s %s' % (the_string,the_file)
>     response = os.popen(command,'r')
>     lines = response.read()
>     print lines
>
> if __name__ == '__main__':
>     import time
>
>     test_file = "/usr/local/devices/motor.10"
>     test_string = "26086594"
>
>     first = time.time()
>     pygrep(test_string,test_file)
>     second = time.time()
>     mygrep(test_string,test_file)
>     third = time.time()
>
>     print "Python: %s" % ((second - first),)
>     print "OS    : %s" % ((third - second),)
>
> I run this and get:
>
> $ ./pygrep.py
>
Punc,202,26086594,SG6,1,2,30days,-9,53,732970,358877,0,67,176761,39326,1,24.
> 94.89.251,
>
>
Punc,202,26086594,SG6,1,2,30days,-9,53,751771,371322,0,68,179118,40654,1,24.
> 94.89.251,
>
>
Punc,202,26086594,SG6,1,2,30days,-9,53,787679,393889,0,68,184387,42710,1,24.
> 94.89.251,
>
>
Punc,202,26086594,SG6,1,2,30days,-9,53,732970,358877,0,67,176761,39326,1,24.
> 94.89.251,
>
Punc,202,26086594,SG6,1,2,30days,-9,53,751771,371322,0,68,179118,40654,1,24.
> 94.89.251,
>
Punc,202,26086594,SG6,1,2,30days,-9,53,787679,393889,0,68,184387,42710,1,24.
> 94.89.251,
>
> Python: 6.7979799509
> OS    : 0.0741490125656
>
> Am I missing a Pydiom that would narrow the gap or is using the OS
function
> the best way?
>
> -Doug-
>




From nick at spam_me_not_videosystem.co.uk  Sun Jun 11 06:16:24 2000
From: nick at spam_me_not_videosystem.co.uk (Nick)
Date: Sun, 11 Jun 2000 11:16:24 +0100
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl>
Message-ID: <39436762@news.xtml.co.uk>

> >>I have read in several places productivity claims of 5 to 10 times over
> >>'C' and "C++". I would like to ask for some feedback from "C++"
> >>programmers who have moved to Python as to whether these estimates are
> >>realistic.

It depends on the problem domain. I write text parsing/generation code and I
would probably say that 5-10 times increase in productivity is almost fair, once
you take into account the higher level data structures, libraries and lack of
obfusticated syntax. I have written equivalent code in C before I learnt Python
and it was horrendous! You could blame my programming style ;-) but writing
parsers in C, even with Lex & Yacc is a pain. But Python is geared towards
pragmatic programming and I do find that.

> >>2. How long did you use Python before you achieved increased
> >>productivity ?

I've been a user for about 6 months. I found productivity gains immediaitely and
am getting better! Its getting out of the C++ mindset and into the Python data
structure one that takes time! That and learning the Python libraries eg.
sys,os,os.path,string,re etc. You can write much more reusable, unit testable
code in Python more easily. I think the lack of typing really aids reusablity
and code productivity.

> When you write a class in C++ and you want to do it *well*, you have to
> think about a horrible amount of stuff; constructor, destructor, copy
> constructor, assignment operator, possibly other operators, public, private,
> protected. Without the STL you'll have to do lots of manual memory
> management as well; reallocating arrays, etc. If you want to write a
> generic class there's a whole set of extra things to consider.
> I found that this took a long time in C++, while whipping up a decent
> class in Python is a piece of cake in comparison. Even generic Python
> classes.
> It's perhaps just a difference in philosophies as well; C++ is a statically
> typed language and its proponents put a lot of emphasis on this kind of
> correctness. In Python you might come up with lots of correctness rules
> as well, but people don't, as the whole approach is different. People
> rarely care about things like public/private, even. This also means a lack
> of getters/setters in many cases; those will only be created if they
> are deemed necessary. This means encapsulation can be broken easily,
> but Python programmers don't care about this so much -- in practice often
> it is more of a problem to strictly preserve encapsulation (and type safety)
> than it is to fix any problems when and if they arise.

I agree. I was thinking about this over the weekend and I think the main reason
why Python is more productive is the syntax. Constructors, overloading, strong
typing, templates, blah, blah in C++. You have to think about your implementaion
far more in C++ than you do with Python, ie. do I use templates or classes etc.
There's more typing (on the keyboard!!) to be done in C++!!! You just get on
with it in Python! The main penalty is speed of execution which I can live with
know how painlessly I wrote the script!

Cheers,
Nick.






From shapr at uab.edu  Mon Jun  5 08:52:49 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 05 Jun 2000 12:52:49 GMT
Subject: Problems with pickler in comination with pygnome/pygtk
References: <393B9006.A553F608@decrc.abb.de>
Message-ID: <393BA93D.93EBBA4A@uab.edu>

Peter Mueller wrote:
<snip>
> The application code and the pickeled object are exacly the same for
> Redhat6.1/6.2
> And I can't leave out the pygnome stuff, because the app's gui is built
> upon this libs.
> 
> Does anyone have an idea?

In py1.5.2 string.atof expects a string, such as '5000.0' or '5000' and
won't work without one. (at least, that's what my tests show)

Did you write the pickle with the same version of python that you're
using to read the pickle?
If that is true, try adding the 'save as binary' switch to the
pickle.dump command.
ex: dump(object,file,1)

I think one of those two might solve your problem, though neither of
them detect or fix the actual bug :)
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From shouldbe at message.com  Fri Jun 23 12:04:30 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 23 Jun 2000 16:04:30 GMT
Subject: Java Objects vs Pythons?
Message-ID: <iUL45.23838$ds.694847@newsread2.prod.itd.earthlink.net>

I want to do a wait object like in java.  I know Python has Lock and RLock,
but I am not sure on how to use it (not in tutorial).  This is for an
embedded app.  What I want to do is in the script it says wait for the C++
app to tell it it is ok to run.  So, I envision something like this

waitObject = ???<What type of Lock should I use
def callbackNotify():
    waitObject.acquire()
    waitObject.notifyAll()
    waitObject.release()

...
#main do some stuff
waitObject.acquire()
waitObject.wait()
waitObject.release()

In the script I don't create any threads or anything, can I do a wait object
without creating a thread (pretty sure you can't do that in java)?  I guess
there kinda is a thread because there is the python script and then there is
the C++ app with the callback(usually waiting for output to finish on a
printer).  What kind of Lock would be best the Lock or RLock or other?

thanx




From emile at fenx.com  Thu Jun 15 19:49:26 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 15 Jun 2000 16:49:26 -0700
Subject: List of functions called in Python source files?
References: <3Xc25.3507$Uw3.210307@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <008f01bfd724$5d5e0320$1906a8c0@fc.fenx.com>

You're not going to find anything that will
give you a definitive answer.  If you don't
use dynamic importing, eval's or exec's in
your source base, you may be able to get
something useful by extracting all imports,
and looking at the functions that may be
used, or at least get a list of the modules
accessed.

Or-not-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: William Dandreta <wjdandreta at worldnet.att.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, June 15, 2000 3:39 PM
Subject: List of functions called in Python source files?


> I need to get a list of the external (ones not in the same
source file)
> functions called in each Python source file. Does anyone
know an easy way to
> do this?
>
> Bill
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From PClaerhout at CREO.BE  Mon Jun 19 09:39:33 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Mon, 19 Jun 2000 15:39:33 +0200
Subject: Getting remote memory configuration
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03013011F7@msgeuro1.creo.be>

Anyone an idea on how you can find out how much memory a remote computer
has installed? The computer is available on the same network and has the same
login as the computer on which you are running the command.

Pieter



From thiele at muc.das-werk.de  Tue Jun 13 13:57:35 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 13 Jun 2000 13:57:35 -0400
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net>
Message-ID: <642A954DD517D411B20C00508BCF23B00129692B@mail.sauder.com>

>   I do have to wonder whether you've yet learned to think in
> Python; it's possible that you're trying to write C++ code in Python,
> the way some Perl people find writing Python difficult.

May be. But what are typical mistakes?
How should a REAL python program looks like.
I design classes like in C++, thats true. But I also use typical python features
like lists, dicts, map etc. I'm not trying to write everything by myself.

How have I to think as a pythonprogrammer?

(P.S. don't misunderstand me: I like python!)





From bob at slim.instantappointment.com  Mon Jun 19 17:51:32 2000
From: bob at slim.instantappointment.com (Bob Erb)
Date: 19 Jun 2000 21:51:32 GMT
Subject: How does __setitem__() work?
Message-ID: <slrn8kt5iu.2cn.bob@slim.instantappointment.com>

Given the following class definition,

class Test:
    
    def __setattr__(self, attribute, value):
        print 'setting', attribute, 'to', value
        self.__dict__[attribute] = value
    
    def __setitem__(self, index, value):
        print 'setting index', index, 'to', value
        ## THE_BLANK

I have two questions.

First, when is __setitem__ called? __setattr__() is
called when I do:

>>> t = Test(); t.list = [1,2,3]
setting list to [1, 2, 3]

I expected __setitem__ to be called when I do:

>>> t.list[0] = 7

That doesn't happen; when does __setitem__ get
called?

My second question is, to pass the assignment in
__setitem__ through, what code goes in 'THE_BLANK'
above?

(What I'm trying to do is ensure that the value of
a variable is a certain type. For instance, I want
to disallow assignment of a string value to an
integer variable. Is this anti-Python?)

Thanks for any help.

 - Bob Erb


From ivanlan at home.com  Fri Jun  2 20:48:40 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Fri, 02 Jun 2000 18:48:40 -0600
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee> <m3g0qv91f2.fsf@wilma.george.org>
Message-ID: <393855E8.DF87D1DB@home.com>

Hi All--
Harry George wrote:
> 
> Install PyXML, then try this.  Read the xml.dom.core for element names
> and the basic functions.  If you are not familiar with xml itself, see
> the XML Bible.  Try this routine on some known-good .xml file (I can
> supply one if need be.)
> 
> I could post one I did to read/write Abiword files.  My heaviest use
> of python+xml is at work, gluing together a bunch of engineering apps.
> 

[snip]

My brain is hurting from trying to understand XML in two days; please
post *anything* you can to help out!

<ly></ly>-ly y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org/
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From shapr at uab.edu  Wed Jun  7 18:49:40 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 07 Jun 2000 22:49:40 GMT
Subject: Why not a Python Package Manager
References: <393f42e4.77383471@news.online.no> <393E8B7E.12751F79@visionart.com>
Message-ID: <393ED13A.26AC0642@uab.edu>

Pete Shinners wrote:
> 
> maybe a step towards what Debian has with its DEB packages?
> 
> some sort of python central library with all packages
> in a standard environment. there would be precompiled
> packages for maintained architectures and/or source.
> each package would list its dependencies
> 
> it would then be trivial to download and choose which
> python packages you wanted since they could automatically
> take care of their dependencies.
> 
> after downloading source and creating the module
> (with distutils in 1.6) you could have the option
> to submit that module back to the repository as a
> precompiled version for your architecture?
> 
> heh, anyways just a wild idea that struck me.
> likely a lot of work, but then again definitely
> worth the investment it seems.
> 
> now i need someone to get inspired and make it happen :]


I agree. I agree with 'something like Debian' so much I think dpkg and
friends is pretty much the way to go. Why not completely use the debian
package manager?

If you're running Debian, try apt-get install dpkg-python
It's old code (1997 I think), but I think it would be a start for a
CPyAN.
I think that the current maintainer of dpkg-python has moved it to
sourceforge, not sure.

The idea of a Python CPAN has come up before, I think it was decided the
distutils-sig got to deal with that problem.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From smst at bigfoot.com  Tue Jun 13 11:31:27 2000
From: smst at bigfoot.com (Steve Tregidgo)
Date: Tue, 13 Jun 2000 16:31:27 +0100
Subject: importing module through variable
References: <8i5e9p$nm0$1@zingo.tninet.se> <20000613160606.W20700@xs4all.nl> <20000613162155.A12328@beta.telenordia.se>
Message-ID: <394653CF.926817E4@bigfoot.com>

Hi Andr?,

Andr? Dahlqvist wrote:
> 
> > You want the __import__ function:
> >
> > module = __import__(var)
> 
> I misphrased the question. Inside my module I have a dictionary that I
> need to access. And the name of this dictionary is stored in a
> variable. If I do:
> 
> import the_module
> 
> How can I then access the dictionary inside this module, if the name
> of the dictionary is stored in a variable? Or is there some
> equivalence to "from the_module import the_dictionary" when you're
> using __import__?

How about getattr?  Like this:

    import my_module
    the_dict = getattr(my_module, 'dict_name')

This should give you the same result as:

    import my_module
    the_dict = my_module.dict_name

HTH,
Steve

-- 
Steve Tregidgo
Software Developer
http://www.businesscollaborator.com



From urner at alumni.princeton.edu  Mon Jun  5 15:34:50 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Mon, 05 Jun 2000 12:34:50 -0700
Subject: Nth digit of PI
Message-ID: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com>

Here's another "math through programming" challenge:

Recent number theory breakthrough:  you can get the
nth digit of PI without doing all the work needed 
to compute all the digits up to the nth digit (same
with some other numbers -- not sure if we know all
the relevant criteria as to which).

The original algorithm worked for PI expressed as 
a hexadecimal number, but I found a web page purporting
to give the algorithm for a base 10 PI, but I couldn't
figure out how to make it work in Python.[1][2]

Anyone?

Another question:  does the algorithm for getting 
the nth digit of PI actually do less work (in terms
of flipflops) than would an algorithm that gets 
all digits up to n inclusive?  I assume so, as I 
read about someone getting a trillionth digit or
something -- but I don't think we've got PI to that
many places.

Kirby

[1] http://www.mathsoft.com/asolve/plouffe/plouffe.html
[2] http://www.lacim.uqam.ca/plouffe/Simon/articlepi.html




From mfletch at tpresence.com  Fri Jun 16 08:54:07 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 16 Jun 2000 08:54:07 -0400
Subject: Python Productivity over C++
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33C0@VRTMAIL>

Actually, the comment would more likely be:
	"one-man's-mess-is-another-man's-perl-ly-y'rs<0.7 wink>".

And-there's-the-essence-of-Python-in-a-nutshell-ly y'rs,
Mike

Kingdom for a nutcracker...


-----Original Message-----
From: Frank V. Castellucci [mailto:frankc at colconsulting.com]
Sent: Friday, June 16, 2000 8:32 AM
To: python-list at python.org
Subject: Re: Python Productivity over C++

...
Like with any language, Python can be messy. There are not less ways to
be messy in Python, there are just different ways. As I am pretty much a
neophyte in the Python ways (but trying to come up to speed), a review
of any code I have written by someone like yourself would probably
elicit the statement "Well, thats a wee bit messy lad". 
...



From katz at Glue.umd.edu  Mon Jun 12 22:40:53 2000
From: katz at Glue.umd.edu (Roy Katz)
Date: Mon, 12 Jun 2000 22:40:53 -0400
Subject: Python Productivity over C++
In-Reply-To: <8i3kbi$3a0$1@slb1.atl.mindspring.net>
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu> <8i3kbi$3a0$1@slb1.atl.mindspring.net>
Message-ID: <Pine.GSO.4.21.0006122230240.2409-100000@z.glue.umd.edu>

Oh of course, agreed!

My point is that I find Python easier use for roughing out a
quick sketch--like how charcoal is to art. program. I'm hooked on it
as a general-purpose language, please understand; however, now 
I have an (working) example for my friends (whom I'm trying to persuade 
to pick up Python). 

The punchline: coding took two hours in Python and two days in C. 


Roey Katz
vu geyt ir?


On 12 Jun 2000, Aahz Maruch wrote:

> The real moral is that, had you invested the same amount of time
> speeding up your Python code (possibly by writing part of it in straight
> C), you almost certainly would have received a speedup of somewhere
> between two and a hundred times -- so at worst your C program would have
> been five times faster and at best your C program would have been ten
> times slower.  (Best and worst being relative to the Python program, of
> course.  ;-)
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
> 
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
> 
> "I love you."
> "Duhhhh!"  --SFJ
> 




From jkraska1 at san.rr.com  Mon Jun 12 02:36:30 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 12 Jun 2000 06:36:30 GMT
Subject: Adding Directory to PYTHONPATH
References: <8i1l7j$d01$1@nnrp1.deja.com>
Message-ID: <3944858D.F38977F2@san.rr.com>

> Probably been asked before but how do you add a new library module
> directory to PYTHONPATH?

Eh? You put the .dll/.so in a directory specified by the python
path. Either that, or you change the python path. How you change
this variable will depend on the environment you're running on.

> I know you can do it dynamically in Python, but in this case I need to
> have a particular module directory available everytime that Python is
> loaded.

I'm confused. Are you attempting to add things to the search
directory of a *running* python?


C/


From quinn at amatho.ugcs.caltech.edu  Wed Jun  7 19:51:52 2000
From: quinn at amatho.ugcs.caltech.edu (Quinn Dunkan)
Date: 7 Jun 2000 23:51:52 GMT
Subject: indirect references
References: <8hha1n$phl$1@nnrp1.deja.com> <393CEBE0.8CEAF61D@sightreader.com> <0d4a01bfcfb1$96479d60$1906a8c0@fc.fenx.com>
Message-ID: <slrn8jto60.3t9.quinn@amatho.ugcs.caltech.edu>

On Tue, 6 Jun 2000 05:20:08 -0700, Emile van Sebille <emile at fenx.com> wrote:
>Well, that calls for a response! ;-)
>
>>>> def spam(a):

[... hackery ...]

>Of course, it's not not going to work except
>in console mode.

There was a small hackery arms-race a while back about finding the name of the
variable a function was called with.  It escalated up until Hudson posted a
bytecodehack.  It'd be easy to extend his code to do what spam() does, but
work more "robustly" :)


From amitp at Xenon.Stanford.EDU  Tue Jun  6 19:01:12 2000
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 6 Jun 2000 23:01:12 GMT
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <393D3948.821E1DB6@prescod.net>
Message-ID: <8hjvro$bmr$1@nntp.Stanford.EDU>

 Paul Prescod  <paul at prescod.net> wrote:
| 
| Actually, Java applets were last year's programming model.  In fact,
| Java applets are something of an endangered species.
| 
| Java is almost entirely used on the server side these days and Python is
| just as good there as Java is. In fact, with Zope, Python is better.
| 
| Client-side Javascript is quite popular but in terms of volume of code
| it is a drop in the bucket compared to what is going on on the server
| side.

And all the server side stuff (like Zope) is easier to make work on
the new generation of wireless devices & set-top boxes.  Does YOUR
cell phone run Java applets, or even Javascript?

     - Amit


-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From loredo at spacenet.tn.cornell.edu  Tue Jun 13 18:19:57 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Tue, 13 Jun 2000 18:19:57 -0400
Subject: "import foo" and "from foo import bar"
References: <slrn8kaa4u.1r0.jajs@Pascual.es>
Message-ID: <3946B38D.17700983@spacenet.tn.cornell.edu>

Hi-

An additional difference, relevant if you are working in IDLE or
interactively and using names you have imported from a module
you may change (i.e., you are debugging it), is that if you
use

import me

then when you edit me.py, you can get the updated contents by using

reload(me)

I don't know of any similarly simple reload method if you are
explicitly loading specific names from me.py.

Peace,
Tom Loredo


From aahz at netcom.com  Mon Jun 19 10:42:01 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jun 2000 14:42:01 GMT
Subject: simple thread problem
References: <394E2F9C.6BCD96B5@iue.tuwien.ac.at>
Message-ID: <8ilbfp$6a8$1@nntp9.atl.mindspring.net>

In article <394E2F9C.6BCD96B5 at iue.tuwien.ac.at>,
Markus Gritsch  <gritsch at iue.tuwien.ac.at> wrote:
>
>I started a very simple program to get familiar with Python threads.
>It works fine, except for some misterious messages which I get every
>time a thread exits:
>
>currentThread(): no current thread for xxxx

I'm not sure how to tell you to fix the problem in your current context;
instead, let me push you strongly to subclass threading.Thread() instead
of using the thread module.  I guarantee this will make your life a lot
simpler.  If you still have trouble, come back here and I'll show you
how to do your example using Thread().
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From root at rainerdeyke.com  Fri Jun  2 12:53:30 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 02 Jun 2000 16:53:30 GMT
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why -- or why  not? (was Re: Damnation!))
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <393326D3.400AA248@cosc.canterbury.ac.nz> <u1z2gl8zu.fsf@cs.uu.nl>
Message-ID: <eERZ4.2206$r7.561638@news1.rdc2.tx.home.com>

<piet at cs.uu.nl> wrote in message news:u1z2gl8zu.fsf at cs.uu.nl...
> I want to add another reason for a/b to mean floating division:
>
> As it is now in Python, you can have a==b and c==d both being true, but
> a/c==b/d being false. Which would be a bad surprise for most people.

But, with floating point numbers, you can have (a + b) - b == a be false.
The only way to avoid this sort of situation is to avoid floating points
entirely.  I don't want my programs to become infested with floats just
because of one innocent division.


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From robin at jessikat.demon.co.uk  Tue Jun 20 13:21:36 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 20 Jun 2000 18:21:36 +0100
Subject: Roman Number Representation
References: <S9Q1YMA7B5T5Ew1U@jessikat.demon.co.uk>
 <TyN35.7946$Q7.118988@news-east.usenetserver.com>
Message-ID: <fHQ8yKAgg6T5EwnH@jessikat.demon.co.uk>

In article <TyN35.7946$Q7.118988 at news-east.usenetserver.com>, Joel Lucsy
<jjlucsy at concentric.net> writes
>I remember saving this from the discussion.
>
>--
>Joel Lucsy (jjlucsy at concentric.net)
>
>
>"Robin Becker" <robin at reportlab.com> wrote in message
>news:S9Q1YMA7B5T5Ew1U at jessikat.demon.co.uk...
>> I could have sworn I had seen a discussion of Roman Numbers somewhere on
>> clp, but can't find it. Anyone got any clues?
>> --
>> Robin Becker
>
>
>
>[ A UUEncoded file (roman.py) was included here. ]
>
Thanks. Any idea if it's a timbot production?
-- 
Robin Becker


From thollowe at opentext.com  Thu Jun  8 08:51:36 2000
From: thollowe at opentext.com (Skip Hollowell)
Date: Thu, 8 Jun 2000 08:51:36 -0400
Subject: How to tell if I can open a file or not
Message-ID: <8ho4to0m8n@enews2.newsguy.com>

I have a indexer that I am writing that goest through a user-specified
directory and all subdirectories and indexes all the .htm* file contained
therein.  This works great until I get to a file that someone has chmod-ed
so that I can't read it.  Then I bomb after the open().

Is there a way I can check the attrbutes of the file before I open it?  This
is easy enough in c or perl, but have yet to find an equivalent in Python.

Thanks for any and all help, and I would appreciate being copied on email as
well as the group as I don't always have access to newsgroups where I am at.

Skip Hollowell
skip_hollowell 'at' yahoo.com




From mikael at isy.liu.se  Tue Jun 20 14:12:39 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 20 Jun 2000 20:12:39 +0200 (MET DST)
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <m3k8fk5stl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <XFMail.000620201239.mikael@isy.liu.se>

On 20-Jun-00 Michael Hudson wrote:
 >  Mikael Olofsson <mikael at isy.liu.se> writes:
 > > Erm, isn't a pt as in 10 pt font the typographic unit 1/72 inch?
 >  
 >  Well, sort of; the thing is the software doesn't know how big your
 >  monitor is physically, so can't work out what's it's own dpi is.
 >  You'd think a task as simple as "draw something this size" would be
 >  pretty simple after fifty years of computing, but you'd be wrong...

I've always thought that a pt is 1/72 inch, as I wrote before, and that
a pixel width or height is referred to as a px. I sounds like we are 
talking about the latter. I guess there is an overall confusion about 
the two measures.

FWIW, both units are used in W3Cs CSS level 1 and 2 recommendations.
However, they give their own definition of px, using arm lengths and 
angles.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jun-00
Time:    19:51:32

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

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


From cut_me_out at hotmail.com  Fri Jun 30 16:14:03 2000
From: cut_me_out at hotmail.com (Alex)
Date: 30 Jun 2000 16:14:03 -0400
Subject: Web server executing Python scripts
References: <2B1262E83448D211AE4B00A0C9D61B030130125E@msgeuro1.creo.be>
Message-ID: <etdk8f7ncz8.fsf@w20-575-116.mit.edu>

I have written a sort of "Hello, database" application in Zope recently.
I built it around the "Boring Product," which you can find on the Zope
web site.  It was almost easy, and I think doing it again will be a
snap.  A couple of things confused me, like the fact that you have to
have doc strings for the methods that Zope calls, but on the whole it
was very easy.

I would strongly recommend checking Zope out.

Alex.


From cjensen at bioeng.ucsd.edu  Fri Jun 23 20:51:52 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Fri, 23 Jun 2000 17:51:52 -0700
Subject: Array Memory Allocation Size
Message-ID: <39540628.B284C15F@bioeng.ucsd.edu>

I'm using NumPy to allocate heaps of arrays.  Is there a command to find
out how much space each array is using in memory?

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


From emile at fenx.com  Mon Jun 12 07:07:37 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 12 Jun 2000 04:07:37 -0700
Subject: lambda and if
References: <8i2c9l$b0r$1@news1.wire.net.au> <0006122253431T.01005@localhost.localdomain>
Message-ID: <07cd01bfd45e$9b0d07c0$1906a8c0@fc.fenx.com>

You could use this:

>>> filenames = map(lambda x:x[2] + "_" + x[0] + "_" +
["","big"][x[1]=="big"] + ".txt" ,items)

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: matt <matt at virtualspectator.com>
To: <python-list at python.org>
Sent: Monday, June 12, 2000 3:46 AM
Subject: lambda and if


> say you had the following
>
> filenames = map(lambda x:x[2] + "_" + x[0] + "_" + x[1] +
".txt" ,items)
>
> but really wanted
> filenames = map(lambda x:x[2] + "_" + x[0] + "_" + (x[1]
in ("big",) ? "big" :
> "") + ".txt" ,items)
>
> where (x[1] in ('big',) ? "big" : "") is a pseudo
c++_pyhton expression that I
> want to implement .....
>
> i.e. the result is to append "big" to the filename being
created if it is the
> 2nd element of item, otherwise don't append anything.
>
> regards
> Matt
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From neilh at scintilla.org  Sun Jun  4 09:37:05 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 04 Jun 2000 13:37:05 GMT
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <BXh_4.4616$Hz.35258@news-server.bigpond.net.au> <memo.20000604135334.59489A@digitig.compulink.co.uk>
Message-ID: <5Ys_4.5311$Hz.39657@news-server.bigpond.net.au>

> > > >>> def bar(foo = None,
> > > ...    # /lots/ of other optional parameters
> > > ...    ):
> > > ...    # some implementation
> > > ...
> > > >>> baz = Foo()
> > > >>> bar(foo = baz)
> >
> >    Wouldn't 'foo' be better used to describe its role rather than its
> > type.
> >
> >    bar(output = baz)
>
> Depends. For instance, I'm working on something at the moment that uses a
> flyweight character class, and being a flyweight the context is separated
> from the class itself. So, for instance, the method that renders needs to
> be passed a Tkinter.Canvas on which to render. It makes sense to me to
> define the method as:
> def render(canvas, ...)

   So it is a required parameter? In the original problem statement it was
an optional parameter, which is why the name is more prominent. I agree that
the primary subject of a call may sensibly follow a different naming scheme
than a bunch of optional arguments which generally indicate 'flavourings' of
a call.

> If I call it "output", "context" or some such then it's not clear what to
> pass; calling it "canvas" makes it self-documenting.

   Mmm. Fixing up Python's lack of strictly typed parameters ;)

> Using "output_canvas"
> or some such is more verbose and IMHO adds nothing at all to clarity -- if
> anything the verbosity detracts from the clarity. I /used/ to use the
> convention "a_canvas", but I found it to be nowhere near as clear &
> maintainable.

   I'm thinking about moving the opposite way (I sometimes currently use
lower case class names for class instances) and starting to write aCanvas
rather than canvas as it doesn't seem much additional verbiage to remove an
opportunity for confusion.

   Neil





From dalke at acm.org  Fri Jun  2 12:22:17 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 2 Jun 2000 10:22:17 -0600
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why -- or why  not? (was Re: Damnation!))
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <393326D3.400AA248@cosc.canterbury.ac.nz> <u1z2gl8zu.fsf@cs.uu.nl>
Message-ID: <8h8mpo$vl2$1@nntp9.atl.mindspring.net>

piet at cs.uu.nl wrote:
>As it is now in Python, you can have a==b and c==d both being true, but
>a/c==b/d being false. Which would be a bad surprise for most people.


At the very least, not to many programmers.  It's the same behaviour
in C:


> cat > tmp.c
#include <stdio.h>
main() {
 int i=7, j=2;
 float x=7.0f, y=2.0f;

 printf("i==x : %d\n", i==x);
 printf("j==y : %d\n", j==y);
 printf("i/j==x/y : %d\n", (i/j)==(x/y));
}
> /usr/local/bin/gcc tmp.c
> ./a.out
i==x : 1
j==y : 1
i/j==x/y : 0
>

                    Andrew
                    dalke at acm.org





From david_ullrich at my-deja.com  Fri Jun 16 13:09:09 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 16 Jun 2000 13:09:09 -0400
Subject: Problems with classes and __add__ and __mul__...
References: <8ibmn8$k5h$1@nnrp1.deja.com> <20000616011643.L26436@xs4all.nl>
Message-ID: <642A954DD517D411B20C00508BCF23B001296D5D@mail.sauder.com>

In article <20000616011643.L26436 at xs4all.nl>,
  Thomas Wouters <thomas at xs4all.net> wrote:
> On Thu, Jun 15, 2000 at 10:52:31PM +0000, abg at saturnsys.com wrote:
>
> > I would like to do this...
> > def __add__(self, other_instance):
> >     return self.list + other_instance.list
>
> > Which returns a list.  I would like to return an object of my class
type, but
> > it's not right obvious to me as to how to do that.
>
> Well, the result of 'x + y' is exactly what you return in x's __add__
> method, or if x doesn't have one, what y's __radd__ returns. So, if
you want
> 'x + y' to return an instance, make __add__ return an instance:
>
> class spam:
> 	def __init__(self, l=None):
> 		self.list = l or []
>
> 	def __add__(self, other_instance):
> 		return spam(self.list + other_instance.list)
>
> If you're worried about 'spam' not being accessible to '__add__',
because
> you are defining the class outside of the global namespace, you can do
this
> as well:
>
> 	def __add__(self, other_instance):
> 		return self.__class__(self.list + other_instance.list)
>
> A bit more clutter, but possibly more obvious.

   Another advantage to doing it this way is it does the
right thing if self is an instance of a subclass of spam.
(I started returning self.__class__(whatever) from
__xxx__ methods a while ago, wasn't sure whether it
would be regarded as evil to do it that way. I guess
not; that's a relief.)

DU


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



From ullrich at math.okstate.edu  Mon Jun 19 11:37:55 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 19 Jun 2000 15:37:55 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <01bfd93c$66c03180$2ace8ad1@daves-dell> <oO935.12054$Xx5.557215@bgtnsc06-news.ops.worldnet.att.net> <01bfd96b$f7ddcea0$85a38ad1@daves-dell> <LOc35.5656$_54.105800@news1.rdc1.md.home.com>
Message-ID: <01bfda04$7a2c2120$55a28ad1@daves-dell>


Eric Hagemann <ehagemann at home.com> wrote in article
<LOc35.5656$_54.105800 at news1.rdc1.md.home.com>...
> [snip]
> Also you might have better luck with the regular expression stuff (rather
> than the find command) and precompiling the string
> [snip]
> 
> I think I am the somebody who said this ;)

	yeah, I wasn't sure whether it was appropriate to include your
name or not...

> My mind was racing faster than my fingers.

	Um, ???

>  The re stuff _is_ slower than
> find. I have measured the slowdown to be a factor of 2 or 3. As also
proven
> below.
> 
> My real thoughts were that most grep's (or findstr in NT) do more than
> simple string matching and thus the re module might be desired or a
better
> comparison.  If all that is desired is simple matching however, then
> string.find is your way to go.

	Right. In fact this is more or less what I took your post to mean
at first glance - had to read it carefully to find what Bill had said you
said.

DU

> Cheers
> Eric


From jkraska1 at san.rr.com  Wed Jun  7 22:04:13 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 08 Jun 2000 02:04:13 GMT
Subject: Can I explicitly request garbage collection from C?
References: <etdem692ba3.fsf@w20-575-109.mit.edu>
Message-ID: <393EFFB7.733758B8@san.rr.com>

Alex wrote:
> 
> Hi.
> 
> I am having a little trouble keeping track of reference counts in C
> extensions.  I thought it would be helpful if I could call garbage
> collection from an arbitrary point in my program and check what was
> destroyed.  Is that possible?

Things are destroyed instantly when their reference counts reach
zero. There's a function somewhere that prints out an object's
reference count; that's probably what you want. Either that, or
if you're writing a C extension, you can examine the object's
reference count directly. In virtually all cases, it will be
one higher than you expect, because the tuple that was sent as
an argument to your external function also happens to own a
reference to your object.

C/


From jkraska1 at san.rr.com  Fri Jun  9 20:13:11 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 10 Jun 2000 00:13:11 GMT
Subject: WIDESPREAD INCOMPETENCE AT BELL ATLANTIC
References: <B60FEC7EB.pduegowihyk@ooobpqvldjd.com>
Message-ID: <394188BF.B39F1255@san.rr.com>

> Perhaps the US Government should take a look at Bell Atlantic in
> much the same way it's currently looking at Microsoft, with a similar
> remedy in mind.

Bell Atlantic just Acquired GTE and doubled in size. It appears
the FTC disagrees....


C/


From dsavitsk at e-coli.net  Sat Jun  3 00:24:23 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sat, 03 Jun 2000 04:24:23 GMT
Subject: Why is my Tkintering so bad?
Message-ID: <XL%Z4.2161$gr4.67106@news1.rdc1.il.home.com>

example 1 does not work, example 2 does. anybody know why?
more importantly, i need to read text from a ScrolledText (Pmw) and i can't
do it.  The code i am attempting to use (or at least the relevant part) is
in Example 3.  i am used to things like...
x = Form.Text1.text (vb) so this is really foreign. essentially there is a
text box and a button. at startup the text box reads from a text file (this
works fine), then when the user pushes the button the contents of the text
box are written to the output window.  It is this step of connecting code to
events to visuals that i am not getting. any suggestions at all will be a
big help and greatly appreciated.

thanks,
doug

# Example 1 -------------------------------------
from Tkinter import *
root = Tk()
def buttonAction():
    print "click"

cmd = Button(root, text="Press").pack(side=TOP)
cmd.bind('<Button-1>', buttonAction)
root.mainloop()

# Example 2 --------------------------------------
from Tkinter import *
root = Tk()
def buttonAction():
    print "click"

cmd = Button(root, text="Press", command=buttonAction).pack(side=TOP)
root.mainloop()

# Example 3 ---------------------------------------
from Tkinter import *
import Pmw

def SaveEdits():
    x = st.get()
    print x

def MakeInterface():
    flt = Frame(root)
    st = Pmw.ScrolledText(flt, labelpos=N, label_text='Stuff',
text_wrap='none')
    st.importfile('TestData.txt')
    st.pack(fill=BOTH, expand=1, padx=5, pady=5)

    flt.pack(side=TOP, expand=YES, fill=X)

    flb = Frame(fl)
    cmdSave = Button(flb, width=20, text='Save Edits', borderwidth=3,
relief=GROOVE, command=SaveEdits).pack(side=TOP, pady=5)
    flb.pack(side=TOP, expand=YES, fill=X)

root = Tk()
root.title('window title')
MakeInterface()
root.mainloop()




From dworkin at ccs.neu.edu  Wed Jun 14 13:29:47 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 14 Jun 2000 13:29:47 -0400
Subject: Waffling between Python and Ruby
In-Reply-To: "Warren Postma"'s message of "Wed, 14 Jun 2000 11:56:13 -0400"
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
Message-ID: <vndd7lktblg.fsf@camelot-new.ccs.neu.edu>

"Warren Postma" <embed at NOSPAM.geocities.com> writes:

> After using Python seriously for 6 months, I must confess I still
> dislike the "whitespace as syntax". 

This can in part be considered a matter of taste, but...

> Having something look correct on the screen, but not work because of
> misuse of tabs versus spaces, or incorrect indentation, is something
> that still continues to bug me.

What do you use to write your code?  This is not a problem when using
any decent editor that I know of.  The "whitespace issue" is what kept
me from trying Python for a couple of years, but once I started really
using it (and in a decent programmer's editor) I have never found it
to be an issue.

No need to get started on this again, though, if you simply don't like
the look of it.

> 2. Python uses ":" to introduce a new scope, something I find extraneous

Yes, it is extraneous to the parser.  It is there to aid readability,
which some people have found it to do.  Some people here have posted
code to make some editors automatically insert ':' at appropriate spots.

> Ruby is a great language. If I had to make it even, by criticizing something
> in Ruby [snip]

If I were to do so, I would criticize the aggressiveness and
spamminess with which the proponents of Ruby have tried to market it
as 'A better X than X' for all X and sending frequent off-topic
evangelical posts to non-Ruby mailing lists.  This is in fact one of
the biggest things keeping me from looking seriously at Ruby.  Its
evangelists are offensively aggressive at times, enough so to make me
not want to be part of the Ruby community.

-Justin

 



From angband at blueyonder.co.uk  Sat Jun  3 19:13:10 2000
From: angband at blueyonder.co.uk (Bablos)
Date: Sat, 03 Jun 2000 23:13:10 GMT
Subject: ftplib storbinary method interception
Message-ID: <39399FBA.MD-1.4.4.angband@blueyonder.co.uk>

Hi There...

As my ISP has recently moved over to a new server, I've discovered
that none of my current software is compatable for FTP uploads and the
like, so I've decided to put my Python knowledge to some use and
actually write my own little FTP client.

I'm currently using the storbinary method for transferring files, but
during the transfer of large files I'm oblivious to how far through
the transfer I currently am.  I'd like to have at least some
indication that I've transferred xx% of the file, or 100,000 bytes so
far.  Obviously, storbinary doesn't have such a capability, but is
there some code buried out there that does?

Please don't tell me I'm going to have to rewrite chunks of ftplib.py
- I'm well and truly a beginner with Python...

Hope someone can help!


Bablos.


From david_ullrich at my-deja.com  Fri Jun  9 13:00:51 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 09 Jun 2000 17:00:51 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no>
Message-ID: <8hr7rn$6qh$1@nnrp1.deja.com>

In article <3940af8b.236270458 at news.online.no>,
  thomas at cintra.no (Thomas Weholt) wrote:
> Yo,
>
> PythonWin has some great stuff, like "partial" code-completion. I too
> have som problems with code gone haywire, sometimes Pythonwin gets
> stuck, but most of the time it`s a good alternative to IDLE.
>
> I use Emacs under Linux and it works just fine. Just wish the
> code-completion-thingy in Pythonwin could somehow be implemented in
> Emacs. That`s probably never gonna happen, but an editor with these
> capabilities would help the python grow bigger, and Windows-converts
> like myself get over the habits gotten from programming in
> Delphi/JBuilder. Code-completion, Code-Insight, parameter-lookups ...
> It would be nice wouldn`t it??!

    Surely this sort of thing is much harder in Python due
to the extremely dynamic nature of the language? At least
parameter lookups seem much trickier, in fact it seems
impossible to do that perfectly in Python: In Delphi the
compiler knows exactly what function a given identifier is
referring to at a given place in the code - otoh in Python
you can do something like

def F(n): return n

def G(n,m): return n + m

F = G

Now when you type 'F(' the parameter-lookup thingie should
prompt you for _two_ parameters, in spite of the "declaration"
of F as a function that takes one parameter.

    The code that maps "F" to G could be anywhere - I don't
see how a perfect parameter-lookup is possible.

DU

> Ah well ...
>
> Thomas
>
>


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


From thomas at xs4all.nl  Sat Jun 24 12:34:18 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 24 Jun 2000 16:34:18 GMT
Subject: Am I the only one to receive the messages several times????
References: <l03102802b577e7881f31@[193.78.237.153]>
Message-ID: <slrn8l9onv.elg.thomas@centurion.xs4all.nl>

>>Perhaps a BeOpen list is mailing to the python list, and the Python list is
>>mailing back to them???? This should cause a loop and make messages appear
>>several times.

>I got one, too. Barry W. told me:

>"Let's just say there was a local configuration error"

>Whatever that means ;-)

It might be a sick sort of joke ;) One of Sendmail's more notorious
errormessages is 'Local Configuration Error' which basically translates to
'I dont know what the f*** went wrong, it could be anything.' ;)

I know dinsdale (& Barry) switched to Postfix a couple of months back, but
perhaps Barry misses Sendmail occasionally ;-)

Thomas.


From mogul at primus.ca  Wed Jun  7 17:45:12 2000
From: mogul at primus.ca (Robin Porter)
Date: Wed, 7 Jun 2000 17:45:12 -0400
Subject: Newbie question....
Message-ID: <jmz%4.822$223.8853@news1.tor.primus.ca>

    How is it that you run code once you enter it?  I am using Python 1.5.2
in Win 9x.  After inputing the code in the Python GUI, and saving it, I have
been trying to run it by pressing F5 or pressing edit (top menu) and then
selecting RUN at the bottom.  Please tell me what I am doing wrong?
    Thank you so much for your help.

Robin

Here is the code:

TaxRate  = 0.23
MinPay   = 300.0
RegHours = 40.0

TaxPayable = 0
Overtime = 0
OvertimeHours = 0.0

HoursWorked = float(raw_input("Please enter the hours worked per week? "))
PayRate = float(raw_input("Please enter the rate of pay? "))

if HoursWorked > RegHours:
    OvertimeHours = HoursWorked - RegHours
    Overtime = (OvertimeHours) * (PayRate*1.5)

GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime

if GrossPay > MinPay:
  TaxPayable = (GrossPay-MinPay)*TaxRate

NetPay = (GrossPay-TaxPayable);

print "The gross pay is %f" % GrossPay
print "The net pay is %f" % NetPay




From ddb at crystal.uwa.edu.au  Mon Jun 12 02:00:54 2000
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Mon, 12 Jun 2000 14:00:54 +0800
Subject: jpython +java module problem
Message-ID: <39447C96.3F3E72F4@crystal.uwa.edu.au>

Hi, 
I have tried to write a small java class (myclass), subclassing
PyObject, 
intended to be used within jpython, where all the java class 
methods take PyObject arguments.
If the incoming PyObject is of the same type as myclass I 
would like to be able to coerce it directly to myclass and continue,
otherwise if it is of type PyFloat, or PyInteger,
I would like to be able to extract the relevant numbers, create a new
 myclass object and continue. 

Unforunately the PyObject I get back is a 
org.python.core.PyJavaInstance which I cannot find any obvious way 
to crack and or convert back to a myclass object. Is the original
myclass object burried in there somewhere or am I barking up
the wrong tree and its gone for good?

Help appreciated
thanks
doug


From dennis_marti at yahoo.com  Tue Jun  6 23:03:05 2000
From: dennis_marti at yahoo.com (Dennis Marti)
Date: Tue, 06 Jun 2000 23:03:05 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <m3wvk2wy1u.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <dennis_marti-298857.23030506062000@news.starpower.net>

In article <m3wvk2wy1u.fsf at atrus.jesus.cam.ac.uk>, Michael Hudson 
<mwh21 at cam.ac.uk> wrote:

> 
> Tangentially, does anybody know of any implementations of JavaScript
> (not necessarily connected to a browser) that let you type interactive
> commands into a listener and genereally muck about with the language?

You can download the source for a JavaScript interpreter from the 
Mozilla site (http://ftp.mozilla.org/pub/js/). One of the build 
configurations is for a "standalone JS engine and shell."

Dennis


From arta at NOSPAMx-stream.nl  Mon Jun 12 06:56:50 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Mon, 12 Jun 2000 12:56:50 +0200
Subject: Client with select.select()
Message-ID: <#F2#YuI1$GA.106@net025s>

Hi,

What's wrong?
I have some simple server programmed for experiments. You can chat with it.
blabla.
But with the code below, I want to connect with it. In the beginning it
prints a welcome message.
Or, it's meant to do. But it only prints the first line from the
welcome-message and not the
following. What am I doing wrong? Because in the logfile from my server,
there's  a line
in which stands that a login-connection has been made. So that's good.
But what am I doing wrong? I want the whole message printed, not the first
line.

Excuse me for my English, blame my teacher :)

A[r]TA

--
-Life can kick your ass, but all you got to do is stand up and move on...



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



from socket import *
from select import *

s = socket(AF_INET, SOCK_STREAM)
s.connect('127.0.0.1', 23)
text = ''

while 1:
    q = [s.fileno()]
    rd, wr, er = select(q, q, [], None)

    for n in rd:
        if n == s.fileno():
            print rd, wr, er, q
            text = s.makefile().readline()
            print text

    for n in wr:
        if n == s.fileno():
            g = 'password' +chr(13)+ '\n'
            s.send(g)





From riaan at e.co.za  Thu Jun 22 19:43:21 2000
From: riaan at e.co.za (Riaan Booysen)
Date: Fri, 23 Jun 2000 01:43:21 +0200
Subject: Boa stalled? (GUI editor for wxPython)
References: <8iqbv4$4p$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <3952A499.1339701@e.co.za>

Hi Bernard,

Bernhard Reiter wrote:
> 
> Is Boa development stalled?

Definitely not, it has progressed quite a bit since the last release.
I'm pretty close to another release.

> It looks like nobody made changes in the last three month.

No changes to the web site, apologies for that.
If you want to keep up to date the mailing list is a little bit busier
than the web site.

> The program still needs developers. It is in competition to the nice
> glade and libglade thing and not mature enough yet.
> Why is nobody picking it up?

I've received no code/patches so far, but a few people have offered 
their help and I intend taking them up on it.

> 
> (Just to bring this to your attention.)
> 
> http://boa-constructor.sourceforge.net/
> 
> --
> Professional Service around Free Software                (intevation.net)
> The FreeGIS Project                                         (freegis.org)
> Association for a Free Informational Infrastructure            (ffii.org)

-- 
Riaan                     >>> a='a=%s;a%%`a`';a%`a`
___________________________________________________
Boa Constructor - RAD GUI building IDE for wxPython
     http://boa-constructor.sourceforge.net


From loewis at informatik.hu-berlin.de  Tue Jun 27 04:52:54 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Jun 2000 10:52:54 +0200
Subject: Final garbage collection?
References: <962028703.1248416950@news.libertysurf.fr> <FwrnM8.MMA@world.std.com> <962045446.1993274611@news.libertysurf.fr>
Message-ID: <p6q7lbbzes9.fsf@informatik.hu-berlin.de>

Jerome Quelin <jerome.quelin at insalien.org> writes:

> I was wondering why Python does not implement such a mechanism, that
> seems to me a Good Thing (tm), because _in final_, every object _is_
> destroyed, either if their reference count is nul or not, since the memory is
> reclaimed by the system.

I think the primary reason for that in Python 1.5 (and before) is that
the interpreter does not keep track of all objects. So if a cycle
becomes garbage, it really *is* garbage - the interpreter does not
even know the cycle is there. To release the cycle at the end of the
program, you'd have to keep track of all objects. That would increase
the size of an object, and slow down object creation.

In Python 1.6 (starting with beta1), there will be 'true' garbage
collection. One of the downsides is that it therefore will have to
keep track of all objects. In your original example, I believe Python
1.6 still wouldn't collect your objects, as they have __del__
finalizers - which pose a big problem to a garbage collector.

As for cleaning up all objects at the end: If the objects have
finalizers, this is very difficult. What if the finalizer choses to
access an object that was already cleaned-up, or choses to bring the
object back into life? If programs are suddenly exposed to
cleanup-at-end, most would probably run in some problem and raise an
exception. That, in turn would invoke cleanup, which would start
cleaning the same object, which would throw an exception, etc....

Regards,
Martin



From mikael at isy.liu.se  Mon Jun 19 11:40:23 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Mon, 19 Jun 2000 17:40:23 +0200 (MET DST)
Subject: split() with a separator set
In-Reply-To: <xjvr99tyac7.fsf@gjallarhorn.ifi.uio.no>
Message-ID: <XFMail.000619174023.mikael@isy.liu.se>

On 19-Jun-00 Igor V. Rafienko wrote:
 > # input looks like this: 'foo:bar|baz/zot'
 > # I want [ 'foo', 'bar', 'baz', 'zot' ]
 >  string.split( src, ":|/" )
 >  
 >  doesn't work.

Igor!

Oleg just gave you a solution using re:s. If you try to avoid these 
animals, like I do, you could do the following:

seps = ':|/'
mystring = 'foo:bar|baz/zot'
newstring = string.join(map(lambda x, seps=seps:(x+seps[0])[x in seps],a),'')
result = string.split(newstring,seps[0])

Good luck!

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    19-Jun-00
Time:    17:34:03

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

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


From bparsia at email.unc.edu  Fri Jun 16 07:48:35 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Fri, 16 Jun 2000 07:48:35 -0400
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <Pine.BSF.4.10.10006152355270.4382-100000@chi.pair.com>
Message-ID: <1ecb5af.1t0clrqdbel98N%bparsia@email.unc.edu>

Michal Wallace <sabren at manifestation.com> wrote:

[snip]
> 
> I pretty much write all of our code, so I've had little experience
> with pair programming.. But I would LOVE to see an editor wired up
> over the net so that two people could work on code at the same
> time. (If I knew anything about GUI, I'd probably try hacking IDLE
> myself)
[snip]

If you don't mind using Emacs, or at least viewing Emacs, it does have a
shared buffer mode.

I find VNC to be a nice lightweight solution. It works best if you take
turns "hosting", i.e., the person who's doing most of the typing isn't
doing it over VNC ;) If the "observer/kibbitzer" is remote, then you can
even get away with VNC over a dialup line (it's like working with a
*really* slow typist :)). It's probably a good idea to combine it with
some sort of messaging (e.g., IRC). Also, switch from VNC-client to
VNC-server typically requires a bit more of a formalized "turn" system.

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From digitig at cix.co.uk  Fri Jun  2 17:08:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Fri, 2 Jun 2000 22:08 +0100 (BST)
Subject: Timeout using tkinter?
References: <qReZ4.6366$wYl.233060864@newsb.telia.net>
Message-ID: <memo.20000602220827.48011A@digitig.compulink.co.uk>

In article <qReZ4.6366$wYl.233060864 at newsb.telia.net>, effbot at telia.com 
(Fredrik Lundh) wrote:

> On the other hand, I'm not really sure what you're asking for.
> What's a "user response" in this case?  A mouse click?  Data
> entry in a dialogue box?

In this particular case, keystroke, but yes, the "after" method looks just 
right -- and a pointer to some real documentation on tkinter, too, so I'll 
go to sleep happy tonight! Thanks.


From rapto at arrakis.es  Sat Jun 24 08:53:48 2000
From: rapto at arrakis.es (Marcos =?iso-8859-1?Q?S=E1nchez?= Provencio)
Date: Sat, 24 Jun 2000 12:53:48 GMT
Subject: Gadfly Question: TypeError on __Init__
References: <395408ae.75890752@news.telus.net>
Message-ID: <3954CC36.EE2D4CC4@arrakis.es>

I believe that __init__ should always return None (what functions do
silently when they don't find a return clause).

Leave the error to propagate, should there be any. Catch it in the
calling code.


From borlak at home.com  Mon Jun 19 18:57:31 2000
From: borlak at home.com (Michael Morrison)
Date: Mon, 19 Jun 2000 22:57:31 GMT
Subject: Capabilities of python?
Message-ID: <vzx35.960$ul.60878@news1.rdc2.tx.home.com>

I'm creating a pretty simple program, but I want to make sure python can
handle everything I need it to.

The program is basically a mixture of a word processor/translator.  Consider
reading in a foreign language and converting it to english, then printing it
to the screen.  This will have a full GUI using (i guess) TK.

Also I will need this program to be able to open a serial port (such as
com2) and read in the data, converting it the same way.  Also once this data
is converted, I need to be able to send it to other hosts via ethernet.

The program must also supply templates, like in a word processor.  I am
pretty sure I can do this already ;-)

I'm guessing this is not a problem, but want to make sure before I jump on
the bandwagon.

...
As a side note, when using IDLE-Python on my p350, the text output is VERY
slow.. Pythonwin seems a lot faster but is still not up to par compared to
your basic notepad.  I need to able to scroll very fast and print out text
very fast, can TK handle this?  Or should I look into another python GUI
addition?


Any help is greatly appreciated!

Michael Morrison






From dale at out-think.NOSPAMco.uk  Mon Jun 26 07:32:56 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 26 Jun 2000 12:32:56 +0100
Subject: Driving Excel from Python
Message-ID: <8j7f1k$f3u$1@supernews.com>

I'm driving Excel from Python and most of it's working fine except for:

>>> xlsh.UsedRange.Address(ReferenceStyle=1)
TypeError: call of non-function (type string)

xlsh is an Excel Sheet object and I'm trying to get the used range in RxCx
format. However, when I try and specify ANY arguments to the Address
property, I get this error message. I've tried both positional and keyword
arguments.

I have the Python Win32 book but this this subject is not quite covered in
sufficient depth.

Any help appreciated.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From maxm at normik.dk  Tue Jun 13 09:05:08 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Tue, 13 Jun 2000 15:05:08 +0200
Subject: Where should I put my HTMLgen directory?
Message-ID: <7BD10B680501D411B9DF009027E06F32011ACA@EXCHANGE>

I am trying to install HTMLGen on my machine.

My python is installed in:
C:\Program Files\Python\Lib\HTMLgen

I have unpacked HTMLGen into:
C:\Program Files\Python\Lib\HTMLgen

But when I try to run:
import HTMLgen

I get the traceback:

Traceback (innermost last):
  File "<string>", line 1, in ?
  File "C:\root\pythonScripts\HTMLgen\test.py", line 1, in ?
    import HTMLgen
ImportError: No module named HTMLgen

So obviously I am doing something wrong. I thought that Python traversed the
Lib diretory recursively to find any modules. But no?

Should I put all the *.py* files in the root of Lib? That seems a bit messy
to me.

Kind regards
	
	Max M



From ge at nowhere.none  Sat Jun 17 19:18:36 2000
From: ge at nowhere.none (Grant Edwards)
Date: Sat, 17 Jun 2000 23:18:36 GMT
Subject: Which linux distros have Python by default?
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial> <%TS15.655$i06.206156@ptah.visi.com> <20000614175918.A5625@dophnic.yi.org> <RVW15.694$i06.216628@ptah.visi.com> <394BE553.99FBC464@libc.org>
Message-ID: <gHT25.1412$i06.318267@ptah.visi.com>

In article <394BE553.99FBC464 at libc.org>, Bill Anderson wrote:

>> >Yeah, but the distribution maintainers will be forced to include it as
>> >a default to be considered seriously.
>> 
>> Do most of the "consumer" distros even include kernel sources
>> by default?  I'm not positive, but I don't think RH includes
>> kernel sources in the standard "workstation" configuration.
>
>In the case of Redhat (and I believe Mandrake), it is basically
>irrelevant, since the installer and many admin tools require python,
>making it a mandatory install. They do not install the devel files, but
>python is available.

Right.  In order to end up with a RH installation without
Python and Tkinter (and snack) you've got to actually know it's
there and then intentionally uninstall it after your system is
up and running.

I just bought a 2.2G removable-media SCSI drive (Castlewood's
ORB) -- which if it works as advertised is going to be really
handy.  So, once I've got that going, maybe I'll try out some
of the other big "Consumer" names (Mandrake, Caldera, Suse,
Corel) just to see what you get if you pick the default answers
to all of the intall questions.

I think I can assume that Debian and Slackware users will know
whether or not they have Python/Tkinter and how to install it
if they want to point-and-click insted of "vi /etc/rpshsi.conf".

-- 
Grant Edwards                   grante             Yow!  HOW could a GLASS
                                  at               be YELLING??
                               visi.com            


From dsavitsk at e-coli.net  Wed Jun 21 20:06:11 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Thu, 22 Jun 2000 00:06:11 GMT
Subject: Python on FreeBSD
Message-ID: <TLc45.12472$dF.523323@news1.rdc1.il.home.com>

i've been doing the python thing for a while on win32, but i finally got my
bsd box up.  i did a make install, and python from the terminal window
works.  i am wondering how to start idle, or some other ide.
also, will i need to make install other stuff to connect to interbase or
postgres?
thanks
doug




From moron at Glue.umd.edu  Tue Jun 27 14:52:27 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 27 Jun 2000 14:52:27 -0400
Subject: float literal too large
References: <ulsntz2kra.fsf@runner.princeton.idaccr.org>
Message-ID: <8jat5b$2t4@stochastic.eng.umd.edu>

Victor S. Miller said:

> float() literal too large.
> 
> The literal in question was '0.23489798479873E-2325' (this was
> generated on a Cray computer).  I couldn't find anything in the
> documentation about this restriction.

Python floats are stored in platform doubles.  On IEEE-754 compliant
hardware (pretty much anything other than a Cray (or VAX, but I don't
think they make those anymore (unless alphas have a VAX math mode)))
that means you get a 53-bit mantissa multiplied by a factor ranging
from 2**-1022 to 2**1023 (or down to 2**-1072 with precision loss).
In decimal that's 14 to 17 digits of mantissa (depending on how you
define precision) and an exponent range of -307 to +307 (or maybe 308,
for some values of the mantissa), and down to -322 with precision
loss.

It is kind of odd that it said it was too big.  One might even go so
far as to call it a bug.

as far as workarounds, what's wrong with

	a[i, j] = eval(s)

?

Warren Focke

-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From nielsenjf at my-deja.com  Thu Jun 29 19:13:32 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Thu, 29 Jun 2000 23:13:32 GMT
Subject: newbie questions about imports, threads, and optimizations
References: <8jd7f1$kic$1@murrow.sp.trw.com> <8jfilg$9sr$1@slb6.atl.mindspring.net>
Message-ID: <8jgl6f$3s9$1@nnrp1.deja.com>

If you download the latest win32 extension for python, you'll see some
documentation in the win32 Overviews sections regarding threads
(specifically w/regards to looking at the eventlog, but the threading
part would be applicable to other environments).

john



In article <8jfilg$9sr$1 at slb6.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <8jd7f1$kic$1 at murrow.sp.trw.com>,
> Bryce Sellers <bryce.sellers at trw.com> wrote:
> >
> >Second, are there any tutorials or even reference manuals out there
on using
> >Python with threads?  I haven't so much as even seen any sample code,
> >although I admit I haven't looked very hard.
>
> I've posted a few things (unfortunately, DejaNews is doing poorly as
an
> archival service these days), but there's nothing up on a web site
yet.
> After I do my presentation at OSCON, I'm planning on whipping up
> something for the web.  If you've used threads in another language,
it's
> pretty easy to use Python threads; even if you've never used threads,
> it's still easy, but you need to know a few tricks.
>
> In any event, take a look at the threading module (*NOT* thread) and
see
> if that's enough to get you started.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> The best way to get information on Usenet is not to ask a question,
> but to post the wrong information.  --Aahz
>

--
nielsenjf at my-Deja.com


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


From aahz at netcom.com  Thu Jun 22 13:32:59 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jun 2000 17:32:59 GMT
Subject: Python website
References: <3951fe9c_2@nnrp1.news.uk.psi.net>
Message-ID: <8itikb$cs7$1@slb6.atl.mindspring.net>

In article <3951fe9c_2 at nnrp1.news.uk.psi.net>,
Simon Brunning <sbrunning at trisystems.co.uk> wrote:
>
>Has <http://www.python.org> disappeared, or is it just me?

It's up currently, but it does sometimes go off the air briefly.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From tim_one at email.msn.com  Fri Jun  9 14:09:38 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 9 Jun 2000 14:09:38 -0400
Subject: Cycles (was RE: Python Productivity over C++)
In-Reply-To: <Pine.LNX.4.10.10006091418020.10231-100000@naxos.skunk.org>
Message-ID: <000801bfd23d$e0c1d2a0$a12d153f@tim>

[Drew Csillag]
> ...
> Good point.  But there are actually two cycle detectors (one may or may
> not be derived from the other) that I know of: plumbo and Cyclops (I've
> used both).  I believe (I may be totally wrong on this) that one of them
> will be included with 1.6.

Cyclops started as an attempt to give plumbo a serious speed boost, but
quickly diverged.  Both require manual setup and some expertise, and
Cyclop's larger set of "features" makes it that much more intimidating at
first.

Neither will be included with 1.6, AFAIK.  There's still reason to hope that
NeilS's GC patch will be, albeit perhaps that enabling it will require
setting a compile flag at Python build time.

running-in-circles-ly y'rs  - tim





From walter111 at structurex.net  Thu Jun 22 04:54:10 2000
From: walter111 at structurex.net (walter hanagriff)
Date: Thu, 22 Jun 2000 03:54:10 -0500
Subject: trouble with tkinter
Message-ID: <3951D431.B0F20B4@structurex.net>

hi all, i havent been programming for long so i started a simple program
to practice, the python part is easy, but tkinter is nothing but trouble

first question, and plenty more to come, lol, i hope yall can help, i
stumped the people at tutor with earlier questions that i thought were
simple, but just in case i really do ask a simple question, is it
allowed here?
i have a class, and some def statements in it, each def is a frame, how
can i call one frame from another def?

if thats not enough info for an answer, should i just show the source
code for that class or is this enough for an answer?




From mfletch at tpresence.com  Sun Jun 25 19:48:30 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 25 Jun 2000 19:48:30 -0400
Subject: smtplib
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33E8@VRTMAIL>

Same here, the Python newsgroup has posts showing up fine, but the list is
only getting messages that are from other subscribed peoples.  Guessing that
when they fixed the repeating messages they wound up shutting down the
gateway :( .  Or possibly the gateway got spam-listed for flooding?

Ah well, means I can spend more time working on the weekend :-/ and less on
playing with Python stuff.

Enjoy,
Mike

-----Original Message-----
From: David Porter [mailto:jcm at bigskytel.com]
Sent: Sunday, June 25, 2000 7:43 PM
To: Python List
Subject: Re: smtplib


* Gregory Tucker <tuckerg at acm.org>:
> 
> P.S. I am getting very little mail off the Python list. Is it something
> wrong with me?
> 

There must be something wrong with me too. My only thought is that the
newsgroup posts aren't making it to the list, but I haven't looked into it
at all.  

  david.

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



From frederic.laurent at sxb.bsf.alcatel.fr  Fri Jun 23 02:42:54 2000
From: frederic.laurent at sxb.bsf.alcatel.fr (Frederic LAURENT)
Date: Fri, 23 Jun 2000 08:42:54 +0200
Subject: trouble with Tkinter installation
Message-ID: <395306EE.E02DE460@sxb.bsf.alcatel.fr>

Hi all,

I compiled Python 1.5.2 with the tcl/tk module and when I start the
python interpreter
to test my installation I'have the following problem :

Python 1.5.2 (#2, Jun 22 2000, 10:49:11)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import _tkinter
>>> import Tkinter
>>> Tkinter._test()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File
"/users/public_w3p/share/python/Python-1.5.2/lib/python1.5/lib-tk/Tkinter.py",
line 1947, in _test
    root = Tk()
  File
"/users/public_w3p/share/python/Python-1.5.2/lib/python1.5/lib-tk/Tkinter.py",
line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories: 
    



This probably means that Tcl wasn't installed properly.

>>> 


Where is the location in which the interpreter tries to find init.tcl
The list is empty.
In fact, I had used tkinter before recompiling and installing a new
python version,
and all was fine !

(I followed the tests defined here :
http://www.pythonlabs.com/pub/www.python.org/topics/tkinter/trouble.html#checking
)


Any idea ?

Cheers,

-- 
Fr?d?ric LAURENT
mailto:frederic.laurent at sxb.bsf.alcatel.fr
T?l : (33) 03 88 55 70 00  poste 57228


From aahz at netcom.com  Thu Jun  1 21:48:01 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jun 2000 01:48:01 GMT
Subject: "Fixing" os.system on Windows
References: <8h71ec$oh4$1@nnrp1.deja.com>
Message-ID: <8h73oh$4no$1@nntp9.atl.mindspring.net>

In article <8h71ec$oh4$1 at nnrp1.deja.com>,
Howard B. Golden  <hgolden at my-deja.com> wrote:
>
>Carey Evans showed a version of system() that gets the exit code under
>Windows in his message:
>
>http://www.python.org/pipermail/python-list/1999-May/009409.html
>
>I hope this could be adopted into the Windows part of the core Python
>system, so that os.system() will work the same under Windows and Unix.
>If there is some reason for not doing this, I'd appreciate it if
>someone would explain it.

Looks like it requires the win32 extensions; I believe those are not
currently part of the core Python distributions.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From jimclark22 at msn.com  Wed Jun 28 06:28:43 2000
From: jimclark22 at msn.com (ZWang Productions)
Date: Wed, 28 Jun 2000 06:28:43 -0400 (EDT)
Subject: Are you searching for a merchant's account?
Message-ID: <20000628102843.D8AC71D086@dinsdale.python.org>


    
I just visited your Web Site from a search engine.

We are sending you this email to tell you about what we do,
as it may be beneficial to both of us. 

We have an interesting concept! 

My company has taken notice of your online presence and
wanted to bring a few things to your attention.

We have spoken with thousands of on-line business individuals like
yourself, who, unfortunately, are not happy. The reasons
for their unhappiness varies, but all seem to have common 
issues.

I invite you to visit my company at www.zwangproductions.com
and, if you have any questions, please email me at Sales at zwangproductions.com
If this message has been sent to you before, please accept my apology. 
My Staff advises me that the Remove Filter was not function properly.  

Again, if you want to be removed from this list, please send me an e-mail at  Sales at zwangproductions.com indicate "remove" in the subject area and it will be permanently removed from my list.  


Thank you for your time and consideration, and I look forward to
being of service to you.

				Sincerely,

				Jim Clark
				ZWang

PS. Unhappy with your Web Hosting company visit www.wtz-productions.com
 
************************************************************************
Under Bill s.1618 Title III, pending approval by the 105th U.S.Congress,
this mail cannot be considered spam as long as we include contact
information and a remove link for instant removal from our mailing list. 
To block any future email from us, please reply to this email and include
the word "remove" in the subject heading and your email address in the
body.  This message contains confidential information and is intended only
for the individual named.  If you are not the named addressee you should not disseminate, distribute, or copy this e-mail.  If you have received this e-mail by mistake, please delete this e-mail from your system.  Please see http://www.cauce.org/pressreleases/pr-hr3113.shtml for additional information. 


From famvandenberg at home.com  Thu Jun 15 13:00:49 2000
From: famvandenberg at home.com (bo Vandenberg)
Date: Thu, 15 Jun 2000 17:00:49 GMT
Subject: Advice on a Learning strategy
Message-ID: <39490B9E.950C2576@home.com>

I'm trying to put together a basic toolkit for general consultant
programming in a win32 environment. (I know this is vague but I have to
be flexible at this point.

My thoughts are to work in Python and Delphi.

Python is going to be for all the client side customization and Delphi
will be my engine to do what doesn't need to be interpreted. I'm
considerring C++ but I don't know if I want the learning curve.


VB vs Delphi    Delphi programs seem more self contained than VB. I also
like the possibility of going to Kylix in the future. If I write it in
Delphi I think the client is less likely to accidentally break it (no
promises for Microsoft though).


I would like to use Tkinter but I think speed concerns will have me move
to wxwindows and pythonwin.


Has anyone gone this route, or a compatible route, before who would be
willing to share some advice???


These are my first programming languages.



From jim at interet.com  Tue Jun 13 14:53:12 2000
From: jim at interet.com (James C. Ahlstrom)
Date: Tue, 13 Jun 2000 14:53:12 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <39468318.9EA637E0@interet.com>

Steve Mullarkey wrote:

> 1. What productivity increase do you achieve ?

I get 3 to ten times productivity increase.

> 2. How long did you use Python before you achieved increased
> productivity ?

Maybe six months.

IMHO, the increase is mostly due to the higher-level
language constructs.  That is, Python has dictionaries and
lists, and so it is more likely a programmer will use
the correct algorithm in Python, rather than suffer through
having to program his own dicts and lists in C.  So the
productivity gain is largest for programs which really
do need dicts and lists.  If the Python resembles plain
low level C, I would not expect as much gain.

Relative to C++, Python is a simpler OO language to learn
and use.

Jim Ahlstrom


From gmcm at hypernet.com  Wed Jun 14 09:22:47 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Jun 2000 13:22:47 GMT
Subject: Threading tips.
References: <X%C15.25733$hp4.604260@newsread1.prod.itd.earthlink.net>
Message-ID: <8F535CB2Fgmcmhypernetcom@199.171.54.194>

Arint? wrote: 

>Its me again with questions about my embedded python app.  My scripts
>are primarily to talk to devices.  

I read this as saying App calls Python reads device.

>If a script is running and the device
>does something, then I want the app to call the py script callback if
>one was set.  

I read this as saying Python calls extension-that-read-devices calls back 
into Python.

>Do I need to block the Python threads?  What other steps
>do I need to take?

If that's an extension reading the device, it needs to 
Py_BEGIN_ALLOW_THREADS and Py_END_ALLOW_THREADS so it doesn't block Python 
while waiting on the device. If the callback is made on the same thread 
that sent the extension off to read the device, everything should be fine.

If, however, you extension has created it's own thread, you need to create 
a valid Python thread object for that new thread before it can call back 
into Python. That is very advanced stuff.

- Gordon


From scoota at my-deja.com  Tue Jun 27 08:43:15 2000
From: scoota at my-deja.com (scoota at my-deja.com)
Date: Tue, 27 Jun 2000 12:43:15 GMT
Subject: Python COM for Linux?
Message-ID: <8ja7gv$687$1@nnrp1.deja.com>

Hi,

I'm writing an automation interface for a large system embedding Python
as the script language.  To keep the scripts as easy to write as
possible (for the end-user) I'm providing a set of extension objects
written in C++.  What I would ideally like is for these objects to fire
events which are then handled in the Python script.  So far I can see
no platform independent way of doing this except using explicit
callbacks (which is messier than I would like).

Has anyone any experience in this area?

Scoota


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


From bparsia at email.unc.edu  Fri Jun 16 07:48:37 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Fri, 16 Jun 2000 07:48:37 -0400
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie>
Message-ID: <1ecb5k1.1s3tce6xg2hfxN%bparsia@email.unc.edu>

Russell Wallace <rwallace at esatclear.ie> wrote:

> > stevemul at ozemail.com.au (Steve Mullarkey) wrote:
> > 
> > > I have read in several places productivity claims of 5 to 10 times over
> > > 'C' and "C++". I would like to ask for some feedback from "C++"
> > > programmers who have moved to Python as to whether these estimates are
> > > realistic.
> 
> As someone who uses and likes both C++ and Python, I think estimates
> like that are excessively simplistic.
[snip]

Actually, the "simplistic" part is not taking into account things like
problem domain, tools, and, most importantly, programmer "type" (skill,
experience, etc.). Strangely enough, I would bet that *temperment*
counts for oodles: If you *really hate* Python, I doubt you'll be wildly
productive in it. On the other hand, if you're like me and there's no
way in this lifetime that you'll learn C++ (granted I'm not a programmer
*per se*), then Python makes me *way* more productive. (Er...ok,
Smalltalk/Squeak makes me way more productive, but actually having
written Python code, and actually being willing to read Python code,
still makes Python a significant increaser of my programming
productivity over C++ code, the mere thought of which makes me more
likely to work on my disseration, or clean the toilet).

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From thomas at xs4all.net  Mon Jun  5 17:50:10 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Jun 2000 23:50:10 +0200
Subject: operator overloading
In-Reply-To: <393C2386.307A160D@sightreader.com>; from 12klat@sightreader.com on Mon, Jun 05, 2000 at 03:02:47PM -0700
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com> <393C2386.307A160D@sightreader.com>
Message-ID: <20000605235010.C469@xs4all.nl>

On Mon, Jun 05, 2000 at 03:02:47PM -0700, Ken Seehof wrote:

> The following more closely conforms to these conventions:

The return code of __cmp__ is hardly a convention! If you dont return one of
those values, in a consistent manner, you're going to confuse the users of
your class greatly. And possibly break things like list.sort()'s default
sort function.

> ...  def __cmp__(self,other):
> ...   if self.a == other.a:
> ...    return 0
> ...   elif self.a < other.a:
> ...    return -1
> ...   else:
> ...    return 1

Good example not withstanding, you can of course write this much shorter,
and more obvious, and less errorprone:

def __cmp__(self, other):
	return cmp(self.a, other.a)

This will even work if __cmp__ would be changed to return different values
;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From quinn at mark.ugcs.caltech.edu  Thu Jun 22 23:29:08 2000
From: quinn at mark.ugcs.caltech.edu (Quinn Dunkan)
Date: 23 Jun 2000 03:29:08 GMT
Subject: Extending Python command line
References: <8iuj9g$f0t$1@news-int.gatech.edu>
Message-ID: <slrn8l5mhl.4pd.quinn@mark.ugcs.caltech.edu>

On Fri, 23 Jun 2000 02:45:42 -0400, Andrew Henshaw
<andrew.henshaw at gtri.gatech.edu> wrote:
>Is there a reasonably simple way to extend the Python command line so that
>instead of typing:
>
>   >>> myspecialfunction('this is some stuff')
>
>I could type:
>
>   >>> myspecialfunction this is some stuff
>
>Everything else should be handled normally.

Sure (warning, completely untested code):

import string, sys, traceback
try:
    while 42 != 6 * 9:
        c = string.strip(raw_input(sys.ps1))
        if not c:
            continue
        if c[-1] == ':' or c[-1] == '\\':
            while 1:
                s = raw_input(sys.ps2)
                if not s:
                    c = c + '\n'
                    break
                if s[-1] == '\\':
                    c = c + ' ' + s[:-1]
                else:
                    c = c + '\n' + s
        special = 'myspecialfunction'
        if string.split(c)[0] == special:
            c = special + '(' + repr(c[len(special) + 1:]) + ')'
        try:
            print repr(eval(c))
        except SyntaxError:
            try:
                exec c
            except:
                traceback.print_exc()
        except:
            traceback.print_exc()
except:
    print

>A nice bonus would be if it could handle line continuations.

Depending on what that bonus is, I might give you an address to send it to.


From effbot at telia.com  Thu Jun  1 03:05:51 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 01 Jun 2000 07:05:51 GMT
Subject: *blink* That's neat.
References: <GXJY4.28606$WS3.320758@typhoon.we.rr.com> <SZKY4.6213$wYl.227247104@newsb.telia.net> <8h07fv$bnc$3@newshost.accu.uu.nl> <UINY4.6227$wYl.223307264@newsb.telia.net> <8h0ne2$70i$2@newshost.accu.uu.nl> <3934242B.ECFDBAAE@endea.demon.nl> <8h1k0q$h04$2@newshost.accu.uu.nl> <wUfZ4.6380$wYl.229799424@newsb.telia.net> <8h4djp$pkc$1@newshost.accu.uu.nl>
Message-ID: <jXnZ4.6415$wYl.227858944@newsb.telia.net>

Martijn Faassen wrote:
> > I'm probably just dense, but how would you design such a
> > double-blind test?
> 
> Construct two Python interpreters (or environments). One is case-sensitive,
> the other is case-insensitive. Then randomly distribute them to your
> newbie programmers.

how do you make sure that the *observers* won't be able to
figure out what kind of interpreter any given user is using?

(if you don't, it's not really a double-blind test, right?)

</F>



From wtanksle at dolphin.openprojects.net  Tue Jun  6 21:59:04 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 07 Jun 2000 01:59:04 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net> <393C81D0.36732896@Lugoj.Com> <slrn8jql5v.6d4.wtanksle@dolphin.openprojects.net> <3dya4itrxf.fsf@amarok.cnri.reston.va.us>
Message-ID: <slrn8jrb37.b4q.wtanksle@dolphin.openprojects.net>

On 06 Jun 2000 17:34:20 -0400, Andrew M. Kuchling wrote:
>wtanksle at dolphin.openprojects.net (William Tanksley) writes:
>> I'm not the one that's claiming that removing case-sensitivity will ruin
>> Python.  My claim is very simple: first, it's a minor change for us to
>> make, and second, it helps newbies.

>Like Aahz, I don't think the change is that minor.

I suppose we'll just have to agree to disagree, then.  Because I can't see
the change causing me to care -- my coding style will hardly change.

Now, the change WILL cause old code to not work -- but in case you missed
it, it's been established that Py3K is going to do that.  Adding or
removing this feature will not fix the code.

>I'm also doubtful of the "helps newbies" claim; unless there have been
>some other citations in this thread that I missed, we have exactly one
>study using exactly one programming environment that reported CS
>(case-sensitivity) was a problem, and even that report says "most [users]
>continued to type case-incorrect tokens in their programs for a short
>period", which implies that they did eventually stop.  The evidence is
>hardly overwhelming, and basically purely anecdotal.

Now here we can make some progress, I think.  Once CP4E is funded, I would
hope that this is something they would do some work in.  Hopefully before
the change is finalised in Py3K.

However, from my experience teaching people programming and typing, people
don't normally think or communicate in a case sensitive way, and a large
portion of people have a really hard time doing so.  I've hypothesised
that this is an artifact of visual versus auditory learning.

>IMHO, if GvR was inventing a completely new language for teaching,
>then it would be worthwhile to debate whether it should be CS or CI.
>But he's not; Python is an implemented language with a sizable and
>growing user base that's used to CS, and a number of published books
>and other materials that will be invalidated by switching to CI. 

And the fact is that they're going to be invalidated anyhow.  Nothing we
do to CS will change that.  Part of the reason for Py3K is to allow Guido
to make large, backwards-incompatible changes.

>A.M. Kuchling			http://starship.python.net/crew/amk/

-- 
-William "Billy" Tanksley


From kenneth at loafman.com  Sat Jun 10 11:04:06 2000
From: kenneth at loafman.com (Kenneth Loafman)
Date: Sat, 10 Jun 2000 10:04:06 -0500
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net> <20000610042548.A9887@bigskytel.com>
Message-ID: <394258E6.6D9D6042@loafman.com>

I wonder if anyone has come up with this as an answer:

I like lower/mixed case because it reminds me of the advances we have
made in computing over the years.  When I first started computing it was
all uppercase.  FORTRAN and COBOL were shouted at the computer to make
it work.  Now we have mixed case and mixed languages.  All advances.  To
go back to all one case is a retreat to the past and I want to move
forwards.

Besides, I sometimes work with fellow Neanderthals that would write in
all upper case if they could.  Having that possibility in a language
would make the language more attractive for their use, and much harder
for me to read.

...Thanks,
...Kenneth




From gmcm at hypernet.com  Mon Jun 26 20:11:55 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 27 Jun 2000 00:11:55 GMT
Subject: Threading +  embedding = System Dump.
References: <8j7udf$15au$1@news2atm.raleigh.ibm.com>
Message-ID: <8F5FCDB88gmcmhypernetcom@199.171.54.154>

Arint? wrote: 

>Me again.  My usual update
>I have a C++ app/ dll.  This is a windows dll is loaded by LoadLibrary. 
>The dll holds all the python stuff while the app talks to devices. 
>Right now if I was to run a script that has a callback, the script can
>be running and in the middle of the script the app can pop up the
>callback's message window. That is fine.  But, I want the option of the
>script waiting for the callback to be called before continuing.  So, I
>(well Gordon did) figured I need and event or the Condition Lock stuff,
>but unfortunately this is causing the app to crash.  It seems to be
>crashing in PyFrame_New() in frameobject.c (this is python code) on the
>1st line of the function 
> PyFrameObject *back = tstate->frame;   <----- tstate is null according
> to 
>VC++ 6.0 debugger.

That sounds very much like there is no multithreading.

So why doesn't "diz[0].write(mx)" block? Or does it?

Is the real problem here that you're using one of the PyRun_ calls when you 
need to go down a level and use PyImport_ etc. so the C++ has some control 
over what's going on?

Or maybe that "diz[0].write(mx)" could become 
  response = diz[0].getresponse(mx)
  if response is "eggs":
    print "without SPAM? NO WAY!"

Without more clues, we're just as lost as you.

- Gordon

[rest of the message]

>This code is after I call the the callback from my app with
>PyEval_CallObject.  Note: this works without threads/event code, but no
>wait.  I consider doing the Event part in my dll, but the script should
>be the one to look at the callback message and determine if that is the
>message it wants to continue the script.
>
>I have this code:
>
>import sys
>import threading
>import PossDevice
>import poss
>devlist = [0,1]
>
>def threadFun():
> LOCK.set()
>
>def testCallback(str):  #the app calls back here sets the event so the
>script can continue
> mthread.start()
> poss.alert(str)
>
>diz = [0,1]
>LOCK = threading.Event()
>mthread=threading.Thread(None,threadFun)
>poss.initialize(PossDevice.Enumerate)
>diz[0] = PossDevice.QueryDevices(1,11,35)
>diz[0].setCallback(testCallback)
>diz[0].open()
>mx="\x1b\x47""EVENT\n"  <--- this will cause a device to send a message
>to the app
>print diz[0].write(mx)
>LOCK.wait()  <---wait on event I sent out above
>poss.alert("Waited")
>diz[0].close()
>
>thanks
>
>



From manni.heumann at gmx.de  Wed Jun 14 14:29:11 2000
From: manni.heumann at gmx.de (Manni Heumann)
Date: Wed, 14 Jun 2000 14:29:11 -0400
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <394b3e7d.284293552@news.online.no> <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net> <8i8ebi$5et$1@nnrp1.deja.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296A8F@mail.sauder.com>

spenrose at my-deja.com wrote:
[SNIP]
>FWIW, I agree with those who think Python an ideal choice for a first
>language--it worked for me! It's just the orientation of the two
>O'Reilly books I'm referring to here.
>
[SNIP]

If anybody cares: there was a discussion on slashdot recently about 
teaching kids how to program, and especially what language they should 
use. Ok, so I'm not a kid and I already know C and C++, but this 
discussion made me take a closer look at python.

http://slashdot.org/askslashdot/00/05/21/1919205_F.shtml


--

Manni



From kaimar at ut.ee  Fri Jun  2 11:01:12 2000
From: kaimar at ut.ee (Kaimar Karu)
Date: Fri, 02 Jun 2000 18:01:12 +0300
Subject: xml and python
Message-ID: <3937CC38.FD1360E@ut.ee>

Hi,

Where can I find some kind of helping material (tutorials etc) about how
to use xml in python?
I need some examples, just pure specification is not enough.

tia,
--
Kaimar Karu | kaimar at ut.ee | 251 13 468
       = What a wonderful world =




From petersc at stud.ntnu.no  Fri Jun  9 01:52:08 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 09 Jun 2000 07:52:08 +0200
Subject: Problem compiling source
References: <Z2%%4.1751$iy.137366@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <39408608.F222A69@stud.ntnu.no>

William Dandreta wrote:
> 
> Can anyone tell me what __FLAT__ is for and which header it is defined in?

Last time I checked with DOS compilers (long ago) FLAT mode was
using a protected mode environment to make the whole memory available
in one large 32-bit segment instead of having to address using
16-bit segment and 16-bit offset (being thereby limited to
1048576 memory (- bios stuf + window to ems etc.).

I think having python run in FLAT mode would be a better idea than not.

Else you might run into 640K problems.

640K-oughta-be-nuff-ly  y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From aahz at netcom.com  Thu Jun  1 15:32:43 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Jun 2000 19:32:43 GMT
Subject: random # errors?
References: <8h5m0n$k1o$1@news2atm.raleigh.ibm.com>
Message-ID: <8h6dor$10r$1@slb3.atl.mindspring.net>

In article <8h5m0n$k1o$1 at news2atm.raleigh.ibm.com>,
Arint? <shouldbe at message.com> wrote:
>I am trying to use whrandom for picture generation and I am getting this
>error?
>I am not sure the list is...  Any hints?
>Traceback (innermost last):
>  File "Beer", line 131, in ?
>  File "Beer", line 37, in biko
>  File "C:\Python\Lib\whrandom.py", line 46, in __init__
>    self.seed(x, y, z)
>  File "C:\Python\Lib\whrandom.py", line 60, in seed
>    t = int((t&0xffffff) ^ (t>>24))
>IndexError: list index out of range

What platform is this on?  What are lines 30-40 in Beer?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From matt at null.net  Thu Jun 22 18:52:41 2000
From: matt at null.net (Matthew Schinckel)
Date: Thu, 22 Jun 2000 16:52:41 CST
Subject: Pickle Question
References: <eviltofu-9E08E0.21184121062000@news.earthlink.net>
Message-ID: <20000622165243.352854@210.9.21.161>

Jerome Chan <eviltofu at rocketmail.com> wrote:
> If I unpickle from a non - pickled file or byte stream what error does
> Pickle throw? I've gotten both KeyError and EOFError, does it throw
> anything else?

Why?


From Gareth.McCaughan at pobox.com  Sun Jun  4 17:53:17 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: 04 Jun 2000 22:53:17 +0100
Subject: GTK broke things! Re: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home>
	<39397FD4.67C7FD73@inc.it> <867lc64a47.fsf@g.local>
	<393A2B6D.4B33A5C8@inc.it> <393A3902.F67CD1F8@stud.ntnu.no>
	<393A73CD.4EAAB457@inc.it>
Message-ID: <86n1l1p0z6.fsf@g.local>

Paolo Redaelli wrote:

> I was WRONG!
> GTK broke things in a way I can't even wonder HOW....
> I attack here two version. One functions and the other not. Little
> problem. The ONLY difference detween them is a "import gtk" at the
> beginning. Delete this line and it functions!!!!

Ah, that would explain why it worked for me. I removed
the "obviously" irrelevant imports at the start.

What you actually did at the start was not "import gtk"
but "from gtk import *". That puts everything into your
global namespace. Maybe gtk defines "float" or something
of the kind.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From davidw at linuxcare.com  Tue Jun 20 07:56:45 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 20 Jun 2000 13:56:45 +0200
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no>
Message-ID: <871z1sr2f6.fsf@eugene.prosa.it>

thomas at cintra.no (Thomas Weholt) writes:

> My mission is to swap expensive, oversized databases solutions like
> MS SQL 7.0 with PostgreSQL 7.0, crappy inconsistent script-languages
> like VBScript with Python and MS SiteServer / IIS with Zope, Linux
> with Windows 2000/ NT 4.0. I argue that the load on the system will
> never be to great for PostgreSQL to handle as supposed to MS
> SQL-server, VBScript, it sucks and they agree to that, Linux instead
> of Windows will save lots of cash, so will the "reduced" cost of
> PostgreSQL instead of MS SQL 7.0.
 
> The scenario above is normal in my experience, and I`d appreciate
> any feedback from people with similar ideas. Related Case-studies
> would also help alot.
 
Yes, you are correct.  Thousands of people are using similar setups.

> The thing is that these people don`t thrust OpenSource. They laugh
> at script-languages in general, and the idea of sharing code is even
> more stupid.
 
If worst comes to worst, there are a lot of other people looking to
employ people with open source coding skills who will appreciate your
knowledge;-)

> I need selling points for OpenSource, the idea and philosophy,
> examples of successful technology and eh .. moral support. I feel
> like I`m in the lions den right now. ( " Think of a happy
> place. Calm blue ocean, calm blue ocean ..." )

Well, www.opensource.org is a good place to start.

There is some stuff on our site that mibht be helpful, WRT to case
studies, and corporate support for Open Source.  For example:

http://www.linuxcare.com/linux-business/case-for-linux/articles.epl

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From tanzer at swing.co.at  Sun Jun 11 02:16:50 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sun, 11 Jun 2000 08:16:50 +0200
Subject: PATCH: Augmented assignment 
In-Reply-To: Your message of "Sat, 10 Jun 2000 15:22:19 EDT."
             <3942956B.2CAB4C9A@yale.edu> 
Message-ID: <m13113C-000wcEC@swing.co.at>

Paul Magwene <paul.magwene at yale.edu> wrote:

> Christian Tanzer wrote:
> > 
> .
> > 
> > It doesn't obfuscate code looking like:
> > 
> >    nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1
> 
> Please tell me you're not REALLY writing code that looks like that, when
> something like:
> 
> 	nested[i][j] = nested[i][j] + 1
> 
> will do just as well (and is a lot easier to read).

I always try to use the shortest possible names. But I avoid using
names which are meaningless.

If `i' and `j' are just plain loop indices (i.e., numbers), I would be
pretty annoyed if someone used `i_loop_variable' (I actually saw that
once in a FORTRAN program).

OTOH, if the indices select from a dictionary and the expression is
part of some complicated algorithm, I'd rather use expressive names --
and this usually are longer than 1 character.

Please-don't-jump-to-premature-conclusions-when-reading-a-quickly-made-up-
example-l'y yours,
Christian

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From bragib at my-deja.com  Thu Jun 22 08:53:23 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Thu, 22 Jun 2000 12:53:23 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand>
Message-ID: <8it27t$vmt$1@nnrp1.deja.com>

Steffen:

Are you sure you need to use re to do this?  What about:


>>> import string
>>> t = 'uid=joe, ou=orgunit, o=some\\org'
>>> string.split(t, ',')
['uid=joe', ' ou=orgunit', ' o=some\\org']



In article <m3k8fiosky.fsf at gondolin.beleriand>,
  Steffen Ries <steffen.ries at sympatico.ca> wrote:
> Hi,
>
> I'm trying use re.split to separate an LDAP Distinguished Name into
> its naming components. The naming components are of the form
> "attr=value" and they are separated by ",".
>
> The part, which I can't solve is, that "value" may contain the
> sequence "\,", which obviously does not separate a naming component.
>
> e.g.
> "uid=joe, ou=orgunit, o=some org"
> should be split into ["uid=joe", "ou=orgunit", "o=some org"]
>
> "uid=joe, ou=org\\, unit, o=some org"
> -> ["uid=joe", "ou=org\\, unit", "o=some org"]
>
> Is there a way to use re.split to achieve this? (I solved it searching
> [^\\], and splitting the string in a loop, but I was wondering if it
> can be done easier.)
>
> TIA,
> /steffen
> --
> steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!
>


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


From root at rainerdeyke.com  Thu Jun 29 15:18:35 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 29 Jun 2000 19:18:35 GMT
Subject: Constants and Enums
References: <Pine.GHP.4.21.0006291234020.9839-100000@don.et.byu.edu>
Message-ID: <fiN65.1659$6y5.1533336@news2.rdc2.tx.home.com>

C. Porter Bassett <porter at et.byu.edu> wrote in message
news:Pine.GHP.4.21.0006291234020.9839-100000 at don.et.byu.edu...
> I am sure that this topic has been beat to death, but I can't seem to find
> the answer to my question.  I am used to programming in C/C++, and using
> many enums and #define ed constants.  In the absense of these, I have
> resorted to using tons of global variables, which makes me
> uncomfortable.  I know that the general concensus is to use dictionaries
> for all this, but I can't figure out how a dictionary would be usefule to
> use in place of an enum.  I am trying to think in a more python way, but I
> need a little nudging to get my brain to go that way.

In C:

enum bob { fred = 1, otto = 2 };
int n = fred;

In Python:

bob = {'fred':1, 'otto':2}
n = bob['fred']

Alternately:

class bob:
  fred = 1
  otto = 2

n = bob.fred


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




From Stuart.Ford at Boeckh.com  Wed Jun 28 18:33:33 2000
From: Stuart.Ford at Boeckh.com (Stuart Ford)
Date: Wed, 28 Jun 2000 17:33:33 -0500
Subject: simple question ... exec in W32?
Message-ID: <DD768E0BEA95D211B27A0008C74C86250A6B92@BEKEXCH1>


Brian Alexander <crystalc at interlog.com> wrote in message
news:<Nxu65.20928$Ip.511269 at cac1.rdr.news.psi.ca>...
> Hello;
> 
> How to I start a W32 application, like Word etc, from a Python script? I'm
> actually trying to rewrite a backup script, which, additionally, does a
lot
> of non-backup processing overnight.
> 

Brian,

Look at the os module..

you can launch a process with the spawnv function..

Ex:  os.spawnv(os.P_WAIT, 'c:\\winnt\\notepad.exe', ())

This will launch notepad, and stop processing instructions until notepad
is closed..

There are several spawn modes available..
(P_WAIT, P_NOWAIT, P_OVERLAY and P_DETACH) 

Stuart



From steffen at biomedcomp.com  Sun Jun 18 10:59:42 2000
From: steffen at biomedcomp.com (David Steffen)
Date: Sun, 18 Jun 2000 09:59:42 -0500
Subject: Problem Installing/Using Windows Extensions
References: <2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1@lp.airnews.net> <BnS25.1127$fw6.20615@news6-win.server.ntlworld.com>
Message-ID: <A407A87E76046B5A.585BB97BB87A45ED.62347D8BDA3615B6@lp.airnews.net>

Because of help from this group, my problem is solved and I am
accessing Access from Python on my Windows95 Laptop.

THANKS!!

In reply to article
<642A954DD517D411B20C00508BCF23B001296EA2 at mail.sauder.com> from
richard_chamberlain <richard_chamberlain at ntlworld.com> :

> I had the same problem with a PC at work running W95, but not on any of the
> others. I finally narrowed it down to be the only one without Internet
> Explorer 5 installed. And from that I presume that DCOM needs to be
> installed.

I installed IE 5.01, and all was well.  You solved my problem.

THANKS!

In reply to article <mdY25.1544$qz4.13082 at news-server.bigpond.net.au>
from Mark Hammond <mhammond at skippinet.com.au> :

> Sheesh - On the "downloads" page (which is where you _must_ have got 132
> from), just under the explanation about 1.5 and 1.6 living together, it
> says:
> 
> 
> Should you have any problems installing these packages, please refer to
> the <a ..>Installation Problems Guide</a>.
> 
> Ive now made it BOLD, and also copied the line to the info about each
> specific build.
> 
> Please check it out, and let me know how I can possibly make it any
> clearer.  Your exact problem is listed there...

You are right and I am stupid :-)

Had I noticed this (which I definitely should have), I could have
solved my problem with a much smaller footprint than resulted from a
full Internet Explorer installation.  What you had on your page before
should have been adequate and the way you changed it makes it even
better.  Honest, I tried, please just chock it up to my aging brain.

Oh.  And thanks so much for the extensions - they are wonderful!

-David-


From jilanik at tin.it  Sat Jun  3 21:45:37 2000
From: jilanik at tin.it (Jilani Khaldi)
Date: Sun, 04 Jun 2000 03:45:37 +0200
Subject: exit(); halt; in pyhton
Message-ID: <3939B4C1.3FACC11D@tin.it>

Hi All,

How can I write this Pascal code in python?
var
  x: double;
begin
   if x = 0 then halt
  else
   do others things
end;

Freeing everything.

Thanks.

Jilani

--
<jk>
___________________________________________________
http://space.tin.it/scuola/jkhaldi
FreeBSD ~ The Power to Serve





From mhammond at skippinet.com.au  Sun Jun 18 00:14:28 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 18 Jun 2000 04:14:28 GMT
Subject: Optimizing Object Instantiation (was Re: How Fast Does Python Run?)
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net> <39499C91.3A1586BE@sightreader.com> <8if21t$4u5$1@slb1.atl.mindspring.net> <20000617120003.S26436@xs4all.nl> <see-0F7889.18054217062000@news.dnai.com>
Message-ID: <E0Y25.1541$qz4.13200@news-server.bigpond.net.au>

"Sam Penrose" <see at message.body> wrote in message
news:see-0F7889.18054217062000 at news.dnai.com...

> 1) Does informed opinion agree that object instantiation is
> noticeably performance-expensive in CPython? (Unfortunately,
> I can't seem to find the article in question.)

Sorry - no informed opinions here :-)

But I would guess that "type" creation (ie, creation of objects
implemented in extension modules) would be quite fast, and mainly
dependent on the native malloc() speed.

I could well believe that "instance" creation (ie, creation of user
defined class instances) would be quite slow, especially if the object in
question has an __init__ function!

Mark.





From jfr at genyinfo.com  Fri Jun 23 10:25:14 2000
From: jfr at genyinfo.com (Jean-Francois Richard)
Date: Fri, 23 Jun 2000 10:25:14 -0400
Subject: Can I connect to a DDE server?
References: <8iu3eu$fnf$1@nslave1.tin.it>
Message-ID: <yrK45.2680$SK2.51865@wagner.videotron.net>

Download pythonwin and look into the win32/demos/dde directory. There is a
sample
of a dde client and server.

"Mario Tola" <mtbros at tin.it> wrote in message
news:8iu3eu$fnf$1 at nslave1.tin.it...
> Hi,
> I'd like to know if it is possible (of course on Win) to send commands to
a
> DDE server (in my case it is ACDSee.exe) from python. And, if yes, how ?
> I think yes as I have read on "Python Programming on Win32" that
"PythonWin
> uses DDE to
> communicate between instances"... but I can't find documentation on how to
> use it myself.
> If someone could send me a small example...
> Many thanks...
> Mario
>
>
>
>




From thomas at xs4all.net  Fri Jun 16 05:10:11 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Jun 2000 11:10:11 +0200
Subject: broken examples in reference? Help??
In-Reply-To: <3949C65F.67A405F1@fizzylab.com>; from paul@fizzylab.com on Thu, Jun 15, 2000 at 11:17:03PM -0700
References: <3949C65F.67A405F1@fizzylab.com>
Message-ID: <20000616111011.M26436@xs4all.nl>

On Thu, Jun 15, 2000 at 11:17:03PM -0700, paul at fizzylab.com wrote:

> I am trying to work with python and it seems quite
> straightforward most of the time, but I fear I may have a
> broken binary or something else wrong.

I don't think so, 

> #!/usr/bin/python
> import httplib
> h = httplib.HTTP('www.python.org')

[..]

You try to connect to www.python.org, which seems to have some connection
troubles right now. At least, it's not accessible from a couple of my
machines, and only barely accessible from a few other, in a different
network. It might be specific to me, maybe alter.net is filtering or
source-routing some of our networks, but it looks like some maintenance is
being done or such.

It's easy to test if that's the problem though: telnet to 'www.python.org',
port 80, from the machine you try to run the script on. Or replace
'www.python.org' with a webserver closeby, like your ISPs webserver.


-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From db3l at fitlinxx.com  Tue Jun 20 14:30:53 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 20 Jun 2000 14:30:53 -0400
Subject: Why whitespace denotation of blocks doesn't work.
References: <65118AEEFF5AD3118E8300508B1248774CB34E@ALTNET>
Message-ID: <uvgz4gq76.fsf@ctwd0143.fitlinxx.com>

Mike Steed <MSteed at altiris.com> writes:

> In Vim it is easy to select a block of code and shift it in or out one
> 'shiftwidth'.  I don't know emacs, but I would be surprised if it did not
> provide a similar facility.

In Python mode, you can use C-c > and C-c < to shift a region right or
left (respectively) by your current indent.

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


From android at micron.net  Wed Jun 14 23:34:44 2000
From: android at micron.net (Peter Immarco)
Date: Wed, 14 Jun 2000 23:34:44 -0400
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my> <20000614200609.A323@activestate.com>
Message-ID: <39484ED4.2D3CBCD7@micron.net>

Prolog's excellent for this kind of problem too.  (Sorry about the null-answer on
doing it in Python. :)

Peter.

Trent Mick wrote:

> On Thu, Jun 15, 2000 at 10:44:03AM +0800, root wrote:
> > newbie question.
> >
> > has anyone wrote the 8 non-attacking queen problem and conway's game of life
> > in python? i try but face a few newbie problems.
> >
> > 1. conway's game of life:
> >    how to have a matrix in python?
> >    the following is not good;
> >
> >            r0=[0,0,0]
> >          r1=[0,0,0]
> >          ........
> >
> >          m=[r0,r1,....]
> >
> >    the matrix is fixed, which is not good. my solution use dictionary (i
> >    know this is not the right way). the key is a tuple of the matrix
> >    indices. the value is '0'. then base on game of life rule, update the
> >    dictionary and print for next generation.
> >
> >          m={}
> >          m[(0,0)] = 0, m[(0,1)] = 1,....
>
> How about this:
>
> >>> m =[]
> >>> for i in range(5):
> ...     m.append([])
> ...     for j in range(5):
> ...             m[i].append(i*j)
> ...
> >>> m
> [[0, 0, 0, 0, 0], [0, 1, 2, 3, 4], [0, 2, 4, 6, 8], [0, 3, 6, 9, 12], [0, 4,
> 8, 12, 16]]
> >>> m[0][0]
> 0
> >>> m[1][4]
> 4
> >>> m[2][4]
> 8
>
> However, if you are doing heavy linear algebra I believe you should check out
> NumPy.
>
> Trent
>
> --
> Trent Mick
> trentm at activestate.com



From pete at visionart.com  Thu Jun  1 14:21:47 2000
From: pete at visionart.com (Pete Shinners)
Date: Thu, 01 Jun 2000 11:21:47 -0700
Subject: Styling Temporary Code
Message-ID: <3936A9BB.FBEF9201@visionart.com>

I've been developing in C for several years and have
really been enjoying python recently.

one thing i do frequenctly in C is to place temporary
"debug/doublecheck/sanity" type code while i'm developing
it. i always put this code in with no indenting, so it's
easy to spot, work with, and cleanup. it looks something
like this...


int myfunc(int nullargs)
{
    int newval = do_fancy_stuff(nullargs);
cout << "newval = " << newval << endl;
    return more_fancy(newval);
}


ok, in this 5 line example it doesn't really sparkly,
but when i use it in big sections of code, it can't
be beat. call it the mark of the beast, chase me with
pitchforks, i simply love it and i don't want to change!

now in python this obviously won't work, but i haven't
been able to settle on a style that works for me yet.
what do other people use for this "working" code?


From rmilner at point-one.net  Tue Jun 20 14:56:59 2000
From: rmilner at point-one.net (Roman Milner)
Date: 20 Jun 2000 13:56:59 -0500
Subject: Python and mSQL
References: <8io97i$d6u$1@newsg1.svr.pol.co.uk>
Message-ID: <590BE2DAD3B7B942.25480992FB1DA19A.830393EE9789A0B1@lp.airnews.net>

>>>>> "JF" == Jim Ford <jaford at watford53.freeserve.co.uk> writes:

    JF> Hi I use mSQL version 1.0.16 on Linux and want to write a
    JF> script to access the database. I'm currently fiddling around
    JF> with Perl and am starting to get somewhere, but what little I
    JF> know about Python attracts me and I'm wondering whether it
    JF> would be a better choice.

    JF> What is the current state of a Python module to access mSQL,
    JF> please?

You can talk to mSQL from python via SQLRelay.

http://www.firstworks.com

^Roman





From jschmitt at vmlabs.com  Sun Jun 18 02:33:03 2000
From: jschmitt at vmlabs.com (John Schmitt)
Date: Sat, 17 Jun 2000 23:33:03 -0700
Subject: writing (Gnu)MAKE in Python
In-Reply-To: <394B266C.42F7EC63@san.rr.com>
Message-ID: <NEBBIOGBDMDPEANLFKGHAEAFCEAA.jschmitt@vmlabs.com>

This is a very good question, and if I'd answered it originally, I would
have generated less traffic.

I've always thought that I was handy enough at writing makefiles on whatever
platform.  Recently I've come across a very complicated makefile (that calls
sed and grep from bash), and under most Windows' it does not work very well.
I can get by with MKS, but we have been told that we cannot require our
customers to install MKS.  (Hey, can I just pull out the GNU binaries and
drop them into our SDK and distribute that?)  (On another note, does anyone
actually do this: grep for the string "CFLAGS" in a C source file, and add
that to your compile rule?  Seems too kludgy to me, but I've seen it.)  Make
bombs out with memory problems under NT and 98 sometimes, I can't quite
predict its foreslash and backslash behaviour (although I think I've got it
down now), and Make's process for finding a shell is incomprehensible.

Another reason is that I'd like to have exactly one development environment
on all platforms, even my Mac.  Sadly, until OS X, the GNU standard UNIX
stuff won't be available on the Mac.

Anyway, I think I might be able to make a strong Python case if I could say
that we would have many fewer problems on all our platforms.  I'm almost
done with this project, and then I'll enforce the cross-platform
requirements.

Not-nearly-as-clever-as-I-imagined-my-self-ly,
more-humbly-yours-ly,

John

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Courageous
Sent: Saturday, June 17, 2000 12:17 AM
To: python-list at python.org
Subject: Re: writing (Gnu)MAKE in Python



> In a smaller way, I would like to have Bash, sed, and grep as Python
> programs rather than compiled C.

Why?



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




From rcm at bcnartdirecte.com  Sun Jun 11 11:55:36 2000
From: rcm at bcnartdirecte.com (Rafael Cordones Marcos)
Date: Sun, 11 Jun 2000 17:55:36 +0200
Subject: 8-bit cleanliness
In-Reply-To: <39430E84.2679DD45@micron.net>; from android@micron.net on Sat, Jun 10, 2000 at 11:59:01PM -0400
References: <39430E84.2679DD45@micron.net>
Message-ID: <20000611175536.A3365@bcnartdirecte.com>

Hi,

I just started to use Python a few days ago because I got fed up of so much convoluted Perl code. ;)
Anyway, I have to read some text files and process the words appearing in them. I have discovered, to
my surprise, that accents like (?, ?, ...) get replaced by a 4 character code. Is there any class/module/option
available to read *text* files with non english characters in them?

Maybe the Python installed in my Debian system is compiled with the wrong options.

Thanks for your time.

Rafa

-- 
Unix IS user-friendly, it just chooses its friends very carefully.



From jkraska1 at san.rr.com  Fri Jun  2 21:59:37 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 03 Jun 2000 01:59:37 GMT
Subject: scoping problem with lambda
References: <3933484B.B471017A@san.rr.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <00060313245601.00960@localhost.localdomain>
Message-ID: <3938672B.32B646C9@san.rr.com>

> class trying_it:
>     def __init__(self):
>         self.tmp = [[['a'],['one','two','three','four']],
>                 [['b'],['one','two','four','five']],
>                 [['c'],['one','two','three','six']],
>                 [['d'],['one','two','two','four']],
>                 [['e'],['one','ten','three','four']]]
> 
>         self.kformat = ['one','two','three','four']
> 
>     def try_it(self):
>         tmp2 = self.tmp[:]
>         for i in (0,2,3):
>             tmp2 = filter(lambda x:x[1][i] in (self.kformat[i],),tmp2)
>         return tmp2


Python lambdas don't have the lexical closure you might expect,
but you can build it. To wit:

             tmp2 = filter(lambda x,i=i,self=self:x[1][i] in (self.kformat[i],),tmp2)

% py test.py

[[['a'], ['one', 'two', 'three', 'four']], [['e'], ['one', 'ten', 'three', 'four']]]


C/


From jon_mclin at attglobal.net  Tue Jun 27 00:10:55 2000
From: jon_mclin at attglobal.net (Jon McLin)
Date: Mon, 26 Jun 2000 22:10:55 -0600
Subject: OO Approach to file parsing with Python?
Message-ID: <3958294F.4B17E6F7@attglobal.net>

Our company has decided to try an initial deployment of Python in a file-parsing
role.  We need to translate some customer-provided structured ASCII files into
our standard schema for use by legacy applications.    A procedural
implementation is obvious, but completely non-reusable, and I'd like to be able
to leverage this in the future.  (different inputs, the same outputs.)  Can
someone point me to examples of OO approaches to this type of problem,
preferrably implemented in Python?

Some data, and thoughts:

The basic file structure is

Multi-Line Header
One or More DataSets
    where each Data Set consists of
    a single line header, and
    one or more lines of tab-delimited data (fixed structure across all sets).

The data sets are identified by the structure of their header lines.

A nested loop processor is the obvious procedural approach, assigning each input
field to a corresponding field in an output object.

One seemingly-OO approach is to create classes for each construct on the input
side.  The constructors would actually parse the file, and call subordinate
constructors (member classes) as appropriate.   (This seems awkward).
The output classes would match our schema, and of course these would be
reusable.

Since there is not a one-for-one correspondence between input and output in
either classes or records, the next step would be to derive the set of  outputs
from the inputs (quantity of output instances).    These would be constructed,
and the input structure would be walked, telling each instance to populate it's
output analogue.  Finally, the output hierarchy would be walked, telling each
instance to "Publish" itself.

With this basic model, the overall structure and the output classes would be
reusable, while the input classes and the translators (if not part of the input
classes) would be application specific.

Conceptual Code Snippet:
#  Parse and Create an object hierarchy
#    MyInput becomes an object containing member objects, dictionaries, etc..
#    SourceData is a class in an ApplicationSpecific
MyInputObject=SourceData("MyFile")

# Translate and create output hierarchy.
#  The custom classes will need to import the standard OutputObject module,
which contains constructors
#    for all output classes.
MyOutputObject=MyInputObject.Translate()

#  Create the desired (standard) output
MyOutputObject.Publish()

But as I noted, parsing with this model seems awkward.

Any pointers/constructive criticisms/suggestions are much appreciated.
Non-constructive criticism will be deflected without ego damage.   After all, I
know that I know far less than I know not, and that furthermore far more is
unknown than is known to be unknown.   Sigh.   ;>

Thanks,
Jon





From klaus.doerfler at stud.fh-regensburg.de  Fri Jun 23 08:09:27 2000
From: klaus.doerfler at stud.fh-regensburg.de (Klaus Doerfler)
Date: Fri, 23 Jun 2000 14:09:27 +0200
Subject: COM question
Message-ID: <39535377.6E6E74FA@stud.fh-regensburg.de>

Hi!

I'm a Python-newbie so I apologize ...
I hope someone could help me with my (simple?) COM question:
I'm writing a simple COM server and I want to use early-binding, so how
exactly does the server know about what the interface in the tlb-file is
like and how do I have to use it in the server? Are there any naming
conventions?

Thanks in advance,


Klaus


From chris at ucmb.ulb.ac.be  Wed Jun 28 04:55:49 2000
From: chris at ucmb.ulb.ac.be (Christian Lemer)
Date: Wed, 28 Jun 2000 08:55:49 +0000
Subject: jpython casting to abstract java class
In-Reply-To: <395820B1.417B380E@crystal.uwa.edu.au>
References: <39570C59.2CBCFAEB@crystal.uwa.edu.au> <395820B1.417B380E@crystal.uwa.edu.au>
Message-ID: <Pine.LNX.4.10.10006280845130.3063-100000@vangogh.ulb.ac.be>

On Tue, 27 Jun 2000, Douglas du Boulay wrote:

> 
> > 
> > Hi,
> > I have a small(ish) code fragment below which fails to
> > cast a python-subclass of an abstract-java-base-class back to that
> > java-base-class when it is passed as input to another
> > java (constuctor) method.
> 
> Further testing suggests that the "abstract" word in the above is
> irrelevant. It should say: 
> 
> any JPython scripted class which inherits from, or subclasses a 
> java based class cannot be cast back to that base class when 
> passed back to another java based method requiring that 
> base class type as an argument.
> (hope that makes sense)

Maybe I miss something, I'm quite new to java but have a two year
experience in Python.

Recently, I made a prototype of a query language tool for a Java
object-oriented database. My idea was to allow the end-user to extend the
query-language by writing some specific 'filters' and 'comparison
functions' from withing JPython. The database backend has to process some
algorithms using the filters and comparsion functionss.

For this I wrote some interfaces in java and implemented them by
subclassing them in python. I could then pass an instance of the
object from the jpython interpreter to the java filtering method
without any problem...

> 
> I have no idea where to start looking to figure out where it is going
> astray. Help appreciated.
> Thanks again
> Doug
> 
> 

-- 
Christian Lemer
Service de Conformation des Macromolecules Biologiques et de Bioinformatique
Universite Libre de Bruxelles                           Brussels Free University
mailto:chris at ucmb.ulb.ac.be   +32.2.648.52.00   http://www.ucmb.ulb.ac.be/~chris
    PGP: 94 40 92 E5 BC 4A C0 B1  64 87 6E 11 58 71 B1 DD 56 F5 AF E7






From kern at caltech.edu  Mon Jun 19 15:31:43 2000
From: kern at caltech.edu (Robert Kern)
Date: Mon, 19 Jun 2000 15:31:43 -0400
Subject: Kjbuckets-Module, Python 1.6 and NT
References: <394E4B02.5F4A0AB4@szs.ira.uka.de>
Message-ID: <394E751F.DE9E5FF4@caltech.edu>

Thomas Freytag wrote:
> 
> Dear all,
> 
> we have severe trouble getting the kjbuckets-module running with Python
> 1.6 and Windows NT. We have tried the following so far:
> 
> 1)Compile the kjbucketsmodule.c as a dll. Fails and produces some
> "Initialiser is not a constant" error, whatever...

Edit the code as mentioned in section 3.24 of the FAQ.

That is, kjbuckets has several extension types.  There will be lines
like the following (taken from the FAQ, not kjbuckets):

  statichere PyTypeObject bstreamtype = {
      PyObject_HEAD_INIT(&PyType_Type)
      0,
      "bstream",
      sizeof(bstreamobject),
  ...


Replace &PyType_Type with NULL.  Then add something like the following
line to the initialization module:

      bstreamtype.ob_type = &PyType_Type;

Repeat for all extension types.

> 2)Re-compile the Python interpreter using Visula C++ 6.0 and adding
> kjbucketsmodule.c to the files to be considered in the project. Plus we
> edited the "setup.in" in the Modules directory.  C++ claims to copile
> "kjbucketsmodule.c", however, when we try  "import kjbuckets" with the
> freshly compiled interpreter the answer is "No module....".

Don't know what's happening here.

> Sooo...has anyone managed to get kjbuckets running with 1.6 and NT? Any
> hints are greatly appreciated.
> 
> Thanks a lot.
> 
> Thomas.

HTH.

--
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From phd at phd.russ.ru  Wed Jun 28 08:33:04 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 28 Jun 2000 12:33:04 +0000 (GMT)
Subject: compilation
In-Reply-To: <8jaht1$rui$5@bmerhc5e.ca.nortel.com>
Message-ID: <Pine.LNX.4.21.0006281232450.15654-100000@fep132.fep.ru>

On 27 Jun 2000, Michael P. Soulier wrote:
> 	Neat. So, once it's compiled into a pyc and pyo files, is there a way to
> run those directly for better performance? I know that freeze links in the
> interpreter. I guess I should use that if I want it. 

   http://www.lyra.org/greg/python/#dev

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From digitig at cix.co.uk  Tue Jun 27 16:01:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 27 Jun 2000 21:01 +0100 (BST)
Subject: boolean true and false values.
References: <3957FA7F.7C982483@cosc.canterbury.ac.nz>
Message-ID: <memo.20000627210104.56821B@digitig.compulink.co.uk>

In article <3957FA7F.7C982483 at cosc.canterbury.ac.nz>, 
greg at cosc.canterbury.ac.nz (Greg Ewing) wrote:

> Tim Rowe wrote:
> > 
> > > >what would you delimit it with? :-)
> > >
> > > {x,y,z} ?
> > 
> > That's going to get me confused...
> 
> Maybe that could be syntactic sugar for a dictionary which
> maps the keys x, y and z to some default value, e.g. 1.
> Then add some operators for doing set operations on
> dictionaries of that form, and hey presto -- a set type
> without adding a new type!

True, albeit a rather inefficient one in terms of space. I don't see 
what's wrong with a set class and doing 's = set(x, y, z)'.


From rjroy at takingcontrol.com  Tue Jun 13 18:41:30 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Tue, 13 Jun 2000 22:41:30 GMT
Subject: unsettling tkinter bug -- or is it just me?
References: <39458AFC.8AD5B850@tricity.wsu.edu>
Message-ID: <3946b62a.507360500@news1.on.sympatico.ca>

On Mon, 12 Jun 2000 18:14:36 -0700, Bob Lewis <bobl at tricity.wsu.edu>
wrote:

>Consider the following code:
>--------
>
>from Tkinter import *
>
>root = Tk()
>img = PhotoImage(file='bobl.gif')
>
>def func():
>    cnvs = Canvas(root, width=img.width(), height=img.height())
>    cnvs.create_image(0,0, image=img, anchor=NW)
>    cnvs.pack(side=TOP)
>    Button(root, text='Quit', command=root.quit).pack()
>
>func()
>root.mainloop()
>--------
>This runs just fine: It creates a root toplevel with "bobl.gif" displayed
>in it and a "Quit" button.
>
>Now we make one wafer-thin change: Move the the "img =" inside "func()":
>--------
>from Tkinter import *
>
>root = Tk()
>
>def func():
>    img = PhotoImage(file='bobl.gif')
>    cnvs = Canvas(root, width=img.width(), height=img.height())
>    cnvs.create_image(0,0, image=img, anchor=NW)
>    cnvs.pack(side=TOP)
>    Button(root, text='Quit', command=root.quit).pack()
>
>func()
>root.mainloop()
>--------
>And now, while the canvas has the right dimensions, the image does
>not appear.  Okay, so "img" went from a global to local scope, but
>that shouldn't have affected execution, right?  Have I missed something?
>
>This is on a Red Hat Linux system running versions 1.5.2-13 of both
>python and tkinter.
>
>	- Bob Lewis
>	  bobl at tricity.wsu.edu

I was bitten by this also. You must persist the PhotoImage instance
independently. I seem to recall that this is documented somewhere, but
I can't remember where.


Bob


From donn at oz.net  Wed Jun  7 00:47:07 2000
From: donn at oz.net (Donn Cave)
Date: 7 Jun 2000 04:47:07 GMT
Subject: Need help on Python/C++
References: <8hifbl$k55$1@gw.localdomain> <8gltp7$f7f$1@gw.localdomain> <p6qitvx4aes.fsf@informatik.hu-berlin.de>
Message-ID: <8hkk4b$n0p$0@216.39.151.169>

Quoth "Sameh chafik pro" <chafik at nevrax.com>:
| That  is okay, but why i do to translate a class write in C++. The
| problem that the standard way is to use the PyTypeObject and define the
| constructor as a function, but python don't recognize the translation as
| class but as type object.morality if i do the translation of a class "X" it
| is'nt posible in python to use it as a derivate class:
| class A(X):
|     def doSomething(self):
|     .....
|     ....
|
| is not possible. Because X is a function.That seem insignificant but i want
| to do an efficient translation between C++ code and Python.

I'll go along with Ken Seehof's response in another thread, where
you posted what appears to be more or less this same question.

What you need is a Python class, written in Python, that binds itself
to the internal type instance that you wrote in C++.  This binding
can support both the C++ functions, invoked from Python, and Python
functions invoked from C++ virtual functions.

I do this for the BeOS API, which is C++.  I got most of the important
ideas from a Jim Ahlstrom paper some time back,

http://www.python.org/workshops/1995-12/papers/

	Donn Cave, donn at oz.net


From ge at nowhere.none  Thu Jun  8 14:14:51 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 08 Jun 2000 18:14:51 GMT
Subject: Tkintr: how to seize focus in a transient?
Message-ID: <voR%4.1057$zm3.238149@ptah.visi.com>

I'm sure it's in Grayson somewhere, but I can't find it.

In a transient dialog window, how do I disable focus for the
rest of the app?  I only want to prevent focus for the rest of
the app, not the whole X11 system.

Know what I mean, Vern?

-- 
Grant Edwards                   grante             Yow!  This PIZZA symbolizes
                                  at               my COMPLETE EMOTIONAL
                               visi.com            RECOVERY!!


From dalke at acm.org  Mon Jun 26 13:29:38 2000
From: dalke at acm.org (Andrew Dalke)
Date: Mon, 26 Jun 2000 11:29:38 -0600
Subject: unexpected behaviour in class variables
Message-ID: <8j83ml$2s7$1@slb3.atl.mindspring.net>

Hello,

I found an unexpected behaviour in which C-based functions as class
attributes are treated different than user-defined (python-based) functions.
This case is documented, so I guess it's my misunderstanding.  However,
the docs also say the behaviour should occur for instance attributes and
it doesn't.  A detailed description is below.

I have a wrapper around a set of C-based extension functions.  Objects
in the C library must be freed when no longer needed, so I wrote a class
which looks like:

import dayswig_python
class smart_ptr:
  dt_dealloc = dayswig_python.dt_dealloc
  def __init__(self, handle):
    self.handle = handle
  def __del__(self):
    self.dt_dealloc(self.handle)

I keep a reference to dt_dealloc in the class namespace because the order
of module cleanup is not defined, and it is possible that dayswig_python
is cleaned up before this smart_ptr module.

This approach works fine, so long as dt_dealloc is not a user-defined
function (that it, it is an extension object in C rather than in Python).

I started writing a emulator dayswig_python.py file which implements just
enough of the extension library to be useful for testing.  Now my smart_ptr
class fails because it converts my user-defined dayswig_python.dt_dealloc
function into a method.  Here's an example:

import math
def eggs(x):
 pass

class Spam:
 builtin_func = math.cos
 def_func = eggs
 def p(self):
  print self.builtin_func
  print self.def_func
  print math.cos

>>> spam = Spam()
>>> spam.p()
<built-in function cos>
<method Spam.eggs of Spam instance at 1064060>
<built-in function cos>
>>> spam.builtin_func is math.cos
1
>>> spam.def_func is eggs
0

When the class attribute is a user-defined function, it is treated as a
member function.  When it is a function from an C extension, no conversion
occurs.

Looking in the manual I see this is documented:
> When a class attribute reference would yield a user-defined function
> object, it is transformed into an unbound user-defined method object
> (see above).

The "(see above)" says:
> User-defined method objects are created in two ways: when getting an
> attribute of a class that is a user-defined function object, or when
> getting an attributes of a class instance that is a user-defined function
> object.
and
> Also notice that this transformation only happens for user-defined
> functions; other callable objects (and all non-callable objects) are
> retrieved without transformation.

So my situation is the documented case of "getting an attribute of a
class that is a user-defined function object".  Don't know why this
must be the case -- is that an indication of how Python internally stores
member functions?

Note, however, that the other case of "getting an attribute [sic] of a
class instance that is a user-defined function object" is not the
implemented
behaviour!

import math

def eggs(x):
    return math.sin(x)

class Spam:
    def __init__(self):
        self.sin = eggs
        self.cos = math.cos

>>> spam = Spam()
>>> print spam.sin(0)
0.0
>>> print spam.cos(0)
1.0

As I read the documentation, spam.sin is an attribute of a class instance
and it is a user defined function, so it should be converted to a member
function.  But it isn't converted.

So:
  1) it is unexpected to me that class attributes which are user-defined
functions get coerced into member functions.  Can someone explain why that
is appropriate?  Not having the special case would be my (so far only)
Python3K wish :)

  2) The docs say class instance attributes which are user-defined functions
should be similarly coerced.  They do not appear to be coerced, so either
the implementation is wrong or the docs are wrong.  Could someone verify
that my interpretation is correct, so I can properly forward this to either
the bugs or the docs email?

Regarding my original problem, the fix is to turn the deallocator into:

  def __del__(self, dt_dealloc = dayswig_python.dt_dealloc):
    dt_dealloc(self.handle)

                    Andrew
                    dalke at acm.org





From m.faassen at vet.uu.nl  Tue Jun 13 09:41:50 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 13 Jun 2000 13:41:50 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au>
Message-ID: <8i5dmu$gac$2@newshost.accu.uu.nl>

Neil Hodgson <neilh at scintilla.org> wrote:
[snip]
> Coming from an RSX/11M+, Mac, OS/2, Windows background I found the syntax
> choices in Perl, which probably appear obvious to native Unixers,

Um, no. :)

While I'm a fairly new Unix user (8 years or so, 5 years of Linux), 
the syntax choices didn't appear obvious to me at all.

Regards,

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


From kens at sightreader.com  Thu Jun 15 13:38:28 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 10:38:28 -0700
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com>
Message-ID: <39491494.671DC3D9@sightreader.com>



"Frank V. Castellucci" wrote:

> Courageous wrote:
> >
> >
> > > As someone who uses and likes both C++ and Python, I think estimates
> > > like that are excessively simplistic.
> >
> > Me, too. And yet I will claim that I've become *more* productive
> > in Python after a brief 4-6 weeks of use than I am in C++ after
> > 15 years of use, and that's no exaggeration. If Python executed
> > as fast as other languages, I'd be tempted to *require* that our
> > staff use it. And please don't take this wrong; I'm no language
> > groupie -- I use C, C++, Java, Python, and Lisp -- all of them both
> > lately and frequently.
> >
> > > I use C++ for commercial programming, because it gives me the power and
> > > control I need.  Productivity isn't the best when measured in time to
> > > get the first prototype running, but I've found nothing better when
> > > measured in time to develop, deploy, maintain and support something a
> > > customer will pay for.
> >
> > Shrug. If it weren't for execution speed, your C++ environment would
> > be toast. Admit it. :)-
>
> While the last line seems rather toungue in cheek, there are other
> things that C++ brings to the table such as encapsulation, access
> control and type strictness.

All three of these items were intentially left out of python because python
would not benefit from them, though C++ does.

"Frank V. Castellucci" might have written (but actually didn't):
> A fish has certain definite advantages over a duck.  For example a fish has
gills.

A duck floats, so it doesn't need gills.  Fish need gills because they can't
float.  Besides, gills would look really ugly on a duck.

Encapsulation, access control and type strictness are often erroneously
referred to as if they were features.  They aren't.  They are aspects of a
design philosophy.  To make a system less error prone, constrain the programmer
in order to prevent the programmer from writing bugs.  Of course, that's
impossible, since it is always possible to write bugs anyway.  Because C/C++
are extremely error-prone languages, they need stuff like encapsulation,
access, control and type strictness.  Because python is less error prone, it
doesn't need these "features".  Python is based on different philosophy: To
make a system less error prone, make it simple.

You have to ask, "what is the actual value of these features?"  The answer is
that they prevent certain kinds of bugs from appearing.  Now ask, "what bugs
would a C++ programmer encounter if encapsulation, access, control and type
strictness were removed from C++."  Undoubtedly it would be a mess.  Get a
picture in your mind of these bugs.  Now ask a python programmer how often
these kinds of bugs come up.  Answer: "can't remember that ever happenning!"

Look at the big picture.  Do encapsulation, access control and type strictness
really make C++ really less error prone than python?

I think not.

> --
> Frank V. Castellucci

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/a8101ad2/attachment.vcf>

From michael at stroeder.com  Wed Jun 14 03:44:52 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 14 Jun 2000 09:44:52 +0200
Subject: Standalone http server, cgi software?
References: <3946EEA9.DFCBAB4D@erols.com>
Message-ID: <394737F4.B3003678@stroeder.com>

"Edward C. Jones" wrote:
> 
> I am a fairly experienced Python programmer but when it comes to
> http servers, etc, I am a newbie. I would like to be able to fill
> out an html form, press the submit button, and execute a Python
> program with data from the form.

Use the module cgi for processing the form data. There is an easy
example in the docs of the Python lib.

> httpserver
> [..]
> keywords 
> [..]
> standalong cgi server python forms.

Do you really need a stand-alone server? If yes, why? If you're
looking for persistent web application programming you have to face
several other problems like secure session managment, concurreny
etc.

For a really full-blown web application environment the standard
answer is http://www.zope.org.

Ciao, Michael.


From mogul at primus.ca  Wed Jun  7 18:27:52 2000
From: mogul at primus.ca (Robin Porter)
Date: Wed, 7 Jun 2000 18:27:52 -0400
Subject: Definite Newbie Question....
Message-ID: <q_z%4.828$223.8755@news1.tor.primus.ca>

    How is it that you run code once you enter it?  I am using Python 1.5.2
in Win 9x.  After inputing the code in the Python GUI, and saving it, I have
been trying to run it by pressing F5 or pressing edit (top menu) and then
selecting RUN at the bottom.  Please tell me what I am doing wrong?
    Thank you so much for your help.

Robin

Here is the code:

TaxRate  = 0.23
MinPay   = 300.0
RegHours = 40.0

TaxPayable = 0
Overtime = 0
OvertimeHours = 0.0

HoursWorked = float(raw_input("Please enter the hours worked per week? "))
PayRate = float(raw_input("Please enter the rate of pay? "))

if HoursWorked > RegHours:
    OvertimeHours = HoursWorked - RegHours
    Overtime = (OvertimeHours) * (PayRate*1.5)

GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime

if GrossPay > MinPay:
  TaxPayable = (GrossPay-MinPay)*TaxRate

NetPay = (GrossPay-TaxPayable);

print "The gross pay is %f" % GrossPay
print "The net pay is %f" % NetPay






From skip at mojam.com  Tue Jun 27 12:50:54 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 27 Jun 2000 11:50:54 -0500 (CDT)
Subject: ANN: cgi-buffer 0.1
Message-ID: <14680.56174.25987.149715@beluga.mojam.com>

Mark Nottingham wrote:

    cgi_buffer is a Python libary that may be used to improve performance of
    CGI scripts in some circumstances, by applying HTTP mechanisms that are
    typically not supported for them.

After taking a quick peek at the referenced web page for cgi-buffer, a
couple things came to mind:

    1. Python 1.6 will support HTTP/1.1.  Don't know if this will affect
       your library or not.

    2. Python 1.6 will have a new module, atexit, that adds a little smidgen
       of protocol on top of sys.exitfunc in the form of a register
       function, thus allowin multiple modules to register exit functions
       without stepping on each others' virtual toes.  If atexit gets loaded
       and sees that sys.exitfunc is already defined (and not its own exit
       func) it will simply add that function to the list of exit functions,
       so your code should work unchanged *if* it gets there before atexit
       does.

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
On Phil Jackson's ability to manage multiple Lakers superstars, Shaquille
O'Neal said: "He's got the rings.  I listen to the man with the rings."



From ax at infocentr.ru  Fri Jun 16 16:41:16 2000
From: ax at infocentr.ru (Aleksey Kishkin)
Date: Sat, 17 Jun 2000 02:41:16 +0600
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com> <394A7763.DE2A771D@infocentr.ru> <jwbnews-21EBF8.13275116062000@news.olympus.net>
Message-ID: <394A90EC.A8410046@infocentr.ru>

"John W. Baxter" wrote:
> 
> In article <394A7763.DE2A771D at infocentr.ru>, Aleksey Kishkin
> <ax at infocentr.ru> wrote:
> 
> > Akira Kiyomiya wrote:
> > >
> > > I wonder if Python runs faster than Perl.
> > >
> > > Could someone give me some comparison and contrast with some languages
> > > such
> > > as Perl, Tcl, JavaScript, etc?
> > >
> > > I am just curious.
> > >
> > > Thanks
> > >
> > > Akria
> >
> > I did. I found the programm (sieve) for java and rewrote this
> > programm to all languages which I had at that moment.
> > It worked 10 seconds and printed how much this task was solved.
> >
> > (it means bigger amount is better)
> >
> > TCL  - 13 times / 10 sec
> > perl - 64
> > python - 36
> > C without optimization - 6468
> > C with optimization  - 19111
> > Java without JIT (1.1.7) - 921
> > Java with JIT (kaffe)  - 8230
> > pascal (fpc)   - 4451
> >
> > Well. that's all
> 
> Having encountered the infamous sieve years ago in an article in "Byte"
> which made Pascal on 6502 (1 mhz) look slower than Pascal on Z80 (2
> mhz), when in fact the reverse was true, I discount all such claims
> unless I see (and understand) the code.
> 
> Part of the problem in the Byte code was that runtime checking was on by
> default in the compiler on 6502, and off by default in the compiler for
> the Z80.  There were other issues as well.  (You also don't indicate
> what compilers are used for any of the above (unless fpc specifies a
> particular Pascal).
> 
> All that is separate from the fact that the sieve isn't real-world code
> (aside from classes and benchmarks, how often is it actually used)?  But
> if properly written, it is indicative.
> 
 Well. I dont think either that my test was really exact estimation.
And it was just "5-minutes work" for fun after a workday..
I used translators "as is", I hadn't change any settings, 
and, speaking frankly, I wanted to know how fast was java, 
not python..
But if you want to know versions of translators, 

tcl/tk 8.0
perl 5.005
python 1.52
C - gcc 2.95, without optim. just " gcc sieve.c ",
   with optimization "gcc -O2 sieve.c "
Java without JIT -JDK 1.1.7 from blackdown.org 
and with JIT - kaffe (I can't remember version, it was from 
  distribution of linux mandrake 7)
pascal - fpc

Computer Pentium 233MMX 80 Mb Ram linux, mandrake 7

Sorry again, I don't think that it was strong enough test, but
I thought it might be interesting....









>   --John
> 
> --
> John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From jvickroy at sec.noaa.gov  Thu Jun 29 10:10:17 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Thu, 29 Jun 2000 08:10:17 -0600
Subject: accessing SQL Server from HP UX
Message-ID: <395B58C9.9F18E2CF@sec.noaa.gov>

Hello all,

Has anyone used Python to access a Microsoft SQL Server database from an
HP-UX platform?  If so, could you please tell me the driver software
being used.

Thanks,

- jv



From mfletch at tpresence.com  Fri Jun  9 11:50:10 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 9 Jun 2000 11:50:10 -0400 
Subject: Finding cycles and garbage, was( RE: Python Productivity over
	 C++)
Message-ID: <CC59868A0A76D311B1F50090277B321C1F339F@VRTMAIL>

Hmm, Neil's patch doesn't seem to be compatible with Stackless? (this app is
stackless-dependent (10s of 1000s of micro-threads of control, huge numbers
of waiting/suspended/manipulated micro-threads)).  Cyclops (in my fumbling
hands) hasn't come up with any notable cycles (there was a single one that
kept a single (small) object in memory, but that was the stackless
scheduler, which is seldom created/destroyed).  I'm looking for something
lower-level, like the VC++ feature where, at the end of the run, the
debugger says "hey, as I was doing shutdown, I noticed that the following
objects (including strings, floats, etceteras) were still in memory, that's
probably not what you wanted".  Neil's patch should be able to do that...
just need to get that and stackless into core Python and we're rocking :) .

Ah well, servers were meant to be restarted every 60,000 hits or so ;o) .

Enjoy,
Mike

-----Original Message-----
From: Michael Hudson [mailto:mwh21 at cam.ac.uk]
Sent: Friday, June 09, 2000 10:56 AM
To: python-list at python.org
Subject: Re: Python Productivity over C++
...
Have you tried Neil Schemenauer's patch?

http://www.enme.ucalgary.ca/~nascheme/python/gc.html

It can help with some of these things.

There's also cyclops, which I think is somewhere on ftp.python.org.
...



From dana at oz.net  Sun Jun 11 00:33:45 2000
From: dana at oz.net (dana_booth)
Date: 11 Jun 2000 04:33:45 GMT
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21> <3942DC44.4A85EEA7@rubic.com> <39430A2E.99CE3FC@yahoo.com>
Message-ID: <8hv4r9$pnt$0@216.39.141.21>

Sam Wun <samwun at yahoo.com> wrote:

SW: I've tried to use command python build.py to build the module, but it
SW: complains the _mysqlmodule.so is not found...

I got ahold of this module, and it compiled just fine. Just make sure that
in the "Setup" file, the paths to mysql are okay. For instance, on mine, I
had to change from "/usr/include" and "/usr/lib" to "/usr/local/..."

I'm using this module to connect, which it's doing just fine. I'm sort of
stumbling through doing much of anything right off the bat, though, since
the only documentation is in the .py module itself.

Also, this module compiled right away on a Windows 98 computer, too, which
is what I really wanted, anyway.

-- 

-----
Dana Booth <dana at oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From jkingry at my-deja.com  Thu Jun  8 13:57:04 2000
From: jkingry at my-deja.com (jkingry at my-deja.com)
Date: Thu, 08 Jun 2000 17:57:04 GMT
Subject: setblocking in Win32
Message-ID: <8homp6$a47$1@nnrp1.deja.com>

Hi, I think I'm going to be told what I don't what to hear, but from
the error below, I think I have surmised that trying to set non-
blocking on a socket in a Win32 environment doesn't work.

Are there any work arounds?

Joe

Traceback (innermost last):
  File "D:\Prog\Python\Lib\threading.py", line 376, in __bootstrap
    self.run()
  File "D:\Prog\Python\Lib\threading.py", line 364, in run
    apply(self.__target, self.__args, self.__kwargs)
  File "io.py", line 80, in run
    sock,(address,port) = server.accept()
  File "D:\Prog\Python\Lib\plat-win\socket.py", line 36, in accept
    sock, addr = self._sock.accept()
error: (10035, 'winsock error')


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


From schorsch at schorsch.com  Mon Jun 26 04:41:10 2000
From: schorsch at schorsch.com (Georg Mischler)
Date: Mon, 26 Jun 2000 08:41:10 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au> <39505457.5223ADC5@san.rr.com> <20000621155338.C11308@fangorn.csse.monash.edu.au> <dj93lsk9jgln6534nmgqnl8m8sq4anqunt@4ax.com> <3951A7F9.A067CBBC@san.rr.com>
Message-ID: <8j74v4$u8a$1@nnrp1.deja.com>

Courageous wrote:
>
> > Very cool.  But as long as you are patiently explaining things,
> > can you tell me how you generated the "?" character inside gvim?
> > The  "Alt-0187" trick did not seem to work.
>
> It's:
>
> <ctrl-V>187

Something related that I just played around with:

If you want vim to warn you when you are mixing spaces and
tabs in a file, you can use the following:

  syn match pyTabSpace "\t\t*  *" contains=pySpaceTab,pyInitTab
  syn match pySpaceTab "  *\t\t*" contains=pyTabSpace,pyInitSpace
  syn match pyInitSpace "^  *"    contains=PySpaceTab
  syn match pyInitTab "^\t\t*"    contains=PyTabSpace

  hi link pyTabSpace    Error
  hi link pySpaceTab    Error
  "hi link pyInitSpace   Error
  "hi link pyInitTab     Error

This will hightlight any combination of tabs and spaces
in an annoying color, so you get motivated to fix them.
If you want to avoid either leading spaces or tabs on
any line altogether, then you can uncomment one of the
last two lines, to your choice (remove the '"').

Typically, those lines go into the vim-5.x/syntax/python.vim
file of your installation, so they take effect every time
you open a python file (the syn and hi[light] sections
are seperated in there, but that's hard to miss).
It would actually be niceto encapsulate this in some ifdef
clauses, so that the user can configure the highlighting
in his .vimrc file. Maybe another time...

Most of this will work fine together with the special tab
display explained in the post that led to the above,
except that only the spaces will get the error highlighting
if you have "listchars=tab:xxx" and "set list" activated.
But then, this means that you prefer tabs anyway, so the
result should still be useful.


Have fun!

-schorsch

--
Georg Mischler  --  simulations developer  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


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


From mal at lemburg.com  Wed Jun  7 07:22:03 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 07 Jun 2000 13:22:03 +0200
Subject: unicode in python
References: <8g3m0b$nb7$1@gw.localdomain> <uhgV4.36$Z9.2056@news> <8gaor8$6kq$1@gw.localdomain> <8gaveq$7fm$1@gw.localdomain>
Message-ID: <393E305B.54E78D47@lemburg.com>

Sameh chafik pro wrote:
> 
> I see that thy are a Unicode support.
> if S is a unicode sitring we can write s = u'thome thing'. that work quite
> well in python mode.
> my problem is how can i do to catch this string in a C/C++ programe with the
> PyArg_ParseTuple fonction.
> can i na?fly write PyArg_ParseTuple(Arg,"U",AString);
> I try that and it dos'nt work.

The above will give you a Unicode object. You may want to
look at the API docs in 1.6 for more infos on how to convert
them to other storage types for use in C/C++. The file
Include/unicodeobject.h also has quite a lot to say :-)

Note that

 Py_UNICODE *v;
 PyArg_ParseTuple(Arg,"u",v);

will give you access to the raw Py_UNICODE buffer in the
Unicode object.
 
> Sam.
> 
> Sameh chafik pro <chafik at nevrax.com> a ?crit dans le message :
> 8gaor8$6kq$1 at gw.localdomain...
> > Okay, i turne my question in an ather way,
> > How is the standard way to print text in Chinese or Japanese with the
> Python
> > API ?
> > In reality i'am looking for a solution for the international display in
> > multi-plate form and i want to use Python for resolve this problem.
> >
> > Sam.
> >
> > Robert <rhicks at rma.edu> a ?crit dans le message : uhgV4.36$Z9.2056 at news...
> > > It is being written into version 1.6...
> > >
> > > Bob
> > >
> > > Sameh chafik pro <chafik at nevrax.com> wrote in message
> > > news:8g3m0b$nb7$1 at gw.localdomain...
> > > > Hello evry body,
> > > > I want to know if there are any Unicode module support in python ansd
> if
> > > yes
> > > > were can i find a document or a sample for using it.
> > > > thank.
> > > >
> > > >
> > >
> > >
> >
> >

-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/



From jcm at bigskytel.com  Wed Jun 28 20:27:16 2000
From: jcm at bigskytel.com (David Porter)
Date: Wed, 28 Jun 2000 18:27:16 -0600
Subject: (Serious?) package namespace problem (and a proposal)
In-Reply-To: <slrn8ll4ak.25h.hzhu@rocket.knowledgetrack.com>; from hzhu@knowledgetrack.com on Wed, Jun 28, 2000 at 11:56:19PM +0000
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <8F61BBD91gmcmhypernetcom@199.171.54.194> <slrn8ll4ak.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <20000628182716.B12278@novara.avenue>

* Huaiyu Zhu <hzhu at knowledgetrack.com>:

> >You could use Greg Stein's imputils, but in this case just packagizing the 
> >whole thing is a more straightforward solution.
> >
> 
> Where can I find it?  

http://www.lyra.org/greg/python/




From jkraska1 at san.rr.com  Wed Jun  7 23:33:49 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 08 Jun 2000 03:33:49 GMT
Subject: Please help!!!!
References: <7WD%4.1099$0l2.8049@news2.tor.primus.ca>
Message-ID: <393F14B9.320983C2@san.rr.com>

> Could someone please tell me what is wrong.  I have re-installed Python
> 1.5.2.  I have tried other code and all seems to lead me to the same
> problem.  Please help!

This code works as is in my version of idle under Windows 98.
I noticed while executing it in cygwin/tcsh, however, that
the prompt never appeared. I then simply typed in the numbers,
pretending the prompt was there, and it worked.

Oddly enough, when I tried running it the way you did, it
didn't work for me. The way I ran it was to run it, type
"import test" (the name I gave the file I pasted your
code into), and it executed perfectly.

Note that if you're running from Windows, for an "import"
command to work like that, the module will have to be in
your path or some other known location. I believe you can
temporarily adjust things by running idle from a command
line or bat file and setting PYTHONPATH to your source
directory.

Tell us if one of those tricks don't work.



C/


From alex at magenta.com  Wed Jun 28 11:14:57 2000
From: alex at magenta.com (Alex Martelli)
Date: Wed, 28 Jun 2000 17:14:57 +0200
Subject: Driving Excel from Python
References: <8j7oqn$e9q$1@supernews.com>
Message-ID: <8jd51g0j9k@news2.newsguy.com>

Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote in message
news:8j7oqn$e9q$1 at supernews.com...
> (Sorry if this appears more than once. I'm getting errors from news
server).
>
> I'm driving Excel from Python and most of it's working fine except for:
>
> >>> xlsh.UsedRange.Address(ReferenceStyle=1)
> TypeError: call of non-function (type string)
>
> xlsh is an Excel Sheet object and I'm trying to get the used range in RxCx
> format. However, when I try and specify ANY arguments to the Address
> property, I get this error message. I've tried both positional and keyword
> arguments.
>
> I have the Python Win32 book but this this subject is not quite covered in
> sufficient depth.
>
> Any help appreciated.

For reasons that escape me, .Address doesn't seem to be seen as an
indexed-property; however, a .GetAddress is also made available that
does seem to behave as an indexed-property like .Address should.  I
sort of suspect that this is to let people use .Address, without any
parenthesis, in the common case where they want to let all arguments
default.  However, I can't find any docs about this guess of mine.

Anyway, you CAN now call

xlsh.UsedRange.GetAddress(ReferenceStyle=1)

and get the same '$A$1' kind of string that you'd get with

xlsh.UsedRange.Address

since, after all, xlA1, aka 1 (win32com.client.constants.xlA1, too),
IS the default value for ReferenceStyle.

A further puzzlement, however, is that

xlsh.UsedRange.GetAddress(ReferenceStyle=-4150)

ALSO appears to return the same '$A$1' string too!  BTW, that funny
-4150 is xlR1C1, aka win32com.client.constants.xlR1C1, who knows
why.  But anyway, while this does allow accessing the Address indexed
properties with 'indices' (arguments), this doesn't appear to be
doing much good at all.  Maybe somebody else understands what's
going on...?

It does work fine from the Immediate windows in VBA inside Excel:
    print Sheet1.UsedRange.Address
$A$1
    print Sheet1.UsedRange.Address(ReferenceStyle=xlR1C1)
$A1

but from Python I always get '$A$1'.  Weird...


Alex






From maxm at normik.dk  Tue Jun 13 06:59:09 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Tue, 13 Jun 2000 12:59:09 +0200
Subject: SV: Python Productivity over C++
Message-ID: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE>

Fra: Ken Seehof [mailto:kens at sightreader.com]

>When I code in C++ or python, I spend most of my time testing and
debugging,
>so the question is: How does python save time on debugging?

I believe that many factors is in play when considering how long time it
takes coding in C/C++ vs. Python.

C/C++ is probably more error prone due to memory managment and pointers this
should make Python faster in itself.

But consider what happens if you find a design flaw in your code and you
have to refactor  all of it. Shorter code will be faster to refactor.
Shorter code will probably also have fewer design flaws.

This leads to a quadratic growth in refactoring time for longer code.
Roughly:

Twice as long code -> Twice as many design faults to refactor (Perhaps even
more).
Twice as long code -> Twice as long refactoring time.

I know that this is oversimplification, and that object oriented programming
should make it unnessecary to refactor all the code, but I do believe that
it is very close to the truth. The difference in productivity should be
problem^2 for C/C++ as compared to Python.

The theory can be tested by writing really long programs in both Python and
C/C++.

:-)

There should be a really HUGE difference in productivity for the Python
programmer as Python is a much easier language to design in.

Max M



From phd at phd.russ.ru  Mon Jun 19 11:23:04 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 19 Jun 2000 15:23:04 +0000 (GMT)
Subject: split() with a separator set
In-Reply-To: <xjvr99tyac7.fsf@gjallarhorn.ifi.uio.no>
Message-ID: <Pine.LNX.4.21.0006191519420.21957-100000@fep132.fep.ru>

On 19 Jun 2000, Igor V. Rafienko wrote:
> I've just started learning Python, so, please bear with me. 

   Welcome!

> But what if I have several one-character separators? Obviously:
> 
> # input looks like this: 'foo:bar|baz/zot'
> # I want [ 'foo', 'bar', 'baz', 'zot' ]
> string.split( src, ":|/" )


   Regular expressions are your friends!

import re
re.split("[:|/]", src )


   (Russian? Can read/write cyrillic?)

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From thomas at cintra.no  Thu Jun 15 02:49:02 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 15 Jun 2000 06:49:02 GMT
Subject: XML-code examples
Message-ID: <394f7bc2.365514882@news.online.no>

Hi,

Is there more simple XML code out there? I mean the HowTos took about
ten minutes, and it didn`t cover much. 

If somebody has some tips or code on how to traverse a DOM-tree,
processing nodes based on tagname, and especially, based on relations
with other nodes ( i.e. who`s your daddy, here`s your sister etc ),
that would be great.

Thanks.

Thomas


From tjs at cs.monash.edu.au  Wed Jun 21 01:53:38 2000
From: tjs at cs.monash.edu.au (Toby J Sargeant)
Date: Wed, 21 Jun 2000 15:53:38 +1000
Subject: Why whitespace denotation of blocks doesn't work.
In-Reply-To: <39505457.5223ADC5@san.rr.com>; from jkraska1@san.rr.com on Wed, Jun 21, 2000 at 05:34:09AM +0000
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au> <39505457.5223ADC5@san.rr.com>
Message-ID: <20000621155338.C11308@fangorn.csse.monash.edu.au>

On Wed, Jun 21, 2000 at 05:34:09AM +0000, Courageous wrote:
> I just spent the last 45 minutes vainly searching the web,
> usenet, comp.editors, and the vim documentation for the
> command to do this. I must not know the magic word. Oh
> please, great Oz, enlighten me.

:set lcs=tab:?\ 
:set list

(note that there's a space on the end of the first line)

Unfortunately, I can't find a way to do the same thing in
emacs, display-table notwithstanding. There doesn't seem
to be a way to keep the variable size nature of a tab,
and still control the way in which it is displayed.

Toby.



From rogerha at ifi.uio.no  Sat Jun 24 05:58:25 2000
From: rogerha at ifi.uio.no (Roger Hansen)
Date: 24 Jun 2000 11:58:25 +0200
Subject: extending python problem
References: <8iu37c$jsa$1@nnrp2.deja.com>
Message-ID: <ya0zoob8kou.fsf@trymheim.ifi.uio.no>

* rdejuana at my-deja.com
> 
> I'm trying to extend python with c.  I trying to
> get the progam at the tutorial to work. but every
> time i try to import it it tells me it cant find
> the init function.  This is the error:
> 
> >>>import libPYSTAF
> Traceback (innermost last):
>   File "./libPYSTAF.py", line 3, in ?
>     import libPYSTAF.so
> ImportError: dynamic module does not define init
> function (initlibPYSTAF)
> 
> This is my code:

[snip]

> I'm new to coding, so i probably forgot something
> stupid.  If you stop it please tell what it is.

Your code worked fine on my computers (Solaris and Linux). Are you
sure you compiled correctly? I did

:-) gcc -fpic -Wall -ansi -I/local/include/python1.5 -c pystaf.c 
:-) gcc -shared pystaf.o -o libPYSTAFmodule.so
:-) python -c 'import libPYSTAF'

The -fpic must be there, or else the code will be relocated when its
loaded. 

HTH,

R


From jamarijr at my-deja.com  Fri Jun  2 11:59:37 2000
From: jamarijr at my-deja.com (Arinté)
Date: Fri, 02 Jun 2000 15:59:37 GMT
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com> <8h8jg3$d66$1@slb6.atl.mindspring.net>
Message-ID: <8h8lkt$rb7$1@nnrp1.deja.com>

You must be using a different version or something.  I use the python
gui thing and did this
print "\x1ba"
got a funny circle
print "\x1b a"
[] a   < not an exact represetation.
But, it still points out the fact that python is picking up the 1 b and
a.


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


From phd at phd.russ.ru  Thu Jun 29 09:26:00 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 29 Jun 2000 13:26:00 +0000 (GMT)
Subject: newbie questions about imports, threads, and optimizations
In-Reply-To: <8jd7f1$kic$1@murrow.sp.trw.com>
Message-ID: <Pine.LNX.4.21.0006291323230.25142-100000@fep132.fep.ru>

On Wed, 28 Jun 2000, Bryce Sellers wrote:
> First, is there a way to do something like this:
> 
> import string, mystring
> m = 'string'
> `m`.captalize('hello')
> m='mystring'
> `m`.capitalize('hello')

   You are very close :)

m = string
m.captalize('hello')

   In python, every name (EVERY NAME!) is just a name of variable
containing a pointer to an object. When you import a module, there is
nothing magical in its name - it is just a variable containing a pointer to
the module. Copy the pointer (m = string) and use the new variable that now
holds the same pointer.

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From darrell at dorb.com  Sat Jun  3 20:47:45 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 3 Jun 2000 17:47:45 -0700
Subject: Automatic reload()?
References: <8hbnt1$62i$1@news2.isis.unc.edu> <m3r9aexydb.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <04e301bfcdbe$81ff0f40$6401a8c0@home>

Michael Hudson wrote:

> def rr(keepem=sys.modules.keys()):
>     for i in sys.modules.keys():
>         if type(sys.modules[i]) is types.ModuleType and i not in keepem:
>             print "reloading",i
>             reload(sys.modules[i])
>
> I can't remember quite what the "keepem" bit's for, but I presume
> there was some reason I put it in...
>
Just guessing..
keepem will have the current list of modules, when this module is run. So
new modules will be reloaded because they aren't in that list.

Cool idea.


--Darrell Gallion




From sblakey at freei.com  Mon Jun 26 20:03:30 2000
From: sblakey at freei.com (Sean Blakey)
Date: Mon, 26 Jun 2000 17:03:30 -0700
Subject: erk, Bug? ( long post )
In-Reply-To: <3957E733.4237B55B@mindspring.com>; from dbroadwell@mindspring.com on Mon, Jun 26, 2000 at 07:28:51PM -0400
References: <3957E733.4237B55B@mindspring.com>
Message-ID: <20000626170330.A18005@freei.com>

I get a different message on my FreeBSD laptop:
OSError: [Errno 2] No such file or directory.

What is happening is that you are using chdir() to move in to each directory you
check, then calling os.listdir() on the relative path from where you started the
script.

I have attatched a modified version of your script that might clarify this.  The
only modification I made was the os.getcwd() call on line 58.

I suggest seeing if you can work without the os.chdir() calls (or maybe seeing
if you can get the job done with os.path.walk()).
        -Sean

On Mon, Jun 26, 2000 at 07:28:51PM -0400, David Broadwell wrote:
> All of my text is at the bottom...
> 
    <<Source of grope.py snipped by Sean Blakey>>
> 
> ERROR:
> Good morning Dave.
> Checking ( dir ): .
> Checking ( files ): .
> ['DLLs', 'Lib', 'libs', 'include', 'Tools', 'Doc', 'source']
> Crossing the bridge of death (while loop that falls off the net).
> Depth now at: 0 (how many directories deep am i? .. fails)
> Scan path for subdir loop ['DLLs', 'Lib', 'libs', 'include', 'Tools',
> 'Doc', 'source']
> Checking ( dir ): DLLs
> Checking ( files ): DLLs
> Checking ( dir ): Lib
> Checking ( files ): Lib
> Checking ( dir ): libs
> Checking ( files ): libs
> Checking ( dir ): include
> Checking ( files ): include
> Checking ( dir ): Tools
> Checking ( files ): Tools
> Checking ( dir ): Doc
> Checking ( files ): Doc
> Checking ( dir ): source
> Checking ( files ): source
> Checking ( dir ): Lib/Plat-Win
> Checking ( files ): Lib/Plat-Win
> Traceback (innermost last):
>   File "C:\PROGRA~1\PYTHON\TOOLS\IDLE\ScriptBinding.py", line 131, in
> run_module_event
>     execfile(filename, mod.__dict__)
>   File "C:\WINDOWS\Profiles\mongoose\Desktop\grope.py", line 112, in ?
>     main()
>   File "C:\WINDOWS\Profiles\mongoose\Desktop\grope.py", line 29, in main
>     file(dir)
>   File "C:\WINDOWS\Profiles\mongoose\Desktop\grope.py", line 63, in file
>     filetemp = os.listdir(filepath)
> OSError: [Errno 3] No such process
> 
> Is there a reason that that wrror message fills me with dread? Why is
> the interpreter punishing me? What did i do wrong?
> 
> It always dies on the first use of the files that calls a directoy that
> has a subpath. I tested that ability alone in the interpreter and it
> worked perfectly well. I am on python 1.5.2 on a windows 98 box. I did
> see aome comments on DejaNews about it, and it was also involving
> os.listdir.. But there was only one instance, and i do not see precisely
> how they relate...
> 
> -- 
>         ~We're all fallen angels who've forgotten how to fly~
>         Remember how: http://www.cyber-action.com/oed/dbroadwell.html
> --
> -- 
> http://www.python.org/mailman/listinfo/python-list

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
The trouble with computers is that they do what you tell them, not what
you want.
		-- D. Cohen
-------------- next part --------------
import os

filelist = []
subdirlist = []
filetype = 'm3u'
output = 'out.txt'

def main():
    stop = 2
    depth = 0
    path = '.'
    scan(path)
    file(path)
    print subdirlist
    if (subdirlist):
       print 'Crossing the bridge of death.'
       while (depth <= stop):
           print 'Depth now at:',
           print depth
           depth = depth + 1
           temp = []
           print 'Scan path for subdir loop',
           print subdirlist
           if (subdirlist):
              temp = subdirlist
              for dir in temp: 
                  scan(dir)
                  file(dir)
              print 'recycling while loop, wish me luck.'
       else: 
#           print 'Stopping that while loop variable readout: '
#           print 'This is where merge will be called once stable'
           print 'Filelist:',
           print filelist
#           print 'Subdirlist:',
#           print subdirlist
#           print 'Last depth:',
#           print depth
           print 'Stop set at:',
           print stop

def scan(scanpath):
    print 'Checking ( dir ):',
    print scanpath
    os.chdir(scanpath)
    dirtemp = os.listdir('.')
    for item in dirtemp:
        if (os.path.isdir(item)):
           if (scanpath == '.'): 
              subdirlist.append(item)
#              print 'Added: ' +item+ ' to subdirlist'
           else: 
              subdirlist.append(scanpath + '/' + item)
#              print 'Added: ' +(scanpath + '/' + item)+ ' to subdirlist'
    if (not(scanpath == '.')): os.chdir('..')

def file(filepath):
    print 'Checking ( files ) from', os.getcwd(),':',
    print filepath
    filetemp = os.listdir(filepath)
    for item in filetemp:
        for x in range(len(item)):
            if (item[x:(x+1)] == '.'):
               if (item[(x+1):] == filetype):
                  if (filepath == '.'):
                     filelist.append(item)
                     print 'Added: ' +item+ ' to filelist'
                  else: 
                     filelist.append(filepath + '/' + item)
                     print 'Added: ' +(filepath + '/' + item)+ ' to filelist'

def merge(filelist):
    if (filelist):
       print 'Sory dave actually merging the functions is disabled'
#       outbuffer=open(output,'w')
#       for item in filelist:
#           inbuffer=open(item)
#           outbuffer.write(inbuffer.read())
#           inbuffer.close()
#       outbuffer.close()
       print 'But if you want to know, I would have:'
       print 'Merged files:',
       print filelist
       print 'With:',
       print './' + output
    else:
       print 'No ' + filetype + ' files to process, Exiting...'

print 'Good morning Dave.'
main()

From timaratz at lisco.com  Sun Jun 25 13:02:49 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Sun, 25 Jun 2000 12:02:49 -0500
Subject: smtplib
In-Reply-To: <20000623154721.B630@sindri.callisia.com>
Message-ID: <000c01bfdec7$30f84b60$bdcb45c6@timaratz>

I'm using smtplib with a CGI script. I'm puzzled as to why this module
doesn't support sending a subject for the e-mail.

And the e-mail I receive doesn't show the to address in my e-mail client.




From pelonmail at yahoo.com  Sat Jun 10 16:54:23 2000
From: pelonmail at yahoo.com (Phylo Kramer)
Date: Sat, 10 Jun 2000 13:54:23 -0700
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net>
Message-ID: <960673850.532727042@news.pacbell.net>

On Fri, 09 Jun 2000, Phil Fraering wrote:
>Guido van Rossum <guido at python.org> writes:
>
>> Yet, here are some of the reasons why I am considering making Python
>> case-insensitive:
>> 
>> (1) Randy Pausch, a professor at CMU, found, when teaching Python to
>> non-CS students in the context of Alice (www.alice.org), that the
>> number one problem his students were having was to remember that case
>> matters in Python.  (The number two problem was 1/2 == 0; there was no
>> significalt number three problem.)
>
>> (2) I've met many people who are experienced and accomplished Python
>> programmers but who still, when referring to Python modules or
>> functions in an informal context (e.g. email), make up their own case
>> conventions.  E.g. people will write me about the String module or
>> about FTPLib.
>> 

Forgive a newbie for jumping in here,
After reading a couple of books and some of the online material I sat down and
rewrote many of my old Perl scripts. I had to refer to the docs on only one
occasion. I was able to write nearly perfect code on the first try. Amazing!
But when I reached into the library for certain modules I had to constantly
refer to the docs because I could not remember if it was "selectColor" or
"SelectColor","BaseHTTPServer" or baseHTTPserver".

Because Python has such a low learning curve, it seems that many developers
see it as a macro language for their particular favorite. Java programmers use
Java conventions, Perl programmers use Perl, TCL for TCL, C++ for C++, etc. In
the long arduous path to learning any of these languages, a new programmer
inherits the traditions of the teacher. But with a language that can be learned
in a day, there is no time to study convention.

Consistency is all that I ask for. 

_____
Pelon
pelonpelon at geocities.com




From stevena at permanent.cc  Sun Jun  4 00:22:54 2000
From: stevena at permanent.cc (Steven D. Arnold)
Date: Sun, 04 Jun 2000 00:22:54 -0400
Subject: [Python-Dev] What's that sound?
In-Reply-To: <3302A334E7FAFD16.1D3E89521AAD6B9B.465172EF8E4C146A@lp.airn
 ews.net>
References: <Pine.LNX.4.10.10005310928270.30220-100000@nebula.lyra.org>
Message-ID: <4.3.1.2.20000604002128.0165a620@phear.dementian.com>

At 09:14 PM 5/31/2000 -0500, Cameron Laird wrote:
>5.  Bruce is starting to release
>     chapters of *Thinking in Python*.

Wow, where are these?  I've got Bruce's `Thinking in C++' and love it, even 
though I don't particularly like C++. ;-)  I'm sure the combination of 
Bruce's writing and a much more pleasant language will be fantastic!


steve



--
Steven D. Arnold            Que quiero sera         stevena at permanent.cc
"Fear: The best way out is through."                     -- Helen Keller




From thor at localhost.localdomain  Wed Jun  7 12:01:40 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 7 Jun 2000 16:01:40 GMT
Subject: BibTeX entries for python Doc's
Message-ID: <slrn8jt3n7.64h.thor@localhost.localdomain>

BibTeX entries : TkInter Life preserver
Currently I'm doing my last year work at university. And 
it's required to make a bibliography.
While including TkInter Life preserver, I was unable to put
a "stable" URL for that doc. 

Anyway, if python is intended to be used at education, some
BibTeX entries in a corner of PythonLabs with all books, electronic
books and the rest of on-line doc would just be fine.


-- 
MGA


From mcalla at home.com  Mon Jun 26 18:53:09 2000
From: mcalla at home.com (Mike Callahan)
Date: Mon, 26 Jun 2000 22:53:09 GMT
Subject: Questions about Installer/Builder/Standalone
Message-ID: <p9R55.17984$Yr4.303276@news1.rdc1.tn.home.com>

I am playing with Gordon McMillan's Installer 0.3f for Windows and have
gotten it to work using freeze, simple, and standalone with non-Tkinter
apps. Good job Gordon! I can make a Tkinter app using simple but I cannot
get it to work using standalone. I know that simple does some magic to get
the app to use the local Tk it installs and Gordon talks about it in the
docs. But I cannot figure out what the magic is. I want to put every thing
into a single directory and then use Install Shield to make a nice install
program.

I did discover the stand-alone Python method for Windows that Neil
Schemenauer invented, (Thanks, Neil) but I like the fact that Installer puts
all the app code into a single file instead of copying the entire python
library to the install directory. I could live with it, however...

So in a nutshell, has anyone figured out a way to get Installer 0.3f to put
all the files into a single install directory including the Tk stuff that
Install Shield could use? Thanks.

Mike Callahan





From wjdandreta at worldnet.att.net  Tue Jun 20 20:40:37 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 00:40:37 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com>
Message-ID: <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net>

Hi Anders,

I am running uner DOS. Any version will be fine, even older ones as long as
I can index it.

Bill
Anders M Eriksson wrote in message ...

>You don't say which version of dBASE and not on which platform you
>are, so let's assume that you are using Windows 9x/NT/2000.
>
>The only real alternative today is to use ODBC!
>
>The win32-all comes with an ODBC modules that works OK, if you want to
>spice it up a bit get mxODBC (seach on Vaults of Parnassus)
>
>// Anders
>
>




From jparkey at nospamrapidbs.com  Thu Jun 22 07:29:57 2000
From: jparkey at nospamrapidbs.com (John Parkey)
Date: Thu, 22 Jun 2000 12:29:57 +0100
Subject: Manipulating specific bits
References: <Brl45.80022$WS3.716680@typhoon.we.rr.com> <m3snu63t2b.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8isthg$qt7$1@phys-ma.sol.co.uk>

Stephen

You can use the bit-wise manipulation, using masks to set and clear bits
(Mike's example dynamically sets up a mask).
To set the fifth and second bits, you need to 'OR' your byte with a mask
which has those bits set.

The mask would be:  00010010 in binary, = 12Hex (0x12) = 22Octal (022), so
you would set the two bits by saying:

byte = byte | 0x12.

To check whether a bit is set, you need to 'AND' it with a mask representing
the bit.  So, to test if bit 3 is set, the mask needed is:  00000100 in
binary, = 0x4(Hex) = 04(Octal), and you can check whether your byte has it
set by saying:

result = byte & 0x4.

To clear a bit, you need a bit mask which is a complement of the bit(s) you
are testing for, which you AND with your test byte.

For example, to clear bit 3, you need to AND with: 11111011 in binary, =
0xFB(hex) =  0373(Octal), so you would say:

result = byte & 0xFB.

To save you having to work out the complement, you can use the '~' symbol,
for bitwise inversion.  So, to clear bit 2, you just need to say:

byte = byte & ~0x04

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m3snu63t2b.fsf at atrus.jesus.cam.ac.uk...
> "Stephen Hansen" <stephen at cerebralmaelstrom.com> writes:
>
> >     Okay, this is a Very Newbie-ish question.
> >
> >     What I need to do is manipulate specific bits in a byte, for
> > transmission. I have absolutely *no* idea how to do this. :)
> >     For example, say I have my 1 byte: 0 0 0 0 0 0 0 0. Say I want to
set
> > the 2nd and 5th bit to 1.. how the heck do I do that? :)
>
> byte = 0
> byte = byte | (1<<4) | (1<<1)
>





From info.jyv at videotron.ca  Sat Jun 10 10:57:56 2000
From: info.jyv at videotron.ca (Jean-Yves Vinet)
Date: Sat, 10 Jun 2000 10:57:56 -0400
Subject: windll
References: <3941ACC5.305B8830@videotron.ca> <3941B6B4.264FA7A1@san.rr.com> <l8j05.833$cQ.10319@news-server.bigpond.net.au> <3941D9BD.46BBEF0C@san.rr.com>
Message-ID: <39425774.9A7400F0@videotron.ca>

I was taking about using dynamic lybrairies from python not
from C or C++. On Unix/Linux theses lybrairies are often
name like "*.so". "windll" is a python module to help
programmer to use these kinds of libraries without
programming in C or C++.

On windows, "windll" helps me a lot. I wrapped
a financial system and the Fame database. I want
to do the same for Sun and Linux.

Is someone know how to do it on Linux/Unix ?

JYV


Courageous wrote:

> > > > Is there someting like to "windll" to use
> > > > dynamic librairy under Linux/Unix plateform.
> > >
> > > What does windll do? "Using" dynamic libraries
> > > under Unix is as easy as linking with them and
> > > then making certain that they are in your
> > > LD_LIBRARY_PATH.
> >
> > Using them from Python is that easy?  Really?
> >
> > windll tries to make using arbitrary DLLs, *from Python* that simple!
>
> Yeah, I had a discussion off line with someone else.
> For some reason I didn't consider that the person meant
> a python module. Anyway, what you're looking for on unix
> is "dl", as in:
>
> import dl
>
> C/



From db3l at fitlinxx.com  Fri Jun 30 22:10:48 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 30 Jun 2000 22:10:48 -0400
Subject: Communicating between two computers
References: <395d48b3$0$21842@wodc7nh0.news.uu.net>
Message-ID: <uem5e38if.fsf@ctwd0143.fitlinxx.com>

"donotspam-jen at personic.com" <dnotspam-jen at personic.com> writes:

> I've read a little about SOAP, but I think that my needs are probably a lot
> simpler than that.  The easiest way to accomplish this is to setup a Web
> server on each of the build machines, too, I guess ... but I was hoping that
> I could just have the scripts that run on the build servers run as NT
> services.

Well, there's probably no end to the number of protocols and methods
by which you could get the machines to talk, but given that it's
Python invoking Python, you might take a peek at Fredrik Lundh's
XML-RPC library.  XML-RPC was sort of the precursor to SOAP.  Similar
in concept, but it's dirt simple to get two Python processes
communicating (calling arbitrary methods with any Python objects and
returning any Python objects).

See http://www.pythonware.com/products/xmlrpc/index.htm.

I imagine you could run a server utilizing this within a service using
the PythonService module included in the Pythonwin extensions without
too many problems.

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


From aa8vb at yahoo.com  Tue Jun 20 10:18:13 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 20 Jun 2000 10:18:13 -0400
Subject: mpeg encoding
In-Reply-To: <8iie3s$epp$1@newsg1.svr.pol.co.uk>; from
 dave@rotwang.freeserve.co.uk on Sun, Jun 18, 2000 at 01:03:10PM +0100
References: <8iie3s$epp$1@newsg1.svr.pol.co.uk>
Message-ID: <20000620101813.B1582140@vislab.epa.gov>

Dave Berkeley:
 |I have written a python extension which allows me to control a Kodak DC215
 |digital camera for use in time lapse photography using the Kodak SDK. I can
 |now grab jpg images every 20 or so seconds, and leave it running all day.
 |
 |Does anyone know how to convert a set of jpg images into mpeg using PIL?

With PIL?  No. 

mpeg_encode is an option:

   ftp://mm-ftp.CS.Berkeley.EDU/pub/multimedia/mpeg/encode/

IIRC it only takes frame speeds of 24, 25, and 30 fps.  You can specify
images multiple times in its param file to fudge the frame rate you want
though.

-- 
Randall Hopper
aa8vb at yahoo.com



From dfan at harmonixmusic.com  Wed Jun 21 09:28:19 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 21 Jun 2000 09:28:19 -0400
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <wk66r3cgek.fsf@turangalila.harmonixmusic.com>

"William Dandreta" <wjdandreta at worldnet.att.net> writes:

| What I am doing is reading a comma delimited text file like this
| 
| line = AB12345,20000621,3.56,...
| 
| fields = splitfields(line,',')
| 
| recognizing that
| 
| fields[Date][:5]  is the year
| 
| is more obvious and less cryptic than
| 
| fields[1][:5]
| 
| Trying to put this into a dictionary seems a contrivance to me that
| doesn't make things clearer and only adds overhead.

I think it's exactly the opposite.  If you want to look things up by a
name rather than a number, look them up by a name, not by a fake name
that actually maps to a number.  Trying to shoehorn your 'Date, Foo,
Bar' fields into an array is the contrivance.

Perhaps you come from a background in languages like C?  Often it
takes a while to get used to writing code the way you think, without
worrying too much about overhead and exactly how the computer is
laying out your data in memory.

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


From shapr at uab.edu  Sat Jun 17 20:47:23 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sun, 18 Jun 2000 00:47:23 GMT
Subject: performance benefits from byte code files?
References: <8idr3p$33h$1@nnrp1.deja.com> <394B70C7.39FB2DB6@uab.edu>
Message-ID: <394C1CA1.A89367D6@uab.edu>

Shae Erisson wrote:

> Check out this url:
> http://www.lyra.org/greg/python/#dev
> Greg Stein gives a method of executing *.pyc files directly for CGI.
> 
> Personally, I'd like to know if *.pyo files also will work the same way.

I emailed greg stein, .pyo files do also work with this trick.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From shapr at uab.edu  Sat Jun 17 08:34:18 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sat, 17 Jun 2000 12:34:18 GMT
Subject: performance benefits from byte code files?
References: <8idr3p$33h$1@nnrp1.deja.com>
Message-ID: <394B70C7.39FB2DB6@uab.edu>

Dan Gindikin wrote:
> 
> I was wondering what the performance benefit is from having a .pyc file
> around.
> 
> Here an example that motivates this quiestion. Lets say I have a
> relatively large script.py CGI script. Since it is always executed
> directly, a script.pyc file is never generated. Is it then beneficial
> to write a tiny wrapper that does nothing but
>    import script
>    script.run()
> because then script.pyc does get generated?

Check out this url:
http://www.lyra.org/greg/python/#dev
Greg Stein gives a method of executing *.pyc files directly for CGI.

Personally, I'd like to know if *.pyo files also will work the same way.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From davec999 at my-deja.com  Fri Jun 30 04:33:07 2000
From: davec999 at my-deja.com (David Currie)
Date: Fri, 30 Jun 2000 08:33:07 GMT
Subject: command-line history in win
References: <395c304e.1224185@news.student.lu.se>
Message-ID: <8jhm03$qnm$1@nnrp1.deja.com>

In article <395c304e.1224185 at news.student.lu.se>,
  invalid at dev.null (Alexander K) wrote:
> aloha!
>
> i did a backsearch for this on deja. found a question but no answers.
> and the question is: isnt there a historyfunction in the python
> command-line interpreter for win?.
>
> you know, the bash-kinda thing.
>
>   thanks in adv. / alex k
>

It works for me (Python 1.5.2)...

Dave


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


From rsteelandt at bucodi.com  Tue Jun 13 11:09:52 2000
From: rsteelandt at bucodi.com (BuCoDI-Rony Steelandt)
Date: Tue, 13 Jun 2000 17:09:52 +0200
Subject: Something for speedfreaks and pythongoeroes
Message-ID: <39464EC0.E4ECEEC2@bucodi.com>

Hello,
 I was wondering if this code could be speed up by using some tricks
# sqlquery.py
# The query object for sqlia
import dbi,odbc
from string_man import *

connectstring = 'Urk/sa/'

class connection:
 def __init__(self):
  self.dbc = odbc.odbc(connectstring)
  self.crsr = self.dbc.cursor()
 def close(self):
  self.dbc.close()
class querry(connection):
 def cmd(self,command):
  self.t=[] ## self.t is the resultset
  numbcol = 0
  self.listcol = [] ## contains the length of the colums
  self.crsr.execute(command)
  ## -- Get columndescription
  for col in self.crsr.description:
   numbcol = numbcol +1
   self.listcol.append(col[3])
  self.result = self.crsr.fetchall()
  # Add the formatted strings to the result list
  for tmp in self.result[0:]:
   self.row=''
   for y in range(0,numbcol):
    self.row=self.row + " " +
formatstr(str(tmp[y]),int(self.listcol[y]))
   self.t.append(self.row)
  self.close()
  return self.t

if __name__ == "__main__":
 from time import *
 numlines = 0
        print "Application started"
        start=time()
        result = querry().cmd("SELECT * FROM TBL_BUYERS")
        for cnt in result:
                #print cnt
                numlines = numlines + 1
 print 'Execution time=',time()-start
 print 'Number of lines=',numlines
 print 'Application ended'


--
======================================================
Bureau de Conseil et de D?veloppement Informatique
2, Rue Henri Avril
22480 Saint - Nicolas du Pelem

T?l : +33.2.96.29.79.08
Fax : +33.2.96.29.79.09

e-mail : bucodi at worldnet.fr
======================================================




From kazan at kazan.  Mon Jun  5 16:01:21 2000
From: kazan at kazan. (kazan at kazan.)
Date: Mon, 05 Jun 2000 20:01:21 GMT
Subject: operator overloading
Message-ID: <f3trf8.08.ln@kazan>

Hello

How can I overload the less than operator?
learning python did mention something about __cmp__ but did
not provide any examples. 


Thanks in advance

Erling


From root at rainerdeyke.com  Mon Jun 12 18:07:26 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 12 Jun 2000 22:07:26 GMT
Subject: tkinter question
References: <39454A0E.1A630514@uniserve.com>
Message-ID: <yad15.25$x_6.8854@news1.rdc2.tx.home.com>

Bob van der Poel <bvdpoel at uniserve.com> wrote in message
news:39454A0E.1A630514 at uniserve.com...

> Third, how can I define a widget command with an arg. For example:
>
> q = Button( text="Quit", command = quitit("sure?") )

One word: lambda.


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




From bvdpoel at uniserve.com  Sat Jun 24 14:42:09 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 24 Jun 2000 11:42:09 -0700
Subject: tkinter, grid and resize
Message-ID: <39550101.AD8C7956@uniserve.com>

I'm having a problem getting widgets managed by grid to resize
themselves when the screen size changes. This is easy to do using the
pack manager and 'expand', however 'expand' doesn't exist in grid and
the docs seem to indicate that the 'news' sticky option should do the
same. The following snippet creates a text widget with scrollbars. If
you resize the main window, the widgets do not resize....

from Tkinter import *

root=Tk()

ys = Scrollbar(root)
xs = Scrollbar(root)

dsp = Text(root, yscrollcommand=ys.set, xscrollcommand=xs.set)
xs.config(orient='hor', command=dsp.xview)
ys.config(orient='vert', command=dsp.yview)
xs.grid(row=1, column=1, sticky=E+W)
ys.grid(row=0, column=0, sticky=N+S)
dsp.grid(row=0, column=1, sticky=N+S+E+W)

root.mainloop()

What am I don't wrong? I've tried to insert columnspan stuff and played
with the weight options--but I don't seem to be getting it.

Thanks.

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



From herzog at online.de  Tue Jun 13 15:53:25 2000
From: herzog at online.de (Bernhard Herzog)
Date: 13 Jun 2000 21:53:25 +0200
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu>
Message-ID: <m3g0qh5pe2.fsf@greebo.nodomain.de>

Alex <cut_me_out at hotmail.com> writes:

> > Do keep digging. Ref-counting is important.
> 
> Ref-counting errors can be such a pain in the neck to track down that I
> have taken to placing assertions at the end of routines to make sure
> the ref-counts are what they ought to be.

That's usually very hard to get right. It's usually very hard to guess
what the reference count of a given object should be.

> This is the sort of thing I've been doing:
> 
> int dict_add_pattern (PyObject *dict, char *string, int length, long pos[DEPTH]) {
>   PyObject *pattern, *coords, *value, *tuple;
>   int return_code, pattern_existed;
> 
>   assert (length <= DEPTH);
> 
>   coords = dict_set_coords (length - 1, pos);
>   if (!coords) {goto fail;}
>   
>   pattern = Py_BuildValue ("s#N", string, length, coords);
>   if (!pattern) {Py_DECREF (coords); goto fail;}
 
>   value = PyDict_GetItem(dict, pattern);
>   if (!PyDict_GetItem(dict, pattern)) {
>     pattern_existed = 0;
>     value = PyInt_FromLong( 1 );
>     return_code = PyDict_SetItem(dict, pattern, value);

You have to DECREF value here.

>   } else {
>     pattern_existed = 1;
>     assert (PyInt_Check( value ));
>     value = PyNumber_Add( value, PyInt_FromLong( 1 ));

Here's a memory leak for the old value and the 'new' intobject returned
by PyInt_FromLong. 

>     if (value == NULL) {goto fail;}

If value is indeed NULL, you leak the reference to pattern

>     return_code = PyDict_SetItem(dict, pattern, value);

You have to DECREF value here.

>   }
>   /* If a tuple equal to pattern was already in dict, then pattern will not be
>      referred to by dict, and should have a refcnt of 0.  If there was no such
>      tuple, then PyDict_SetItem increased the refcnt of dict by one.  In
>      either case, the refcnt has to be decremented, now.
>      */
>   Py_DECREF (pattern);

The DECREF of pattern may delete pattern so pattern should not be
dereferenced after this point. Since pattern has the only reference to
coords, coords may also have been deleted.

>       
>   if (return_code == -1) {
>     goto fail;
>   } else {
>     if (DEBUG_REFCOUNTS) {
>       printf ("coords: %i\npattern: %i\nvalue: %i\n",
> 	      coords->ob_refcnt, pattern->ob_refcnt, value->ob_refcnt);
>     }
>     if (DEBUG_LIST_PATTERNS) {
>       PyObject_Print (dict, stdout, 0); printf ("\n");
>     }
>     goto pass;
>   }
> 
> fail:
>   return 0;
> pass:
>   assert (pattern->ob_refcnt == (!pattern_existed));
>   assert ((coords->ob_refcnt == (!pattern_existed)) | (length == 1));

You dereference potentially free()'ed pointers here and in the debug
messages above. To make things worse, if DEBUG_LIST_PATTERNS is true,
PyObject_Print might execute pretty arbitrary Python code which might
result in the memory being reused again.

>   return 1;
> }

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


From ge at nowhere.none  Tue Jun 27 11:53:09 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 27 Jun 2000 15:53:09 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <Gq365.6003$Sn2.40878@news-server.bigpond.net.au>
Message-ID: <F5465.1869$iN5.400512@ptah.visi.com>

In article <Gq365.6003$Sn2.40878 at news-server.bigpond.net.au>, Neil Hodgson wrote:

>> In case you haven't heard Microsoft is coming out with a massive new Web
>> stratagy that, in part, uses a new language called C# (That sharp, like in
>> music).  ...
>
>   I read the intro and skimmed the language reference today. It doesn't
>look like a bad language but it doesn't have anything that jumps out as a
>good reason for moving to it. I'd call it a safer superset of C++. 

A superset of a set that already has an order of magnitude more
members than it should.  There's progress...   ;)

Your car is too heavy, slow, and can't corner?  MS solution:
add more chrome, lights, and two more colors of paint!

-- 
Grant Edwards                   grante             Yow!  It's the RINSE
                                  at               CYCLE!! They've ALL IGNORED
                               visi.com            the RINSE CYCLE!!


From hat at se-46.wpa.wtb.tue.nl  Wed Jun 21 10:37:26 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 21 Jun 2000 14:37:26 GMT
Subject: boolean true and false values.
References: <3947f922.425016156@nntp.interaccess.com>
Message-ID: <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl>

On Wed, 14 Jun 2000 21:32:34 GMT, Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
>Does python have true and false values.
>I would like to do something like
>
>class A:
>     def _init_(self):
>           self.processed=false
>     def process(self):
>	self.processed=true
>
>at the moment I am using
>false=(1==0)
>true=(1==1)
>but would like something more elegant, or at least less sloppy.

I read a number of replies, but didn't read an answer that seems satisfactory.
I ran in the same problem, and was also not able to find a elegant solution.
Using 0 and 1 or `true` and None is not correct imho, there should be a native
way to use true and false for indicating boolean values.

Note that this is an advanced topic in the sense that beginning programmers
don't even realize that (1==2) is an expression with a value. On the other
hand, I use bool and its true and false values quite often in C++ rather than 0
and 1 which leaves way to much room for mistakes like assigning 2 to it.

Imho Python should have the data type boolean, with the values true and false.
Obviously, it can be build (or maybe already exists), but it would be real nice
if it would be a native data type.

Also, I am missing the data type 'set'. It can be faked with a dictionary, but it
is just not the same. Any one else got this problem ?

Albert
---
Look ma, windows without Windows !!


From etoffi at softhome.net  Wed Jun 14 02:25:34 2000
From: etoffi at softhome.net (etoffi)
Date: Wed, 14 Jun 2000 02:25:34 -0400
Subject: tuple question
Message-ID: <642A954DD517D411B20C00508BCF23B001296A2E@mail.sauder.com>

*** post for free via your newsreader at post.newsfeeds.com ***

admittedly this is a very [insert descriptive word here] question to ask, but
what is a tuple and why is it used (as opposed to an array, etc).  i suppose the
answer could be found in some comp sci book somewhere, but i dont have access to
those books.

ps: i know:

	(x, y) = point.get()

i just want to know what was the reason for creating/using tuples, more from a
language implementor standpoint than a witing-a-program-in-Python standpoint...

--
Hopefully somebody will understand what I'm trying to say

  **** Post for FREE via your newsreader at post.newsfeeds.com ****

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
*** Newsfeeds.com - The #1 Usenet Newsgroup Service on The Planet! ***
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
  http://www.newsfeeds.com       |         http://www.newsfeeds.com
                                 |
* Anonymous posting server!      | * Totally Uncensored!
* SUPER Servers!                 | * Over 80,000 Newsgroups!
* BINARIES ONLY Servers!         | * 16 seperate Newsgroup Servers!
* SPAM FILTERED Server!          | * Instant access!
* ADULT ONLY Server!             | * Multiple OC 3's and OC 12's!
* MP3 ONLY Server!               | * 99% Article Completion!
* MULTIMEDIA ONLY Server!        | * Months of Retention!
* 7 UNCENSORED Newsgroup Servers | * Lightning FAST downloads!
                                 |
  http://www.newsfeeds.com       |         http://www.newsfeeds.com
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
        **** Point your newsreader to post.newsfeeds.com ****
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 



From a at a.a.a  Tue Jun 27 20:04:29 2000
From: a at a.a.a (Douglas Taylor)
Date: Wed, 28 Jun 2000 00:04:29 GMT
Subject: Help with launching acrobat
Message-ID: <hib65.8$dm4.455036@news0.optus.net.au>

Hi!

I am currently faced with a problem -

My company has a windows product which needs to launch a PDF.

Previously, we were putting the path to the acrobat executable into the
autoexec.bat.

This was fine, until some of our customers decided to run it in NT.

My question is :-

How can I query the windows registry for the Acrobat path?

**OR**

How can I use com to launch Acrobat reader?




From db3l at fitlinxx.com  Wed Jun 14 16:04:22 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 16:04:22 -0400
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de>
Message-ID: <uya48t4ft.fsf@ctwd0143.fitlinxx.com>

Thomas Thiele <thiele at muc.das-werk.de> writes:

> But there is no mechanism to ensure that the behaviour of your
> memfileobjectsis like a fileobject. For instance if you don't note
> that you forgot to write the write()-function.  In a worst case
> scenario it works for many months until a user calls a function in
> which the write() -function is needed.

I think there are probably both good and bad points to this very fact.

On the bad side I agree that there is no explicit mechanism to support
the verification of interfaces (e.g., it's not like inheriting from an
abstract base class in C++ where you'll get a compiler error if you
don't provide an implementation for each method).  So if you want that
validation, there's no built-in method to ensure it.  You could roll
your own pretty easily with introspection and some definition of what
you needed to validate as supported by the class, but that's extra work.

On the good side, it means you aren't absolutely wedded to the
interface in cases where you know you won't need it, and can produce a
useful partial implementation to get the ball rolling.  For example, I
don't have to implement every file object method to create or use an
object to replace sys.stdout - just having a "write" is enough.  This
contributes to more efficiency during prototyping and interative
development.  And as you continue to build out, it isn't really that
much effort to ensure that you conform to the interface.

But I do admit that some additional built-in method of indicating
interfaces supported by a class might be nice.  Of course you can
probe for whatever set of functions you may need if you are the user
of the class, but it would be nice to encapsulate a set of probes
based on an interface definition - or something along those lines.
I'm not that familiar with it, but Ruby's approach of including
modules into classes (where the module is basically the interface
definition) seems interesting.

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


From invalid.address at 127.0.0.1  Fri Jun 23 21:08:16 2000
From: invalid.address at 127.0.0.1 (David)
Date: Sat, 24 Jun 2000 01:08:16 GMT
Subject: Gadfly Question: TypeError on __Init__
Message-ID: <395408ae.75890752@news.telus.net>

What on earth is going wrong with this?  I *HAD* it working, and then I
tried to make it a singleton class, and now I can't even get it back to
original form.   ARRRRGH, how frustrating!
 
Please tell why I keep getting a TypeError: __Init__ should return None!
 
And if you can do this as a singleton, or make it otherwise better (ie. I
don't think I should have to create/delete a table just to get the .GFD
file to appear -- the docs indicate that as soon as the .startup method is
called, files will be created ... but not that I noticed in practice.)

====
import gadfly, os, os.path


class DBase:
    def __init__(self, path, dbase):
        try:
            if not os.path.isdir(path):
                print "creating database directory: "+path
                os.mkdir(path)

            if not os.path.isfile(path+"/"+dbase+".gfd"):
                print "creating database file: "+dbase
                connection = gadfly.gadfly()
                connection.startup(dbase, path)

                cursor = connection.cursor()
                cursor.execute("CREATE TABLE deletemenow (i integer)")
                cursor.execute("DROP TABLE deletemenow")
				
                connection.commit()

            connection = gadfly.gadfly(dbase, path)

        except:
            return 0

        else:
            return connection


# stand-alone testing
if __name__ == "__main__":
    connection = DBase("test", "testdb")
    cursor = connection.cursor()
    cursor.execute("create table ph (nm varchar, ph varchar)")
    cursor.execute("insert into ph(nm, ph) values ('arw', '3367')")
    cursor.execute("select * from ph")
    for x in cursor.fetchall():
        print x
        print ("arw", "3367")
    connection.commit()




From thomas at xs4all.net  Mon Jun  5 19:21:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 6 Jun 2000 01:21:19 +0200
Subject: indirect references
In-Reply-To: <XFMail.20000605161330.shaleh@valinux.com>; from shaleh@valinux.com on Mon, Jun 05, 2000 at 04:13:30PM -0700
References: <20000605160915.G11994@activestate.com> <XFMail.20000605161330.shaleh@valinux.com>
Message-ID: <20000606012119.F469@xs4all.nl>

On Mon, Jun 05, 2000 at 04:13:30PM -0700, Sean 'Shaleh' Perry wrote:

> > I don't think I am sticking my foot in my mouth: Strings are immutable in
> > Python and passed/assigned by value (not by reference, which is what you are
> > asking for here). Arrays and dictionaries are passed by reference. Maybe you
> > could setup you data to use an array, pass a reference to it and change its
> > value, like this:

> I knew that a string itself was immutable, I guess I was looking for a
> reference object of some kind (a la C pointers, C++/perl references).

Well, if you wanted a reference, you got it :) 

foo = "string"
bar = foo

Makes foo and bar refer to the same object, the string "string". But bar
doesn't point to foo, it merely points to the same object foo points to, so
if you change foo to point somewhere else, foo and bar wont be the same
anymore.

However, if you change the object foo points to, the object bar points to
will change, too -- because it's the same object. But since strings are
immutable, you *can't* change the object foo or bar points to. The only
thing you can do is make foo or bar point to a different string, and you
can't do that for both foo and bar, magically, at the same time.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From ayinger1 at my-deja.com  Fri Jun  2 19:35:02 2000
From: ayinger1 at my-deja.com (yinger)
Date: Fri, 02 Jun 2000 23:35:02 GMT
Subject: ILU Question...
Message-ID: <8h9gb5$g54$1@nnrp1.deja.com>

How do I have an ILU method return the equivalent of a Python StringIO
object in ISL file?

example:

I have something like this in my ISL file:

TYPE DeploymentServer = OBJECT
     COLLECTIBLE
     DOCUMENTATION "basic ISL wrapper for distributed deployment server"
     METHODS
          Pull(tag:ilu.CString, sourceDir:ilu.CString,
atype:ilu.CString) : StringIO "cvs checkout tag, return StringIO",
     END;




So I guess what I am asking is: is there some ilu object that is
exactly like a python StringIO object?  Like ilu.CStringIO or
something?  If not, how do I return my own Python object?  Do I have to
use that ILU pickle crap or something?  How does that work?

Thanks in advance,

andrew at mail.idrive.com


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


From skip at mojam.com  Fri Jun 30 11:08:25 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 30 Jun 2000 10:08:25 -0500 (CDT)
Subject: Passing variable-length keyworded arguments list to base class constructor?
In-Reply-To: <962379165.85859735@news.libertysurf.fr>
References: <962352947.103057778@news.libertysurf.fr>
	<8jhum101vku@news2.newsguy.com>
	<962367281.1198008296@news.libertysurf.fr>
	<8ji6qg$8oq$1@news1.sunrise.ch>
	<962379165.85859735@news.libertysurf.fr>
Message-ID: <14684.47081.590051.102076@beluga.mojam.com>

    >> Starting with Python 1.6a[12], you can call Parent.__init__(self,
    >> **args).
    Jerome> And with python 1.5.2?

You must still use apply():

    class Child(Parent):
        def __init__(self, *args, **kwargs):
	    apply(Parent.__init__, args, kwargs)
	    ...

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
On Phil Jackson's ability to manage multiple Lakers superstars, Shaquille
O'Neal said: "He's got the rings.  I listen to the man with the rings."



From horst at proceryon.at  Wed Jun 14 07:36:52 2000
From: horst at proceryon.at (Horst Gassner)
Date: Wed, 14 Jun 2000 13:36:52 +0200
Subject: Tag file for Visual Slick Edit
Message-ID: <39476E54.4673C04B@proceryon.at>

Hi all!

Does anyone use Visual Slick Edit for Python development?
If yes, does anyone konw if it is possible to get a so called tag-file
for Python?

Thanx in advance
Horst


From m.faassen at vet.uu.nl  Tue Jun 20 09:03:58 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jun 2000 13:03:58 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no>
Message-ID: <8inq3u$p0i$1@newshost.accu.uu.nl>

Thomas Weholt <thomas at cintra.no> wrote:

[scary stuff about 'closed-source fanatics' -- these exist outside
companies which sell software? :)]

> To be more exact :

> My mission is to swap

> expensive, oversized databases solutions like MS
> SQL 7.0 with PostgreSQL 7.0,

> crappy inconsistent script-languages like
> VBScript with Python

> and MS SiteServer / IIS with Zope,

This all sounds great!

> Linux with Windows  2000/ NT 4.0.

Why switch from Linux to Windows NT/2000, though? ;)

[snip]
> The thing is that these people don`t thrust OpenSource. They laugh at
> script-languages in general,

Probably due to exposure to various crappy scripting languages. Perhaps
just not use the word scripting. Use words like 'byte-codes' and 'virtual
machine'. Mention 'portable'. Thanks to Java these words have good hype-value.

Application specific hacked-up scripting languages are generally pretty
crappy, so they may have all the wrong association setup.

Of course there's probably also the 'real men use compilers, not
interpreters' aspect. The response to that should be reason: 'wimps
using interpreters can produce the same application in less time and
end up with a more robust and flexible system. And *perhaps* it runs
slower, but usually that's fast enough'.

> and the idea of sharing code is even more stupid. 

Perhaps point at successful open source projects?
Or perhaps cite from personal experience. Or have them read something by ESR.

> The growth of OpenSource, Python, Linux etc. depends on how well it
> can be a suitable substitution for ClosedSource-alternatives. A more
> agressive form of propaganda, showing off successful solutions, WITH
> benchmarks, results of stress-tests etc. is vital. Right now there are
> alot of talk about how great Python are, but few actual test of how
> well Python performes under pressure, and not just Python alone, but
> in a more complex environmet, like in Zope

Zope appears to survive slashdotting. There were various discussions about
Zope under stress on the zope lists recently. There's also a wiki; perhaps
you can find more info there.

> or using Python as interface to a RDBMS.

> I need selling points for OpenSource, the idea and philosophy,
> examples of successful technology and eh .. moral support. I feel like
> I`m in the lions den right now. ( " Think of a happy place. Calm blue
> ocean, calm blue ocean ..." )

I think Python is relatively easy to sneak in. Start with
a few 'insignificant scripts' that after a while turn out to do 
quite important stuff; i.e, bind them with some subtle spells.
Python can do neat stuff on Windows, so you might want to exploit that.
Use it to convert or adjust existing databases; I've used Python
quite a bit for that.  Or have it generate PDFs from whatever data you have.

Also perhaps sneak in PostgresSQL through one of the various PostgresSQL
on Windows solutions; the database runs on Linux but something bridges
it to Windows (perhaps through ODBC).

Zope is harder to introduce, as it's such an extensive package. Maybe as a
way to integrate various company data sources into an intranet site. Some
Zopists are working on stuff like that. 

Good luck!

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


From johngrayson at home.com  Fri Jun  9 07:14:49 2000
From: johngrayson at home.com (John Grayson)
Date: Fri, 09 Jun 2000 11:14:49 GMT
Subject: Can this be stopped from leaking memory?
References: <393FFFE7.1887276C@ebeon.com>
Message-ID: <8hqjj1$ml4$1@nnrp1.deja.com>

In article <393FFFE7.1887276C at ebeon.com>,
  Niall Smart <niall.smart at ebeon.com> wrote:
> Hi,
>
> I'm wondering if this can be stopped from leaking memory?
>
> import imp
> import time
>
> for i in range(0, 20000):
>         fn = "mod.pyc"
>         m = imp.load_module(str(time.time()), open(fn, "rb"), fn,
(".pyc", "rb", imp.PY_COMPILED))
>         m.main()
>
> Where mod.py is:
>
> def main():
> 	print "foo"
>
> or something else trivial.
>
>

The first argument to load_module is the module NAME,
so I don't see how 'str(time.time()) satisfies the module
name.

*This* does not leak...

import imp

fn = "mod.pyc"
for i in range(0, 20000):
    m = imp.load_module('Mod', open(fn, "rb"), fn, (".pyc", "rb", \
                                                imp.PY_COMPILED))
    m.main()

However, this leaves the file descriptor dangling (further reading of
the documentation reveals:

Important: the caller is responsible for closing the dile argument,
if it was not None, even when an exception is raised. ...

So, this is probably better:

import imp

fn = "mod.pyc"

for i in range(0, 20000):
    fd = open(fn, "rb")
    m = imp.load_module('Mod', fd, fn, (".pyc", "rb", imp.PY_COMPILED))
    fd.close()
    m.main()



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


From hniksic at iskon.hr  Fri Jun  9 15:57:15 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 09 Jun 2000 21:57:15 +0200
Subject: Python for Windows -- what to install?
Message-ID: <dpbt1ay6ec.fsf@mraz.iskon.hr>

I'd like to install Python for my younger brother (age 16) on the
Windows machine that he uses.  I've never installed Python on Windows,
I'm not sure what components I need to prepare.

I'd like him to be able to use the stuff that is normally distributed
with Python 1.5.2, plust IDLE under Tkinter.

I assume I'll need `py152.exe' for the basic functionality.  Do I need
`win32all.exe' for the above mentioned things to work?  If so, does it
supersede py152.exe?

Do the above files include documentation, or do I have to download it
separately?

Thanks for your help!


From jbauer at rubic.com  Tue Jun  6 17:34:27 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 06 Jun 2000 16:34:27 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun  5)
References: <BA7ED618087F752B.EE3239CC079D7980.5169D95D2EE3FD08@lp.airnews.net> <060620001708472558%pecora@anvil.nrl.navy.mil>
Message-ID: <393D6E63.1C2AA631@rubic.com>

>> It's finally public; Guido, Fred Drake, Jeremy Hylton, 
>> and Barry Warsaw are now working for BeOpen.

> Is the above good, bad or neither?

Good.  Now that the guys are finally employed,
the PSA can get these slackers off the dole rolls.

-Jeff



From matt at null.net  Mon Jun 19 20:30:31 2000
From: matt at null.net (Matthew Schinckel)
Date: Mon, 19 Jun 2000 18:30:31 CST
Subject: executing os commands
References: <394C0A4D.E2B9DBD8@uniserve.com>
Message-ID: <20000619183041.090632@210.9.22.17>

Bob van der Poel <bvdpoel at uniserve.com> wrote:
> Is getstatusoutput() the perfered method to run os commands from
> within a python script or is there something more generic? This works
> fine, but I'm just thinking that since this is part of a unix only
> module....



Try using os.system(command)

Matt.





From azuresky at my-deja.com  Sat Jun 24 16:59:55 2000
From: azuresky at my-deja.com (Acorn)
Date: Sat, 24 Jun 2000 20:59:55 GMT
Subject: Also looking for DBI info
References: <j454jscp2mdk4pkj8nfeb35l4ugc64bso7@4ax.com>
Message-ID: <8j37g5$btq$1@nnrp1.deja.com>

I too am in the same boat.  Someone with some knowledge on the subject,
please give Mat (and me!)  a hand.

My work involves both DBA work as well as application programming (front-ends
for DB's).  In most cases I'm working with MS-SQL Server or Centura SQLBase. 
I'd like to try my hand at developing python interfaces for my own projects
going against the same database backends.

What are my options?  (I don't mind developing a solution, but an outline on
how to proceed from someone here would be much appreciated.)


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


From rol9999 at attglobal.net  Thu Jun 15 08:05:48 2000
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Thu, 15 Jun 2000 12:05:48 +0000
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu>
Message-ID: <3948C69C.7D37E1BF@attglobal.net>

How about this;

>>> class A:
...     def __init__(self):
...         self.x = 1
...     def __setattr__(self, name, value):
...         if name == "x":
...             self.__dict__[name] = value
...         else:
...             raise AttributeError
...
>>> a = A()
>>> a.x = 10
>>> a.y = 20
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 8, in __setattr__
AttributeError

Roland Schlenker



From tismer at tismer.com  Thu Jun  8 10:11:22 2000
From: tismer at tismer.com (Christian Tismer)
Date: Thu, 08 Jun 2000 16:11:22 +0200
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com>
	 <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us> <20000608073337.A324967@vislab.epa.gov>
Message-ID: <393FA98A.5699E035@tismer.com>


Randall Hopper wrote:
> 
> Andrew Kuchling:
...
>  |So, what would people want on top of distutils?  A Debian-style "do
>  |apt-get, upgrade" to update currently installed packages?  A
>  |CPAN-style "fetch and install this module for me, please"?  Simple
>  |version checking?
> 
> Someday, hopefully:
> 
>          setup.py install
> 
> or some tool will fetch, build, and install, not just the module dist, but
> all of its dependencies too.
> 
> I guess I've been spoiled by the FreeBSD ports system.  Fetch the small
> port for a package and just kick off "make install".  Under the hood, it
> fetches source dists from ftp/www sites, and patches, configures, builds,
> and installs the package _and_ all its dependent packages, all
> automagically.  Simple (particularly good for the newbies).

Sounds great.
Do you have an idea how they handle the security
part of the game?
If this kind of thing, with multiple ftp servers involved
and scripts in several packages being auto-called, gets
tampered by a moderate level hacker, (s)he could do quite
everything nasty.

> Wonder if Python could have something similar.  Folks could post URLs to a
> "getit.py" (or setup.py) script, and users could literally cut-and-paste
> this URL into a Python install tool that would just "do the install".  That
> is, don't bother them with where to FTP or surf, how to get the files,
> where to put the packages, what the name of the distdir in the tarball is,
> etc.  Just install the package and do what it takes to make it happen.
> That'd be cool.

And it would be cool to host this service on the new
Starship site which is being built at the moment.
The machine is very powerful.

ciao - chris

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



From tim.hochberg at ieee.org  Thu Jun 22 19:12:47 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 22 Jun 2000 23:12:47 GMT
Subject: Countdown
References: <3952952D.88E3588D@asme.org>
Message-ID: <m2snu52wez.fsf@spiff.home.com>

Jeff Sandys <sandysj at asme.org> writes:

> Python is really elegant when counting up:
> 
> for i in range(len(thing)):
>     print thing[i]
> 
> But when counting down looks ugly:
> 
> for j in range(len(thing)-1,-1,-1):
>     print thing[i]
> 
> Advice so far:
> Stop doing things backwards:
>     (customer request)
> Use reverse first:
>     (destructive, need to unreverse after)
> Live with it:
>     (bytecode doesn't care, but it is still ugly)
> Try range(len(thing)).reverse():
>     (doesn't work but):
>         rthing = range(len(thing))
>         rthing.reverse()
>         for k in rthing:
>             print thing[i]
>     (works, but also ugly)


This is completely overkill for this, but you could get Numeric and use:

from Numeric import arange
for j in arange(len(thing))[::-1]:
    print thing[j]

-tim


From dan at control.com  Wed Jun 28 10:58:29 2000
From: dan at control.com (Dan L. Pierson)
Date: Wed, 28 Jun 2000 10:58:29 -0400
Subject: Tutorials at IPC9
References: <39591877.535A96BC@sage.att.com>
Message-ID: <DBJaOdIZt=rHMEreIVh+jZXbEJqI@4ax.com>

Garry Hodgson <garry at sage.att.com> wrote:
> I'm not looking for specific topics per se (although I'm open to
> suggestions), but rather what kind of classes would be useful.
> To get things going, here are some thoughts I had on what might
> be good:

I'd love to have Phillip and/or Ty give an in depth tutorial on
applying ZPatterns!  This would be in the "tutorials for experienced
Zope developers" category, but I bet that a lot of us would appreciate
it.

Dan Pierson
Dan Pierson, Control Technology Corporation
dan at control.com


From db3l at fitlinxx.com  Wed Jun 14 19:15:34 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 19:15:34 -0400
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net>
Message-ID: <uitvbluqx.fsf@ctwd0143.fitlinxx.com>

"Jin" <jin.choung at gte.net> writes:

> but i think the reason the C++ book is working so well for me is that it
> assumes nothing about prior knowledge in an OO language because C++ is the
> base upon which everything else is built.  the book realizes this and so all
> such concepts are explained in detail from scratch.

Just a friendly comment - if what you've read so far has led you to
believe that C++ is the origin of all OOP concepts and the base upon
which all OO languages have been built, I'd suggest you could use a
few more books at some point and a little broader exposure :-)

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


From mikael at isy.liu.se  Fri Jun 16 07:59:03 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 16 Jun 2000 13:59:03 +0200 (MET DST)
Subject: more on python and GUI speed
In-Reply-To: <jGl25.2323$3E6.42734@news1.alsv1.occa.home.com>
Message-ID: <XFMail.000616135903.mikael@isy.liu.se>

On 16-Jun-00 Daniel Kramer wrote:
 >      The tkinter and wxPython apps were very choppy when the cpu was under
 >  load from another process.. where the tcl/tk app saw almost no performance
 >  hit at all.. it was still just as fluid as when my computer's cpu was idle.

I'm no expert here, but it sure sounds like your different programs are
given different priorities. 

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    16-Jun-00
Time:    13:57:00

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

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


From m.faassen at vet.uu.nl  Wed Jun  7 11:55:07 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 15:55:07 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com>
Message-ID: <8hlr8r$77a$1@newshost.accu.uu.nl>

Courageous <jkraska1 at san.rr.com> wrote:
[snip]
> Well, yeah, but [0:3] as "element zero through three, inclusive"
> is about as intuitive as you can get. While I understand completely
> how the current implementation works (reading the source code and
> all, dontcha know), I still fail to see *why* this was chosen.

a[0:i] + a[i:] == a

a[0:len(a)] == a

Inserting a list at index i:

a[:i] + b + a[i:]

In practice, no need to think about a *lot* of off-by-one errors.

Regards,

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


From johnvert at my-deja.com  Mon Jun 19 16:06:01 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Mon, 19 Jun 2000 20:06:01 GMT
Subject: {Q} OOP Animal class.
Message-ID: <8iluf0$le$1@nnrp1.deja.com>

Greetings,

I just the example and explanation of the classic Animal class in the
tutorial in the end of Programming Python by Orielly, and I have a few
questions.

This is the class (irrelevant parts snipped):

class Animal:
  def reply(self):  self.speak()
  def speak(self):  print 'spam'

class Mammal(Animal):
  def speak(self):  print 'huh?'

class Primate(Mammal):
  def speak(self):  print 'Hello world!'

class Hacker(Primate):
  pass

>>> data = Hacker()
>>> data.reply()
Hello world!

Programming Python says: ``Hacker has no methods of its own, but
inherits from Primate, which inherits from Mammal, which inherits from
Animal. When we call the reply method of the Hacker instance, it runs
Animal's reply, which in turn calls the Primate's speak method
(self.speak finds the first speak by looking at Hacker and above).
Finally, our message gets printed.'', p. 846

I want to make sure I understand the sequence of events right:

1. I call the method reply()
2. The interperter sees reply() does not exist in class Hacker
3. It goes one class up the hierarchy to class Primate
4. reply() doesn't exist in class Primate
5. It goes one class up the hierarchy to class Mammal
6. reply() doesn't exist in class Mammal either
7. It goes one class up the hierarchy to class Animal
8. reply() exists in class Animal, it calls it, and self.reply() gets
translated to `Primate.reply()' since Hacker is of type Primate
9. Primate.reply() is called
10. `Hello world!' is printed

Is that about right? The thing that I'm most unsure of is step #8..
hoping you could clarify.  I really want to have a solid understanding
of this because it's important to understand OOP.

Thanks,
--
John




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


From nikolai.kirsebom at siemens.no  Tue Jun 20 04:15:19 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Tue, 20 Jun 2000 08:15:19 GMT
Subject: win32all-132 / Word Automation
References: <394e2b8d.515788975@news.mch.sni.de> <lAy35.3131$qz4.24329@news-server.bigpond.net.au>
Message-ID: <394f269c.580059221@news.mch.sni.de>

Mark,
Thanks for your quick response.  It might be obvious ... but not
really to me.

Here are the results from your suggested commands:

>>> print doc
Document2
>>> print doc.Variables
<win32com.gen_py.Microsoft Word 8.0 Object Library.Variables>
>>> print doc.Variables.Item
<method Variables.Item of Variables instance at 1f79250>
>>> print doc.Variables.Item(1)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File
"win32com\gen_py\00020905-0000-0000-C000-000000000046x1033x8x0.py",
line 11402, in __str__
    return str(apply( self.__call__, args))
  File
"win32com\gen_py\00020905-0000-0000-C000-000000000046x1033x8x0.py",
line 11398, in __call__
    return apply(self._ApplyTypes_, (0, 2, (8, 0), (), "Value", None)
)
  File "D:\Python\win32com\client\__init__.py", line 341, in
_ApplyTypes_
    return self._get_good_object_(apply(self._oleobj_.InvokeTypes,
(dispid, 0, wFlags, retType, argTypes) + args), user, resultCLSID)
AttributeError: 'tuple' object has no attribute 'InvokeTypes'


>
>This is different from VB, due to the different handling of byref args.
>
>Try:
>>>> print doc
>>>> print doc.Variables
>>>> print doc.Variables.Item
>>>> print doc.Variables.Item(1)
>
>And it should become obvious...
>
>Mark.
>
>
>



From jerome.quelin at insalien.org  Mon Jun 26 10:11:43 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Mon, 26 Jun 2000 16:11:43 +0200
Subject: Final garbage collection?
Message-ID: <962028703.1248416950@news.libertysurf.fr>

Hi all,

I would like to know if there is really a final garbage-collection in
Python: it seems that it does not handle cross-references if we do not
break cycles...

$ cat test.py
#!/usr/bin/python
class Test:
    def __init__(self, name):
        self.name = name
    def __del__(self):
        print "Ouch! (", self.name, ")"
a = Test('a')
b = Test('b')
c = Test('c')
a.cross_ref = b
b.cross_ref = a
$ ./test.py
Ouch! ( c )
$

It seems like only the c instance gets destroyed?!
Yet when the interpreter stops, it is supposed to give all the memory
back to the system, therefore destroying all objects created during
the run of the program. And if the objects are destroyed, aren't the
destructors supposed to be called?

But in the example shown above, the only instance for which the
destructor gets called is 'c'. What happened to the 'a' and 'b'
instances? They sure are destroyed by the system, but does Python
notice this "memory leak" (ok, it isn't a memory leak since the system
gets the memory back)? Why do their destructors don't get called? Is
it that Python does not have a _final_ garbage-collection that can
handle cycles and break them?

Thanks,
Jerome
--
jerome.quelin at insalien.org


From borlak at home.com  Sat Jun  3 22:43:20 2000
From: borlak at home.com (Michael Morrison)
Date: Sun, 04 Jun 2000 02:43:20 GMT
Subject: User Input?
Message-ID: <cnj_4.3811$r7.919327@news1.rdc2.tx.home.com>

My friend was talking about a programming asignment and I was telling him
about python, so we decided to make programs in our different languages (his
being QBasic :P) ...

Anyway, it dealt with user input, and from only having read Learning Python,
I didn't know how to do it (and still don't).
Is there an easy way to do user input in python?  I found a non-standard
module, but it looked too complicated (not something I'd expect from the
ease-of-use I've had so far).

Example: python blah.py
Name? (type in name)
Address? (type in address)
etc....

Any help would be appreciated, thanks!






From aahz at netcom.com  Wed Jun  7 12:22:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 16:22:34 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news> <8hlrkn$5t2$1@nnrp1.deja.com>
Message-ID: <8hlssa$r3$1@nntp9.atl.mindspring.net>

In article <8hlrkn$5t2$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com> wrote:
>
>    Not if they never hear about it. I thought he went out of his
>way to be polite, to make it clear that he wasn't saying Python
>bad Ruby good. Sounds to me like something that might be of
>interest to a lot of Python people.
>    Maybe it doesn't matter to people who are somehow instantaneously
>informed of everything, but the rest of us appreciate a little bit
>of marginally on-topic information. (I first heard about Delphi when
>I was lurking in a VB group trying to figure out whether it was more
>awful or less awful than the BASIC I was using at the time...)

Well, Ruby comes up often enough naturally.  I think the objection here
comes from the explicit advertisement combined with the repetition (this
isn't the first time).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From tgfuchs at my-deja.com  Thu Jun 15 02:10:02 2000
From: tgfuchs at my-deja.com (Thomas Fuchs)
Date: Thu, 15 Jun 2000 06:10:02 GMT
Subject: Standalone http server, cgi software?
References: <3946EEA9.DFCBAB4D@erols.com> <394737F4.B3003678@stroeder.com> <39479D02.42B5AD06@erols.com>
Message-ID: <8i9rvj$796$1@nnrp1.deja.com>

In article <39479D02.42B5AD06 at erols.com>,
  "Edward C. Jones" <edcjones at erols.com> wrote:
> I am looking for a system that resides entirely on one PC. The idea is
> to pass arguments to a Python program via a HTML form. The program
> would look things up in a database, process them, etc. and return to
> the browser a page of html containing the answers.
>
This is also my preferred method for a simple GUI.
Here is a minimal sample program:
"""
simple http server
it executes a system command and returns stdout of command
or return code as plain text

arguments: -port portnumber
default port: 8000
accepted urls:
      /command?cmd=command       # perform command
      /command/form/get          # send form for get
      /command/form/post         # send form for post

usage: start the program and in the browser type the address
    http://localhost:8000/command/form/get

This server is intended to be a starting point for writing any simple
http server.
Modify the part between -- edit begin -- and -- edit end --
supported methods are GET, HEAD and POST,
where POST content-type must be application/x-www-form-urlencoded
Two functions must be provided:
analyse(pathlist, d, headOnly = None)
    pathlist is the path part as list splitted by '/'
    d is the query string as dictionary
    headOnly is set, if method is HEAD
    the function must return value a triple:
    typ       content-type, if not set, the http return code is 404
    size      optional
    info      forwarded to do function
do(info, fout)
    info is the result of analyse
    fout the output stream
    the function must write the content to fout

"""

import os
import sys
import string
import SocketServer
import BaseHTTPServer
import urllib
import socket

class RequestHandler(BaseHTTPServer.BaseHTTPRequestHandler):

    def do_GET(self):
        info = self.send_head()
        do(info, self.wfile)

    def do_HEAD(self):
        self.send_head(headOnly = 1)

    def do_POST(self):
        ctyp = self.headers.getheader('content-type')
        if ctyp != 'application/x-www-form-urlencoded':
            print 'POST ERROR: unexpected content-tpye:', ctyp
            return
        clen = self.headers.getheader('content-length')
        if clen:
            clen = string.atoi(clen)
        else:
            print 'POST ERROR: missing content-length'
            return
        data = self.rfile.read(clen)
        dummy = self.rfile.read(2)
        self.path = '%s?%s' % (self.path, data)
        self.do_GET()

    def send_head(self, headOnly = None):
        pathlist, d = urlParse(self.path)
        typ, size, info = analyse(pathlist, d, headOnly)
        if typ:
            self.send_response(200)        # http return code: ok
            self.send_header('Content-type', typ)
            if size:
                self.send_header('Content-length', size)
        else:
            self.send_response(404)        # http return code: not found
        self.end_headers()
        return info

# url handling

def urlParse(url):
    """ return path as list and query string dictionary
        strip / from path
        ignore empty values in query string
        for example:
            if url is: /xyz?a1=&a2=0%3A1
            then result is: (['xyz'], { 'a2' : '0:1' } )
            if url is: /a/b/c/
            then result is: (['a', 'b', 'c'], None )
            if url is: /?
            then result is: ([], {} )
    """
    x = string.split(url, '?')
    pathlist = filter(None, string.split(x[0], '/'))
    d = {}
    if len(x) > 1:
        q = x[-1]                  # eval query string
        x = string.split(q, '&')
        for kv in x:
            y = string.split(kv, '=')
            k = y[0]
            try:
                v = urllib.unquote_plus(y[1])
                if v:               # ignore empty values
                    d[k] = v
            except:
                pass
    return (pathlist, d)

# --------------  edit begin ----------------------------

defaultPort = 8000                 #   set default port for server

def analyse(pathlist, d, headOnly = None):
    size = None
    info = ''
    typ = None
    #print pathlist, d    # test only
    if pathlist == ['command'] and d.keys() == ['cmd']:
        typ = 'text/plain'
        if not headOnly:
            info = syscmd(d['cmd'])
            size = len(info)
    elif pathlist[:2] == ['command', 'form'] and d == {} and \
         pathlist[2:] in [['get'], ['post']]:
        typ = 'text/html'
        if not headOnly:
            info = buildform(pathlist[2])
            size = len(info)
    return (typ, size, info)

def do(info, fout):
    fout.write(info)

# task specific helper functions
def syscmd(cmd):
    f = os.popen(cmd, 'rb')
    rsp = f.read()
    rc = f.close()
    return rsp or 'RC: %s' % rc

def buildform(method):
    # get hostname of server
    try:
        host = socket.gethostname()
    except:
        host = "???"
    doc = '''\
<TITLE>command %(host)s</TITLE>
<H1>command %(host)s</H1>
<P>
<FORM ACTION="/command" METHOD="%(method)s">
command:  <INPUT TYPE=TEXT NAME="cmd" SIZE=80>
<BR>
<INPUT TYPE=SUBMIT>
</FORM>
''' % locals()
    return doc

# --------------  edit end ----------------------------

if __name__ == '__main__':
    # argument handling
    try:
        i = sys.argv.index('-port')
        port = string.atoi(sys.argv[i+1])
        del sys.argv[i:i+1]
    except:
        port = defaultPort

    # start server
    print 'Serving on port', port
    server = SocketServer.TCPServer(('', port), RequestHandler)
    server.serve_forever()


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


From thomas at xs4all.net  Wed Jun 14 09:51:25 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 14 Jun 2000 15:51:25 +0200
Subject: tuple creation in C extensions
In-Reply-To: <140620000716512992%pecora@anvil.nrl.navy.mil>; from pecora@anvil.nrl.navy.mil on Wed, Jun 14, 2000 at 11:16:50AM +0000
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil>
Message-ID: <20000614155125.A26436@xs4all.nl>

On Wed, Jun 14, 2000 at 11:16:50AM +0000, Louis M. Pecora wrote:

> > Wow, thanks for looking that code over, Bernhard.  What a horrible bunch
> > of holes.

> And who could have seen all that so clearly without an expert on hand. 
> Even then I couldn't understand all the reasoning.  In Python we are all
> happy, but the ugly side is when we return to C.  It's obvious to me:

> CAN'T SOMEONE UPDATE THE EXTENSIONS PDF MANUAL, ADDING A BETTER TUTORIAL
> ON REFERENCE COUNTS AND (!!) LISTING Py_XXXX FUNCTIONS AND THEIR REFERENCE
> EFFECTS ??

Sure, anyone can. The address to submit your new sections is
python-docs at python.org, I believe. Note that you shouldn't send in PDF
documents. The actual documentation is written in TeX, which is used to
generate PostScript, PDF, HTML, info and what not. I'm sure someone will
volunteer to rewrite your plaintext-version to TeX if you write a good
manual on reference counts ;-)

> Sorry to yell, but it's obvious that many people like me are stumbling
> with this one.  I'd hate to have all that time I save using Python eaten
> up by debugging C extensions.

The API interface documentation is pretty clear in which functions give or
take a reference. *most* functions do not touch references, they take the
object you give to them, do their thing, and return a result. Those that
return objects are documented as returning either a new reference or a
borrowed reference (in which case, if you keep the object around, you have
to INCREF yourself) and those that steal a reference (take an Object
argument without returning it, and without INCREFing it) are documented to
do so as well.

The only trouble I had with refcounting (while writing the
augmented_assignment patch) was not bothering to see wether PyDict_SetItem()
stole a reference or not. It doesn't, but I did INCREF, so I was leaking ;-P

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From info.jyv at videotron.ca  Fri Jun  9 22:49:41 2000
From: info.jyv at videotron.ca (Jean-Yves Vinet)
Date: Fri, 09 Jun 2000 22:49:41 -0400
Subject: windll
Message-ID: <3941ACC5.305B8830@videotron.ca>

Is there someting like to "windll" to use
dynamic librairy under Linux/Unix plateform.


JYV.





From gmcm at hypernet.com  Thu Jun 15 14:03:45 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jun 2000 18:03:45 GMT
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com> <01bfd2ed$42b67880$29cf8ad1@daves-dell> <39463BA7.43A3CD65@gssec.bt.co.uk> <39474D1D.F959969C@prescod.net> <8F545A89Agmcmhypernetcom@199.171.54.154> <8ial2l$opl$1@nnrp1.deja.com>
Message-ID: <8F54830BDgmcmhypernetcom@199.171.54.195>

echuck3 at my-deja.com wrote:

>In article <8F545A89Agmcmhypernetcom at 199.171.54.154>,
>  gmcm at hypernet.com (Gordon McMillan) wrote:
>> Paul Prescod wrote:
>>
>> >Try double clicking a PDF document with a space in its path. Adobe
>> >Acrobat 4 will not find it. It's annoying as hell because Internet
>> >Explorer puts files from the Web in the "Temporary Internet Files"
>> >directory.
>>
>> That's the installer's fault, not Acrobat 4's. Just (double)quote the
>%1 in
>> the open action for the file type.
>>
>> I get a big kick out of MS's habitual use of paths with embedded
>spaces.
>> Sort of a barometer of hubris.


>Perhaps they do it on purpose so that programmers will get their act
>together and support pathnames with spaces. As someone already pointed
>out, this has been supported for many years on UNIX and I guess about 5
>on Windows.

>I don't care for Microsoft, but I have to say it's the developers who
>have screwed up this time.

Snort. Try the system() call. If the arg has spaces in it, it has to be 
within double quotes. Fine. But if it *doesn't* have spaces, lovely Windows 
will give you an error if you *do* double quote it. And the error msg is 
completely incomprehensible (of course). [This on NT 4 - other forms of 
Windows may blow different colored chunks.]

>What's wrong with the developers at Adobe that [a] they can't handle a
>space and [b] their pre-release testing didn't find this?

I doubt there's any problem in opening a file with a space in the path. The 
issue is that the file association says:

<some.exe> %1

where it should say:

<some.exe> "%1"

(which it does on my system, but maybe I fixed it; or maybe Paul screwed it 
up ;-)).

- Gordon


From rupole at compaq.net  Mon Jun 19 18:10:06 2000
From: rupole at compaq.net (Roger Upole)
Date: Mon, 19 Jun 2000 22:10:06 GMT
Subject: Getting remote memory configuration
References: <2B1262E83448D211AE4B00A0C9D61B03013011F8@msgeuro1.creo.be>
Message-ID: <2Tw35.7372$Zh1.638419@news.easynews.com>

Try the win32pdh module from the win32 extensions.  The NT performance
monitor can access memory stats on another machine if you have sufficient
privilege.  The hard part is getting the path string right, but there is a
path
browser to help with that.  I think it's win32/lib/win32pdhutil.py.
                       Roger Upole

"Pieter Claerhout" <PClaerhout at CREO.BE> wrote in message
news:2B1262E83448D211AE4B00A0C9D61B03013011F8 at msgeuro1.creo.be...
> Forgot to mention that I was talking about Windows NT machines.
>
> Pieter
>
> Pieter Claerhout
> Application Support - CreoScitex Europe
>
> > -----Original Message-----
> > From: Pieter Claerhout
> > Sent: Monday, June 19, 2000 3:40 PM
> > To: 'python-list at python.org'
> > Subject: Getting remote memory configuration
> >
> > Anyone an idea on how you can find out how much memory a remote computer
> > has installed? The computer is available on the same network and has the
same
> > login as the computer on which you are running the command.
> >
> > Pieter
>




From mal at lemburg.com  Fri Jun 23 04:34:52 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 23 Jun 2000 10:34:52 +0200
Subject: SAP R/3 and Python ?
Message-ID: <3953212C.84E3127F@lemburg.com>

Hi everybody,

does anyone have an idea of how to interface to SAP R/3 from
Python ?

If there are no such Python interfaces, how much effort would it
take to add orders from a web system into an R/3 system ?

[I know, my ignorance shines through here: I've never used
 R/3, but a customer has confronted me with that need.]

Thanks,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/



From tom-main at REMOVEME.home.com  Tue Jun  6 13:39:48 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 06 Jun 2000 17:39:48 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au>
Message-ID: <EHa%4.165801$55.3514980@news2.rdc1.on.home.com>

Mark,

Thanks for the response.

I'm interested in your comments on MFC because I'm an MFC developer who is
looking for an alternative.

I agree that MFC seems to be on the way out, but there are two things that
prevent me (and some other MFC developers) from switching:
1) The CodeTools & CodeGuru sites contains TONS of peer reviewed code (in
the form of messages at the end of every article) that is free, even for
commercial use.  80% of this code is in the form of MFC extensions.
2) What's next?
- STL?  Yes, but that's not a GUI.
- ATL & WTL?  Too MS only.
- wxWindows?  Only has a tiny fraction of the number of users & code that
the other options have.
- TCL/TK?  I've heard this is slow, and it can't create native windows' apps
(users want the windows look & feel).
- Java?  Again, too proprietary.

Since I'm starting to learn Python (using your book), it would be nice to
use something that is in common use from both Python and C++.  I guess that
leaves me with wxWindows.

BTW, I think I'm part of the target market for your Visual Studio Python
product.  I know VS quite well, so it would be nice (ie. worth paying for)
to be able to develop for Python in it.  If this meant that I could use the
windows' resources that I'm familiar with than that would also be
convenient, but then the cross-platform character of Python would be lost -
so I don't know which I would prefer.

Tom.

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:m03%4.6579$Hz.48365 at news-server.bigpond.net.au...
> "Tom" <tom-main at REMOVEME.home.com> wrote in message
> news:w3Q_4.160138$55.3354931 at news2.rdc1.on.home.com...
>
> > Studio 7, but what does this all mean for the future of the PythonWin?
>
> Dunno.  Not much will change.  As usual, no one will be paying me to work
> on it! :-)
>
> > Will PythonWin (the IDE and the MFC wrapper library) continue to be
> actively
> > and publicly developed?
>
> Dunno.  I like it :-)  If Active State end up developing an IDE that I
> would personally prefer to use over Pythonwin, then I doubt I will spend
> too much time on Pythonwin as an IDE.  I seriously doubt that will be MSVC
> based to qualify.  Certainly, the extra resources AS can dedicate should
> mean the Mozilla based IDE ends up qualifying, but Pythonwin exists today
> and is quite slick on Windows (as opposed to the cross-platform focus of
> AS) meaning I doubt it will qualify in the short term.
>
> The good side of this is that is doesnt "compete" directly with any AS
> products.  One IDE is focussed on existing MSVC (and maybe new) owners,
> while the other will be a great cross-platform solution.  Pythonwin doesnt
> try and be either of these!  So while it is unlikely that AS will fund
> further development of Pythonwin, they also will certainly not discourage
> me from continuing to tinker.
>
> Re MFC: Microsoft dont seem too interested in MFC any more.  Ive certainly
> heard nothing about it, and everything new I see has alternatives that
> compete directly with it.
>
> Unfortunately, my personal opinion is that MFC itself will die pretty
> soon, kept alive only by the weight of all the old code written for it.
> So with that in mind, I dont believe that MFC will undergo enough
> significant enhancements that keeping up will be a problem!
>
> Mark [and obviously not speaking for ActiveState]
>
>




From duncan at rcp.co.uk  Mon Jun  5 07:41:37 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 5 Jun 2000 11:41:37 GMT
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com> <8h8jg3$d66$1@slb6.atl.mindspring.net> <8h8lkt$rb7$1@nnrp1.deja.com> <8h8qqs$l5a$1@slb6.atl.mindspring.net>
Message-ID: <8F4A84753duncanrcpcouk@194.238.50.13>

aahz at netcom.com (Aahz Maruch) wrote in
<8h8qqs$l5a$1 at slb6.atl.mindspring.net>: 
>Ah.  Python appears to allow three-digit hex numbers.  I guess you'll
>have to paste the string together, character by character.

The best way to avoid problems with hex escapes is to use string
concatenation. Unless you know that the following character is safe, you
should always follow each hex escape with "". So "\x1b""a" will give you
the desired effect. (Or use single quotes if you prefer: '\x1b''a'). 


From aa8vb at yahoo.com  Thu Jun  8 14:29:25 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 08 Jun 2000 14:29:25 -0400
Subject: CPAN for python?
In-Reply-To: <393FA98A.5699E035@tismer.com>; from tismer@tismer.com on Thu,
 Jun 08, 2000 at 04:11:22PM +0200
References: <393fbb15.100053125@nntp.interaccess.com>
 <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us>
 <20000608073337.A324967@vislab.epa.gov> <393FA98A.5699E035@tismer.com>
Message-ID: <20000608142925.A328070@vislab.epa.gov>

Christian Tismer:
 |Randall Hopper wrote:
 |> Andrew Kuchling:
 |>  |So, what would people want on top of distutils?  A Debian-style "do
 |>  |apt-get, upgrade" to update currently installed packages?  A
 |>  |CPAN-style "fetch and install this module for me, please"?  Simple
 |>  |version checking?
 |> 
 |> I guess I've been spoiled by the FreeBSD ports system.  Fetch the small
 |> port for a package and just kick off "make install".  Under the hood, it
 |> fetches source dists from ftp/www sites, and patches, configures, builds,
 |> and installs the package _and_ all its dependent packages, all
 |> automagically.  Simple (particularly good for the newbies).
 |
 |Sounds great.
 |Do you have an idea how they handle the security
 |part of the game?
 |If this kind of thing, with multiple ftp servers involved
 |and scripts in several packages being auto-called, gets
 |tampered by a moderate level hacker, (s)he could do quite
 |everything nasty.

I'm not sure I fully appreciate your question, but here goes.

On UNIX, packages are typically installed in common directories only
writable by root.  That protects the integrity of Python, distutils, and
friends from tampering unless your root account is compromised.

The real vulnerability is in the downloaded software (I think that's what
you were referring to).  Did it come from a trusted source, or has it been
verified by a trusted authority?

Two things come to mind with regards to this verification.  All ports are
fetched from ftp.freebsd.org (or an official FreeBSD CD, etc.).  So you
know that someone working for the project has signed off and committed
their name to public CVS as checking the port in as a working, useful port.

That leaves the actual source distribution fetched and built by the port.
The port contains the full URLs to the source dist files, so you can have
confidence those haven't been tampered with.  The port also contains an md5
checksum for the package the port was built against.  If that doesn't
match, the port build is failed.  This guards more against the port
maintainer changing the package after the port was cooked, rather than
being a security measure.  But it could help.  A full digital signature for
the package stored in the port would be better for security.

Key point:  All the "get this package from here", "build this dependent
package", etc. except the user's initial "make install" command comes from
the ports, and these come from the trusted distribution authority.

It's certainly not fullproof but it's pretty good.  If someone hacks DNS
and routes your FTPs/HTTPs to an alternate host, you could be hosed.  If
someone compromises the package author's download machine or ISP web server
where the package lives, you could be hosed.  However all this is also true
for the case where Joe Developer mails out a URL to a Python module so you
can fetch and build it by hand.

I think the key is having a central distribution point for
Python module distributions (or those "getit.py" scripts I was talking
about -- whichever).

For module dists/getit.py scripts not fetched from this central repository,
users of course have to be just as careful with what they fetch and give
control of their machine to as they do any other package.  Did it come from
starship?  python.org?  Some dark site deep in Libya. ;-)

 |And it would be cool to host this service on the new
 |Starship site which is being built at the moment.
 |The machine is very powerful.

Agreed.  Also a good distribution authority since folks could trust module
dists pulled from the site.  Ideally they'd be trial-installed once before
being put up for grabs, or maybe there's a testing record attached to the
download page for each (e.g. installed by tismer at tismer.com on 6/8; seems ok).

-- 
Randall Hopper
aa8vb at yahoo.com



From root at rainerdeyke.com  Mon Jun 19 19:36:47 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 19 Jun 2000 23:36:47 GMT
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net>
Message-ID: <j8y35.1071$ul.67488@news1.rdc2.tx.home.com>

Jeff Massung <jmassung at magpiesystems.com> wrote in message
news:riu35.36$JM5.4708 at news.uswest.net...
> I'm still learning, but lemme take a crack and let others correct me ;)
>
> __name isn't a member of MyDialog class, so you can't use self with it,
> because then it looks for MyDialog.__name, which doesn't exist. But
perhaps
> you can use Goption.__name instead of self.__name?

This would indeed work if it wasn't for name mangling.  Don't call it __name
if you want to access it from a nested class.


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




From jovecka at ethicalhackerz.com  Wed Jun  7 12:12:35 2000
From: jovecka at ethicalhackerz.com (Jim Ovecka)
Date: Wed, 07 Jun 2000 11:12:35 -0500
Subject: Python under VMware/NT?
References: <r3h%4.23$cT1.480@news> <14653.56846.80327.424290@anthem.python.org> <393E446C.665CA10A@first.gmd.de>
Message-ID: <393E7473.41AB9E0B@ethicalhackerz.com>


Thilo Ernst wrote:

> "Barry A. Warsaw" wrote:
>
> > I very briefly ran Linux under NT-native VMware and of course ran
> > Python inside the Linux vm.  Worked fine.  I had the general
> > impression that networking was much slower for the entire OS but spent
> > no time benchmarking or otherwise investigating.  I do plan to play
> > with NT running inside Linux-native VMware pretty soon, but I don't
> > expect it to be all that different than running in native NT.
> >
>
> I could almost repeat Barry's words, just with the OSes reversed. I tried
> some Python stuff on a virtual NT4.0 under Linux-native VMware recently,
> and noticed no severe performance degradations. My Python programs
> weren't network-intensive, though.
>
> The general network performance (i.e. with non-Python apps) is, ahem,
> acceptable, with some (reproducible) exceptions in the "absolutely lousy"
> range. E.g. saving a PowerPoint document to a Samba share on the Linux
> "mother" machine takes AGES as soon as the presentation has more
> than around 10 slides. This is getting off-topic, but ... hints anybody?
> (Perhaps I should simply upgrade to VMware 2.0; still running 1.1)
>
> Nevertheless, by and large I just love having this NT window around in a
> corner  of my KDE desktop. Especially, the setup was great for Python-based
> interoperability experiments I did some months ago, e.g. remote-controlling
> MS-Office apps from Linux via DOPY and the win32 extensions.
>
> Best regards, Thilo

You should use VMware 2.0 and have a large amount of memory 128meg or more.
Networking is some what slower in VMware, but there where some major
improvements in 2.0 . I use RH linux as the host and NT as the guest OS.



From thomas at xs4all.net  Sat Jun 17 09:03:33 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 17 Jun 2000 15:03:33 +0200
Subject: Python 3000
In-Reply-To: <394c2b15.634411296@nntp.interaccess.com>; from olczyk@interaccess.com on Sat, Jun 17, 2000 at 07:49:49AM +0000
References: <394c2b15.634411296@nntp.interaccess.com>
Message-ID: <20000617150333.T26436@xs4all.nl>

On Sat, Jun 17, 2000 at 07:49:49AM +0000, Thaddeus L. Olczyk wrote:
> Is there a link where I can look up what's being considered for Python
> 3000?

Nope, because Py3K is just a name and an idea. A vision, so to speak, and a
place to dump great ideas [for Python] that aren't possible without breaking
old code. But, now that Guido & a couple of hardcore insides have moved/are
moving to PythonLabs, we might just start seeing concrete ideas ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From frankc at colconsulting.com  Thu Jun 15 07:00:54 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Thu, 15 Jun 2000 07:00:54 -0400
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com>
Message-ID: <3948B766.4C256CAE@colconsulting.com>

Courageous wrote:
> 
> 
> > As someone who uses and likes both C++ and Python, I think estimates
> > like that are excessively simplistic.
> 
> Me, too. And yet I will claim that I've become *more* productive
> in Python after a brief 4-6 weeks of use than I am in C++ after
> 15 years of use, and that's no exaggeration. If Python executed
> as fast as other languages, I'd be tempted to *require* that our
> staff use it. And please don't take this wrong; I'm no language
> groupie -- I use C, C++, Java, Python, and Lisp -- all of them both
> lately and frequently.
> 
> > I use C++ for commercial programming, because it gives me the power and
> > control I need.  Productivity isn't the best when measured in time to
> > get the first prototype running, but I've found nothing better when
> > measured in time to develop, deploy, maintain and support something a
> > customer will pay for.
> 
> Shrug. If it weren't for execution speed, your C++ environment would
> be toast. Admit it. :)-

While the last line seems rather toungue in cheek, there are other
things that C++ brings to the table such as encapsulation, access
control and type strictness. 

-- 
Frank V. Castellucci


From darrell at dorb.com  Sun Jun  4 19:00:49 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 4 Jun 2000 16:00:49 -0700
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk> <393A3A9A.FB9F2E37@stud.ntnu.no> <393A4C60.MD-1.4.4.angband@blueyonder.co.uk> <m3d7lxxngf.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <033501bfce78$ba0d62a0$6401a8c0@home>

Check out os.py in the standard libs.
It configures for each OS.
>>> os.name
'nt'

--Darrell Gallion

From: "Michael Hudson" <mwh21 at cam.ac.uk>
> 
> Because I suspect the behaviour common to all the platforms that
> Python runs on is so restricted as to be useless.
> 





From kc5tja at garnet.armored.net  Thu Jun  8 21:32:50 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 8 Jun 2000 18:32:50 -0700
Subject: Testing -- not a new message in a while
Message-ID: <slrn8k0hv9.pq3.kc5tja@garnet.armored.net>

Just testing; I haven't seen any new posts in this newsgroup for quite a
while.  Thanks for your patience.

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From annis at biostat.wisc.edu  Tue Jun  6 14:35:21 2000
From: annis at biostat.wisc.edu (William Annis)
Date: 06 Jun 2000 13:35:21 -0500
Subject: popen in real time?
References: <8hjdn4$6kf$1@bmerhc5e.ca.nortel.com>
Message-ID: <jtln0iy7x2.fsf@wazor.biostat.wisc.edu>

msoulier at nortelnetworks.com (Michael P. Soulier) writes:

> 	Hey guys. I realize that you can loop on the output of popen
> if you want to read from a pipe, but it would seem from experience
> that the output of the child process is not available until the
> process exits. I've tried this with the output of long-running
> processes like find. Is there a way to get the output of a child in
> real time, as it's printed?

        I've found using the Popen3 class in popen2 to be very helpful
in these cases... turn off the buffering by setting the bufsize option
to zero.

-- 
William Annis - System Administrator - Biomedical Computing Group
annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
Mi parolas Esperanton - La Internacian Lingvon  www.esperanto.org


From it at onestopent.com.au  Tue Jun 13 02:39:23 2000
From: it at onestopent.com.au (Jason)
Date: Tue, 13 Jun 2000 16:39:23 +1000
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com>
Message-ID: <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au>

wow thanks for the quick reply!
but can you explain what you wrote for me pls...i am new to the syntax...
where would  i put the original text file to run this script on it? etc...
i am running on windoze 98 at work....

thanks
jason

Courageous <jkraska1 at san.rr.com> wrote in message
news:3945D0AA.E8F51A9C at san.rr.com...
>
> Q: "Can python do this?"
> A: "Python was born for it, baybee!"
>
> import string
> file = open ("file")
> lines = file.readlines()
> for line in lines:
>     pair = string.split(line,":")
>     print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
> file.close()
>
> --
>
> >>> import process
> FORM: ['Popular Recording']
> CMTYPESWITCH: ['Popular']
> RELOCORIGINAL: ['55864A2603AC94ACCA2568F200224388']
> CMCOUNTRY_ORIGIN: ['']
> CMGENRE_CODE: ['']
> **CPTITLE: ['INFINITE POSSIBILITIES']
> **CPARTIST: ['AMEL LARRIEUX']
> CPOARTISTS: ['']
> **CMCATNUMBER: ['4948792']
> **CMAPN: ['9399700067507']
> **CMARIAPRICECODE: ['C9921']
> **CMARIAMEDIACODE: ['21']
> **CMARIADISTRIBUTORCODE: ['SMA']
> CMARIAPACKAGECODE: ['']
> **DMRELEASE: ['19/06/2000']
> cMDateFormat: ['19/06/2000 12']
> DMDELETE: ['']
> cMDeleteStatus: ['No']
> **CMLISTPTRACKS: ['GET UP', 'I N I', 'SWEET MISERY', "SEARCHIN' FOR MY
SOUL", 'EVEN IF', 'INFINITE
> POSSIBILITES', 'SHINE', 'DOWN', 'WEATHER', 'MAKE ME WHOLE', 'GET UP
(THREAD HAD FUN MAIN MI', 'GET
> UP (MV MIG MIX (RADIO))']
> CMLISTPARTISTS: ['', '', '', '', '', '', '', '', '', '', '', '']
> RELOCMEDIA: ['F73C80183A01A8D7CA2568F2002A5517']
> **CURMPRICE: ['15.91']
> **CMTYPE: ['Compact Disc']
> CMRECORDCOMPANY: ['SONY MUSIC']
> CMPACKAGE: ['']
> CMARIADISTRIBUTORHOUSECODE: ['']
> cDistributorHouse: ['']
> CMLOCALE: ['Y']
> $UpdatedBy: ['CN=PPT/OU=AEROS/O=JUKEBOX']
> $Revisions: ['02/06/2000 05']
> >>>




From dwhite2 at blue.seas.upenn.edu  Wed Jun 28 14:15:02 2000
From: dwhite2 at blue.seas.upenn.edu (David White)
Date: 28 Jun 2000 18:15:02 GMT
Subject: list.pop([i]) and list.remove(x) w/ for loops
Message-ID: <8jdfb6$ugh$1@netnews.upenn.edu>

I was getting some unexpected behavior when using list.remove(x), wondering if
someone can tell me what's going on.

I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].  I wan't to
delete all the a objects out of the list while preserving the order of the b
objects.  My code looks like this:

for o in olist:
    if (o.type == a):
       olist.remove(o)

Now for some reason I get a left over a object that wasn't removed.  Anyone
know why?  Using python 1.5.2.

thanks,
david


From scorder at incigna.com  Mon Jun 12 13:02:50 2000
From: scorder at incigna.com (scorder at incigna.com)
Date: Mon, 12 Jun 2000 13:02:50 -0400
Subject: OT: nederlands (Ans something else).
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <PCp%4.109780$Qp2.291904@afrodite.telenet-ops.be> <m3pupth63m.fsf@chinon.cnrs-orleans.fr> <393E9A33.AD5282C1@dulcesol.es>
Message-ID: <6k5aksc7p07le8923ubtvmn7s94ko98i3s@4ax.com>

There is a pdf generating module at www.reportlab.com I have been
experimenting with it now for a few days and it looks great.  It works
well and is fast.  Basically all you do is call graphics methods on a
canvas.  There is also a higher level library built on top called
platypus which allows you to easily do tables and other things that
would be common to a report.  Its a little hard to understand at first
but there are plenty of examples and adequate documentation.  The best
part is that you can take advantage of OOP while designing your
reports.

On Wed, 07 Jun 2000 19:53:39 +0100, Pablo Prieto
<pablo.prieto at dulcesol.es> wrote:

>In VB (Please, put the knife down!) I can use an object called Printer.
>Well...
>I've made a pretty class with this methods (some of them, and translated
>to Python and English)
>
>def setBold(value):
> ....
>
>def setItalic(value):
> ...
>
>def print (text,x,y):
> ...
>
>def newPage():
> ...
>
>and so on.
>
>Is there a module or something that I could do this with python?. I
>guess in VB, Printer is an interfaz to the printer, so could be
>Windows-dependent, but doesn't matter (yet). 
>I use this for generating reports not seen in screen (So wxPython (great
>package, I say!) facilities are not considered, 'cos I don't want to
>print the contents of a widget but lines generated from the program).




From dsavitsk at e-coli.net  Sun Jun 25 13:13:21 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sun, 25 Jun 2000 17:13:21 GMT
Subject: verifying date values
References: <PcQ45.16280$dF.645882@news1.rdc1.il.home.com> <My355.514$F01.10888@news6-win.server.ntlworld.com>
Message-ID: <R4r55.19690$dF.774126@news1.rdc1.il.home.com>

i needed to accept both a string of characters '12042000' as well as a list
of formatting parameters 'MMDDYYYY' and see if the date when formatted by
the params was a date, and if not why. i ended up just writing a little
function myself.  thanks though.


"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote in message
news:My355.514$F01.10888 at news6-win.server.ntlworld.com...
> Hi,
>
> Marc Lemburg's mxDateTime module (do a search at www.vex.net/parnassus)
> returns a RangeError if you create a date that doesn't exist. I've created
a
> checkDate method below which returns 0 or 1.
>
> mxDateTime is an excellent module if you're going to be doing anything
> complicated with dates or times.
>
> Richard
>
> import DateTime
>
> def checkDate(y,m,d):
>     "year,month,day"
>     try:
>         checkDate=DateTime.DateTime(y,m,d)
>     except DateTime.RangeError:
>         return 0
>     else:
>         return 1
> def testDates(y,m,d):
>     print '%d/%d/%d' %(y,m,d)
>     if checkDate(y,m,d):
>         print 'is ok'
>     else:
>         print 'is not valid'
>
> def main():
>     testDates(2000,3,10)
>     testDates(2000,13,3)
>
> if __name__=='__main__':
>     main()
>
>
> dsavitsk <dsavitsk at e-coli.net> wrote in message
> news:PcQ45.16280$dF.645882 at news1.rdc1.il.home.com...
> > is there a quick way to see if a string is a valid date? something like
> >
> > >>> isDate('5/22/1999')
> > 1
> >
> > thanks
> >
> >
> >
>
>




From shaleh at valinux.com  Mon Jun  5 19:13:30 2000
From: shaleh at valinux.com (Sean 'Shaleh' Perry)
Date: Mon, 05 Jun 2000 16:13:30 -0700 (PDT)
Subject: indirect references
In-Reply-To: <20000605160915.G11994@activestate.com>
Message-ID: <XFMail.20000605161330.shaleh@valinux.com>

> I don't think I am sticking my foot in my mouth: Strings are immutable in
> Python and passed/assigned by value (not by reference, which is what you are
> asking for here). Arrays and dictionaries are passed by reference. Maybe you
> could setup you data to use an array, pass a reference to it and change its
> value, like this:

I knew that a string itself was immutable, I guess I was looking for a
reference object of some kind (a la C pointers, C++/perl references).



From yawyi at gmx.de  Mon Jun  5 07:02:57 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Mon, 5 Jun 2000 13:02:57 +0200
Subject: stripping out unwanted characters in string
References: <393b1f58$0$95987@news.execpc.com>
Message-ID: <393b8ede$0$23346@personalnews.de.uu.net>

Jerry F. Davis <jfdecd at execpc.com> wrote:
>I have a string that has some unwanted characters in it.
>Octal 240 for instance.
>
>I want to actually strip them out of my string.
>
>I have looked at string.replace, but can't seem to figure out how to do
>this.

table = string.maketrans("", "")
sOut = string.translate(sIn, table, "\240")

Johannes



From neilh at scintilla.org  Mon Jun  5 20:01:17 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 06 Jun 2000 00:01:17 GMT
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <5Ys_4.5311$Hz.39657@news-server.bigpond.net.au> <memo.20000605230416.25437B@digitig.compulink.co.uk>
Message-ID: <hbX_4.6274$Hz.46116@news-server.bigpond.net.au>

> ... Of course, you may work in a
> different way and find the opposite, but there seemed to be an opinion in
> this thread that I should be prevented from working in a way /I/ find
> reliable and effective, to no benefit that I can see. By all means have
> editors or static checkers with the /option/ of looking for this, but no
> enforcement in the language, please!

   Its often hard to argue against freedom because we have a natural
tendency to desire liberty. I see the big benefit in having programming
language restrictions is that they facilitate reasoning about what code does
by eliminating the need to consider additional realms of variability.

   "OK, so sacrifice your freedom but leave mine alone!"

   By making the restrictions optional, the code is made harder to
understand because of the additional factor that then goes into
understanding it. Because of the optional strict mode of Visual Basic, you
have to remember when reading code whether it is in a strictly typed module
or not.

   Neil




From aahz at netcom.com  Sat Jun 17 20:10:48 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jun 2000 00:10:48 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com> <39520785.690843718@nntp.interaccess.com>
Message-ID: <8ih428$oen$1@nntp9.atl.mindspring.net>

In article <39520785.690843718 at nntp.interaccess.com>,
Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
>
>The point is not that I can copy and paste with out reindenting.
>The point is that sometimes I have to. let me show a generic example
>
>if a:
>   for b in B:
>      if c(b):
>          block statement.
>
>Assume that you are editing another part and you realise that you
>are about to essentially retype the block statement. So you do
>
>def fun():
>  block statement
>
>if a:
>   for b in B:
>      if c(b):
>           fun()
>you have to reindent the block statement. This is not copy and paste
>reuse. It is cut and paste refactoring, and you have to reindent.

Sure.  But if you're Doing It Right, you'd have to do that for any other
language; Python simply *forces* you to do it.  While it's sometimes a
PITA for debug code, I think the consistent readability is worth it.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From mal at lemburg.com  Tue Jun 13 06:32:44 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 13 Jun 2000 12:32:44 +0200
Subject: ANN: mxCGIPython Project - Version 0.2.3
Message-ID: <39460DCC.FFFE6C94@lemburg.com>

ANNOUNCING:

                     mxCGIPython Version 0.2.3

           An easy to install one-file Python interpreter


WHAT IT IS:

mxCGIPython is a project to provide a complete ready-to-run
Python interpreter in one single executable file. The main
intent for supplying such a single file executable is to 
ease deployment of Python for CGI scripting even when the ISP
is not willing or able to install Python for you.

The mxCGIPython archive contains everything needed to build
the one file interpreter. The web page provides a set of links
to contributed binaries for direct use in addition to setup
instructions to get the binaries up and running at your ISP.

See the documentation page for a list of available binary
packages. Binaries for new platforms are always welcome :-)

WHAT'S NEW ?

Version 0.2.3 adds new versions of a few helpers and fixes
the URLs for the binaries.

WHERE CAN I GET IT ?

The full documentation and instructions for downloading and installing
can be found at:

        http://starship.python.net/~lemburg/mxCGIPython.html

WHAT DOES IT COST ?

It comes with a Python-style license, but is otherwise free for
commercial and non-commercial use.

REFERENCE:

<P><A HREF="http://starship.python.net/~lemburg/mxCGIPython.html>
mxCGIPython 0.2.3</A> - One-file easy install Python interpreter.
(2000-06-13)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From scarblac-spamtrap at pino.selwerd.nl  Wed Jun  7 05:43:27 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Jun 2000 09:43:27 GMT
Subject: Spurious coercing? [was: Re: 'if foo' calls getattr?!?]
References: <Pine.GSO.4.10.10006062045150.1650-100000@sundial> <oqsnuqbomf.fsf@titan.progiciels-bpi.ca>
Message-ID: <slrn8js67e.npl.scarblac-spamtrap@flits104-37.flits.rug.nl>

Fran?ois Pinard wrote in comp.lang.python:
> I saw many emails in which people were surprised by Python behaviour, and I
> interpret the surprise as a sign that predictability suffers.  For example,
> A.__cmp__(B) will not be called for an equality comparison when A "is" B.
> Yet, given the meaning of __cmp__, this sounds absolutely correct to me.
> 
> Also given the meaning of __coerce__, I would consider just equally correct
> to avoid calling A.__coerce__(B) when type(A) "is" type(B).  The spontaneous
> predictable thing, here, is that __coerce__ should *not* be used when it is
> evident it should not.

But when both types are InstanceType, __coerce__ may still be needed, right?


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From bjorn at roguewave.com  Wed Jun 28 20:53:17 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 28 Jun 2000 18:53:17 -0600
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194>
Message-ID: <395A9DFD.1C6C508A@roguewave.com>

Gordon McMillan wrote:
> Sean Blakey wrote:
> >On Wed, Jun 28, 2000 at 09:56:27PM +0000, Huaiyu Zhu wrote:
> [snip]
> >OK, I think I see what you mean.  You are objecting to the way that
> >python searches first for modules in the local directory before
> >searching the normal PYTHONPATH, correct?
> 
> >If you don't want python to search the current directory for imports,
> >there is a very simple solution: put this snippet above all of your
> >other imports: import sys
> >sys.path.remove('.')
> 
> Not a real solution. If module x is part of a package, and does an "import
> y", then the package is searched first. This has nothing to do with  '.'
> being on sys.path. It's called a "relative import", and it means that if
> you have "package.os", then the real "os" is unavailable anywhere in
> "package". Period.
> 
> I think the "relative import" feature should be shot in the head (2 posts
> today), but unfortunately most packages rely heavily on it.

Well, I get the problem too now (bang my head hard enough and I'll
eventually understand ;-)

The relative import feature probably should be "shot in the head", but
since it's in use (and presumably useful), how do we go about loading
the global modules?

The current suggestions are "import magicroot.os" and "import .os". I
don't like either, but would probably prefer the first since it is more
explicit...

-- bjorn



From jar at mminternet.com  Tue Jun 13 14:49:24 2000
From: jar at mminternet.com (jim roush)
Date: Tue, 13 Jun 2000 11:49:24 -0700
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com> <2af14c7e.01880953@usw-ex0104-028.remarq.com>
Message-ID: <ge0dks0erbc0ki9g4pdelolgn17fcmvnon@4ax.com>

On Tue, 13 Jun 2000 05:44:24 -0700, Richard Chamberlain
<richard_chamberlainNOriSPAM at ntlworld.com.invalid> wrote:

>Jim,
>
>It works fine for me on Windows - except in IDLE. So either run
>it from the command prompt or from PythonWin.
>
>In PythonWin you'll get the result back to the interactive
>window, which you may not be expecting.
>

I upgraded to IDLE 0.5 and it works fine.  What is PythonWin and where
do I find it?

Thanks.

jar at mminternet.com
James A Roush


From petersc at stud.ntnu.no  Sat Jun 10 05:34:39 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 10 Jun 2000 11:34:39 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net>
Message-ID: <39420BAF.D86DA2BF@stud.ntnu.no>

Phil Fraering wrote:
> 
> Personally, I simply think that lowercase is more readable than uppercase,
> but I don't really know the reason why.

That's quite easy to explain.

IF YOU HAVE TO READ ALL-UPPERCASE TEXT IT QUICKLY BECOMES BORING
BECAUSE ALL THE UPPERCASE LETTERS FIT INTO RECTANGLES OF THE SAME
HEIGHT AND SO WORDS ARE DIFFICULT TO DISCRIMINATE.

if you write in lower-/mixed-case you see some letters sticking
out of the top or the bottom. Studies have shown that words are
not recognized letter by letter but as a complete pattern.

SHOUTING-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From alan.gauld at gssec.bt.co.uk  Tue Jun 13 09:58:39 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Tue, 13 Jun 2000 14:58:39 +0100
Subject: Tkinter: Entry widget with history/completion
References: <20000613071201.A7987@foxboro.com>
Message-ID: <39463E0F.97B65778@gssec.bt.co.uk>

Hans-Joachim Widmaier wrote:
> I'd like to have an Entry(-like) widget that has a history
...
> find any widget method that gives me the current index of the
> cursor 

> How would you recommend doing it?
>  - By subclassing Entry
>  - By creating a new Pmw MegaWidget

I'd probably go for the former but...

have you considered using a single line Text widget 
instead of the Entry? It does have methods for accessing 
the curent index position.

Alan g.

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


From g_will at cyberus.ca  Thu Jun 22 07:47:24 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: 22 Jun 2000 11:47:24 GMT
Subject: Win32 Serial Port Setup
Message-ID: <01bfdc3f$b19a6800$6642c3d1@Cyberus>

Hi,

I would like to know how to set up the serial port using the win32 package
by Hammond.  I see some stuff in the MS SDK on it but dont see the
equivalent in the win32 package or docs.  I think that I am using build 125
if that makes a difference.

I know there is a Sio module but I would like to do it directly.

A code snip would be useful as I am new to this.

Thanks

Gordon Williams


From nospam at bitbucket.com  Thu Jun 22 18:30:48 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Thu, 22 Jun 2000 15:30:48 -0700
Subject: Access violation when no memory
References: <olD35.82$_%.4481@newsfeed.avtel.net> <slrn8kvpml.flp.perly@typhoon.xnet.com> <o9e45.18$FG1.1680@newsfeed.avtel.net> <395222DD.353DDD8F@inrialpes.fr>
Message-ID: <Vtw45.36$FG1.3244@newsfeed.avtel.net>

Vladimir Marangozov wrote in message <395222DD.353DDD8F at inrialpes.fr>...
>Phil Mayes wrote:
>>
>> No, I understand that - my program is explicitly designed to consume
>> all memory by progressively consuming smaller and smaller blocks,
>> throwing exceptions at each size.
>
>Whatever your task is, I strongly doubt that consuming all the memory
>of your machine is really what you want to do. Not from Python, at least.

No, of course not; I have a 15,000-line program in beta that crashes on
several users' machines, and the sample I gave is the distillation of
weeks of painful debugging.

>The net effect of your program is zeroing the memory the OS is kind enough
>to allocate for Python. It's not really Python's fault in such extreme
>situations. Tim Peters reported (and one could try it easily with a
>malloc(BIG) call) that Linux overbooks it's memory, that is, the malloc
>succeeds, but accessing everything subsequently segfaults because the
>requested mem size is not physically present (RAM + swap included).
>
>So this boils down on how the OS handles such extreme cases. Python
>usually does not trap OS signals, etc. and it cannot handle OS-level
>mem overflows. All it does is trusting C malloc and checking the return
>values for success or failure.

Yes, a page-fault on a committed malloc(BIG) is an OS issue, but that's
not what is happening, which is:
  a) a MemoryError exception is thrown
  b) handling the MemoryError involves getting a traceback object
  c) there's no memory for it, so NULL is pushed
  d) the pop uses its address

>> So I need a ceval.c expert to vet my change.
>
>Forget about it <wink>. Nobody is convinced that checking for NULL the
>top of stack after popping its value would solve any problem. Your example
>crashes on my combo in the builtin dict methods (& elsewhere).

Yes, I am concerned that I have not found all the places that Python
can fail under low-memory conditions.

> This is
>really low-level OS stuff, and Python has not builtin emergency procedures
>for these cases. I don't know why you want to eat all your memory, but
>I'd suggest changing the approach or write a C module that deals with
>the memory issues you're looking for.

In my experience, error-handling is one of the hardest tasks in software
development, and ends up being about 20% of the development effort.
If Python is to be used for widely-distributed software, it has to be
robust and handle such extreme cases gracefully, and I suspect that the
uses to which it has traditionally been put have not fully exercised
some error situations.

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

--
Phil Mayes    pmayes AT olivebr DOT com






From timr at probo.com  Thu Jun  8 00:40:33 2000
From: timr at probo.com (Tim Roberts)
Date: Wed, 07 Jun 2000 21:40:33 -0700
Subject: win python and graphics
References: <393692FB.8610EBCA@dallas.net> <393D7BFD.F50F9932@sightreader.com> <Bah%4.22817$Rq.2324142@pouncer.easynews.com> <0006071724260A.00959@localhost.localdomain>
Message-ID: <gk8ujskos0ic0j1qc27n95lsagvoten7v6@4ax.com>

matt <matt at virtualspectator.com> wrote:

>Hi, I pretty much only use Linux, but have to do some things on windows.  In
>particular I need to take screen shots of the desktop or particular windows.  I
>would like do this using python rather than in borland C++.  Can anyone help out
>here ?

Not in straight python, even with the Win32 extensions.  You can locate
windows and even get a DC, but the necessary GDI calls aren't available.

If you're willing to include a manual step, it might be possible.  The
PrntScrn key on PC keyboards takes a snapshot of the current desktop to the
clipboard, from which many graphics apps can process it.  Alt-PrntScrn
snapshots the topmost window to the clipboard.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From richard_chamberlain at ntlworld.com  Sat Jun  3 03:50:05 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 3 Jun 2000 08:50:05 +0100
Subject: Why is my Tkintering so bad?
References: <XL%Z4.2161$gr4.67106@news1.rdc1.il.home.com>
Message-ID: <rF2_4.235$JO2.3303@news2-win.server.ntlworld.com>

Example 1 doesn't work because when you create and pack in the same
statement the pack method actually returns none.

So in your example cmd is actually None, which obviously hasn't got a bind
method.

try

cmd=Button(root,text='Press')
cmd.pack()

cmd.bind etc.

Richard
dsavitsk <dsavitsk at e-coli.net> wrote in message
news:XL%Z4.2161$gr4.67106 at news1.rdc1.il.home.com...
> example 1 does not work, example 2 does. anybody know why?
> more importantly, i need to read text from a ScrolledText (Pmw) and i
can't
> do it.  The code i am attempting to use (or at least the relevant part) is
> in Example 3.  i am used to things like...
> x = Form.Text1.text (vb) so this is really foreign. essentially there is a
> text box and a button. at startup the text box reads from a text file
(this
> works fine), then when the user pushes the button the contents of the text
> box are written to the output window.  It is this step of connecting code
to
> events to visuals that i am not getting. any suggestions at all will be a
> big help and greatly appreciated.
>
> thanks,
> doug
>
> # Example 1 -------------------------------------
> from Tkinter import *
> root = Tk()
> def buttonAction():
>     print "click"
>
> cmd = Button(root, text="Press").pack(side=TOP)
> cmd.bind('<Button-1>', buttonAction)
> root.mainloop()
>
> # Example 2 --------------------------------------
> from Tkinter import *
> root = Tk()
> def buttonAction():
>     print "click"
>
> cmd = Button(root, text="Press", command=buttonAction).pack(side=TOP)
> root.mainloop()
>
> # Example 3 ---------------------------------------
> from Tkinter import *
> import Pmw
>
> def SaveEdits():
>     x = st.get()
>     print x
>
> def MakeInterface():
>     flt = Frame(root)
>     st = Pmw.ScrolledText(flt, labelpos=N, label_text='Stuff',
> text_wrap='none')
>     st.importfile('TestData.txt')
>     st.pack(fill=BOTH, expand=1, padx=5, pady=5)
>
>     flt.pack(side=TOP, expand=YES, fill=X)
>
>     flb = Frame(fl)
>     cmdSave = Button(flb, width=20, text='Save Edits', borderwidth=3,
> relief=GROOVE, command=SaveEdits).pack(side=TOP, pady=5)
>     flb.pack(side=TOP, expand=YES, fill=X)
>
> root = Tk()
> root.title('window title')
> MakeInterface()
> root.mainloop()
>
>




From newsbunny at noether.freeserve.co.uk  Wed Jun 21 05:07:44 2000
From: newsbunny at noether.freeserve.co.uk (Martin P Holland)
Date: Wed, 21 Jun 2000 10:07:44 +0100
Subject: Is there no inversion of gmtime()?
References: <20000620211050.29059.00000981@ng-da1.aol.com>
Message-ID: <slrn8l11f0.16n.newsbunny@noether.freeserve.co.uk>

On 21 Jun 2000 01:10:50 GMT,
Dparsavand <dparsavand at aol.com> wrote:
>Looking throught the time module, I'm surprised that there doesn't seem
>to be a
>way to go from the known date in UTC to the number of seconds since the epoch.
 
There was a thread about this in this newsgroup around 05/03/1999 and
G. van Rossum posted a readable solution (which does what you'd expect)
which I use in my ftpupdater app
http://www.noether.freeserve.co.uk/kftpupdater/
 
IIRC there were also a faster (but less readable) solution posted by
the F-bot. Either check out deja (does it go back that far?) or snarf
Guido's solution from my app.
 
atb
 
Martin             

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


From bwarsaw at python.org  Thu Jun 15 00:28:05 2000
From: bwarsaw at python.org (Barry A. Warsaw)
Date: Thu, 15 Jun 2000 00:28:05 -0400 (EDT)
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
	<8i8bi2$1s8$1@news.wrc.xerox.com>
	<slrn8kg937.328.hzhu@rocket.knowledgetrack.com>
Message-ID: <14664.23381.540661.482460@anthem.concentric.net>

>>>>> "HZ" == Huaiyu Zhu <hzhu at knowledgetrack.com> writes:

    HZ> If you use emacs:

    HZ> C-x h M-x tabify

    HZ> will tabify the whole buffer.  I always do this when doing
    HZ> copy and paste from news groups, and do an untabify the other
    HZ> way round.  This is because the default tab space is 4 in
    HZ> emacs (which is sane) but 8 everywhere else.

Actually, by default tab-width is 8 in Emacs, as RMS/GOD intended. :)

The answer is basically never to use tab characters in Python source
files.  Use 4 spaces per indentation level.  Two levels == 8 spaces,
not 1 tab character.  python-mode essentially sets all this up for
you.

-Barry



From dan at control.com  Mon Jun 12 14:24:36 2000
From: dan at control.com (Dan L. Pierson)
Date: Mon, 12 Jun 2000 14:24:36 -0400
Subject: Mozilla, XUL and...Python?
References: <39442012.692F02C2@bigpond.com>
Message-ID: <qSdFOReERLcu8vr=OApYGSZtUHjT@4ax.com>

Tim Churches <tchur at bigpond.com> wrote:

> The whole architecture of the Mozilla browser is very impressive,
> especially the ability to create applications using an XML-based
> specification (XUL). The only wrinkle is that the only client-side
> scripting language mentioned is JavaScript. Clearly support for
> JavaScript is essential, but is it possible for other scripting
> langauges, in particular Python, to be embedded in Mozilla? 

This is all from my (mis)understanding of messages on the Zope lists,
so take a grain of salt with it :-)

The bad news: It is evidently very difficult to change Mozilla to use
a langauge other than Javascript as its scripting language.  They'd
like to fix this, but it isn't going to happen soon.

The good news: Mozilla is built around a COM copy called XPCOM.  David
Ascher's work will let you write XPCOM objects in Python (or Perl) and
use XPCOM objects from Python (or Perl).  This means that you could do
almost everything in Python plus a thin Javascript glue layer.

Dan Pierson, Control Technology Corporation
dan at control.com


From steffen.ries at sympatico.ca  Thu Jun 22 07:36:48 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Thu, 22 Jun 2000 11:36:48 GMT
Subject: re.split question
Message-ID: <m3k8fiosky.fsf@gondolin.beleriand>

Hi,

I'm trying use re.split to separate an LDAP Distinguished Name into
its naming components. The naming components are of the form
"attr=value" and they are separated by ",".

The part, which I can't solve is, that "value" may contain the
sequence "\,", which obviously does not separate a naming component.

e.g.
"uid=joe, ou=orgunit, o=some org"
should be split into ["uid=joe", "ou=orgunit", "o=some org"]

"uid=joe, ou=org\\, unit, o=some org"
-> ["uid=joe", "ou=org\\, unit", "o=some org"]

Is there a way to use re.split to achieve this? (I solved it searching
[^\\], and splitting the string in a loop, but I was wondering if it
can be done easier.)

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


From swun at esec.com.au  Fri Jun 23 00:48:13 2000
From: swun at esec.com.au (Sam Wun)
Date: Fri, 23 Jun 2000 14:48:13 +1000
Subject: How to convert a string to hex?
Message-ID: <3952EC0D.D5CF9066@esec.com.au>

Suppose here is:

b = "6b584"

I would like to b to hex. Does anyone know how to do that in python?


Thanks
Sam.




From dparsavand at aol.com  Tue Jun 20 21:10:50 2000
From: dparsavand at aol.com (Dparsavand)
Date: 21 Jun 2000 01:10:50 GMT
Subject: Is there no inversion of gmtime()?
Message-ID: <20000620211050.29059.00000981@ng-da1.aol.com>

Looking throught the time module, I'm surprised that there doesn't seem to be a
way to go from the known date in UTC to the number of seconds since the epoch.
There is an inverse function of localtime() which I don't think is what I want
in order to compare two dates such as:
07 May 2000 03:45:33 -0800
07 May 2000 04:33:33 -0400 
for example.  I was just going to convert these dates to gmtime in seconds
since the epoc. Sure, I could write my own program to do this but I figured
Python would make it easier for me. Am I missing something?

In playing with gmtime(), I was also very suprised that it won't accept
negative arguments - why? 

Thanks, Dara (who definitely preferred the name 2.0 to 3000)


From Gareth.McCaughan at pobox.com  Tue Jun  6 14:22:00 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: 06 Jun 2000 19:22:00 +0100
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <86pupud60n.fsf@g.local>

Richard P. Muller wrote:

> What else should we be doing to make Python *the* programming language
> rather than *a* programming language?

We don't (well, shouldn't) want Python to be *the* programming
language. I would be very sorry to wake up one morning and find
that suddenly there was no more Common Lisp or no more C,
unless Python had somehow mutated into something that could
do all the things they do. (Including running as fast as
they can.) I don't think that's likely to happen any time
soon.

I think Python is becoming steadily more popular for all
sorts of purposes. I don't think using Python leads to
marginalisation. (Especially for those like you who know
10-15 languages, assuming you mean "know well enough to
do real work in".)

What can you do to make Python more popular and more mainstream?
Pick some computing task you want to do moderately often
but that most people do with something other than Python.
If it's not easy to do in Python, make it easy to do in
Python (write a module, or some documentation, or whatever
it takes). If it's easy once you know how but not easy to
find out how, write up how to do it in a way that anyone
can understand. (If it's already just plain easy, you picked
a sub-optimal task.) The aim, in every case, is to make the
learning curve as non-existent as possible. Now put your code,
documentation, and whatever else on a web page, and announce
the fact in c.l.py . Repeat weekly. After a couple of months,
you'll have something that could make a real difference.

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction


From aahz at netcom.com  Mon Jun 26 10:23:01 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jun 2000 14:23:01 GMT
Subject: Passing variables between scripts
References: <8j745m$qqj$1@oslo-nntp.eunet.no>
Message-ID: <8j7p05$6oo$1@slb1.atl.mindspring.net>

In article <8j745m$qqj$1 at oslo-nntp.eunet.no>,
Olav Viken <ovi at autornica.no> wrote:
>
>Can anyone tell me if there is a good way to pass a variable between
>two CGI scripts?

Depends what you're doing.  If the two scripts run in sequence, the
first script should print out a form that submits to the second script,
with the form containing all the necessary data for the second script.
Note that your second script should probably do data validation, because
some screwy user might muck with the form.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From JamesL at Lugoj.Com  Tue Jun  6 00:45:04 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Mon, 05 Jun 2000 21:45:04 -0700
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net>
Message-ID: <393C81D0.36732896@Lugoj.Com>

William Tanksley wrote:
> And for case sensitivity it _just doesn't matter_.  The only people who
> are strongly helped or hurt by case sensitivity appear to be newbies, and
> they're being hurt.

Physically hurt? Emotionally hurt? Psychologically scarred? After thousands
of programmers have learned Python doesn't anyone find it strange that this
issue should suddenly become THE hot issue that prevents Pythonic world
domination? Is this a plot by Perl advocates?

If God had meant for languages to be case insensitive, He would not have
invented both upper AND lower case letters. If it is good enough for Him, it
is good enough for me!

P.S. Actually I'm an agnostic....


From hinsen at cnrs-orleans.fr  Tue Jun 13 06:14:05 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 13 Jun 2000 12:14:05 +0200
Subject: NumPy arrays: how to remove extra axes
References: <slrn8kaov1.1ar.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3u2exdh1u.fsf@chinon.cnrs-orleans.fr>

hzhu at rocket.knowledgetrack.com (Huaiyu Zhu) writes:

> For arrays in Numeric, is there a general way to remove axes of length 1?
> For example I have an array of shape (1,3,1,4) I want to change it to (3,4).

For a specific known shape, just assign a new one:

 array.shape = (3, 4)

For suppressing all axes of length one, no matter where, you can write

 array.shape = Numeric.repeat(array.shape, Numeric.not_equal(array.shape, 1))

Note that this changes the array, it does not create a new one.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From olafb at pvv.org  Fri Jun  9 07:51:24 2000
From: olafb at pvv.org (Olaf Trygve Berglihn)
Date: 09 Jun 2000 13:51:24 +0200
Subject: Swig: typemap for string out parameter
Message-ID: <dn1g0qnm5s3.fsf@verden.pvv.ntnu.no>

I guess some of you swig'ers can answer this easily:

I want to create a typemap for out parameter strings similar to what
swig does for int with the "int *OUTPUT". I know the length of the
string.

The C prototype would look like this:

int getNewString( const char *pInParameter,
                        char *pOutString,
	                int  iOutStringLen );

Thanks!

Olaf

-- 
Olaf Trygve Berglihn <olafb at pvv.org> <olafb at trustix.com> 

                     


From sholden at bellatlantic.net  Tue Jun 27 13:29:37 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Tue, 27 Jun 2000 17:29:37 GMT
Subject: multiple replacements
References: <8ja7ok$6be$1@nnrp1.deja.com>
Message-ID: <3958E473.AB773145@bellatlantic.net>

siva1311 at my-deja.com wrote:
> 
> Hi,
> I am a newbie Python programmer and was wondering the best way to make
> multiple replacements of text strings in a file. I have 10-15 strings to
> replace and right now I am doing this like:
> 
>   source = open(source_file,'r')
>   contents = source.read()
>   source.close()
> 
>   contents1 = string.replace(contents, 'foo1', 'bar1')
>   contents2 = string.replace(contents1, 'foo2', 'bar2')
>   ...
>   contents10 = string.replace(contents9, 'foo10', 'bar10')
> 
>   dest = open(dest_file, 'w')
>   dest.write(contents10)
>   dest.close()
> 
> I know this must be a terrible kludge.
> 
Even so, if you only need to do this infrequently it's probably good
enough.  An engineering approach... :-)

> Using sed I can use a single statement to carry out the replacements:
>     sed "s/foo1/bar1/g
>          s/foo2/bar2/g
>          ...
>          s/foo10/bar10/g" source_file > dest_file
> 
> so I assume there must be a much better method than I am currently using
> in my Python version.
> 
> Any suggestions for improvement to the Python code are appreciated.
> 
> -Brad
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Well, there's cleaning the source up and making it easier to extend,
and there's efficiency: you don't say which is your final goal.

You cold clean the source up a bit with:

   foobar = ( ('foo1', 'bar1'),
              ('foo2', 'bar2'),
              ('fooN', 'barN') )
   source = open(source_file,'r')
   contents = source.read()
   source.close()
 
   for foo, bar in foobar:
	contents = replace(contents, foo, bar)
 
   dest = open(dest_file, 'w')
   dest.write(contents10)
   dest.close()

For speedup, you could write the whole loop as a single statement,
but it will get horrible quickly:

   contents = replace(
                 replace(
                 replace(contents,
                 'fooN', 'barN'),
                 'foo2', 'bar2'),
                 'foo1', 'bar1)

and, of course, ths code is much less easy to maintain.

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


From richard_chamberlain at ntlworld.com  Sat Jun 24 17:15:48 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 24 Jun 2000 22:15:48 +0100
Subject: books????
References: <sla4ago0jev12@corp.supernews.com>
Message-ID: <iy955.968$F01.20298@news6-win.server.ntlworld.com>

There a lots,

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

for a list.

The manning and oreilly books have sample chapters online you could look at.

Richard

SugarKing <sugaking at gis.net> wrote in message
news:sla4ago0jev12 at corp.supernews.com...
> Are there any good beginner books for Python? I'm itching to get my hands
on
> any. Thanks
>
>




From bparsia at email.unc.edu  Tue Jun 13 18:00:50 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Tue, 13 Jun 2000 18:00:50 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au> <394484C0.A5E412D9@san.rr.com> <slrn8ka8kq.1r0.jajs@Pascual.es>
Message-ID: <1ec56t9.1jok9br1setrlaN%bparsia@email.unc.edu>

Juanjo ?lvarez <jajs at NOSPAMretemail.es> wrote:

> Courageous dijo sobre Re: Python Productivity over C++,
> en fecha: Mon, 12 Jun 2000 06:33:05 GMT
> 
> |Python is a simply wonderful prototyping environment, IMO.
> |Now only if it had an optimizing compiler...
> 
> I think every Python fan dream with that... 
> 
> It's really so difficult?

1) Look at Vyper, a Python clone+extensions. I haven't heard much about
it in a while. It looked very promising to me. It intended to do a lot
with whole module/program analysis, espeically wrt to type inferencing.
For compiling to machine code executables, this seems like a very good
approach. Stalin (a scheme compiler) takes this path, plus giving up
some runtime dynamaticity (well, you kinda have to if you're going to
analyize the *whole* program) and it achieves quite amazing code.
Outpaced C for certain numeric code, where reasonably idiomatic C and
Scheme were used, and even when someone skanky C was thrown in, IIRC.

2) Common Lisp style compiler. Big grab bag of tricks there, including
optional type declarations. Probably not quite the right match for
Python.

3) Smalltalk/Self style VMs. Improving method/function dispatch alone
can do a hell of a lot, as simple hand optimization shows (e.g., binding
a function to a local variable in an inner loop). Self style is more
ambitious, and probably not worth the implementation effort.

My understanding is the CPython VM is designed for simplicity,
portability, embedability, and easy of using C for extendability. Not
speed. My guess is that one of these has to be reduced for speed to go
up. Acutally, more than one, since simplicity will always be compromised
as you at optimization, and if you add trying to preserve all the
others, you get more complicated quickly.

Vyper shows, in a preliminary way (as does JPython), that there may be
serious gains made by moving to a different, higher level implementation
language. There are indeed some whispers of YA implementation...

...but if I told you about it, you'd tell me back, and then you'd have
to kill me, and my life insurance isn't paid up.

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From majuhe at kanto.jyu.fi  Mon Jun 26 05:39:33 2000
From: majuhe at kanto.jyu.fi (Matti Juhani Heinonen)
Date: 26 Jun 2000 12:39:33 +0300
Subject: Tkinter cannot find tcl.init
Message-ID: <8j78cl$7l8@kanto.jyu.fi>

Hi all,

I am running NT 4.0 SP6a, Python 1.5.2 and Tcl/Tk 8.0.5 on several
machines which are identical in configuration (same users, same software,
same user rights etc.). On almost all the machines Tkinter runs ok, 
but on some of them I get an error:


<quoting>

Traceback (innermost last):
  File "d2hGUI.py", line 1342, in ?
    root = Tkinter.Tk()
  File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} ./lib/tcl8.0 ./lib/tcl8.0 ./library ./library ./tcl8.0/library ./tcl8.0/l
ibrary



This probably means that Tcl wasn't installed properly.

</quoting>


 
When logged in as Administrator everything works out fine. When 
logged on as any other user I get the above error, although on the
other machines everything is ok for that user. This suggests that 
the problem could be user rights related, but the all machines should be 
similarly configured (at least I could not find any differences).

There are no excess tk*.dll or tcl*.dll floating around. Setting 
TCL_LIBRARY and TK_LIBRARY enviroment variables did not help. 
Uninstalling & re-installing the whole Python package did not help.
When I search tcl from the registry of the machine I find the same 
settings as on those machines that work.

Copying C:\Program Files\Tcl\lib\* under C:\Program Files\Python\Lib
did not help, but copying C:\Program Files\Tcl\lib\* to 
<current working directory of the user>\lib\* did work!

I am getting utterly confused. Anybody have any clue what is going on
or how this could be fixed?

Matti
-- 
--
majuhe at math.jyu.fi <A HREF="http://www.math.jyu.fi/~majuhe/">minun sivuni</A>
?l? h?t?ile! yIlImQo ! Don't panic!


From jbauer at rubic.com  Mon Jun  5 11:43:07 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 05 Jun 2000 10:43:07 -0500
Subject: How to prevent a web client from timing out
References: <8hgfn8$3uq$1@nnrp1.deja.com>
Message-ID: <393BCA8B.FC062262@rubic.com>

Alex,

One solution would be to run your number-crunching
in a separate process (e.g. xmlrpc server) that
your cgi script accesses.

Jeff Bauer
Rubicon Research

aiglauer wrote:
> I have written a number-crunching app in python. which is being
> accessed on our intranet.  All pages generated on-the-fly in python.
> Some of the calculations can take a while though (calc time anything
> from 5secs to 30 minutes) and I get the client timing out and the
> server will then kill the calculation process.
> 
> Can I rig it so that the calculation process will still complete and
> write its output to a file so that I can use it later?  (Writing to a
> file is not a prob - preventing the thing getting slapped is).



From pablo.prieto at dulcesol.es  Wed Jun 21 05:08:01 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Wed, 21 Jun 2000 10:08:01 +0100
Subject: remote execution
References: <394FF019.CC0A13BE@visionart.com>
Message-ID: <395085F1.94F33B15@dulcesol.es>

Pete Shinners escribi?:
> 
> i pair of related questions here.
> 
> i'm interested in finding a way to run remote commands
> on different IRIX/Linux machines on our LAN. Usually
> i will be running commands that run between 5 minutes
> to several hours. (ie, not quick little routines)

> actually the ability to run remote python code would
> not be that beneficial, since 99.9% of the time it would
> simply be something like os.system(command)
> 
> thanks!

Just an idea, I don't know if it fits to you, but try a look at pyro
http://www.xs4all.nl/~irmen/python.html


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



From anonymous at abc.com  Sat Jun 17 12:38:58 2000
From: anonymous at abc.com (anonymous at abc.com)
Date: 17 Jun 2000 16:38:58 GMT
Subject: Euro2000 football divination service and  more.....
Message-ID: <8ig9j2$22dg$348@news.cz.js.cn>

    China five thousand year culture,be born I ching,the book of change,it be turn record china physics,mathematics,relativity and more universe algebraical basic knowledge,and provide in line with standard divination method.

   let your intuition  talking,chinese i-ching divination anything.

   About your future:
   I provided service: divination Fortune/Stock Exchange/Contract Signing/Partnership/Official/Sport Race/ Lawsuit / Weather/ Calamity   and more..... 

    http://iching.126.com

    Regards,
       lei




From jbauer at rubic.com  Mon Jun  5 17:06:40 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 05 Jun 2000 16:06:40 -0500
Subject: How to prevent a web client from timing out
References: <8hgfn8$3uq$1@nnrp1.deja.com> <393BCA8B.FC062262@rubic.com> <8hh066$vor$10$1@news.t-online.com>
Message-ID: <393C1660.6A22F08D@rubic.com>

Alexis Iglauer wrote:
> The job is different every time, so a cron solution 
> wouldn't work.  The problem is that when the client 
> times out the server kills the python process doing 
> the number crunching

I didn't suggest cron.  (Someone else may have.)

You can basically have a separate process already 
running, waiting for a number-crunching request.
It is independent of the client, so if the client
times out, nothing happens to the long-running 
process.

Assume the long-running process (LRP) is threaded,
and it can report back the status of the operation 
when your client requests it.  

Scenario:

1. Client (cgi script) submits a number-crunching 
request to LRP.

2. LRP responds back to client, "I'm working
on it.  Press HERE to check on status."

3. User presses HERE button and another cgi script
calls the LRP, which replies, "Gimme a minute,
will ya?  Jeez."

4. User presses HERE button again.  Cgi script
calls the LRP which has now finished the calculation.
It responds, "The answer is 42."

> So how do I stop apache from killing the process?  

Because Apache didn't launch the process, Apache
won't kill the process.  You do the LRP independent
of Apache.

> How well does xmlrpc work in the real world?

I don't know about the "real" world, but I've
created an artificial world where it works pretty
well.  

But for the technique described above, you don't 
*have* to use xmlrpc. Any similar technique (e.g. 
DOPY, Fnorb, ILU) should also prove satisfactory.

Jeff Bauer
Rubicon Research



From clive at pisquaredoversix.force9.co.uk  Fri Jun  9 06:38:10 2000
From: clive at pisquaredoversix.force9.co.uk (Clive Tooth)
Date: Fri, 9 Jun 2000 11:38:10 +0100
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca>
Message-ID: <8hqhhi$396$1@mail.pl.unisys.com>

Robert Israel wrote in message <8houlu$j9s$1 at nntp.itservices.ubc.ca>...

>[...]
>There are two different algorithms here, for two different problems.
>
>The BBP algorithm can find a hexadecimal (or more generally, base
>2^k) digit of pi, faster than any known algorithm that would find
>all the digits up to this one.  This is the exciting one.

Although BBP do mention that "This algorithm is, by a factor of
log(log(log(n))), asymptotically slower than the fastest known algorithms
for generating the n-th digit by generating all of the first n digits of
log(2) or pi."


>Plouffe has another algorithm that can find a digit in an arbitrary
>base.  This takes C*n^3*log(n)^3 time for the n'th digit, so it is not
>faster than other methods that would compute the first n digits, but it
>does take very little memory.  This one has "a more theoretical rather
>than practical interest".
>
>Whether there is an algorithm for the n'th decimal digit that performs
>as well as BBP is still an open question.

An O(n^2) algorithm is described at
http://www.stud.enst.fr/~bellard/pi/pi_n2/pi_n2.html

--
Clive Tooth
http://www.pisquaredoversix.force9.co.uk/
End of document





From gmcm at hypernet.com  Wed Jun 14 17:10:30 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Jun 2000 21:10:30 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de>
Message-ID: <8F53A14B0gmcmhypernetcom@199.171.54.194>

Thomas Thiele wrote: 

>Aahz Maruch wrote:
>
>> Suppose you needed a file object to connect to some chunk of memory. 
>> In other languages, you would need to inherit from the Python file
>> object in order to ensure that random functions that expect file
>> objects would still work.  In Python, that's completely unnecessary. 
>> It's probably a lot simpler to create your own class, complete with
>> open(), close(), read(), and write() methods -- and because you've
>> maintained *interface* consistency, you can use your new class
>> *ANYWHERE* a file object would go.
>> --

>But there is no mechanism to ensure that the behaviour of your
>memfileobjectsis like a fileobject. For instance if you don't note that
>you forgot to write the write()-function.
>In a worst case scenario it works for many months until a user calls a
>function in which
>the write() -function is needed.

And there's no mechanism to ensure that you have the behavior right in C++, 
even if the compiler accepted your interface.

Except good test cases, that is.

- Gordon



From moshez at math.huji.ac.il  Wed Jun 14 16:43:19 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 14 Jun 2000 23:43:19 +0300 (IDT)
Subject: SV: Python Productivity over C++
In-Reply-To: <3947C453.CF24E30F@estinc.com>
Message-ID: <Pine.GSO.4.10.10006142339350.3119-100000@sundial>

On Wed, 14 Jun 2000, Eric Lee Green wrote:

> Michal Vitecek wrote:
> >  this is highly debatable - properly written C/C++ source is pretty well
> >  readable and self-documenting. similarly, badly written python source
> >  makes it as unreadable as in other languages - there's not a point in
> >  trying to prevent people from writing (un)readable source code.
> 
> It's harder to write unreadable Python source because you can't write code
> like, hmm,
> 
> opendir(FOO,\"%s\");@files=readdir(FOO);foreach (@files) { if
> (/^[^\\.].*\\.txt$/) { print \"%s/$_\\n\";};};closedir(FOO);

import os,re
for _ in os.path.listdir("%s"): if re.match("^[^.].*\.txt$': print(
"%%s/%(_)s" % vars())

writing-obfuscated-python-isn't-that-easy-ly y'rs, Z.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From sdm7g at virginia.edu  Wed Jun 28 19:33:27 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Wed, 28 Jun 2000 19:33:27 -0400 (EDT)
Subject: Darwin (Not Python, yet...) [OT]
In-Reply-To: <Pine.A32.3.90.1000628114111.14339A-100000@elvis.med.Virginia.EDU>
Message-ID: <Pine.A32.3.90.1000628191421.13732B-100000@elvis.med.Virginia.EDU>


[1] un unzip command for Darwin is available in the zlib distribution
 in the contrib/minizip directory. Downloaded the sources from
 the Darwin web page ... got some build errors after configure and make ...
 got a more recent version via CVS -- that built with no problems -- 
 had to add "-DHAVE_UNISTD_H -DUSE-MMAP -Dunix" to the makefile CFLAGS
 to compile contrib/minizip (still got an error, but after building 
 miniunz, which was the target I needed. 

[2] unziped the MacOSX python.1.5.2 -- import test.testall seemed to work. 
    'make test' and 'make install' successful. 


[3] Need to compare this distribution with the original  -- supposedly
  only the Mmakefiles have been changed -- and look at the shared library
  build problems. 


Tcl (but not Tk) and Perl already come installed with Darwin -- it would
be nice to get Python packaged in there! 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---

"There's a new programming language called C+++. It's 50% better than C++, 
         but every time you compile, your modem hangs up."





From robin at jessikat.fsnet.co.uk  Tue Jun  6 07:04:07 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 6 Jun 2000 12:04:07 +0100
Subject: Nympy for win
References: <Pine.LNX.4.21.0006061109170.30720-100000@chelin.chtf.stuba.sk>
Message-ID: <fAdMjhAnqNP5Ew3P@jessikat.demon.co.uk>

In article <Pine.LNX.4.21.0006061109170.30720-100000 at chelin.chtf.stuba.s
k>, radoslav hodnicak <hodnicak at bigfoot.com> writes
>Hello.
>
>I would like to use the Numpy package with my windoze/python1.5.2 system,
>but it needs distutils and distutils need m$ visual c++ which I don't own.
>Is here anybody who has it compiled and could send me the
>binaries? Thanks.
>
>Regards, radoslav
>
>ps: I don't know if any distutils-developer reads this, but how about
>adding support for free c compilers for win, like lcc-win32?
>(http://www.cs.virginia.edu/~lcc-win32/)
>
I have built the latest CVS, but it still has some test failures. At one
point the numpy developers said they would provide a precompiled build
for win, but I guess it needs a volunteer or somesuch.
-- 
Robin Becker


From matthias.huening at univie.ac.at  Mon Jun 26 09:51:37 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 26 Jun 2000 15:51:37 +0200
Subject: SMTP with Python 1.6
References: <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com>
Message-ID: <8j7n8d$4u64$1@www.univie.ac.at>

> #!c:\python\python.exe
>
> import cgi, smtplib
> s=smtplib.SMTP("localhost")

'Localhost' will not work.
You have to make this the address of a valid SMTP-server.

> s.quit;
>

Should be s.quit()


Hope this helps.

Matthias

--
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/dt/






From dsavitsk at e-coli.net  Tue Jun  6 02:56:15 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Tue, 06 Jun 2000 06:56:15 GMT
Subject: Local Databases with Python
References: <iT0%4.36661$WS3.441157@typhoon.we.rr.com>
Message-ID: <jg1%4.7291$gr4.288442@news1.rdc1.il.home.com>

I use tkinter for the interface and ms access for the db to run on win2000.
there is a tutorial at the python site on using both DAO and ADO from python
to connect to access (also in mhammond's book with a typo or two). it is
quite easy.
doug

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:iT0%4.36661$WS3.441157 at typhoon.we.rr.com...
>     I'm writing my current project, i'm writting a program primarily in
> Python, but am embedding it in C++ for a couple reasons. Primarily, I want
> the native, fast and flexible GUI that I can get out of using a RAD tool
> like C++Builder, but also, I am not aware of any easyish ways to have a
> local database with Python, so will provide the data-access needed via BDE
> or some other such tool in C++Builder.
>
>     A majority of my actual functionality will be done in Python, and i'm
> wondering if I really need C++Builder to get the database functionality I
> want. Being Cross-platform doesn't concern me at the moment -- but
> working -good- and -prettily- (heh) on Win98 is very important to me. Most
> of the database interfaces I saw on the Vaults, deal with various
> Client/Server SQL database engines that are available... but I don't need
> that. There will only ever be one connection -- my program -- to this one
> database.
>
>     Are there any options for a local database which is accessible via
> Python? Gadfly seems like one, except the idea of storing it all in memory
> terrifies me. There may be a LOT of information in this database as time
> goes on :)
>
> Thanks,
>
> --S
>
>




From rhicks at rma.edu  Wed Jun 21 08:33:46 2000
From: rhicks at rma.edu (Robert)
Date: Wed, 21 Jun 2000 12:33:46 GMT
Subject: Boa stalled? (GUI editor for wxPython)
References: <8iqbv4$4p$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <KC245.51$RB1.10531@news>

All my queries on the subject have been answered with "it is being
updated"...

"Bernhard Reiter" <breiter at usf.Uni-Osnabrueck.DE> wrote in message
news:8iqbv4$4p$1 at newsserver.rrzn.uni-hannover.de...
> Is Boa development stalled?
> It looks like nobody made changes in the last three month.
>
> The program still needs developers. It is in competition to the nice
> glade and libglade thing and not mature enough yet.
> Why is nobody picking it up?
>
> (Just to bring this to your attention.)
>
> http://boa-constructor.sourceforge.net/
>
> --
> Professional Service around Free Software                (intevation.net)
> The FreeGIS Project             (freegis.org)
> Association for a Free Informational Infrastructure            (ffii.org)




From invalid.address at 127.0.0.1  Thu Jun  1 13:02:33 2000
From: invalid.address at 127.0.0.1 (David)
Date: Thu, 01 Jun 2000 17:02:33 GMT
Subject: ANNOUNCEMENT: PythonWorks 1.0 now available (may 31, 2000)
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com> <39356936.7115883@news.telus.net>
Message-ID: <39369663.495889@news.telus.net>

On Wed, 31 May 2000 19:36:30 GMT, invalid.address at 127.0.0.1 (David) wrote:

>Your CSS is nasty.  Fonts are displaying far too small.  I suggest you
>validate against Opera (version 3 or 4) or the latest Mozilla, both of
>which are extremely compliant (quite unlike, say, MSIE or Netscape).  You
>may also wish to enlarge your browser-detecting Javascript, to recognize
>that there is more than just MSIE out there.

Lemme apologise for being so blunt/nasty, and for posting this here instead
of sending it to the webmaster!  I just wasn't thinking--I'd read the URL
on Usenet and never made the mental transition to using Web EMail for the
website comments...

To clarify a bit further, though, "far too small" is that the text is about
four pixels high.  It's *really* uncommon for me to encounter unusable web
pages, so it was a bit of a shocker that a professional Python group could
create one that is hostile to everything but MSIE... quite unexepected.

Neat looking software, though.



From m.faassen at vet.uu.nl  Wed Jun 21 07:45:56 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 21 Jun 2000 11:45:56 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl> <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu> <20000620143255.F26436@xs4all.nl>
Message-ID: <8iq9tk$n8q$1@newshost.accu.uu.nl>

Thomas Wouters <thomas at xs4all.net> wrote:
[function method caching]

> I know next to nothing about Smalltalk, let alone how it optimizes, but I
> can tell you why it would be complicated in Python ;) Lets take a simple
> example:

> def spam(x, y):
>     if x in range(1, y):
>         return 1
>     if x in range(y, 1):
>         return 1
>     return 0

> Might not seem very useful, but it's just an example ;) This can easily be
> optimized, right ? We lookup 'range', it's a builtin, we store the address,
> and call it two times. Well, no. We dont *know* range a is a builtin, it
> could be a global that points elsewhere. If the 'spam' function contained a
> 'from module import *' or 'exec' statement, it could even be a local.

> So, we have to check if range is still the same function. Which means we
> have to look it up and compare identity -- and we've already lost any
> advantage the cache could give us. Or we could place a marker on the
> __builtin__ module, and invalidate the cached entry of 'range' whenver
> 'range' is changed. But then we also have to add a marker to the global
> namespace, because something can *add* a range there.

You could do some analysis of local and global namespace, and see 
that it is possible that 'range' is altered. If such a possibility
exists at all (assignment to the name, or by from import), then range
is simply never cached at all.

If 'from module import *' is used, the user deserves what they
asked for and *nothing* in that module is cached. I'm not sure
how 'exec' can be used to screw things up, but let's say we can apply
the same rule to that. Though that is possibly more tricky..

Something along the lines of this approach could work, I think? The
approach would be to forbid the caching of certain names if it's
detected certain things could happen to them that would make them
uncachable.

> And don't think it can't happen:
[snip things with range() that could happen but almost never *actually*
happen]

[__getattr__]

If __getattr__ is defined in a class, we might still be able to cache
the methods that do exist. Anything that exists only due to
__getattr__ isn't cached, obviously.

> Also, we need a way to invalidate the cached entry of egg.spam if any of egg or
> any of its parent up to the one defining 'spam', have something done to
> their 'spam' attribute.

If the possibility exist that a method is changed, don't cache that
method. Of course setattr() will be a major problem; very hard to determine
whether setattr() won't affect some class at some point. And you won't
know what class, or what name. Hmm.

> And that means than whenever you change an attribute
> on a class or instance, you need to traverse the entire tree of subclasses
> and instances (up to one that 'masks' the change) looking for cached entries
> and invalidating them.

Far cheaper would be simply to invalidate the entire cache in such
a case. That is, if class attributes are changed. I'm not sure why
you'd need to invalidate things when you change something on an instance,
though? (except the thing that's changed itself)

> And of course, if you do something like:

>   egg.ham.viking.spam.plate.drop()

> you have to consider *each lookup* cached ;-P I don't think lookups are
> *that* expensive. I'm not sure what the penalty for a failed lookup is, I
> seem to recall something about a successful lookup being relatively cheap,
> compared to a failed one, and a 2nd lookup is not likely to fail ;)

I don't seem to understand this bit. What does it mean, considering
each lookup cached? Could you explain?

[snip]
>> Hmm. Ok. I think. Maybe. Why it has to be do each time one accesses the
>> *instance*'s method I find a tad confusing. I mean ok, you change the
>> class def, you'll want to update the instances. But surely it's ok to
>> make class updates more expensive than method access?

> Yes, but this works, and caching & invalidation doesn't :-) It would be no
> little ammount of code to add it, even if you discard the ideas about
> invalidation (say, if you add the new code under a -OOO option, and tell
> people not to f*** around with base classes after instances have been
> created, which I doubt Guido or any of the other core developers would
> allow.)

I think you need the ideas of invalidation. I think you also need the
idea of "don't ever cache this" (or at least "don't cache this until I
tell you to"). I think that approach can help with some of the tricky bits.

Regards,

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


From fiona at sitegnome.com  Thu Jun  1 11:03:43 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 1 Jun 2000 15:03:43 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 1st, 2000
Message-ID: <20000601150343.32494.qmail@synop.com>

Greetings,

Only a couple of entries into http://python.faqts.com tonight.

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
I'd like to create a subclass of list that loads its contents from a file when you instantiate it. What do I subclass to do this?
http://www.faqts.com/knowledge-base/view.phtml/aid/3424
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich

You can't directly subclass types, like lists and integers. This is 
commonly viewed as a flaw in Python, maybe this will be changed once.

However, you can subclass UserList.UserList, a Python wrapper around a 
list (see Lib/UserList.py for its source).


-------------------------------------------------------------
(win32api) how to enumerate processes?
http://www.faqts.com/knowledge-base/view.phtml/aid/3428
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond

You need to use the performance monitor functions - check out
win32pdhutil.py for a demo.


-------------------------------------------------------------
What's the event name for catching mouse movement without a button pressed?
http://www.faqts.com/knowledge-base/view.phtml/aid/3427
-------------------------------------------------------------
Fiona Czuczman
John Grayson

>From John Grayson's book "Python and Tkinter Programming"

I think Example_6_2.py demostrates this, and the event is also noted on 
page 618.

However, this snippet tracks the motion of the mouse without a button 
down...

from Tkinter import *

root = Tk()

def motion(event):
    print 'Mouse: x=%d, y=%d' % (event.x, event.y)

frame = Frame(root, width=250, height=250)
frame.bind('<Motion>', motion)
frame.pack()

root.mainloop()






From mirko at picard.inka.de  Thu Jun 22 22:27:25 2000
From: mirko at picard.inka.de (Mirko Zeibig)
Date: 23 Jun 2000 02:27:25 GMT
Subject: Cold Fusion and IBM Websphere using Python as CGI ?
References: <00ad01bfdb88$1a5e8320$8235a8c0@pc-p2457.patni.com> <slrn8l285i.7mp.andyl@azaal.dircon.co.uk>
Message-ID: <slrn8l5imn.afk.mirko@localhost.localdomain>

Andy Leighton <andyl at azaal.dircon.co.uk> wrote:
>On Wed, 21 Jun 2000 19:23:39 +0530, Anil Singh <anil.singh at patni.com> wrote:
>>whether Python Scripts=20
>>are compatible to Cold Fusion Server 4.0 & IBM IBM Web Server. If it =
>>require changes=20
>>what will be the nature of changes ?
>
>I am not sure what you mean by IBM
>Web Server 1.0 so I cannot comment.
Hello Anil,
IBM HTTPServer 1.3.6 is just a customized version of Apache (Webinterface for
editing httpd.conf, own 56-bit-ssl-version):

--- snip rpm-info ---
Name        : IBM_HTTP_Server              Relocations: (not relocateable)
Version     : 1.3.6                             Vendor: International
Business Machines Corporation ( IBM )
Release     : 1                             Build Date: Don 29 Jul 1999
21:59:11 CEST
Install date: Fre 21 Apr 2000 02:13:25 CEST      Build Host:
calderabuild.raleigh.ibm.com
Group       : Networking/Daemons            Source RPM:
IBM_HTTP_Server-1.3.6-1.src.rpm
Size        : 3724906                          License: (C) Copyright
International Business Machines Corp. 1995, 1996, 1997, 1998, 1999. All
rights reserved.
Packager    : Shane Owenby <shane at raleigh.ibm.com>
URL         : http://www.software.ibm.com/webservers/httpservers/
Summary     : IBM HTTP Server Webserver
Description :
IBM HTTP Server powered by Apache is based on the Apache HTTP
Server, which is the most popular HTTP server on the Web. IBM has
embraced and enhanced the server. For example, IBM has added SSL for secure
transactions and offers full support, as it would with any other IBM
product.

IBM supplies the Apache-powered HTTP server with products that run
on AIX, Solaris, Linux, and Windows NT. Initially, you'll find the IBM HTTP
Server powered by Apache in IBM WebSphere Application Server V2.0
(Standard Edition and Advanced Edition).

--- snap ---

So installing your python-scripts to /opt/IBMHTTPServer/cgi-bin should work.
Don't know wether this is true for Version 1.0 as well ...

Regards
Mirko


From akuchlin at mems-exchange.org  Tue Jun  6 17:34:20 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 06 Jun 2000 17:34:20 -0400
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net> <393C81D0.36732896@Lugoj.Com> <slrn8jql5v.6d4.wtanksle@dolphin.openprojects.net>
Message-ID: <3dya4itrxf.fsf@amarok.cnri.reston.va.us>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:
> I'm not the one that's claiming that removing case-sensitivity will ruin
> Python.  My claim is very simple: first, it's a minor change for us to
> make, and second, it helps newbies.

Like Aahz, I don't think the change is that minor.  I'm also doubtful
of the "helps newbies" claim; unless there have been some other
citations in this thread that I missed, we have exactly one study
using exactly one programming environment that reported CS
(case-sensitivity) was a problem, and even that report says "most
[users] continued to type case-incorrect tokens in their programs for
a short period", which implies that they did eventually stop.  The
evidence is hardly overwhelming, and basically purely anecdotal.

IMHO, if GvR was inventing a completely new language for teaching,
then it would be worthwhile to debate whether it should be CS or CI.
But he's not; Python is an implemented language with a sizable and
growing user base that's used to CS, and a number of published books
and other materials that will be invalidated by switching to CI. 

CI buys nothing for me, because practically none of my bugs stem from
case errors, and because most other systems I use are CS.  (Good thing
I don't use SQL much.)  CI will cost me, though, because I'll need to
convert existing code and learn new habits; given that it'll probably
be straightforward to restore case-sensitivity to the Python3000 code
base, it's probably lower cost to just use a patched version of the
interpreter than to rewrite everything.  I suspect many experienced
users will agree, judging from this thread.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
All this narcissism is rather boring, isn't it?
  -- Philip, in Peter Greenaway's _8 1/2 Women_ (1999)






From ullrich at math.okstate.edu  Tue Jun 13 10:35:18 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 13 Jun 2000 14:35:18 GMT
Subject: No strptime in Windows?
References: <000901bfd3de$ee54d280$27a2143f@tim>
Message-ID: <01bfd546$58159660$21a28ad1@daves-dell>


Tim Peters <tim_one at email.msn.com> wrote in article
<000901bfd3de$ee54d280$27a2143f at tim>...
> [David Ullrich]
> > ...
[...]
> 
> "The platform" is the combination of hardware, compiler, operating
system, C
> library, and (less so) Python build-time configuration options.
> 
[...] 
> > ...
> > exactly what is it that's missing strptime, so that my Python
> > doesn't have one available to wrap?
> 
> OK, but the answer isn't going to help you <wink>:  msvcrt.dll (that's
the
> MicroSoft Visual C RunTime DLL).

	Thanks. No, it doesn't help me p my strtime but I'd already done
that - it does help me understand what the heck various things mean.
(I guess what it comes to (roughly) is that in this case there _is_
such a thing as "Windows C", at least in the present context.)

> >     (When I asked the question I was assuming that
> > most of the time.this/that functions were done "by
> > hand" in the Windows version of Python.)
> 
> They're not -- most of them are standard ANSI C library functions (for
which
> see the C standard), and "even Microsoft" <wink> takes a stab at
supporting
> those.  Since MS didn't see fit to support strptime (which is not std
ANSI
> C), the only way it will show up in an MS Windows version of Python is if
> someone contributes implementation code for it using straight C.

	It's not very hard to do in Python; surely it wouldn't be much harder
in C (nudge nudge). Luckily I obviously don't know any C myself...

	(Well, it's not hard except for the Daylight Savings Time part - having
no idea how to do that I didn't.)

> note-that-strptime-is-missing-under-some-unices-too-ly y'rs  - tim

	Why would that matter? (haha)

DU


From thomas at xs4all.net  Tue Jun 13 18:40:16 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 14 Jun 2000 00:40:16 +0200
Subject: SV: Python Productivity over C++
In-Reply-To: <85zoopfaxb.fsf@one.net.au>; from neurocrat@one.net.au on Wed, Jun 14, 2000 at 08:55:44AM +1000
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <ur9a1mmdk.fsf@ctwd0143.fitlinxx.com> <85zoopfaxb.fsf@one.net.au>
Message-ID: <20000614004015.Y20700@xs4all.nl>

On Wed, Jun 14, 2000 at 08:55:44AM +1000, Neurocrat wrote:

> for (int i = 0; i < 10; ++i);
>         cout << "Hello" << endl;
> 
> This, of course, prints "Hello" only once, not 10 times as we
> intended. In a toy example it's easy to spot the offending ; at the
> end of the loop construct, but is there a C or C++ programmer alive
> who hasn't spent hours trying to track down insidious problems caused
> by that single legal semicolon? I know I have.
> 
> If I made the corresponding mistake in Python:
>         for x in range(10)      # forgot the :
>                 print "hello"
> I'd know it straight away.

This reminds me of a funny piece of code I saw not so long ago. A year or so
ago I inherited a codebase for a MUX, a Multi User Experience (similar to a
MUSH, which is similar to a MUD, a Multi User Dungeon ;) This particular MUX
is the engine for an online game of Battletech. It's also code that has been
around for a long, long time, and has been hacked by literally dozens of
people, either directly or indirectly.

A couple of years back, someone hacked in a new feature. Instead of always
displaying the same 'welcome' message before a login, the MUX now chose one
at random from a directory of such files. While I was doing a major cleanup
of the whole MUX, a couple of months back (and fixing a *lot* of bugs, by
the way) I came into the vicinity of that piece of code because of an unused
variable, and this is what I saw:

#if 0
  /* Done this way IF things worked right */
  while (1)
    {
      if (!(de = readdir(d))) break;
      if (de->d_name[0] == '.');
        continue;
      if (!strstr(de->d_name, ".txt"))
        continue;
      if (*cnt == max) continue;
      sprintf(buf, "%s/%s", dir, de->d_name);
      fcache_read(&(foo[*cnt].fileblock), buf);
      (*cnt)++;
    }
  closedir(d);
#else
  sprintf(buf, "perl -e 'opendir(FOO,\"%s\");@files=readdir(FOO);foreach (@files) { if (/^[^\\.].*\\.txt$/) { print \"%s/$_\\n\";}; };closedir(FOO);'", dir, dir);
  f = popen(buf, "r");
  while (!feof(f))

	[ ... code to read the pipe ... ]

#endif

I kid you not, this feature, which had been in for a couple of years,
running between 2 and oh, 5, public MUXes and probably lots more private
ones, at any one time, started *perl* to read a directory contents. And why?
Because the code that should've worked, using opendir() and all, didn't, and
the author itself (*not* one of those other thirty-odd hackers) couldn't
figure out why it didn't work.

Of course, none of the people who could have written that dare admit it now
;) so the real doofus will remain anonymous forever. I have to admit, too,
though, that I wouldn't have caught it that fast if it wasn't for 'indent'.
But then, that's why there are tools like that ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From Gordon.McLeod at wolfson.co.uk  Fri Jun 16 12:28:42 2000
From: Gordon.McLeod at wolfson.co.uk (Gordon McLeod)
Date: Fri, 16 Jun 2000 17:28:42 +0100
Subject: backslash in regular expression problem
Message-ID: <394A55BA.69269793@wolfson.co.uk>

Apologies if this has been asked before -  I'm a newbie. I am trying get
rid of backslashes from a string using re.sub in python 1.5.2:

import re
string="\END\\"
x=re.sub('[\]','',string)
print x

and expect x to be 'END' but I get:

Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "C:\Program Files\Python\Lib\re.py", line 47, in sub
    pattern = _cachecompile(pattern)
  File "C:\Program Files\Python\Lib\re.py", line 33, in _cachecompile
    value = compile(pattern, flags)
  File "C:\Program Files\Python\Lib\re.py", line 79, in compile
    code=pcre_compile(pattern, flags, groupindex)
pcre.error: ('missing terminating ] for character class', 3)

I am trying to follow the advice of the Python Library Reference Manual,
section 4.3 which says:

     However, if you want to include a literal backslash in a
     regular expression represented as a string literal, you have
     to quadruple it or enclose it in a singleton character class.
     E.g. to extract L A T E X `\section{. . . }' headers from a
     document, you can use this pattern: '[\]section{\(.*\)}'.

It looks like Python sees the [\]  as [\\]. [\_] doesn't crash but
doesn't sub either.  Can anyone help?

Thanks,

Gordon McLeod,



From hschroeder at gmx.net  Fri Jun 16 04:11:47 2000
From: hschroeder at gmx.net (Henning Schroeder)
Date: Fri, 16 Jun 2000 10:11:47 +0200
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru> <8ibqbk$o8j$1@agate.berkeley.edu>
Message-ID: <3gnci8.vt.ln@odin.henning-schroeder.de>

Edward Jason Riedy <ejr at lotus.CS.Berkeley.EDU> wrote:
>  - GET:  filename, headers = 
>	urllib.urlretrieve("http://my.host.name:8080/loginscript?%s" % 
>	urllib.urlencode({"param": value}))
>  - POST: filename, headers = 
>	urllib.urlretrieve("http://my.host.name:8080/loginscript", 
>	urllib.urlencode({"param": value}))
To: python-list at python.org
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta3
Precedence: bulk
List-Id: General discussion list for the Python programming language <python-list.python.org>

Cool. I wasn't aware of this feature using urlencode. (Although I have
been using urllib heavily...)
Path: news!uunet!ffx.uu.net!newsfeed.easynews.net!easynews.net!unlisys!news.snafu.de!fu-berlin.de!uni-berlin.de!pc19eb635.dip.t-dialin.NET!not-for-mail
Xref: news comp.lang.python:99882

> You only answered half of his request.  The other half is _not_
> supported by urllib: adding HTTP headers.
You can add headers with urllib:
	con = urllib.FancyURLopener()
        con.addheader("Referer", "http://www.python.org")
        f = con.open("http://www.yahoo.com", postdata)
	...
        con.close()


Henning

-- 
henning.schroeder at informatik.uni-oldenburg.de
ich at henning-schroeder.de  *  http://www.henning-schroeder.de


From shouldbe at message.com  Mon Jun 26 13:00:09 2000
From: shouldbe at message.com (Arinté)
Date: Mon, 26 Jun 2000 12:00:09 -0500
Subject: Threading +  embedding = System Dump.
Message-ID: <8j7udf$15au$1@news2atm.raleigh.ibm.com>

Me again.  My usual update
I have a C++ app/ dll.  This is a windows dll is loaded by LoadLibrary.  The
dll holds all the python stuff while the app talks to devices.  Right now if
I was to run a script that has a callback, the script can be running and in
the middle of the script the app can pop up the callback's message window.
That is fine.  But, I want the option of the script waiting for the callback
to be called before continuing.  So, I (well Gordon did) figured I need and
event or the Condition Lock stuff, but unfortunately this is causing the app
to crash.  It seems to be crashing in PyFrame_New() in frameobject.c (this
is python code) on the 1st line of the function
 PyFrameObject *back = tstate->frame;   <----- tstate is null according to
VC++ 6.0 debugger.
This code is after I call the the callback from my app with
PyEval_CallObject.  Note: this works without threads/event code, but no
wait.  I consider doing the Event part in my dll, but the script should be
the one to look at the callback message and determine if that is the message
it wants to continue the script.

I have this code:

import sys
import threading
import PossDevice
import poss
devlist = [0,1]

def threadFun():
 LOCK.set()

def testCallback(str):  #the app calls back here sets the event so the
script can continue
 mthread.start()
 poss.alert(str)

diz = [0,1]
LOCK = threading.Event()
mthread=threading.Thread(None,threadFun)
poss.initialize(PossDevice.Enumerate)
diz[0] = PossDevice.QueryDevices(1,11,35)
diz[0].setCallback(testCallback)
diz[0].open()
mx="\x1b\x47""EVENT\n"  <--- this will cause a device to send a message to
the app
print diz[0].write(mx)
LOCK.wait()  <---wait on event I sent out above
poss.alert("Waited")
diz[0].close()

thanks




From jpet at eskimo.com  Fri Jun 23 00:30:13 2000
From: jpet at eskimo.com (Jeff Petkau)
Date: Fri, 23 Jun 2000 04:30:13 GMT
Subject: boolean true and false values.
References: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <pJB45.1682$Kd.323179@paloalto-snr1.gtei.net>

Tony J Ibbs (Tibs) <tony at lsl.co.uk> wrote in message
news:000d01bfdc23$c8aefae0$f0c809c0 at lslp7o.lsl.co.uk...
> There seem to be two (equally valid, I hasten to add!) traditions of
> handling booleans in programming languages. The (arguably) most obvious is
> to have a discrete boolean type, and thus the logical operators return a
> result of that type, and so on. Algol based languages tend to take this
> approach, and I have always found it more natural (as a pedant).
>
> The other tradition is that used by (for instance) the various Lisp
> derivatives, which is to have "empty" quantities represent false, and
> everything else represent true. This has various practical benefits, and I
> hesitate to argue it is less "theoretically" good.
>
> A language designer has to choose one or the other. Language users then
get
> to grumble if they don't like it. But both are eminently respectable, with
a
> long history.

I think Scheme uses a third method--the special value #f is false,
everything else is true. (Python could use None for the purpose).
This is a nice method for a dynamically typed language,
especially in combination with short-circuiting 'and' and 'or' that
return the true value instead of 'true'.

    x = getfoozle() or some_default_value

Short and clear, but a bit dangerous in Python or Perl or CLisp,
since getfoozle() might return a meaningful value that happens
to be false. Or this:

.    while 1:
.        foo = readline()
.        if not foo: break
.        dostuff(foo)

This works because readline() always returns at least a newline
until EOF is reached. But if you switch to some other sequence
where empty strings can appear, oops.

Of course you shouldn't do that, but how many scripts out in
the wild do you think have that exact bug?

Picking-the-trivial-things-to-rant-about,
--Jeff





From digitig at cix.co.uk  Fri Jun 16 19:34:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 17 Jun 2000 00:34 +0100 (BST)
Subject: How Fast Does Python Run?
References: <961107112.124995831@news.libertysurf.fr>
Message-ID: <memo.20000617003444.42607C@digitig.compulink.co.uk>

In article <961107112.124995831 at news.libertysurf.fr>, 
jerome.quelin at insalien.org (Jerome Quelin) wrote:

> In this report, it is stated as one final conclusion that Python and
> Perl are equivalent, with maybe an advantage for Perl :

Which is interesting, because that doesn't seem to be what the numbers 
show.

> But the report also states that you can't issue conclusions on language 
> speed,
> advantages and drawbacks :
> [snip]
> For all program aspects investigated, the performance variability due to
> different programmers (as described by the bad/good ratios) is about as 
> large
> or even larger than the variability due to different languages on 
> average.
> [snip]
> 
> So you can't tell which is faster between Perl or Python. Let's just 
> stop those
> flamewars.

I'm reminded of the case many years ago, in the days when Rolls Royce 
didn't publish technical specs for their cars. Someone faxed them to ask 
what the horsepower of a roller was, and allegedly got back the fax 
"Adequate". I reckon that language -- or at least, compiler/interpreter 
speed /can/ be compared, and I think that paper did it the right sort of 
way, but I'm not much fussed by the results anyway. Last night I wrote 
some code to help my son with his maths homework. I was surprised by how 
quickly I got it working in Python, and pleased with how quickly it ran. 
Then I realised that it was a variation of the Knapsack problem, and so 
was NP complete! Fortunately there were not many items to deal with each 
time (6 or 7), but still, to be satisfied with the running time of an 
NP-complete program then I think Python's speed is "adequate"!

> But not for any kind of applications : you are to be clever enough to 
> turn
> yourself to C/C++ (or even ASM) if you need to perform low-levels 
> operations on
> your system, or to speed up your code.
> 
> Unless you are to develop a 1'000'000+ lines applications ? Then choose 
> ADA or
> another language that will allow you to manage your code efficiently...
> 
> Do not blindly choose Python just because you're Python addicted. Make 
> your
> choice with clever information. And if finally it appears that Python 
> is the
> best suited for the problem you want to solve, you will have the 
> satisfaction to
> know why (Q: "Why are you coding in Python ?" A: "Because it's the 
> best."
> Q: "And why is it the best ?" A: "Hermm....").

Yes, I argued the same in the Python/C flamewar recently. My favourite 
languages are Python, Ada and Eiffel, I'm competent in C++ and Pascal, and 
I do /most/ of my programming in Visual Basic, which I loathe but which is 
the best language for the task because the customer says that I /have/ to 
use it...


From db3l at fitlinxx.com  Wed Jun 14 03:08:47 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 03:08:47 -0400
Subject: TCP packet size?
References: <CC59868A0A76D311B1F50090277B321C1F33BA@VRTMAIL> <3948AC51.98E5EAEC@rpgarchive.com>
Message-ID: <ur9a07n9c.fsf@ctwd0143.fitlinxx.com>

chris <chris at rpgarchive.com> writes:

> The messages just aren't reassembled the way I thought TCP worked.  I'm
> hoping to find a way to received one message at a time in its entirety,
> or determine the start and end of my messages.

The wrinkle you're running into is trying to overlay your "message"
concept on top of TCP's "byte stream" concept.  TCP sees your message
as a bunch of bytes, nothing more.  It's perfectly willing (and likely
depending on network conditions) to only get some of those bytes
through before it hands some to the receiver, or to get multiple sent
pieces of data through at the same time.

I think that latter bit - determining the start and end of your
messages - is the way to go.  It sounds like you just need to impose
some structure on top of your raw data being exchanged.  Or, if you
already have some (perhaps to identify the changes at some higher
level), just ensure that you encode it in a way that can be recognized
in a constant stream of data.

Really dumb and simple example - if you're sending XML then you know
that the data stream is a fairly limited character set.  Pick some
byte that is invalid for XML (say a NUL, ASCII 0) or anything else
that won't show up in your data stream.  Use that byte to divide
messages.  On the receiving side, parse through any buffer you receive
(and keep receiving) until you find that character, and only return
the stuff before it.  If you have other data received that's after the
byte, keep it in the buffer until the next time your receiving routine
is called.

Other approaches include having a small header on each message that
includes the length count (TCP is guaranteed delivery, so you can be
sure you'll get that amount of data if it was sent unless the
connection breaks down), or if you want to use a dividing byte that
may occur in the stream, just make sure you quote it (an easy approach
is to double it) within the stream itself.

It takes a little more setup in the beginning when writing your basic
transmit and receive functions, but the important point is to hide
this processing down in those fairly low level routines.  That way,
you only have one place that worries about this - the rest of your
system can just make a request to send or read one of your "messages"
and it will always work at the full message level, no matter how many
individual sends or receives the lower level routines have to do.

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


From del.vento at libero.it  Thu Jun 29 06:01:11 2000
From: del.vento at libero.it (Davide Del Vento)
Date: 29 Jun 2000 12:01:11 +0200
Subject: Shoud I use Java or Python?&References=<ya0d7l4uu6a.fsf@trymheim.ifi.uio.no>
Message-ID: <00062911572500.00867@localhost.localdomain>

Thank you (and others that write me in mailbox) very much.
;Davide Del Vento

--
Posted from [212.41.208.21] 
via Mailgate.ORG Server - http://www.Mailgate.ORG


From ajcom at biz.netvigator.com  Fri Jun  9 13:00:15 2000
From: ajcom at biz.netvigator.com (a & j trading company)
Date: Fri, 9 Jun 2000 19:00:15 +0200 (MET DST)
Subject: CDR on sale
Message-ID: <419.436687.03793981ajcom@biz.netvigator.com>

A & J Trading Company
P.O. box 88590, sham shui po post office, kln, hk.
Fax : 852 - 27777499
e-mail : ajcom at biz.netvigator.com

Dear Customer,
We are the supplier of a wide variety of computer 
prepherial. Herewith pls find and check the CDR 
quotation for your reference:
Quotation

CD-R74 silver/silver,A-grade
Price in FOB HK USD0.46/pc.
CD-R80 silver/silver,A-grade
Price in FOB HK0.48/pc.
CD-R74 silver/lite blue,A-grade
Price in FOB HK0.40/pc.
CD-R80 silver/lite blue,A-grade
Price in FOB HK0.42/pc.
CD-R74 silver/lite blue,B-grade
Price in FOB HK0.35/pc.
CD-RW74 silver/blue,A-grade
Price in FOB HK0.76/pc.
Packing : in spindle packing.
Min.Qty. : 10000 pcs/item.
Delivery : prompt delivery.

CD-R74 silver/silver w/arita logo,A-grade
Price in FOB HK0.63/pc.
CD-R74 silver/blue w/arita logo,A-grade
Price in FOB HK0.62/pc.
Packing : full-packed (1 pc/jewel case + inlay paper)
Min.Qty. : 10000 pcs/item.
Delivery : 7 days after receipt of your firm order.

CD-R74 silver/lite blue ,A-grade
50000 pcs
Price in FOB HK0.54/pc.
CD-R74 silver/lite blue ,A-grade
100000 pcs
Price in FOB HK0.53/pc.
Logo : tarsan or copyall.
Packing : full-packed (1 pc/jewel case + inlay paper)
Delivery : 7 days after receipt of your firm order.

DVD-R, 3.95,A-grade
Price in FOB HK12.10/pc.
Packing : full-packed (1 pc/jewel case + inlay paper)
Delivery : 10-15 days after receipt of your firm order.

Mini disk, 74 min,A-grade
Price in FOB HK1.00/pc.
Packing : 5 pcs of storage box or single sleeve.
Min.Qty. : 10000 pcs.
Delivery : 7 -14 days after receipt of your firm order.
ALL THE PRICE IS SUBJECT TO OUR FINAL CONFIRMATION.
Remarks :
1/ payment : by irrevocable & drawable L/C at sight or T/T.
2/ logo : please advise if you need to print your own logo.
3/ validity : subject to our final confirmation.

Your interesting to our products is highly appreciated.  
For inquiry, kindly fax to us at 852-27777499 or e-mail 
AJCOM at BIZ.NETVIGATOR.COM to us directly. 
Thanks again for your kind attention and looking forward 
to serving you in the near future.





From Linus_Chang1 at bigpond.com  Sun Jun 18 01:50:10 2000
From: Linus_Chang1 at bigpond.com (Linus Chang)
Date: Sun, 18 Jun 2000 15:50:10 +1000
Subject: win32gui question - is my window the foreground window?
References: <39437A27.1020403@bigpond.com> <85g0qkices.fsf@one.net.au>
Message-ID: <394C6312.2000206@bigpond.com>




Neurocrat wrote:

> Linus Chang <Linus_Chang1 at bigpond.com> writes:
> > 
> > I'd like to detect if the window of my wxPython application is currently 
> > the foreground window.
> > 
> 
> Could you elaborate on this?  Are you trying to determine whether your
> application's main window is at the top of the Z-order? Are you trying
> to determine this from *within* your application, or from another
> process?

Sorry I wasn't more specific. I'm trying to tell from within my own 
application, whether my application is the currently foreground 
application. (ie. the one that's got focus; it's "rectangle" [I can't 
think of the proper term] in the task bar is the one that's depressed.) 
It doesn't really matter whether my app's main window or some child 
window of my app is at the top of the Z-order. But as far as I can tell, 
the foreground app always has its main window (or a child) at the top of 
the Z-order, but I'm not a low level windows gui buff!!

eg. If you've ever used Microsoft Instant Messenger, when a message 
arrives and Instant Messenger is not the currently active window, then 
the app's taskbar "rectangle" flashes and a sound plays. I'd like 
similar functionality in my application. I haven't found an API call to 
make it flash (under Win98 and Win2000 (not NT4) it can be done 
automatically by SetForegroundWindow, since Win98 and Win2000 doesn't 
let you SetForegroundWindow under certain circumstances), but I'm taking 
this one step at a time - detecting if my app is the foreground one is a 
good start!

Thanks,

Linus






From joshuar at fujitsu.co.nz  Mon Jun  5 17:15:00 2000
From: joshuar at fujitsu.co.nz (joshuar at fujitsu.co.nz)
Date: 06 Jun 2000 09:15:00 +1200
Subject: win32Service.OpenService not working
References: <8hggdu$4fa$1@nnrp1.deja.com>
Message-ID: <d7lv96ej.fsf@fujitsu.co.nz>

This is because Windows services have two names. A display name e.g 
"Windows Installer" and a service name e.g. MsiInstaller. You need 
to be careful which name is used for managing services. Usually the
Service Name does not have spaces in it.   


From robin at jessikat.demon.co.uk  Thu Jun  1 04:23:48 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 1 Jun 2000 09:23:48 +0100
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com>
 <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use>
 <3935A24C.35198018@san.rr.com>
Message-ID: <FSDqFCAU2hN5Ewll@jessikat.demon.co.uk>

In article <3935A24C.35198018 at san.rr.com>, Courageous
<jkraska1 at san.rr.com> writes
>
>> I wish the default implementation for lists were deques, so
>> that I didn't have to worry about which end of a list I add
>> to or remove from.
>
>Well, as you saw from my notes, this matters not at all for
>lists <100 elements.
>
>If you want my deque implementation, let me know. It's fairly
>complete, albeit doesn't implement many of the sequence
>operations (slicing, indexing, mapping).
>
>
>C/
Isn't there somewhere we could put it so others can get at it?
Possibly with the benchmarking code.
-- 
Robin Becker


From robin at jessikat.co.uk  Thu Jun 29 05:28:07 2000
From: robin at jessikat.co.uk (Robin Becker)
Date: Thu, 29 Jun 2000 10:28:07 +0100
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>
 <8F61BBD91gmcmhypernetcom@199.171.54.194>
 <slrn8ll4ak.25h.hzhu@rocket.knowledgetrack.com>
 <8F61DA247gmcmhypernetcom@199.171.54.194>
Message-ID: <K1Sz6AAnaxW5Ew+H@jessikat.demon.co.uk>

Several people want the relative import taken out and shot; I disagree.
Many times I want to move/make use of parts of a package eg I might want
to use part C of A.B.C in another context. If everyone uses absolute
package names this becomes a problem. The only problems I have come
across relate to the ambiguity of python classes etc when imported via
two different routes; if we could fix that then relative importing
wouldn't seem so horrid.
-- 
Robin Becker


From eviltofu at rocketmail.com  Sat Jun  3 02:45:43 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 03 Jun 2000 06:45:43 GMT
Subject: Erm... MacOS X?
Message-ID: <eviltofu-1AEA18.02454303062000@news.earthlink.net>

Has anyone compiled Python for MacOS X?


From balrog at webtv.com  Sat Jun 17 13:45:54 2000
From: balrog at webtv.com (balrog)
Date: Sat, 17 Jun 2000 17:45:54 GMT
Subject: Python 3000 -> No, but there's 'Ruby'
References: <394c2b15.634411296@nntp.interaccess.com>
Message-ID: <394BB8DE.AE1D6A7A@webtv.com>

> Is there a link where I can look up what's being considered for Python
> 3000?

Half-seriously, Ruby is nicer looking & acting;  it's part Python,
part Eiffel / Sather, with the quick & easy string handling of Perl.

I'm liking it, currently.  It has a Tk binding, a net lib, etc.

No static binding though.  

I have to say Ruby's more what I was looking for out of a clean & easy
scripting language than Python;  I wrote a simple text processing
thing in Python and I might as well have written it in C.  The only
time saver was the 'split' function, which can be done in C, too.
My impression of Ruby is that it's cleaner, more sophisticated (has 
iterators, full closures) & is safer feeling than Python (has 'end' 
like Modula-2).  Python is just too full of arbitrariness & ugliness.  
Maybe Python 3000 will clean it up, but, 

check Ruby out at:

www.rugy-lang.org/en


From emile at fenx.com  Mon Jun 12 07:31:42 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 12 Jun 2000 04:31:42 -0700
Subject: lambda and if
References: <8i2c9l$b0r$1@news1.wire.net.au> <0006122253431T.01005@localhost.localdomain> <07cd01bfd45e$9b0d07c0$1906a8c0@fc.fenx.com> <0006122328441W.01005@localhost.localdomain>
Message-ID: <07d501bfd461$ced14280$1906a8c0@fc.fenx.com>


----- Original Message -----
From: matt <matt at virtualspectator.com>

> sneaky ... it's nice.
> might be nice to extend that with
> ["","big"][x[1] in ("big",)]  so we can extend the 'yes'
set
>
> but extending it to using the actual match is a little
more difficult,
> i.e.
>  x[1] in ("big","little","wide",)  where the matched item
would be appended if
> it existed or else a blank

Wouldn't ["",x[1]][x[1] in ("big","little","wide",)] do it?



>
>
> thanks
> matt
>

Emile van Sebille
emile at fenx.com
-------------------






From hzhu at knowledgetrack.com  Wed Jun 21 20:24:25 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 22 Jun 2000 00:24:25 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <LOz35.14014$Xx5.685990@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <slrn8l2nce.q2.hzhu@rocket.knowledgetrack.com>

On Tue, 20 Jun 2000 01:30:19 GMT, William Dandreta
<wjdandreta at worldnet.att.net> wrote: 

>
>My error was that I did the following:
>
>from strop import joinfields, splitfields
>from string import replace
>
>I erroneously assumed that replace would use the joinfields, splitfields
>from strop, but it really used the joinfields, splitfields from string which
>are very slow.


If you really want to replace an internal method in an imported module,
maybe you can try this (not tested):

import string, strop
string.joinfields = strop.joinfields
string.splitfields = strop.splitfields
from string import replace

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From echuck at mindspring.com  Thu Jun 15 09:43:25 2000
From: echuck at mindspring.com (echuck at mindspring.com)
Date: Thu, 15 Jun 2000 13:43:25 GMT
Subject: Just-In-Time importing
References: <Pine.BSF.4.10.10006132236080.23561-100000@chi.pair.com>
Message-ID: <8iamho$q38$1@nnrp1.deja.com>

In article <Pine.BSF.4.10.10006132236080.23561-100000 at chi.pair.com>,
  Michal Wallace <sabren at manifestation.com> wrote:

> Now, if zikebase.Content were usually a subclass of zikebase.Node (it
> isn't), creating a zikebase.Content() would produce something that was
> a subclass of Vertex instead. This can come in really handy if you
> want to add something to every class in an object hierarchy and don't
> want to (or can't) actually change the original top-level code. (It
> sure beats subclassing every item in the tree, or writing a
> wrapper-factory)
>
> Cheers,
>
> - Michal
>

Michal, you can also accomplish that by adding the base class at run
time. I use this:

def MixIn(klass, newBase):
    if newBase not in klass.__bases__:
        klass.__bases___ = klass.__bases__ + (newBase,)


Now I can say:

MixIn(ModelObject, ModelObjectInterface)


to add rendering methods to ModelObject in my web and/or gui app.

-Chuck


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


From m.faassen at vet.uu.nl  Tue Jun  6 08:19:25 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 6 Jun 2000 12:19:25 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA>   <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net>
Message-ID: <8hiq8d$ngm$1@newshost.accu.uu.nl>

William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
> On 5 Jun 2000 12:01:04 GMT, Martijn Faassen wrote:
>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>> On 30 May 2000 10:59:24 GMT, Martijn Faassen wrote:
>>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>[snip]
>>>>> Your rhetorical questions answer themselves.  Why should a programmer
>>>>> enforce these things?  Because it's stupid to NOT enforce them.  So
>>>>> enforce them already!

>>>>That's nice to say, but you don't agree with that yourself if you think
>>>>Python's use of indentation is a good thing. It can help a lot if the
>>>>language makes it impossible to be inconsistent. Unless you think the
>>>>_only_ advantage of Python's indentation strategy is the lesser line 
>>>>count.

>>> This is a classical example of false dichotomy ;-).

>>It's not.

> It actually is.  You're claiming that I hold one of two opinions; I hold
> neither one.

I'm claiming you haven't shown yet that consistency has nothing to do
with 'code looking good'. Anyway, you should've said something like that
in your first post -- now you explain what you meant. Now I'll explain
what I meant. :)

>>> I liked Python's
>>> indentation not because it enforced consistency, but rather because it
>>> makes code look good.  In fact, it could be argued that Python does _not_
>>> enforce consistency; the programmer is free to choose any indentation for
>>> each block, so long as each block can be parsed (which does require
>>> internal consistency).

>>"""
>>I liked Python's
>>case sensitivity not because it enforced consistency, but rather because it
>>makes code look good. In fact, it could be argued that Python does _not_
>>enforce consistency; the programmer is free to choose any case-spelling for
>>each variable name, as long as each variable name can be parsed.
>>"""

> But now you've converted my paragraph from a true statement to a false
> one.  The programmer is not in fact able to choose any case-spelling for
> each variable name; they must use the same one which was originally used,
> unless they wish to create or access another variable.

Perhaps that's unclear, but my reading of 'for each variable name' is for
each unique variable name -- I mean, a name can be referred to multiple
times, but it's not a different name each time.

>>Consistency in style happens to one of the most important factors that
>>makes code look good (on the syntactic level). Your indentation looks
>>better in Python because Python complains loudly if you don't indent properly.
>>That's not the most important thing though; you can indent properly in
>>any language. It's just that everybody indents properly in Python, and
>>more or less in the same way.

> Again, Python code looks good not because consistent indentation is
> enforced, but rather because indentation alone is used to draw a tree
> structure.

But indentation is *always* used to draw a tree structure. Python *enforces*
this tree structure, but it's not as if you can't draw tree structures
in C.

> Consider the following example of inconsistent indentation:

> if something:
>        do_it()
> else:
>  dont_do_it()

> Python will happily accept this.  It'll even happily accept this for long
> blocks, here it can be inconvenient.  This makes the code a bit harder to
> read, but it'll still not bad -- because Python expresses a 2-dimensional
> structure (the syntax tree) in a 2-dimensional form.

> Most ugly code I've seen isn't ugly because it has poor indentation
> (although I have seen some); most of it's ugly because of horrible naming,
> or bad structuring, or ...

Well, obviously indentation can't help in case of horrible naming and
bad structuring. But you say indentation makes code look better, and that's
not because it improves horrible naming or bad structuring but because
it improves something else. So either you're saying you like Python's
indentation because it doesn't have braces (and that's the only reason),
or you're buying into the consistency argument after all. I believe I
gave you the braces option before, speaking of dichotomies.

>>> More recently I've come to realise that the reason Python indentation
>>> looks good is that Python, like most modern languages, is a
>>> tree-structured language, and so is best represented in two dimensions.
>>> Most modern languages, however, attempt to represent themselves in one
>>> dimension: an ASCII stream which ignores lines and whitespace.

>>Well, why shouldn't a programmer enforce such indentation in any
>>language?

> Of course they should.  Lisp programmers get really mad if people post
> non-standard formatted code on c.l.lisp.  EMACS enforces it when writing,
> and angry programmers enforce it when reading ;-).

So in your case there's no reason to prefer Python's way of indenting,
*except* that there are no delimiters. Okay, thanks for making that 
clear. Or am I still missing something? :)
 
>>Indentation looks good in any language, not just Python, so
>>that's not really an argument for Python's indentation style.

> Well, there you go -- my point exactly.  I was saying the same for case
> (which Python enforces much more consistently than it enforces
> indentation).

Yes, you were saying the same for case. Now, if I understand you right,
you like Python's indentation strategy more than the delimiter strategy.
The only two reasons I can come up with is:

  * less redundancy because there are no delimiters

  * less space taken by the delimiters

(because the two-dimensional structure argument makes sense for any language
if only you indent I'm not including that)

All right.

> In fact, the nice thing isn't that Python enforces indentation consistency
> (because it doesn't really, any more than the minimum): the nice thing is
> that Python doesn't throw anything more in your way than you really need
> to parse the tree.

Right, so that's the redundancy argument. Of course, if you used Lisp or C
without any indentation, you'd have that same feature too. Why do you
prefer the indentation approach to the delimiter approach? According to
your opinions on case-spelling one would think you'd prefer the former, as
it offers more freedom while the programmer still has the option just
fine to enforce any indentation strategy they want, if they want it.

>>Besides, if you ignore case you ignore that words in Python are spelled
>>consistently in case, which is *also* a property of lots of modern languages.
>>Sometimes not enforced, but still a desired property of the programmer.

> But in case-sensitive languages, words can be spelled *differently* in
> case: NO is not the same as No or no.  If the language were
> case-insensitive this would be impossible.

Hm, I'd say those would be different words, which vary only due to 
case-spelling. Not that the same word can be spelled differently in 
case.

Ignoring that point, spelling the same word in different case-spellings
is only possible in case *insensitive* languages!
So I'm not quite sure what you mean here?

> In fact, it's conceivable to have a case-insensitive language which prints
> an error whenever an alternate case spelling is used.

That's not a case-insensitive language. That's a case-sensitive language
which forbids words which vary only in case-spelling.

> I think that would
> be silly, but it would seem to fit what you want FAR better than a
> case-sensitive language ever could.

No, because I as a programmer happen to like the freedom different
case-spellings happen to offer. It's up to the *programmer* to
enforce the right conventions here. Throwing your arguments back at you
again. :)

> So if you wrote:

>   no = 1
>   No = 2 # error!
>   print NO # error!

> I don't like that -- but it proves that case insensitivity can be more
> powerful in terms of enforcing style.

Ahum? Is this case-insensitivity or case-enforcing? This is currently
what happens in Python, as long as you haven't assigned to 'No' or 'NO'
elsewhere. The case enforcing addition would then reject assigning to
'No' or 'NO' after the assignment to 'no'.

> It also seems to indicate to me at
> least that I don't want a language enforcing THAT much style, any more
> than I'd want Python to require exactly one tab or two spaces or any other
> fixed amount for indentation.  Thank you, I'll decide that on my own.

Right, but it turns out to be the case that the current setup doesn't enforce
that much style (because the language becomes harder to write). It does
give you plenty of errors if you spell things case-inconsistently, but
you can if you want to. A few more warnings would be nice, though this would
require some code analysis. The recent patches for 'NameError' seemed nice,
though.

>>> So I don't really care about languages enforcing consistency; if I need
>>> it, I'll enforce it myself.

>>A computer language would be worth less if it brought no consistency at
>>all.

> Granted.  But this doesn't seem to justify your slippery-slope treatment --
> we don't need to enforce your amount of consistency just because no
> consistency is bad.  We already enforce _some_.

In the case-sensitive version we have now, we already enforce some, but
not to the point the language becomes unwritable. It's also a simple
rule, which helps fit it in people's minds. The case-insensitive rule
would also be simple, but the code becomes less readable and not 
significantly more writable (in my opinion). The case-enforcing rule
makes code even more readable (possibly -- there are arguments about
this by the people who like different case-spellings), but it becomes
far less writable, and the rule has a bit of extra complexity.

[snip examples of consistency that are nice]

> And I like polymorphism (which implies typechecking).

I'm not quite sure what you mean by 'polymorphism' in this context. We
seem to have polymorphism as far as I understand it already?

> And a lot of other
> things, all of which I hope to see.  I'm curious to see whether other
> people like them, and I'll defend them -- but unlike most of the people
> arguing about case sensitivity, an ENORMOUSLY minor issue,

Note that you're arguing about it as well. It's a minor issue with a lot
of effects on readability. And it's easy to argue about. So that's why
we do. :)

> I won't
> consider them prerequisites to using the new Python -- I'll use the new
> Python if it feels like an improvement.

Oh, I'll likely use the new Python if it's case-insensitive. But I'd
prefer it would not be so, and I'm just trying to spread a meme here;
the link with indentation. Your own meme-complex seems to reject it,
so I'll stop trying with you. :)

>>According to your argument, you will just enforce this consistency
>>yourself. The problem is that, *even* if everybody enforces self
>>consistency, people will enforce *different* consistency and their
>>coding style will vary over time. From a maintenance and communication
>>point of view, this is bad. 

> And for case sensitivity it _just doesn't matter_.  The only people who
> are strongly helped or hurt by case sensitivity appear to be newbies, and
> they're being hurt.

I'd like some more evidence about how the newbies are being hurt, but
if we grant that, I still disagree that case-sensitivity only matters
for newbies. How does it not matter for case sensitivity? If Foo writes
a module and you use it in your module, but you use a different case
spelling for common variables (but consistently!), and then I get to
maintain it while I've been using another case spelling when I used
that module all the time, wouldn't that hurt me? 

It's easy to say that it won't matter for communication and maintenance,
but could you try backing that up?
 
>>> Take a lesson from the people complaining
>>> about Python's indentation: one of the leading false accusations is that
>>> it's inflexible, that it won't allow people to choose thier own
>>> indentation.

>>I don't see how this maps to the case sensitivity argument at all. 
>>Python's indentation *is* less flexible in what code-layouts are
>>allowed, and it *is* a silly accusation.

> It may or may not be silly -- but it's _definitely_ false.  Python is
> _not_ inflexible; you can lay your code out in any way you want, so long
> as it's in the same tree shape as its semantic structure implies.

It's more inflexible. I can't do:

  start()
    do_a()
    do_b()
  commit()

And occasionally that would be nice. There are other examples, which I'm
sure the complainers can come up with. Where the complainers are wrong
(though it's their personal taste, but they should actually *try* it
before they complain) is that they don't see the significant advantages
which outweigh these minor disadvantages.

> I don't think it's silly, either; I've complained enough about languages
> which don't do this right, such as Make.

Right -- Make is rather unwritable. There are no advantages there.
I still don't see how these people relate to our present case;
these people are complaining (without having actually tried it; anyone else
complaining is of course in his or her right to complain) about decreased
freedom you get with Python's indentation scheme. Now you say we should
cater to people like that, offering them more freedom (in case spelling),
while nobody of *those* people appears to be complaining about that, and
while the reason many people (not you) like Python's indentation *is*
increased consistency and therefore increased readability.

> There are two obviously terrible errors: inconsistency and inflexibility.
> There's no reason to assume that avoiding one will force everyone into the
> other, and I'm not making that mistake about you.

I'm not quite sure I understand the language. Right now we avoid
inconsistency and we also avoid inflexibility in the case-spelling
domain. There are minor disadvantages which apparently confuse newbies,
so we ought to think about how to improve matters. But case-insensitivity
gives up consistency and case-enforcing gives up flexibility; so it
seems we shouldn't go anywhere. :)

>>So I don't see why I should take a lesson from this complaint or those
>>people. Besides, people who make this silly accusastion generally use
>>languages which *are* case sensitive. So I don't see how this relates to
>>anything.

> Are you trying to miss the point?  I don't think it's working, because you
> previously argued on the point.  I'll confess that you're trying hard,
> though -- three consecutive sentances which miss the point in three new
> ways.

There must be some error in communication here. I genuinely miss the
point. Next time you say I miss the point would you please be so
kind as in explaining what the point is, even though you suspect I'm
deliberately missing the point? It's generally kinder to assume
stupidity and not malice. :)

Regards,

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


From garabik-news at spam.melkor.dnp.fmph.uniba.sk  Fri Jun  9 08:11:55 2000
From: garabik-news at spam.melkor.dnp.fmph.uniba.sk (Radovan Garabik)
Date: 9 Jun 2000 12:11:55 GMT
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <960552651.445451@cdwork.cvt.stuba.sk>

Lawrence Kesteloot <lk at veriomail.com> wrote:

 : If one of the goals of Python 3000 is to make it easy for new
 : programmers to learn, how about removing the distinction
 : between = and ==?  The only kinds of examples I can think of
 : where this is useful is something like:

 :     a = b = 5

 : vs.

 :     a = b == 5

 : But this isn't worth the trouble it gives new programmers.  The
 : syntax could be changed to make = and == equivalent tokens
 : and have the parser figure out which is meant from the context,
 : just as it's done in Basic.  The documentation would only ever
 : mention =, but the language would support == for backward
 : compatibility.

 : (One problem I can see with this proposal is that "a = b = 5" would
 : probably mean "set A to whether B is equal to 5 or not", and this
 : is the less useful of the two interpretations.  I guess the parser
 : could be written to use the other.  Or the whole construct could be
 : disallowed since it's not terrifically useful anyway.)

AFAIK, pre-1.0 python did not have == operator.
The problem is, how to interpret a=b in interactive python?
Should it return 1 or 0, depending on whether a equals b, or should it
assign b to a?


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From fwang2 at yahoo.com  Wed Jun 21 14:20:12 2000
From: fwang2 at yahoo.com (oliver)
Date: Wed, 21 Jun 2000 14:20:12 -0400
Subject: snmp support in Python
Message-ID: <Pine.LNX.4.10.10006211416200.7273-100000@worf.mcnc.org>

Hi,

Can someone update me the current status of SNMP support in Python? I
remember a while ago, there were a few messages saying an development
website setup on sourceforge, when I checked it, nothing was there yet.

Are there any workable solution? how mature is it?

Thanks for help

oliver





From Cedric.Lefeuvre at tcomhp43.epfl.ch  Fri Jun 23 06:01:22 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Fri, 23 Jun 2000 12:01:22 +0200
Subject: python & excel
Message-ID: <39533572.AB90FD1F@tcomhp43.epfl.ch>

do you know, if it's possible
to generate an excel like file (or gnumeric ...something like that).

thanks


--
C?dric Lefeuvre
SUPAERO 2A-Stagiaire EPFL
0041 (0)21 693 26 16
Fax: 0041 (0)21 693 26 83
 _ _ _/_  _
( (-(// /(


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000623/f78e4102/attachment.html>

From m.faassen at vet.uu.nl  Thu Jun  1 18:49:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 1 Jun 2000 22:49:37 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk>
Message-ID: <8h6pa1$ecq$2@newshost.accu.uu.nl>

Jacek Generowicz <jmg at ecs.soton.ac.uk> wrote:
> Peter Schneider-Kamp wrote:

>> Guido van Rossum wrote:
>> >
>> > Oh, and to top it all off, I'm going on vacation.  I'm getting married
>> > and will be relaxing on my honeymoon.  For all questions about
>> > PythonLabs, write to pythonlabs-info at beopen.com.
>>
>> Gefeliciteerd bij de eerste stap naar de echtscheiding!

> So cynical for one so young !

This again proves my Python-Dutch connection conspiracy theory: all
Python programmers really also speak Dutch.

First an individual with a german-sounding name posting from Norway
posts Dutch, and then someone with an eastern european sounding name
posting from the UK understands it.

The-PSU-just-has-to-be-behind-it-ly yours,

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


From boud at rempt.xs4all.nl  Wed Jun 28 15:19:25 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 28 Jun 2000 19:19:25 GMT
Subject: Following a file, or cloning tail
Message-ID: <8jdj3t$or4$1@news1.xs4all.nl>

I'm trying to write an application that follows the output of another
app, analyzes it, and serves it up as html. Kind of filtering tail,
really. Is there any accepted 'best' way of 'tailing' a file?

I thought about having two threads, one doing the tailing, one
doing the servering (instantiating new threads as needed), but I
can't even get the tailing right!

-- 

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


From robin at jessikat.demon.co.uk  Thu Jun 22 03:32:54 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 22 Jun 2000 08:32:54 +0100
Subject: Python class tree viewer (with wxPython)
References: <3951AA42.13155B89@coolest.com>
Message-ID: <yBGXTvAmEcU5Ew62@jessikat.demon.co.uk>

In article <3951AA42.13155B89 at coolest.com>, Alexander Schonfeld
<alex at coolest.com> writes
>Hi,
>
>I wrote (I mean hacked a demo...for wxOGL) to show class trees of bunches of
>python files.  You can then move them around and take screen shots and bask in
>the glory that is your python classes.
>
>To use:
>       python pyinherit.py d:\cvs\Zope2\lib\python\Products\LoginManager
>       python pyinherit.py [specific .py files and dirs... whatever you want]
>Dirs are recursively traversed with 'walk'... looking for py files.
>etc...
>
>It requires wxPython (a gui lib) from:
>       http://wxpython.org/
>
>Please expand on this if it looks interesting, and email me a copy.
>
>Thanks,
>
>Alex.
>
...
Wow! Now all we need is to apply that algorithm which tries to draw the
diagram decently in planar form.
-- 
Robin Becker


From jeremy at beopen.com  Tue Jun 20 18:22:27 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 20 Jun 2000 18:22:27 -0400
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <LOz35.14014$Xx5.685990@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <23B9004DDD25EC59.63D8A67ABEF9C562.CF071FB1FFFA42EE@lp.airnews.net>

"William Dandreta" <wjdandreta at worldnet.att.net> writes:

> My error was that I did the following:
> 
> from strop import joinfields, splitfields
> from string import replace
> 
> I erroneously assumed that replace would use the joinfields, splitfields
> from strop, but it really used the joinfields, splitfields from string which
> are very slow.
> 

This is an important lesson to learn!  In Python, assignment is just a
name binding operation.  The import statement will load a module,
provided it hasn't already been loaded, and then do one or more
assignments to local names.

The "from clinic import argument" version you used loads the clinic
module, looks up the name argument in that module, and binds the
object it finds to the name argument in the current namespace.  From
this point forward, the only way to change the binding for argument in
the current namespace is to explicitly change it in the current
namespace.  If you change the binding in the clinic module, e.g.
    clinic.argument = clinic.abuse
then future lookups of argument or abuse in the clinic module will
return the same object.  But existing references, like the one created
by from clinic import argument will still point to the old object.

Jeremy


From kaimar at ut.ee  Fri Jun  2 19:11:25 2000
From: kaimar at ut.ee (Kaimar Karu)
Date: Sat, 03 Jun 2000 02:11:25 +0300
Subject: zope example at /url/
References: <39383EE4.C87460EE@ut.ee>
Message-ID: <39383F1D.2EC3F459@ut.ee>

Hi,

> **
> Error Type: NameError
> Error Value: Eval
> **

btw, it happens after pressing the 'find' button.

--
Kaimar Karu | kaimar at ut.ee | 251 13 468
       = What a wonderful world =




From aahz at netcom.com  Mon Jun 19 10:39:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jun 2000 14:39:00 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <3948D25D.7C83CD6@micron.net> <39499FC1.8C31F7B5@sightreader.com> <8il8hg$bfn$1@newsy.ifm.liu.se>
Message-ID: <8ilba4$t49$1@slb7.atl.mindspring.net>

In article <8il8hg$bfn$1 at newsy.ifm.liu.se>,
Paul Svensson <paul at svensson.org> wrote:
>Ken Seehof <kens at sightreader.com> writes:
>>
>>LOC (lines of code) is also a measure of programmer output.  The idea is
>>that you evaluate your programmers by the sheer quantity of code they
>>crank out.  The one who cranks out the most lines of code per week is
>>clearly the best programmer (yeah right!).  Anyway, it is sometimes used
>>as a rough extimate of productivity.
>
>Very rough;
>like estimating the productivity of a welder by the amount of acetylene used.

Hey!  That's a good one.  Mind if I stick this in my .sig file:

"Using LOCs as a measure of programmer productivity is like estimating the 
productivity of a welder by the amount of acetylene used."  
--Paul Svensson, comp.lang.python
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From digitig at cix.co.uk  Thu Jun  8 16:36:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 8 Jun 2000 21:36 +0100 (BST)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <393D94F8.712427E6@san.rr.com>
Message-ID: <memo.20000608213632.55497B@digitig.compulink.co.uk>

In article <393D94F8.712427E6 at san.rr.com>, jkraska1 at san.rr.com 
(Courageous) wrote:

> 
> > > GOD CREATED UPPERCASE LETTERS. everything else is the work of man.
> > 
> > god doesn't need to shout...
> 
> Adding to my long list of politically correct terms, I
> now enter into the dictionary: "humor impaired".

Ah, yes, I had been told that some cultures tend to miss irony...


From noreply at no-spam.org  Wed Jun 14 15:08:39 2000
From: noreply at no-spam.org (Brian)
Date: Wed, 14 Jun 2000 19:08:39 GMT
Subject: Python class browser
Message-ID: <XKQ15.1$oL1.1552@paloalto-snr1.gtei.net>

Anyone know of a class browser for Python?  Preferrably written using
wxPython?
Thanks!





From breiter at usf.Uni-Osnabrueck.DE  Fri Jun 23 11:12:37 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 23 Jun 2000 15:12:37 GMT
Subject: Boa stalled? (GUI editor for wxPython)
References: <8iqbv4$4p$1@newsserver.rrzn.uni-hannover.de>
    <3952A499.1339701@e.co.za>
Message-ID: <8ivup5$dsq$1@newsserver.rrzn.uni-hannover.de>

In article <3952A499.1339701 at e.co.za>,
	Riaan Booysen <riaan at e.co.za> writes:
> Bernhard Reiter wrote:

>> Is Boa development stalled?
> 
> Definitely not, it has progressed quite a bit since the last release.
> I'm pretty close to another release.

>> It looks like nobody made changes in the last three month.
> 
> No changes to the web site, apologies for that.

I did not find any progress in the CVS.
Is there a reason for not leeting people see the development drive?

So can I download a CVS version with all the improvements?

> If you want to keep up to date the mailing list is a little bit busier
> than the web site.

Yes there is some activity on the mailinglist. I shouild have looked
there, too. :)

>> The program still needs developers. It is in competition to the nice
>> glade and libglade thing and not mature enough yet.
>> Why is nobody picking it up?
> 
> I've received no code/patches so far, but a few people have offered 
> their help and I intend taking them up on it.

Alright, sound good. 
>> http://boa-constructor.sourceforge.net/
>> 
>> --
>> Professional Service around Free Software                (intevation.net)
>> The FreeGIS Project                                         (freegis.org)
>> Association for a Free Informational Infrastructure            (ffii.org)
> 

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From hzhu at knowledgetrack.com  Wed Jun 28 18:42:33 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 28 Jun 2000 22:42:33 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <vndwvj9bgzh.fsf@camelot-new.ccs.neu.edu>
Message-ID: <slrn8ll00c.25h.hzhu@rocket.knowledgetrack.com>

On 28 Jun 2000 18:03:46 -0400, Justin Sheehy <dworkin at ccs.neu.edu> wrote:
>hzhu at knowledgetrack.com (Huaiyu Zhu) writes:
>
>> from AAA import *
>> from BBB.AAA import * 
>
>[various convoluted issues]
>
>"Doctor, it hurts when I do this."
>"Don't do that, then."

What it hurts is to import BBB/AAA.py instead of AAA.py.  I'd be interested
in how do you "Don't do it".


>Import your modules directly and you don't have to deal with this kind 
>of problem.  As a side effect, your code is usually clearer.

This has nothing to do with my code.  How about

import AAA

Does it solve the problem at all? 

<rant>
It appears that if someone is objecting to a certain idiom then he'll
automatically think that's the cause of whatever problem anyone else is
writing about, as long as it appears somewhere.
</rant>

Please read my original post.  I was discussing a completely different issue.

To avoid any further unnecessary digressions, I state my actual problem
here.  I'm developing the MatPy pacakge which works quite well on Linux and
Solaris, but some Windows user's keep complaining about name clashes.  It
turns out that I have, in the file MatPy/gnuplot.py, this line

from Gnuplot import Gnuplot

Since on Windows Gnuplot.py and gnuplot.py is indistinguishable, this line
tries to import itself!  

Now could any one that think the problem can go away suggest a solution that
does not need to change file names and still works on Windows.

Huaiyu


-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From moshez at math.huji.ac.il  Tue Jun 13 01:53:58 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 08:53:58 +0300 (IDT)
Subject: Testing for a file in python?
In-Reply-To: <394db29a.275888218@nntp.interaccess.com>
Message-ID: <Pine.GSO.4.10.10006130850160.20303-100000@sundial>

On Tue, 13 Jun 2000, Thaddeus L. Olczyk wrote:

> >I don't know what you mean by "most".
> For all those who emailed me ( especially the nasty ones ) suggesting
> that I'm slamming Python in favor of Perl ( which I know well enough
> to know, I don't like it )  by most I mean:
> 
> sh,csh,bash,tcsh,zsh ( and OK Perl )

IOW, sh and csh variants. And most of those *don't* have -f: it's the
test program (often symlinked to "[") that has that optin. Python can
run other executables to, so you could write

if os.system("[ -f %s ]" % filename)==0:
	pass # file exists

So Perl is the only language that has it. I don't think awk has it, 
and I'm certain neither guile nor Tcl have it. You have a really strange
definition of "most". 

> Ones that I'm sure don't do it
> Python, sed ( and I'm not sure about sed, it's just not likely to be
> there ).

sed is hardly a scripting language (well, it's probably turing complete,
but it is quite hard to run other executables from it)

BTW:
You didn't mention Ruby (which I think does not have it), Rebol (ditto), 
and others.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From amitr at w-o-i.com  Thu Jun 22 04:02:10 2000
From: amitr at w-o-i.com (amitr at w-o-i.com)
Date: Thu, 22 Jun 2000 08:02:10 GMT
Subject: zope, python vs apache, perl
Message-ID: <8ish5s$kd4$1@nnrp1.deja.com>

hi

I want to evaluate zope, python vs apache, perl.

which one can be used where!
strong points and pitfalls.

thanks in advance


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


From porter at et.byu.edu  Fri Jun 16 17:22:34 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Fri, 16 Jun 2000 15:22:34 -0600 (MDT)
Subject: Frozen socket. 
Message-ID: <Pine.GHP.4.21.0006161522100.3877-100000@don.et.byu.edu>

I am very new at this sort of this, so I am hoping that there is a simple
and obvious (if you know it) solution to my problem.  I looked in the FAQ
and tutorials, but to no avail.

I have written a simple python program on linux that uses sockets.  My
program connects to an existing server.  There is some data exchanged for
handshaking purposes, and then I am supposed to start sending a long list
(not a python list) of strings to the server.  I don't get any responses
from the server - I keep sending out these strings until I am done, at
which time I disconnect the socket.

At least, that's what is supposed to happen.  Somewhere between string #
1000 and 1020 (it varies somewhat), my program freezes. It does this
whether I send the strings out as fast as I can, or if I space send out
the strings to one per second.  
I do a ctrl-C to escape, and the last line of the exception printout is:
    self.sock.send(messages.packMM(request))
This seems to me that the program is getting stuck in the
sock.send() funciton. 

I have looked around, but I haven't been able to figure out why a program
would choke on a send() command.  I don't even know if it is because of
something I am doing on my end or something happening on the server
end.  If it is my fault, how do I fix it?  If it is because of something
happening on the server side, how do I compensate?

Thank you for your help.  







From richard_chamberlain at ntlworld.com  Sun Jun 25 12:20:23 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 25 Jun 2000 17:20:23 +0100
Subject: Reloading modules
References: <8j5atk$ijs$1@newsg4.svr.pol.co.uk>
Message-ID: <9jq55.1938$NE5.42685@news2-win.server.ntlworld.com>

Hi Duncan,

To reload something you need to have imported it in the first place. You've
not imported the module you've import the contents of the module. You could
probably do something like:

import Graph
reload(Graph)
from Graph import *

or go the more sensible route and use import Graph to begin with.

Richard


Duncan Smith <buzzard at urubu.freeserve.co.uk> wrote in message
news:8j5atk$ijs$1 at newsg4.svr.pol.co.uk...
> >>> from Graph import *
> .
> .
> .
> >>> reload(Graph)
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
> TypeError: reload() argument must be module
>
> What do I need to do to successfully reload a module?  I'm currently
wasting
> a lot of time re-starting PythonWin each time I want to test a changed
> module and I can't find the answer.  Thanks
>
> Duncan
>
>




From rjroy at takingcontrol.com  Tue Jun 13 19:03:29 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Tue, 13 Jun 2000 23:03:29 GMT
Subject: telnetlib strange behavior?
References: <8i39kg$fla$1@nnrp1.deja.com>
Message-ID: <3946bb29.508639203@news1.on.sympatico.ca>

On Mon, 12 Jun 2000 18:20:11 GMT, bragib at my-deja.com wrote:

>I am trying to telnet from a UNIX box to an NT to get the doskeys.
>The NT that I login to runs a couple of .bat scripts which clear the
>screen set up some stuff etc.  When I try to do a read_all() it fails
>the first time with the exception:
>Traceback (innermost last):
>  File "telnet.py", line 17, in ?
>    tn.read_all()
>  File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
>229, in read_all
>    self.fill_rawq()
>  File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
>367, in fill_rawq
>    buf = self.sock.recv(50)
>socket.error: (131, 'Connection reset by peer')
>
>
>That is why I put the first try in a try: except: clause.  The second
>time I try to do a read_all() it works fine but I don't get all the keys
>I am missing some of the first keys.  I assume that the first read_all()
>actually did read some stuff and then the second read got the rest.
>
>Can anyone see anything obviously wrong with this.
>
>
>Thanks, Bragi
>
>import sys
>import telnetlib
>HOST = 'shadow'
>USER = 'bragi'
>tn = telnetlib.Telnet(HOST)
>tn.open(HOST)
>tn.read_until("login: ")
>tn.write(USER+'\r')
>tn.read_until("password: ")
>tn.write('my_password\r')
>tn.write('doskey /macros\r')
>tn.write('exit\r')
>print 'done'
>try:
>    all1 = tn.read_all()
>except:
>    pass
>all = tn.read_all()
>print all
>

The documentation for telnetlib is confusing. What is does not state
is that if you supply a hostname when you instantiate an instance of
Telnet() that it will open the instance. What it does tell you however
is not to try to re-open a connected instance... (see the source code
for __init__ and the docstring for open.

You are doing :
>tn = telnetlib.Telnet(HOST)
>tn.open(HOST)

Perhaps that is causing the problem. I can't get telnetlib to work on
either of my machines.

Bob



From yawyi at gmx.de  Tue Jun 27 07:29:59 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Tue, 27 Jun 2000 13:29:59 +0200
Subject: chroot'ed demon process
References: <395727E3.D4C1388A@stroeder.com> <39575bc1$0$18726@personalnews.de.uu.net> <3957E2B2.C6A8CFEA@stroeder.com>
Message-ID: <3958a3ac$0$4202@personalnews.de.uu.net>

Michael Str?der <michael at stroeder.com> wrote:
>Johannes Stezenbach wrote:
>> 
>> Swigging chroot() or adding it to module posix would be trivial.
>
>Hmm.

Probably the not so trivial thing about adding chroot() to module os
is getting the autoconf stuff right.

But maybe someone from the Python community can be motivated to
add it anyway by mentioning that Perl has chroot()?

>With the chroot utility I have to put all files needed into the
>chroot jail including the Python interpreter. Well, it's a start but
>not really what I was looking for.

I guessed so.

Johannes



From msoulier at nortelnetworks.com  Mon Jun 26 13:37:19 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 26 Jun 2000 17:37:19 GMT
Subject: compilation
Message-ID: <8j84cf$6bj$1@bmerhc5e.ca.nortel.com>

	Hey people. I tried the FAQ, and I can't find this. I'm a Perl 
programmer exploring Python, and I'm wondering how to do a simple compile pass 
on python source to check syntax. In Perl, I'd just do a perl -cw sourcefile
and it would do a quick compile pass with warnings on. I know I can start the
interpreter and import the file as a module, but that's not always desirable.
Is there a simple command-line way to do this?

	Thanks,

	Mike

-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From bparsia at email.unc.edu  Mon Jun 19 14:15:16 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Mon, 19 Jun 2000 14:15:16 -0400
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu>

William Dandreta <wjdandreta at worldnet.att.net> wrote:

> Mike Fletcher wrote in message ...
> >
> >Also, don't discount the expense of an extra function call.  Function calls
> >were (and still are to a lesser extent) extremely heavy in Python five
> years
> >ago.
> 
> I did discount it because in assembly language a function call is little
> more than a jump. I thought that even if it were extremely inefficiently
> implimented in Python, it couldn't reduce speed by a factor of 6.

Oh you'd be surprised :)

> In
> addition, there are 3 function calls in each replace. Logic says eliminating
> one improve speed by 1/3.

Er.. no logic I know of :) Remember that "obvious" reasoning typically
leads you astray when thinking about optimization. The key, as I have
written, is the *lookup* of the method/function. These are *not*
resolved at compile time, but at run time. They are *not* cached (unless
you do so manually), they are performed from scratch each time.

> But on second thought, the one I eliminated was a call to a python.py
> library module, the other 2 function calls are to functions in strop, a
> compiled C 'module'. I'll have to test it to be sure.

Try saving the function to a local var and using it. I'd be interested
to see what happens.

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From skip at mojam.com  Tue Jun 27 14:45:46 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 27 Jun 2000 13:45:46 -0500 (CDT)
Subject: ANN: cgi-buffer 0.1
In-Reply-To: <962131217.3958f51128dbc@mail.mnot.net>
References: <14680.56174.25987.149715@beluga.mojam.com>
	<962131217.3958f51128dbc@mail.mnot.net>
Message-ID: <14680.63066.376896.558608@beluga.mojam.com>

    Mark> The question is, what do you mean by 'support HTTP/1.1'? That
    Mark> covers a lot of ground. My library probably won't be affected,
    Mark> because it improves things for 1.0 clients as well, and the 1.1
    Mark> mechanisms are probably not implemented. I'll be writing up a
    Mark> paper explaining this When I Get Time (ha).

I don't know all the details.  You can grab a CVS snapshot from SourceForge, 
however, and have a look at what's there now.

    >> 2. Python 1.6 will have a new module, atexit...

    Mark> COOL! I was pondering on that just the other day.

I'm not sure if it's been incorporated into the CVS tree yet.  I will send
you the latest copy without the python-list cc (I'm the author).  If anyone
else would like to see it before it gets into the CVS repository, drop me a
line.

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
On Phil Jackson's ability to manage multiple Lakers superstars, Shaquille
O'Neal said: "He's got the rings.  I listen to the man with the rings."



From mhammond at skippinet.com.au  Tue Jun 13 14:35:23 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 13 Jun 2000 18:35:23 GMT
Subject: Problems installing win32com
References: <3103135DA599D111B12B00805FE63E7703EBD24B@lonim26407.uk.ml.com>
Message-ID: <8i5uta$b1h$1@nnrp1.deja.com>

In article
<3103135DA599D111B12B00805FE63E7703EBD24B at lonim26407.uk.ml.com>,
  "Jones, Michael W (London)" <JonesMiw at exchange.uk.ml.com> wrote:
> When attempting to install the Python Interpreter COM server I get the
> following error??

I assume you got the bits from my starship page?  If so, the link
to "Installation Problems" should have been fairly obvious - did you
look?  Your problem is the first one listed!

Mark.


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


From andres at corrada.com  Thu Jun  8 21:51:10 2000
From: andres at corrada.com (andres at corrada.com)
Date: Thu, 8 Jun 2000 21:51:10 -0400
Subject: Why can't I restore my redirected stdout?
In-Reply-To: <TjW%4.9350$r7.1981176@news1.rdc2.tx.home.com>; from Rainer Deyke on Thu, Jun 08, 2000 at 11:51:15PM +0000
References: <8hoo05$fc$1@news5.jaring.my> <20000608150314.A26601@corrada.com> <TjW%4.9350$r7.1981176@news1.rdc2.tx.home.com>
Message-ID: <20000608215109.A25096@corrada.com>

On Thu, Jun 08, 2000 at 11:51:15PM +0000, Rainer Deyke wrote:
> 
> If sys.stdout and sys.__stdout__ point to the same object, and you assign
> some other object to sys.stdout, it is be possible to restore it by
> assigning sys.__stdout__ to sys.stdout.
> 

You are correct. My solution was appropriate but my explanation was not.

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From wjk at wjk.mv.com  Thu Jun 29 20:01:57 2000
From: wjk at wjk.mv.com (wjk)
Date: Thu, 29 Jun 2000 20:01:57 -0400
Subject: list.pop([i]) and list.remove(x) w/ for loops
References: <Pine.GSO.4.10.10006290852230.27161-100000@sundial>
Message-ID: <395BE374.EA6A81EF@wjk.mv.com>

Hi -- I usually use pop when iterating a list in reverse:
after i make a copy of it

y=[1,2,3,4,3,5,6]
copy=[:]

for x in range(len(y)-1,-1,-1):
     if y[x] == 3:
          y.pop(x)

I am unsure of list.remove(something) since I think it
removes the 1st occurance in a list thus iterating in
reverse might present a problem there as well.

other alternative:
lista = [1,2,3]
listb = [1,2,3,1,2,3,4,5,1,2,3]

for x in lista:
    try:
        while listb.index(x):
            listb.remove(x)
    except:
        pass

(kinda ugly huh?)
never thought of using "del listb[x]" that I will add somewhere
in my head..



Moshe Zadka wrote:

> On 28 Jun 2000, David White wrote:
>
> > I was getting some unexpected behavior when using list.remove(x), wondering if
> > someone can tell me what's going on.
> >
> > I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].  I wan't to
> > delete all the a objects out of the list while preserving the order of the b
> > objects.  My code looks like this:
> >
> > for o in olist:
> >     if (o.type == a):
> >        olist.remove(o)
>
> Short answer: never ever modify a list you're iterating on. Make a copy
> first.
>
> running-faster-is-pointless-if-you-get-the-wrong-answer-ly y'rs, Z.
>
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> http://www.oreilly.com/news/prescod_0300.html
> http://www.linux.org.il -- we put the penguin in .com



From offer at sgi.com  Wed Jun 28 19:31:25 2000
From: offer at sgi.com (Nhi Vanye)
Date: 28 Jun 2000 23:31:25 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <slrn8ll17c.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <8je1sd$3t3q8$1@fido.engr.sgi.com>

$ from NO_SPAM_hzhu at knowledgetrack_dot_com -#96726 | sed "1,$s/^/> /"
>
>
>
>>If you don't want python to search the current directory for imports,
>there is a
>>very simple solution: put this snippet above all of your other imports:
>>import sys
>>sys.path.remove('.')
>>
>
>This might be fine for personal use.  But I'm developing a general package
>that is being used by many people.  I don't think it's safe to fiddle with
>that.  And as I said, the functionality of packages can not (and absolutely
>should not) depend on the order of sys.path.  Otherwise packages expecting
>different ordering of sys.path cannot be used together.

It does on unix... If I type sh, the shell I get is going to completely
depend on what my path is.

If I want (as a user) to control which version I want I can use
/bin/sh. As an system admin I can't enforce that that is the only version
of sh on the system.

The same logic applies with python imports.

>
>
>Huaiyu

richard.
-- 
Richard Offer           Widget FAQ --> http://reality.sgi.com/widgetFAQ
MTS-Core Design (Motif)
___________________________________________http://reality.sgi.com/offer



From anders.eriksson at morateknikutveckling.se  Wed Jun 21 04:30:20 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 21 Jun 2000 10:30:20 +0200
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com>

On Wed, 21 Jun 2000 00:40:37 GMT, "William Dandreta"
<wjdandreta at worldnet.att.net> wrote:

>Hi Anders,
>
>I am running uner DOS. Any version will be fine, even older ones as long as
>I can index it.
>
DOS! ;-)

Well, AFAIK there isn't any ODBC in DOS. So you need something else...

Do you REALLY need to use dBASE? Are you interacting with some other
programs that uses the same database?

There is a commercial C library called CodeBase that can do the things
you want, and I still think they have a DOS version. 

http://www.sequiter.com/products/cbase6.htm

Maybe you could use SWIG to create a Python module for using it.
If you're not in a hurry I could maybe do it, but not until late
august.

// Anders



From wjdandreta at worldnet.att.net  Wed Jun 28 11:53:58 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 28 Jun 2000 15:53:58 GMT
Subject: PyType_Type?
References: <Ebo45.13372$Uw3.865124@bgtnsc04-news.ops.worldnet.att.net> <slrn8l9p1v.elg.thomas@centurion.xs4all.nl>
Message-ID: <qcp65.13064$kK.349754@bgtnsc05-news.ops.worldnet.att.net>

Hi Thomas,

>The type of a type object is not defined in a module, but in an Object
file.
>'Objects/typeobject.c' to be exact.
>
>Do you get any other error messages ? What version of Python ?

Sorry for the module confusion, my C compiler (BC4.5) refers to .c files as
modules .

I am trying to compile Python 1.5.2 for DOS without a DOS extender. To do
that I need to use overlays. I started out by including all the .c files in
the source and made several changes and deletions. Currently I am left with
the following modules, which all compile without errors and four linking
errors:

Linking pyth152.exe:
Linker Error: Undefined symbol _initstruct in module SOURCE\PC\CONFIG.C
Linker Error: Undefined symbol _initsignal in module SOURCE\PC\CONFIG.C
Linker Error: Undefined symbol _initnt in module SOURCE\PC\CONFIG.C
Linker Error: Undefined symbol _crypt in module SOURCE\MODULES\CRYPTM~1.C

Any suggestions you might make would be appreciated.

Bill
----------------------------------------------------------
Compiling source\python\SIGCHECK.C:
Compiling source\python\MYSTRT~1.C:
Compiling source\python\STRUCT~1.C:
Compiling source\python\TRACEB~1.C:
Compiling source\python\PYSTATE.C:
Compiling source\python\SYSMOD~1.C:
Compiling source\python\PYTHON~1.C:
Compiling source\python\MODSUP~1.C:
Compiling source\python\ATOF.C:
Compiling source\python\BLTINM~1.C:
Compiling source\python\CEVAL.C:
Compiling source\python\COMPILE.C:
Compiling source\python\ERRORS.C:
Compiling source\python\FMOD.C:
Compiling source\python\FROZEN.C:
Compiling source\python\FROZEN~1.C:
Compiling source\python\GETARGS.C:
Compiling source\python\GETCOM~1.C:
Compiling source\python\GETCOP~1.C:
Compiling source\python\GETMTIME.C:
Compiling source\python\GETOPT.C:
Compiling source\python\GETPLA~1.C:
Compiling source\python\GETVER~1.C:
Compiling source\python\GRAMINIT.C:
Compiling source\python\HYPOT.C:
Compiling source\python\IMPORT.C:
Compiling source\python\IMPORTDL.C:
Compiling source\python\MARSHAL.C:
Compiling source\python\MEMMOVE.C:
Compiling source\pc\CONFIG.C:
Compiling source\pc\GETPATHP.C:
Compiling source\parser\ACCELER.C:
Compiling source\parser\BITSET.C:
Compiling source\parser\FIRSTS~1.C:
Compiling source\parser\GRAMMAR.C:
Compiling source\parser\GRAMMAR1.C:
Compiling source\parser\INTRCH~1.C:
Compiling source\parser\LISTNODE.C:
Compiling source\parser\METAGR~1.C:
Compiling source\parser\MYREAD~1.C:
Compiling source\parser\NODE.C:
Compiling source\parser\PARSER.C:
Compiling source\parser\PARSETOK.C:
Compiling source\parser\PGEN.C:
Compiling source\parser\PRINTG~1.C:
Compiling source\parser\TOKENI~1.C:
Compiling source\objects\ABSTRACT.C:
Compiling source\objects\BUFFER~1.C:
Compiling source\objects\CLASSO~1.C:
Compiling source\objects\COBJECT.C:
Compiling source\objects\COMPLE~1.C:
Compiling source\objects\DICTOB~1.C:
Compiling source\objects\FILEOB~1.C:
Compiling source\objects\FLOATO~1.C:
Compiling source\objects\FRAMEO~1.C:
Compiling source\objects\FUNCOB~1.C:
Compiling source\objects\INTOBJ~1.C:
Compiling source\objects\LISTOB~1.C:
Compiling source\objects\LONGOB~1.C:
Compiling source\objects\METHOD~1.C:
Compiling source\objects\MODULE~1.C:
Compiling source\objects\OBJECT.C:
Compiling source\objects\RANGEO~1.C:
Compiling source\objects\SLICEO~1.C:
Compiling source\objects\STRING~1.C:
Compiling source\objects\TUPLEO~1.C:
Compiling source\objects\TYPEOB~1.C:
Compiling source\objects\XXOBJECT.C:
Compiling source\modules\_LOCAL~1.C:
Compiling source\modules\ARRAYM~1.C:
Compiling source\modules\AUDIOOP.C:
Compiling source\modules\BINASCII.C:
Compiling source\modules\CMATHM~1.C:
Compiling source\modules\CPICKLE.C:
Compiling source\modules\CRYPTM~1.C:
Compiling source\modules\CSTRIN~1.C:
Compiling source\modules\ERRNOM~1.C:
Compiling source\modules\FPECTL~1.C:
Compiling source\modules\FPETES~1.C:
Compiling source\modules\GETBUI~1.C:
Compiling source\modules\IMAGEOP.C:
Compiling source\modules\MAIN.C:
Compiling source\modules\MATHMO~1.C:
Compiling source\modules\MD5C.C:
Compiling source\modules\MD5MOD~1.C:
Compiling source\modules\NEWMOD~1.C:
Compiling source\modules\OPERATOR.C:
Compiling source\modules\PARSER~1.C:
Compiling source\modules\PCREMO~1.C:
Compiling source\modules\PUREMO~1.C:
Compiling source\modules\PYPCRE.C:
Compiling source\modules\PYTHON.C:
Compiling source\modules\REGEXM~1.C:
Compiling source\modules\REGEXPR.C:
Compiling source\modules\RGBIMG~1.C:
Compiling source\modules\ROTORM~1.C:
Compiling source\modules\SHAMOD~1.C:
Compiling source\modules\SOUNDEX.C:
Compiling source\modules\STROPM~1.C:
Compiling source\modules\TIMEMO~1.C:
Compiling source\modules\XXMODULE.C:
Compiling source\modules\YUVCON~1.C:





From rturpin at my-deja.com  Sun Jun  4 23:50:49 2000
From: rturpin at my-deja.com (rturpin at my-deja.com)
Date: Mon, 05 Jun 2000 03:50:49 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com> <8F49E9873gmcmhypernetcom@199.171.54.154>
Message-ID: <8hf82l$87p$1@nnrp1.deja.com>

In article <8F49E9873gmcmhypernetcom at 199.171.54.154>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> I suspect you haven't used many repetitive slicing
> algorithms. It boils down to being nice because
>
>  lst == lst[:x] + lst[x:]

Or to put it another way, the same value serves as
both the end of the previous segment, and the start
of the next. This is really, REALLY useful when
doing list algorithms.

Praising-this-neat-convention-of-Python-ly yrs,
Russell


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


From digitig at cix.co.uk  Mon Jun 26 17:10:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 26 Jun 2000 22:10 +0100 (BST)
Subject: boolean true and false values.
References: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <memo.20000626221037.43263C@digitig.compulink.co.uk>

In article <slrn8lf3q9.us.hat at se-46.wpa.wtb.tue.nl>, 
hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:

> I understand that now.
> However, I am not quite sure that that is what one wants.

Maybe it isn't; then another language may be more appropriate. When it 
/is/ what one wants then Python may be your choice.

Folks don't seem to be able to get away from the notion of a "perfect" 
language, ideal for all applications. I don't believe it exists, and if 
you "improve" Python for some you may well break it for others. Surely the 
competent programmer has a number of languages under his/her belt?

> >Sounds like you're making a case that all newbies should be taught 
> haskell or
> >scheme.  After all, invariants should be preserved.  The best way to 
> do that
> >is to make them not vary.
> 
> I think I am (is that bad ? :-) ).

No, not at all, IMHO, though I'd suggest adding Eiffel to the list. The 
first language I was taught was FORTRAN and would consider INTERCAL to be 
an improvement over /that/! I suspect teaching Python to newbies may be a 
bit like introducing people to basic woodwork by giving them a router 
before they know how to use a saw or chisel!

> In other words, variables are in reality quite tightly connected to 
> their
> values because a meaning is given to the variable.
> For example, pointing the variable to a value of another type does not 
> fit in
> the idea of invariants. For example, by executing 'total="kjerg"', the
> invariant is not longer true, since "kjerg" is not the sum of elements 
> (it is
> not even a number).

But that's /exactly/ right! If you execute 'total="kjerg"' then your 
invariant is indeed violated!

> Thus, to a good programmer, the dynamic nature of variables is not 
> needed. To
> bad programmers, it is a tool which can be used quite abusively.

As can any tool, of course. The dynamic typing is indeed never /needed/ 
(probably one of the things Dijkstra proved!), but it's often very /very/ 
convenient indeed. If it's likely to be a problem to you -- or to the 
newbies, use a different language -- but please don't try to take it away 
from those of us who want it.


From s2mdalle at titan.vcu.edu  Wed Jun 21 17:58:33 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Wed, 21 Jun 2000 16:58:33 -0500
Subject: Python sockets/select
References: <8ip6kt$1lc$1@bob.news.rcn.net> <8ipl0u$9l3$0@216.39.151.169> <20000621082622.G26436@xs4all.nl>
Message-ID: <8ire01$dip$1@bob.news.rcn.net>

> Actually, there are some possibilities to get 0 bytes on a return
> without hitting EOF: if you are using a wrapper object that possibly
> eats up some bytes (for protocol negotiation, for instance.) Telnetlib
> is a good example of this:

Hmm.  That makes sense.  The information sent over the network could
be > 0 bytes, but nothing but headers, no "body" data, right?

> If you do something similar in your own sockets, this may well be the
> case.

I'll have to reexamine my code very carefully.  There is no protocol
stripping going on at all. 
-- 
David Allen
http://opop.nols.com/
----------------------------------------
Can't sleep...Clown will eat me...



From jkraska1 at san.rr.com  Wed Jun  7 22:02:06 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 08 Jun 2000 02:02:06 GMT
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
Message-ID: <393EFF38.C023F10F@san.rr.com>

> 1. Everything is an object of a class that can be subclassed.

You want to subclass integers, why?
 
> 2. Augmentation operators: +=, *=, etc.  Dot operators .*, ./, etc (for
>    numerical computation to distinguish matrix and elementwise operations).

Fair enough, albeit I can see the dot operators not selling.
 
> 3. The := operator for deepcopy as oposed to reference assignment.  Therefore
>    - the assignment (a = b) guarantees (a is b)
>    - the assignment (a := b) guarantees (a == b and a is not b).

Hard. Deep copy isn't trivial, although as long as python dispatches
to external modules which don't implement deep copy fail gracefully,
I suppose this is all right.
 
> 6. Have a clean mechanism to work with other programming languages other
> than C.  (I have matlab/octave in mind.)

What would you propose, exactly?

> 7. NOT case insensitive.  (No followup on this please.  Enough said already.)

The internet doesn't work like that. But I agree. :)
 



C/


From wjdandreta at worldnet.att.net  Fri Jun 23 11:23:42 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 23 Jun 2000 15:23:42 GMT
Subject: Python and HTML forms?
Message-ID: <2iL45.19443$Xx5.1074706@bgtnsc06-news.ops.worldnet.att.net>

We use a web based DB that is updated with HTML forms. If a large number of
updates are needed it takes a long time to type all the data into the forms.

Can Python be used to automate the updating?

If so, what do I need to know to do it?

Bill




From mwh21 at cam.ac.uk  Sat Jun  3 18:25:47 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Jun 2000 23:25:47 +0100
Subject: Automatic reload()?
References: <8hbnt1$62i$1@news2.isis.unc.edu> <m3r9aexydb.fsf@atrus.jesus.cam.ac.uk> <04e301bfcdbe$81ff0f40$6401a8c0@home>
Message-ID: <m3og5ixuz8.fsf@atrus.jesus.cam.ac.uk>

"Darrell Gallion" <darrell at dorb.com> writes:

> Michael Hudson wrote:
> 
> > def rr(keepem=sys.modules.keys()):
> >     for i in sys.modules.keys():
> >         if type(sys.modules[i]) is types.ModuleType and i not in keepem:
> >             print "reloading",i
> >             reload(sys.modules[i])
> >
> > I can't remember quite what the "keepem" bit's for, but I presume
> > there was some reason I put it in...
> >
> Just guessing..
> keepem will have the current list of modules, when this module is run. So
> new modules will be reloaded because they aren't in that list.

That's what it does, yes.  But ISTR something broke if it wasn't
there... ah yes, this is it:

>>> def rr(keepem=[]):
...     for i in sys.modules.keys():
...         if type(sys.modules[i]) is types.ModuleType and i not in keepem:
...             print "reloading",i
...             reload(sys.modules[i])
... 
>>> rr()
[snip]
reloading __main__
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 5, in rr
ImportError: Cannot re-init internal module __main__

Cheers,
M.

-- 
42. You can measure a programmer's perspective by noting his 
    attitude on the continuing vitality of FORTRAN.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From bparsia at email.unc.edu  Thu Jun  8 01:37:04 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 8 Jun 2000 01:37:04 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <7wc%4.7928$r7.1560935@news1.rdc2.tx.home.com> <393DDDF8.DE0401BB@sightreader.com>
Message-ID: <1ebvvjt.1u66b4uho5fk0N%bparsia@email.unc.edu>

Ken Seehof <12klat at sightreader.com> wrote:

[snip]
> Because of it's dynamic nature, Python simply can't be optimized all that much
[snip]

A common misunderstanding. Python is no more dynamic than Smalltalk (and
several others languages). I'm perfectly willing to bet that, for
example, a method cache would speed lookups enormously. This is hardly a
difficult or novel technology. Once we start talking Self/Hotspot stuff
the possible gains are higher.

IIRC, Python 1.5 approximate doubled the speed of Python 1.4. But surely
Python 1.4 was no *less* dynamic that Python 1.5! So why couldn't one
expect further doublings?

(Of course, there may be *implementation* details, like avoiding certain
complexities in the VM, embedding the interpreter issues, and making C
extentions simple, which make various sorts of optimizations undesirable
and that, for it's user base, that's ok.)

Heck, it just might be that average appication speed is better served by
strategic C extentions rather than improving dispatching. If you have a
lot of C modules (as Python does) and a large group of C programmers
churning them out (as Python does), it may be the case that *most*
typical programs end up in C code for critical stuff anyway.

Of course, the lack of speed *encourages* writing code that is more
likely to avoid dispatch cost, etc.

Er...but my main point is that dynamic nature alone doesn't entail
"pure" Python's "slowless". Whether you feel you have to apologize for
this is up to you :)

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From nikolai.kirsebom at siemens.no  Mon Jun 19 10:23:32 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Mon, 19 Jun 2000 14:23:32 GMT
Subject: win32all-132 / Word Automation
Message-ID: <394e2b8d.515788975@news.mch.sni.de>

I've installed win32all-132 (on python 1.5.2) and it seems to work
partially.  I've developped a small module I use to test a Word/VBA
project.  When I run my scripts now, it seems to fail in an area
related to the object model of document (Word 97).  The commands
listed below produces the enclosed error message.

-----------------------
Commands:
>>> import win32com.client
>>> apl = win32com.client.Dispatch("Word.Application")
>>> apl.Visible = 1
>>> doc = apl.Documents.Add()
>>> doc.Variables.Add(Name="abc", Value="def")
<win32com.gen_py.Microsoft Word 8.0 Object Library.Variable>
>>> print doc.Variables.Item(1).Name
-------------------------
Error Result:

Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "D:\Python\win32com\client\__init__.py", line 347, in
__getattr__
    return apply(self._ApplyTypes_, args)
  File "D:\Python\win32com\client\__init__.py", line 341, in
_ApplyTypes_
    return self._get_good_object_(apply(self._oleobj_.InvokeTypes,
(dispid, 0, wFlags, retType, argTypes) + args), user, resultCLSID)
AttributeError: 'tuple' object has no attribute 'InvokeTypes'

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

Using the VBA debugger I'm able to inspect the variable - so it has
successfully been set.

Does anyone know what I'm missing.

Nikolai



From mwh21 at cam.ac.uk  Wed Jun  7 10:03:11 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 07 Jun 2000 15:03:11 +0100
Subject: __getattr__ and __setattr__ troubles
References: <8gehq1$kq6$1@news1.xs4all.nl> <8ggo6m$2v6h$1@news.aros.net> <oq1z2ad4wi.fsf@titan.progiciels-bpi.ca> <m3u2f6wowf.fsf@atrus.jesus.cam.ac.uk> <oqem6abcpf.fsf@titan.progiciels-bpi.ca>
Message-ID: <m3itvlwpuo.fsf@atrus.jesus.cam.ac.uk>

=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> writes:

[snip] 

> > > * By the way, and a bit unrelated, why the space before the two
> > >   asterisks?
> 
> > Probably to do with softspace.  Not sure.
> 
> It should go between printed element, not before the first one,
> should it not?

Here's how it works:

when you execute

print C()

the code for PRINT_ITEM pops the object off the stack, and calls

PyFile_SoftSpace(w, 1)

where w is sys.stdout.  This checks to see if a softspace is required,
which it isn't (so it returns 0, and a space isn't printed) and then
sets sys.stdout.softspace to 1 (that's what the "1" parameter above
is).

Then the code in ceval.c calls PyFile_WriteObject with the C() object
which winds up calling your __getattr__ method which calls "print
'__str__'"; this gets to the point just mentioned in the code for
PRINT_ITEM, but this time when it calls PyFile_Softspace,
sys.stdout.softspace *is* 1, so it returns 1 and a space is printed,
followed by the string '__str__'; then the PRINT_NEWLINE that follows
sets sys.stdout.softspace to 0, so if you're __getattr__ method
returned appropriately (ie. with a callable or by raising
AttributeError), the printed representation of C() would appear on a
line on it's own, without a space.  But it returns None, so
PyObject_Print blows up, and you get a traceback instead.

Does that help?

> > Your wish is my command [...]  Not sure it's a sufficiently common error
> > to justify this though.  I'll send it to patches at python.org and see what
> > they think of it.
> 
> I hope they will accept them.  Each and every diagnostic should be as
> clear as possible, and _especially_ those diagnostics which are uncommon!

No word yet.  Not too surprising, really...

> > Overriding __getattr__ is good for "oo, ain't that cool" value, but it
> > is probably best to try to avoid it - if you need it, fine, but consider
> > alternatives first.
> 
> Of course, a "need" is a pretty subjective matter.  We can do a great
> deal with very little, and there are not many so-called needs that would
> really stand scrutiny.  In my case, I have a working program already,
> and implementing __getattr__ is part of an intermediary cleanup, before
> attempting to move further, so to simplify code all over and gain better
> maintainability.  So in a sense, I could say that I considered at least one
> alternative, which is already implemented and working.  I think that the
> increase in legibility is worth the slowdown, which the project can afford.

Fair enough.  
 
> P.S. - I hope that Guido will do his best so future Python stays legible.

Out of curiosity, here do you mean Python source, or the C source of
Python?  Both are almost uniquely readable, but which did you mean?

Cheers,
M.

-- 
  Just point your web browser at http://www.python.org/search/ and
  look for "program", "doesn't", "work", or "my". Whenever you find
  someone else whose program didn't work, don't do what they
  did. Repeat as needed.    -- Tim Peters, on python-help, 16 Jun 1998


From aahz at netcom.com  Thu Jun  1 13:21:33 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Jun 2000 17:21:33 GMT
Subject: Interprocess comunication
References: <8F456BAC0lminutieidit@192.6.1.10> <3934f634.704981920@news.u-psud.fr> <8F45B7D7Clminutieidit@192.6.1.10> <39366A38.262E0E07@earth.ox.ac.uk>
Message-ID: <8h662t$avd$1@slb0.atl.mindspring.net>

In article <39366A38.262E0E07 at earth.ox.ac.uk>,
Nick Belshaw  <nickb at earth.ox.ac.uk> wrote:
>
>You might want to try the Medusa stuff  - asyncore/asynchat.py - in the
>standard library.
>
>These modules allow fully asynchronous comms between independant progs on
>the same or different machines via  pc-name:port-number channels. - pretty
>powerful stuff.

I've gotten out of the habit of recommending asyncore except for
explicit server applications.  The problem is that you get blocks in
your application unless you either thread the asyncore part or make sure
that everything takes at most a few seconds to process.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From just at letterror.com  Tue Jun 20 04:19:06 2000
From: just at letterror.com (Just van Rossum)
Date: Tue, 20 Jun 2000 09:19:06 +0100
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <XFMail.000620085230.mikael@isy.liu.se>
References: <l03102800b574c9a15f7a@[193.78.237.171]>
Message-ID: <l03102802b574d8b1eb10@[193.78.237.135]>

At 8:52 AM +0200 20-06-2000, Mikael Olofsson wrote:
>On 20-Jun-00 Just van Rossum wrote:
> >  (Due to historic narrowmindedness, a point is the same as a pixel under
> >  MacOS, no matter what physical resolution the display has; I believe
> >  Windows has (had?) the same problem, except that a pixel is 72/96th
> >  points...)
>
>Erm, isn't a pt as in 10 pt font the typographic unit 1/72 inch?

That's what it indeed should be. But the dumb thing is that the logical
resolution of MacOS display devices is hard-wired to 72dpi... (Just like
Windows' logical resolution is hard-wired to 96dpi. Does anyone know if
Win2000 fixes this?)

Just





From petersc at stud.ntnu.no  Tue Jun 13 11:21:10 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 13 Jun 2000 17:21:10 +0200
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no> <3946232F.61B28793@sightreader.com>
Message-ID: <39465166.6A7E261A@stud.ntnu.no>

Ken Seehof wrote:
> 
> This is a portability feature.  I'm pretty sure the converse, backslashes
> in unix, works too.

Does not work for me.

but-then-I-use-linux-and-not-unix-ly y'rs
Peter

--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From erikd at zip.com.au  Sat Jun 10 01:23:41 2000
From: erikd at zip.com.au (Erik de Castro Lopo)
Date: Sat, 10 Jun 2000 05:23:41 +0000
Subject: SOLVED (was Re: Distutils: Problem with install_path?)
References: <3941CC9A.7E88788B@zip.com.au>
Message-ID: <3941D0DD.3D41E114@zip.com.au>

Erik de Castro Lopo wrote:
> 
> Hi all,
> 
> I've got Python 1.5.2 installed in /usr/local and I'm trying
> to install NumPy.

That ALWAYS happens. As soon as I post a question to Usenet I 
figure it our for myself 30 minutes later.

My problem was caused by an older version of NumPy (ver14)
which required an older version of DistUtils. 

The combination of DistUtils-0.8.2 and Numerical-15.3 worked
perfectly.

Erik
-- 
+-------------------------------------------------+
     Erik de Castro Lopo     erikd at zip.com.au
+-------------------------------------------------+
     Customer:  "I'm running Windows '95.
Tech. Support:  "I see."
     Customer:  "My computer isn't working now."
Tech. Support:  "Yes, you already said that."


From morten at src.no  Thu Jun 29 08:19:33 2000
From: morten at src.no (Morten W. Petersen)
Date: Thu, 29 Jun 2000 14:19:33 +0200
Subject: Deleting mails with imapib
In-Reply-To: <8itd35$a7lc$1@nntp6.u.washington.edu>
References: <Pine.LNX.4.21.0006221223060.22204-100000@lade.src.no> <8itd35$a7lc$1@nntp6.u.washington.edu>
Message-ID: <Pine.LNX.4.21.0006291419090.25358-100000@lade.src.no>

> | The imaplib.IMAP connection doesn't seem to have a method that lets you
> | delete mails? only expunge etc..
> 
> Of course expunge is what really deletes them - the delete option
> in the mailer just sets a flag on the message.  Here's how I do it -
> 
>     ok, stuff = self.service.store(number, 'FLAGS', '(\Deleted)')
>     if ok != 'OK':
>         raise self.error(stuff)

Where did you take the self.service from?

-Morten



From tjg at exceptionalminds.com  Thu Jun  8 18:51:08 2000
From: tjg at exceptionalminds.com (Timothy Grant)
Date: Thu, 8 Jun 2000 15:51:08 -0700
Subject: Pythonwin is terrible!
In-Reply-To: <8hp2ku$a25@news.or.intel.com>; from no_spam@umich.edu on Thu, Jun 08, 2000 at 02:19:23PM -0700
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <20000608155108.A8398@reepicheep.avalongroup.net>

On Thu, Jun 08, 2000 at 02:19:23PM -0700, Jeff wrote:
> 
> If there are no answers to these questions, someone needs to work on
> pythonwin's "development" capabilities.

I vote for you to do it. 

-- 
Stand Fast,
    tjg.

Timothy Grant                         tjg at exceptionalminds.com
Chief Technology Officer              www.exceptionalminds.com
Red Hat Certified Engineer                           MIG #1433
Avalon Technology Group, Inc.           
>>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<



From cut_me_out at hotmail.com  Mon Jun 26 21:08:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Jun 2000 21:08:54 -0400
Subject: stopping threads
References: <3957C1DC.E7C77E2E@sec.noaa.gov> <8j8id2$v3h$1@nntp9.atl.mindspring.net>
Message-ID: <etdg0q0hqvt.fsf@w20-575-109.mit.edu>

> I'm currently reading up on Java threads, and while such a function
> exists, the authors recommend very strongly against doing so.
> Instead, your thread should watch for some condition and return/exit
> from its main function/method when that condition becomes true.

Do they say why?

Alex.


From liwen_cao at my-deja.com  Wed Jun 14 05:56:12 2000
From: liwen_cao at my-deja.com (liwen_cao at my-deja.com)
Date: Wed, 14 Jun 2000 09:56:12 GMT
Subject: [Q:] hash table performance!
Message-ID: <8i7kro$i04$1@nnrp1.deja.com>

Greetings,

I'm doing a project on large volume information processing. One of the
tasks is to find out the duplicated files under a directory. I believe
Python would be a good powerful tool for that(yes it is, I've
implemented it in 40 lines of codes). However, performance IS a
problem! Since I'm using directory as the hash table (hash_table={}...),
I doubt the bottle neck is in the hash table: How can the generic hash
table fits every cases?

Is there a way of customize the length, hash algorithm of the hash
table in python?  Or anyone can describe how the python hash table
works.

Thanks in advance and any hints will be helpful.

P.S. Background of duplicate files check:
My way of doing it is simple: walk the directories and files, do MD5
coding for every file, use the MD5 code as the hash key, insert the
file name into the hash table. When two files has the same key, check
the contents by bytes.


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


From mhammond at skippinet.com.au  Tue Jun  6 04:57:51 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Tue, 06 Jun 2000 08:57:51 GMT
Subject: embedding pythonwin: deadlock on PyWinInterpreterLock_Acquire
References: <ECEPKNMJLHAPFFJHDOJBOEEKCKAA.mhammond@skippinet.com.au> <393BC9EA.95B2442F@gmx.net>
Message-ID: <j23%4.6581$Hz.48206@news-server.bigpond.net.au>

*sigh*  I didnt notice the CC to the list, so I replied in personal mail
:-(

"Robert Kiendl" <rkiendl at gmx.net> wrote in message
news:393BC9EA.95B2442F at gmx.net...

> shouldn't there be code in PyWinInterpreterLock_Acquire/Release to take
> care if the same windows thread just does a recursion Win -> Python -> C
> (->Win) -> Python.
> or am I completely wrong?

The short answer is that Pythonwin/CEnterLeavePython can not handle this -
it is the responsibility of the caller (ie, the extension module author)
to ensure they release the lock that could cause a callback to in-turn
attempt to acquire it.  This is already done extensively in Pythonwin,
COM, etc.

Mark.





From kellyk at my-deja.com  Fri Jun 23 16:17:57 2000
From: kellyk at my-deja.com (kellyk at my-deja.com)
Date: Fri, 23 Jun 2000 20:17:57 GMT
Subject: CurrentUser for remote machine under NT
References: <2B1262E83448D211AE4B00A0C9D61B0301301217@msgeuro1.creo.be>
Message-ID: <8j0gl2$i9k$1@nnrp1.deja.com>

In article <2B1262E83448D211AE4B00A0C9D61B0301301217 at msgeuro1.creo.be>,
  Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> > Hello,
> >
> > how can you get the current logged in user for a remote nt system
using Python?
> >
> > Pieter
>

You can try this. It does not always work.

import win32pipe, string

workstation_name= "machinename"
workstation_userid= ""
lines= win32pipe.popen("nbtstat -a " + workstation_name).readlines()
for line in lines:
  fields= string.split(line)
  if len(fields) < 3:
    continue
  if fields[1] == "<03>" and string.upper(workstation_name) !=
fields[0]:
    workstation_userid= fields[0]
# prints blank if could not find
print workstation_userid


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


From fiona at sitegnome.com  Thu Jun 15 09:49:27 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 15 Jun 2000 13:49:27 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 15th, 2000
Message-ID: <20000615134927.5296.qmail@synop.com>

Hi Guys,

Below are the entries that made it into http://python.faqts.com tonight.

Cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Which linux distros have Python by default?
Is there a handy list somewhere of which Linux distributions can be expected to have Tkinter installed?
http://www.faqts.com/knowledge-base/view.phtml/aid/3774
-------------------------------------------------------------
Fiona Czuczman
Thomas Weholt, John W. Baxter, William Park, Michael Str?der, Fran?ois Pinard, Dana Booth

Mandrake 7.0 and 7.1 has it installed, and 7.1 has PIL included too. A
very good distro.

RedHat 6.1 and 6.2 install Python, at least in the way we install it.

Slackware-7.0 has Python package in D series.

S.u.S.E. comes with packages for Python, Tkinter and other handy Python 
modules. Well, the person who installs it has to choose it in the 
install application called YaST (series d).

[Yes, but you choose Python explicitly only for simpler profiles.
Python gets installed automatically in more sophisticated profiles.
If you tune a simple profile yourself (which is what I usually do), you
merely confirm once the installation of Python, when a dependency of any
package you add to the profile.  This is more and more likely, as Python
gets more often needed in packages.  In my last SuSE installation, a few
days ago, I did not have to explictly select neither Python nor 
`pygtk'.]

The new Debian potato has also packages for Python and several modules.

-----------

It's best to choose not to put on Python at install time, and then just
retrieve the latest version from the Internet. Uncomment the lines in 
the Modules/Setup file before you build it pertaining to Tkinter, then 
you're assured of having the latest version.

Of course, you'd need to make sure that you have TCL/TK installed... But
then, you should do that yourself, too.

This way makes it easier to keep track of when you want to upgrade. By 
doing it yourself, you know exactly where the files went. When you 
upgrade after an auto install, you don't know if the distribution's 
install put the files in weird places, so that you'll have conflicting 
crap all over your drive. For instance, I installed Mandrake 6.0 once, 
and it put a ton of KDE junk in /usr/bin. What a stupid place, and what 
a clutter. With Python, if you follow the configuration file defaults 
before you make, it'll always be nice and cozy in 
/usr/local/lib/Pythonxx. Wanna upgrade? You can just move the old 
directory out of the way, and then move your homemade modules directory 
back once you've put a new version in.


-------------------------------------------------------------
I'm searching information regarding the use of pointers for linked and double linked lists.
http://www.faqts.com/knowledge-base/view.phtml/aid/3770
-------------------------------------------------------------
Fiona Czuczman
Martijn Faassen

Linked lists aren't hard:

class Node:
    def __init__(self, next):
        self.next = next

linked_list = Node(Node(Node()))

Every name in Python's a reference, so no pointers are needed. Just 
think of every name in Python as a pointer (to an object), if you like.

Doubly linked lists are along the same pattern, but have the trouble 
that they introduce circular references, which is bad for Python's 
reference counting based garbage collection scheme. You have to break 
one of the references yourself for it to work:

class Node:
    def __init__(self, prev, next):
        self.prev = prev
        self.next = next

node1 = Node(None, None)
node2 = Node(None, None)
node1.next = node2
node2.prev = node1

# and now to clean up so that refcounting works:
node2.prev = None


-------------------------------------------------------------
Where can I find info on combining C, Assembler, and Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3771
-------------------------------------------------------------
Fiona Czuczman
Martijn Faassen

Take a look here:

Extending and Embedding the Python Interpreter

http://www.python.org/doc/current/ext/ext.html

And here:

Python/C API

http://www.python.org/doc/current/api/api.html

These are part of the standard Python documentation. You'll have to do
the assembler calls yourself, from C.


## Edited Entries ##############################################


-------------------------------------------------------------
Where can I best learn how to parse out both HTML and Javascript tags to extract text from a page?
http://www.faqts.com/knowledge-base/view.phtml/aid/3680
-------------------------------------------------------------
Paul Allopenna, Matthew Schinckel
Python Documentation

If you want to (quickly) strip all HTML tags from a string of data, try using the 
re module:

import re

file = open(filename,'r')
data = file.read()
file.close()

text = re.sub('<.*?>', '', data))

This will also strip any javascript, but only if the page has been made 'properly' 
- that is, the javascript is within HTML comments.


If you want to know how it works, read the 're' chapter in the library reference, 
as it discusses the usefulness of 'non-greedy' regular expressions.


-------------------------------------------------------------
Is there a HTML search engine written in Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3105
-------------------------------------------------------------
Fiona Czuczman, Matthew Schinckel
Dale Strickland-Clark, Michal Wallace, Robert Roy, JRHoldem

If you're running it on NT, there's a free search engine you can tap 
into that comes as part of the NT 4.0 option pack.

Otherwise:

Check out http://ransacker.sourceforge.net/ .. There's an Index
class that lets you index arbitrary chunks of text.. But you'll have
to write the program that actually reads the HTML files (and strips
the HTML tags, if that's what you mean by "text content")... 

It also does a ranked searches, but you'll have to wrap that, too, if
you want the output to show up on the web.

A full featured full text indexing solution is not trivial. It all
depends on what kind of queries you want to perform. If all you want
to do are queries such as "find all files which contain the word
'dog'" that can be done quite easily, probably under 200  lines of
code for a trivial solution using sgmllib and gdbm. However if you
want to do phrase searching or stem searching or wild-card searching,
then it gets really complicated in a hurry. 

Another factor is how many files you are dealing with. Indices often
run 4-8X the size of the indexed files. And do you want to dynamically
update the index or are you happy just re-indexing the whole works
periodically. A static index is somewhat easier to build than a fully
dynamic one.

An interesting GPL'd indexing package is SWISH++
see:
http://www.best.com/~pjl/software/swish/

A good tactic might be to use this for your indexing, and running the
search engine as a daemon, building a python interface to talk to it
via Unix domain sockets or alternately shelling out and capturing and
parsing the return values.

You also might want to try using Index Server/ASP combo before going to 
any third party solution...full text searching is no trivial matter and 
chances are it'll give you all the tinkering options you could want.

Additionally:

There is a really simple search engine (single word, really only works with small 
sites), available:

<http://www.chariot.net.au/~jaq/matt/search.tar.gz>

(or look on Parnassus if it's moved :-)


-------------------------------------------------------------
Can I combine a "select" call on some of my file objects with the Tkinter event loop?
http://www.faqts.com/knowledge-base/view.phtml/aid/3728
-------------------------------------------------------------
Rob Hooft, Fiona Czuczman
Grant Edwards, Russell E. Owen

Perhaps one can use file events instead of select. Here is a recent 
exchange that may be relevant -- my initial posted question followed by 
Grant Edwards' detailed and helpful reply. (Note: his snippet has some 
unix-specific bits, but one can ignore those). Also, something not 
stated in the example: the function by the file event handlers receives 
two (or possibly three) arguments:
- the socket
- the flags
- perhaps an optional user-defined third argument (this is supported by 
vanilla Tk, but I don't know if it's supported by Tkinter).

>David Beazley's excellent "Python Essential Reference" says in the 
>section on threads: "In addition, many of Python's most popular 
>extensions such as Tkinter may not work properly in a threaded 
>environment."
>
>I assume it's true, but it was quite a bombshell. I was hoping to write 
>a networked GUI client, hence:
>- read data from a socket and fill in a GUI display
>- accept input from the user and write data to the socket
>I assumed I'd use two threads, one for input, one for output. Now I 
have 
>no idea what to do. Any suggestions?

In Tk, you can assign read handlers to file objects.  Anytime
there is data available to be read, the handler will be called.

Just open the socket connection and assign a read-handler to
it.  Piece-o-cake.  Here's an excerpt from a program that uses
that technique to handle data from a popen2'd child process:

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

if cmd is None:
    exceptString = 'no executable specified'
    raise exceptString, cmd
                                        
self.__returnCode = -1
self.__child = popen2.Popen3(cmd)
self.__fd = self.__child.fromchild.fileno()

fcntl.fcntl(self.__fd, FCNTL.F_SETFD, FCNTL.O_NDELAY);


Tkinter.tkinter.createfilehandler(self.__child.fromchild,
                                  Tkinter.tkinter.READABLE,
                                  self.__stdoutHandler)
                                  
------------------------------------------------------------

self.__child.fromchild      is a file object connected to the "read" end 
of a pipe.
Tkinter.tkinter.READABLE    is a constant that tells Tk what you care 
about.
self.__stdoutHandler        is a function to call when the file object 
has data available.






From android at micron.net  Wed Jun 14 23:38:37 2000
From: android at micron.net (Peter Immarco)
Date: Wed, 14 Jun 2000 23:38:37 -0400
Subject: Python so far
Message-ID: <39484FBD.CAE467A0@micron.net>

Well I've been getting my feet wet with "Python programming in Win32", a
great book.  I'm a long-time C/C++/VB/Assembler programmer and I must
admit, I'm going to enjoy the "straightforwardness" of Python.  Seems to
take a lot of the pain out of writing applications.

However my C++ friends are badgering the crap out of me and they were
already ribbing me about VB. :)

Peter.




From eric at estinc.com  Thu Jun 29 11:20:32 2000
From: eric at estinc.com (Eric Lee Green)
Date: Thu, 29 Jun 2000 15:20:32 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155>
Message-ID: <395B6960.26AB52C1@estinc.com>

Gordon McMillan wrote:
> 
> Eric Lee Green wrote:
> 
> >Steve Juranich wrote:
> >> I am *brand* new to python (as of Sunday), and I just got to the
> >> section in the tutorial about classes.  I was wondering why there
> >> really isn't such an idea as a "private" member of classes?
> >
> >Mostly because Python doesn't "really" have classes. What Python
> >"really" has are hash tables whose entries are objects (either methods
> >or other data objects), and some syntactic sugar [...]
> 
> By this logic, all C++ has is pointers and some syntactic sugar.

Exactly :-).

C++ is syntactic sugar for "C". As should be obvious from the fact that most
early C++ compilers generated "C" as their output :-). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From jcm at bigskytel.com  Wed Jun  7 20:15:39 2000
From: jcm at bigskytel.com (David Porter)
Date: Wed, 7 Jun 2000 18:15:39 -0600
Subject: I need a lesson.
In-Reply-To: <8hmn80$4vu$1@ionews.ionet.net>; from fgbrown@ionet.net on Wed, Jun 07, 2000 at 06:56:19PM -0500
References: <8hmn80$4vu$1@ionews.ionet.net>
Message-ID: <20000607181539.A22543@bigskytel.com>

* F. G. Brown <fgbrown at ionet.net>:

> The following is a module called howfast.py that I can't get to repeat.

> howfast.py
> 
> a = input("What is the distance you traveled in miles?  ")
> b = input("How many seconds did it require?  ")
> c = a*3600.00/b
> print "You were going %d mph!" %c
> if c > 75:
>  print "Hey you better slow down!"
> d = raw_input ("Would you like to do another? (y)es or (n)o")
> if d == 'y':
>  print 'Here we go!'
> else:
>  print 'See ya!'
> 

How about:

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

while 1:
    a = input("What is the distance you traveled in miles?  ")
    b = input("How many seconds did it require?  ")
    c = a*3600.00/b
    
    print "You were going %d mph!" %c
    
    if c > 75:
        print "Hey you better slow down!"
    
    d = raw_input ("Would you like to do another? (y)es or (n)o")
    if d == 'y':
        print 'Here we go!'
        continue	
    else:
        print 'See ya!'
        break

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

  david.
	
	
-- 
Incline thine ear you glassworks of Murano. -- e.e.cummings in a blender



From tcgsales at tcgfinancial.com  Fri Jun 23 12:46:04 2000
From: tcgsales at tcgfinancial.com (The Commonwealth Group)
Date: Fri, 23 Jun 2000 12:46:04 -0400
Subject: python email system help
Message-ID: <3953944C.1934AD89@tcgfinancial.com>

Hi, Im an 18 year old network kid, who knows nothing about programming.
I accidently made a suggestion at a meeting and now I must implement
that suggestion, and I know nothing about progamming.  I need to write
this program by Monday.  I started learning python about a month ago,
but I've been too busy to make real progress. I need to write a program
that:
1st, dial internet,(if busy, or someone else is using, dial again later)

2nd, get email for several accounts and put on local file server

I'm not sure If I will be dialing from winNT or Linux yet(not sure if
what OS I'm dialing from makes a difference).  Also, I need this program
to also be instantly executable if someone needs to get their mail that
instant.  Also, can I write the email program in a fahion so that after
retrieving the mail, there are still full email copies on the server.  I
would aappreciate if someone that isn't busy this weekend could sort of
guide me in writing this, maybe via irc, multiple news postings, emails,
or others.  Or if someone is real nice, they could write it for me.
Thanks for the attention and help.  Regardless of if I get help or not,
I have to have this by Monday and I'd rather not pull my hair out trying
to write something in a language I don't know.  Its like trying to have
a conversation in a foreign language you don't know.
Jason Toy
gee308 at mediaone.net
btw, the name you see for this posting is the work name, so you can gind
me as toy or through email.  Thanks again.



From rwallace at esatclear.ie  Thu Jun 15 13:17:13 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Thu, 15 Jun 2000 18:17:13 +0100
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com>
Message-ID: <39490F99.2161@esatclear.ie>

Frank V. Castellucci wrote:
> 
> Courageous wrote:
> >
> >
> > > As someone who uses and likes both C++ and Python, I think estimates
> > > like that are excessively simplistic.
> >
> > Me, too. And yet I will claim that I've become *more* productive
> > in Python after a brief 4-6 weeks of use than I am in C++ after
> > 15 years of use, and that's no exaggeration.

*For the sort of things I use Python for*, I was more productive in
Python after two days than in C++ after 12 years if you include C.

> > > I use C++ for commercial programming, because it gives me the power and
> > > control I need.  Productivity isn't the best when measured in time to
> > > get the first prototype running, but I've found nothing better when
> > > measured in time to develop, deploy, maintain and support something a
> > > customer will pay for.
> >
> > Shrug. If it weren't for execution speed, your C++ environment would
> > be toast. Admit it. :)-

Actually, execution speed isn't the issue.  I write business software
for a living and for that, the time spent executing application code is
hardly ever the performance bottleneck.

> While the last line seems rather toungue in cheek, there are other
> things that C++ brings to the table such as encapsulation, access
> control and type strictness.

Those aren't big issues for me either; static type checking is handy,
but I could live without it.

What I can't live without, for code that's going to ship to customers
and be installed on machines I didn't set up, is low-level control.  I
find always at the end of the day I need to be able to decide exactly
what the system will do, rather than issue high-level directives and
trust the library/virtual machine/whatever to handle the details, and
I've yet to see anything (except assembler, of course) come close to
matching C/C++ for that.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From MSteed at altiris.com  Fri Jun  9 12:52:24 2000
From: MSteed at altiris.com (Mike Steed)
Date: Fri, 9 Jun 2000 10:52:24 -0600 
Subject: Python Productivity over C++
Message-ID: <65118AEEFF5AD3118E8300508B1248774CB333@ALTNET>

> From: nbecker at fred.net [mailto:nbecker at fred.net]
> Sent: Friday, June 09, 2000 9:38 AM
> To: python-list at python.org
> Subject: Re: Python Productivity over C++
> 
> 
> >>>>> "Bill" == Bill McClain <wmcclain at salamander.com> writes:
> 
>     Bill> What frustrates me about C++ is the enormous declarative
overhead required
>     Bill> to get anything done. Every class requires me to instruct the
compiler on
>     Bill> how to copy, assign, compare and reference objects, and how
subclasses
>     Bill> are allowed to inherit from the parent. STL works well for data
structures,
>     Bill> but the advanced features are very cumbersome.
> 
>     Bill> Much of this burden simply vanishes with Python.
> 
> I rarely need to do this for my classes.  Using STL heavily is a big
> help.

This is a surprising thing to say.  STL requires that the types used to
instantiate its various templates be well-behaved in certain ways, e.g., for
copying, assigning, and comparing (depending on the "concept" to which the
template conforms).  This means you have to implement your own copy
constructor, assignment operator, and/or comparison operators for each
(non-trivial) class.  It may not be difficult, but it is overhead that
Python does not impose.

--
M.



From just at letterror.com  Fri Jun  9 17:04:30 2000
From: just at letterror.com (Just van Rossum)
Date: Fri, 9 Jun 2000 22:04:30 +0100
Subject: Stackless Python for the Mac?
In-Reply-To: <zoouk6v7.fsf_-_@SNIPE.maya.com>
References: Ian Maurer's message of "Fri, 09 Jun 2000 18:30:32 GMT"
Message-ID: <l03102801b5670bc57ecc@[193.78.237.159]>

At 3:10 PM -0400 09-06-2000, Jeff Senn wrote:
>Has anyone contemplated (or started...or finished) a port of stackless
>to the Mac?  Other than syncronizing the sources, there shouldn't be
>any complications, right?

Almost none at all... It's only for stock 1.5.2 at the moment:

   http://www.petr.nl/just/PythonCoreStackless.sit.hqx

Instructions:
- create a new folder inside the Python 1.5.2 folder.
- drag the old PythonCore to it
- place the new (stackless) PythonCore to the Python 1.5.2 folder
- run ConfigurePython
- done.

Good luck!

Just





From thomas at xs4all.net  Wed Jun 21 03:12:57 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 21 Jun 2000 09:12:57 +0200
Subject: scope hacks
In-Reply-To: <8ipka0$gpn$1@nnrp1.deja.com>; from anamax@earthlink.net on Wed, Jun 21, 2000 at 05:37:05AM +0000
References: <8ipka0$gpn$1@nnrp1.deja.com>
Message-ID: <20000621091257.H26436@xs4all.nl>

On Wed, Jun 21, 2000 at 05:37:05AM +0000, Andy Freeman wrote:

> I like local functions that can be recursive AND have access
> to the enclosing function's state. 

Why ? Isn't it easier to do something like this:

def _outers_inner(i):
    if i:
        return i + _outers_inner(i)
    else:
        return 1

def outer(j):
    _outers_inner(j)

Or do you need more state from the 'outer' function than the inner function
alone (if so, your example isn't good enough ;)

> I tried to use locals()
> in the following way and ran into a bug; is it me or Python 1.5?

Probably a little bit of both :)

> With the tagged line, everything works as expected.  Without
> the tagged line, inner's second print is {'j' : <value>};
> it doesn't include inner's definition.  (It behaves like a
> bogus optimization, namely, not adding local fns to the local
> scope unless there's some "not call" reference or an explicit
> locals(), the bogosity being that the default for v is an
> explicit use of locals().)

The bogosity is dual: at the time of the 'v = locals()', 'inner' is not
defined in locals, so you won't see it. The 'dictionary' representation
doesn't get updated, apparently, at each change in locals, so you have to
'force' it, by using locals() again. Python 1.6 doesn't (yet) fix this, by
the way.

The problem is that the local dictionary isn't really a dictionary (if
Python can help it, that is) but rather a vector, containing the values, and
the names refering to the 'keys' are translated to straight indexes into
that vector. So, yes, part of it is 'bogus' optimization :)

There are a couple of possible solutions. The easiest is "don't do that
then", as nested functions are not as useful or efficient as they might
seem. (They are still compiled at compile time, wether they'll get used at
runtime or not(*), and they'll get 'assembled' at runtime each time you
enter outer.)

Another solution is to do what you already did, work around the bug by using
the extra locals(). Not very pretty, but it works.

Or you can do something like this:

def outer(j):
     def inner(i, callerlocals=None, v={}):
        if not v and callerlocals:
            v.update(callerlocals)
        print "inner: " + str(i)
        print v
        if i:
            return i + v['inner'](i-1)
        else:
            return 1
 
    print "outer"
    inner(j, locals())

which utilizes the 'static' behaviour of mutable default arguments. You
shouldn't use this trick if you intend to call the inner function from
several different functions, though, you should probably delete v's keys
before you do that. Also note that this introduces a circular reference ;)
so you won't collect any of the objects local or global to outer() or
inner(), unless you fiddle with namespace dicts directly. 

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From aparna at bellatlantic.net  Fri Jun  9 22:16:30 2000
From: aparna at bellatlantic.net (raj srinivasan)
Date: Sat, 10 Jun 2000 02:16:30 GMT
Subject: tkinter -- why doesn't this work?
References: <jl66rj3cqa.fsf@endor.bbb.caltech.edu> <3940EF34.1A04405C@ucdavis.edu>
Message-ID: <3941A4FB.3DB8285E@bellatlantic.net>

Hi,
    I think adding a frame.focus() to the code should enable frame to receive
KeyPress events.

raj




From karl-sve at dsv.su.se  Mon Jun 26 07:09:18 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Mon, 26 Jun 2000 13:09:18 +0200
Subject: Passing variables between scripts
In-Reply-To: <8j745m$qqj$1@oslo-nntp.eunet.no>
References: <8j745m$qqj$1@oslo-nntp.eunet.no>
Message-ID: <Pine.GSO.4.20.0006261255040.14565-100000@luna.dsv.su.se>

On Mon, 26 Jun 2000, Olav Viken wrote:

> Can anyone tell me if there is a good way to pass a variable between two CGI
> scripts?

That depends on what you mean by good... :)
Probably, no. Sending them through forms, in the query string or using
files are the only ways to do it with CGI. Which one that is the best in
your case, I don't know.
I suppose that if you control your own server, you could install
mod_python, which I think gives you better methods for passing variables
around (or at least better performance :)). But I don't know anything
about that.

HTH,
  Kalle Svensson



From karl-sve at dsv.su.se  Wed Jun 14 04:17:25 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Wed, 14 Jun 2000 10:17:25 +0200
Subject: Which linux distros have Python by default?
In-Reply-To: <uKs15.437$i06.133733@ptah.visi.com>
References: <uKs15.437$i06.133733@ptah.visi.com>
Message-ID: <Pine.GSO.4.20.0006140956480.27166-100000@luna.dsv.su.se>

On Tue, 13 Jun 2000, Grant Edwards wrote:

> I've got a Tkinter-based GUI interface used to do
> administrative stuff for a device driver under Linux.  I know
> RedHat installs Python and Tkinter by default.  Is there a
> handy list somewhere of which Linux distributions can be
> expected to have Tkinter installed?
> 
> I know it's available on almost all of the distributions, but
> I'd like to know which ones install it in the "normal"
> configuration that joe-windows-user is going to get if he does
> a default workstation or server install.

It isn't default on Slackware, but I think this isn't a very big
problem (for you). Most slack users would know how to install it, and
it's not a very big distro anyway.

> Had I the spare time and disk space I'd love to install a half
> dozen distributions, but resources don't permit that at the
> moment.

Tsk, tsk... It's all a matter of priorities! :)

HTH,
  Kalle Svensson
--
 * http://www.unix-shells.com/~kalle/ *



From boud at rempt.xs4all.nl  Wed Jun  7 12:54:49 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 7 Jun 2000 16:54:49 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com>
Message-ID: <8hluop$mlo$1@news1.xs4all.nl>

Tom <tom-main at removeme.home.com> wrote:

> I agree that MFC seems to be on the way out, but there are two things that
> prevent me (and some other MFC developers) from switching:
> 1) The CodeTools & CodeGuru sites contains TONS of peer reviewed code (in
> the form of messages at the end of every article) that is free, even for
> commercial use.  80% of this code is in the form of MFC extensions.
> 2) What's next?
> - STL?  Yes, but that's not a GUI.
> - ATL & WTL?  Too MS only.
> - wxWindows?  Only has a tiny fraction of the number of users & code that
> the other options have.
> - TCL/TK?  I've heard this is slow, and it can't create native windows' apps
> (users want the windows look & feel).
> - Java?  Again, too proprietary.

There's another option, of course. You could use Qt (and PyQt), which
works on Windows and Unix/X11, offers a great GUI and a good design.
And there are tons of programs using it you can look at & learn from.
(Not to mention the great documentation.)

-- 

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


From jeremy at beopen.com  Fri Jun  9 12:59:42 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 09 Jun 2000 12:59:42 -0400
Subject: TypeError: unhashable type
References: <393d1c1a_2@news1.prserv.net> <20000606181546.H469@xs4all.nl>
Message-ID: <F8476B9163F138C3.8DBC9B4CCE38EBEA.9FFEEE60294594AF@lp.airnews.net>

It seems that "unhashable type" is a pretty obscure error message.  If
you understand hashes, types, and dictionary implementation, it's
fine.  If not, you're left scratch your head.

It would be helpful to produce an error message like:
"dictionary key must be hashable type, not <type>"

It's a tad tricky to implement because the error message is generated
in PyObject_Hash.  The error would need to be caught in
PyDict_SetItem, cleared, and a new one set.

Any opinions?

Jeremy


From neilh at scintilla.org  Wed Jun 14 22:14:28 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 15 Jun 2000 02:14:28 GMT
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com> <dCL15.4673$cQ.38267@news-server.bigpond.net.au> <8i8m11$bni$1@nnrp1.deja.com>
Message-ID: <8_W15.4949$cQ.40252@news-server.bigpond.net.au>

> As other people start making additional/competing Python distributions
> for Windows, we will (hopefully) find many of these win32 extensions as
> standard.  If we dont, then someone who _does_ bundle them is likely to
> have a more successful distribution, IMO.
>
> Any thoughts on that?

   I'd like to see the Python download smaller rather than larger so I'm not
in favour of adding large chunks of the Win32 extensions - just the bits I
like, OK :) Lua manages a full script interpreter in 160K, and less than 80K
if compressed with UPX, so Python looks bloated to me.

   All that is really needed is for Sam Rushing's calldll/windll to be
included in the standard distribution. Then any required OS calls could be
implemented in Python rather than requiring an extension written in C. I
know it opens Python up to the world of untyped, unchecked hardware pointers
but the increase in capability is large.

   Why are Unix specific modules like commands included in the Win32
installation? It just leads to the confusion which caused this thread.

   Neil





From sholden at bellatlantic.net  Tue Jun 27 13:02:57 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Tue, 27 Jun 2000 17:02:57 GMT
Subject: boolean true and false values.
References: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl> <memo.20000626221037.43263C@digitig.compulink.co.uk> <slrn8lgsjq.us.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <3958DE33.E4E7AA56@bellatlantic.net>


Albert Hofkamp wrote:
> 
> On Mon, 26 Jun 2000 22:10 +0100 (BST), Tim Rowe <digitig at cix.co.uk> wrote:
> >In article <slrn8lf3q9.us.hat at se-46.wpa.wtb.tue.nl>,
> >hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:
> >
> >> I understand that now.
> >> However, I am not quite sure that that is what one wants.
> >
> >Maybe it isn't; then another language may be more appropriate. When it
> >/is/ what one wants then Python may be your choice.
> 
> :-)
> Changing Python is not my goal, I am only trying to really understand the
> differences between C-like declarations, and the Python approach.
> 
[snip]
> >
> >As can any tool, of course. The dynamic typing is indeed never /needed/
> >(probably one of the things Dijkstra proved!), but it's often very /very/
> >convenient indeed. If it's likely to be a problem to you -- or to the
> 
> Hmm, why is it convenient then ?
> Without declarations you can simply pick another identifier when you need a new
> variable, rather than re-use existing identifiers, which may cause all kind of
> bugs and/or hamper readability.
> 
> Albert
> ---
> Look ma, windows without Windows !!

Dynamically-typed variables are NOT to allow re-use because you want
to save storage by using a variable formerly holding an integer for
a string!

In the C style (as opposed to C++, which like Python was designed to
support object orientation), variables are indeed of specific type,
determinable at compile time.  This allows language-processing rigor
to be used to yield a more efficient executable, ofr example, and
also provides a mental straight-jacket for those who like things to
be just so (:-0).

In Python, other things are interesting about the content of a variable
- such as, for example, "does the object in this variable have a xyzzy()
method?".  In this respect Python and C++ differ, in that C++ is more
disposed to go in for as many compile-time checks as possible that the
method called actualy exists.  Python, using late-binding (the
meaning of a name is interpreted at run-time, and methods are sought
by ascending the class hierarchy at run-time), is a bit freer and easier,
and is prepared to put up with run-time rather than compile-time errors.

Thus, static typing in Python (this variable will hold only this certain
type of thing) while it might offer the ability to optimize aspects of
performance, would (if introduced carelessly) limit the class of useful
programs.  There is no current indication it will be iplemented
carelessly, and I imagine that people like me will be able to proceed
along their happy way never specifying the type of value to which a
variable must refer.

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


From db3l at fitlinxx.com  Thu Jun  1 00:23:38 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Jun 2000 00:23:38 -0400
Subject: Dictionary size/speed limit?
References: <f3lZ4.557$r7.287786@news1.rdc2.tx.home.com>
Message-ID: <uk8ga112t.fsf@ctwd0143.fitlinxx.com>

"Michael Morrison" <borlak at home.com> writes:

> So, is using the regular Python dictionary just an insane idea? :)  Should I
> create something in C to do this for me, or are there modules to do this
> sort of thing?

It's certainly not insane, and I'd bet it'll do fine for your application.
Whether or not Python would serve for a final application is going to
be somewhat application and machine dependent, but as a really quick
example - here's something I just typed in interactively:

   Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
   Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
   >>> import time
   >>>
   >>> start = time.time()
   >>> dict = {}
   >>> file = open('words')
   >>> while 1:
   ...   line = file.readline()
   ...   if not line: break
   ...   dict[line] = line
   ...
   >>> load = time.time()
   >>> file.seek(0)
   >>> print len(dict.keys())
   109582
   >>> while 1:
   ...   line = file.readline()
   ...   if not line: break
   ...   value = dict[line]
   ...
   >>> value
   'zyzzyvas\012'
   >>> stop = time.time()
   >>> print start, load, stop
   959832750.636 959832779.768 959832820.086
   >>> print load-start, stop-load, stop-start
   29.1319999695 40.317999959 69.4499999285

This loaded a Python dictionary with a words file with about 100000
english words.  For simplicity I made the dictionary value the same as
the key (and didn't bother stripping the trailing NL).  I then re-read
the same file doing a lookup for each key.  And note - the timing
includes the time to type the information in (which was actually
probably a half to two thirds of each period, even though I type quickly).

The dictionary values themselves were about 1MB in pure character data,
so I've got 2MB of raw data in the in-memory dictionary.  To be fair, my
Python process (under Windows NT) showed growth to about 19MB, so
whether that's a problem depends on environment.  I may have also
artificially increased that by my len() check since that had to create a
list with the same 1MB of data and I'm not positive when that reference
would get cleaned up.  And of course, this is with 100000 entries.

At a minimum, it should be easy to use Python for any initial development,
leaving you to be able to be concerned with overall design and algorithms
and not data management.  If it proves to be a real problem, it can always
be pushed off to a C implementation later.

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


From anamax at earthlink.net  Wed Jun 21 01:37:05 2000
From: anamax at earthlink.net (Andy Freeman)
Date: Wed, 21 Jun 2000 05:37:05 GMT
Subject: scope hacks
Message-ID: <8ipka0$gpn$1@nnrp1.deja.com>

I like local functions that can be recursive AND have access
to the enclosing function's state.  I tried to use locals()
in the following way and ran into a bug; is it me or Python 1.5?


With the tagged line, everything works as expected.  Without
the tagged line, inner's second print is {'j' : <value>};
it doesn't include inner's definition.  (It behaves like a
bogus optimization, namely, not adding local fns to the local
scope unless there's some "not call" reference or an explicit
locals(), the bogosity being that the default for v is an
explicit use of locals().)

def outer(j):
    def inner(i, v = locals()):    # v should get outer's locals()
        print "inner: " + str(i)
        print v
        if i:
            return i + v['inner'](i-1)
        else:
            return 1
    locals()             # why is this line necessary?

    print "outer"
    inner(j)

thanks,
-andy


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


From matthewm at zebrasoft.co.nz  Sun Jun  4 17:37:30 2000
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Mon, 05 Jun 2000 09:37:30 +1200
Subject: Thanks!
References: <3939D364.2EF22B13@zebrasoft.co.nz> <B_u_4.111323$E85.2192336@news1.rdc1.md.home.com> <slrn8jl0rq.18e0.kain@cableadmin.com>
Message-ID: <393ACC1A.C920EDC8@zebrasoft.co.nz>

Thanks for the help. As my project is mostly Python and not COM/Excel, I
was hoping for a quick fix. Looks like I'll have to become a bookworm
for abit. Thanks again, matthew.

Scott wrote:
> 
> As a side note, I also recommend this book.  I really wouldn't even try
> any python-com programming without reading it first.  If you can't afford
> to buy it, it seems you can view it at http://www.netlibrary.com .  It's
> a neat site that allows you to "check out" books for a period of time
> (usually 24 hours at a time) and read them online.  The only drawback is
> they only allow so many "check outs" of each book so it may or may not
> be available to read.
> 
> -Scott
> 
> PS.
> Here is an exact URL:
> http://www.netlibrary.com/SUMMARY.ASP?EV=1239265&ID=24666&advquery=
> 
> and chapter 9 is the main Excel chapter.
> 
> In article <B_u_4.111323$E85.2192336 at news1.rdc1.md.home.com>,
> Eric Hagemann wrote:
> >Matthew,
> >    Read Hammonds "Programming Python in Win32".  There is a very good
> >headstart there with an example of a class that eases the interaction
> >w/Excel via COM.  Next get "Writing Excel Macros" by O'Reilly.  It is all
> >about how to program Excel via Visual Basic but translates very well into
> >Python COM.
> >
> >Cheers
> >Eric Hagemann
> >
> >"Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
> >news:3939D364.2EF22B13 at zebrasoft.co.nz...
> >> hi everyone,
> >>
> >> A couple o0f questions:-
> >> 1)
> >> can you access an excel (.xls) file via com without having excel loaded
> >> on that machine?
> >> 2)
> >> once you've done:-
> >>  blah = win32com.client.Dispatch("Excel.Application")
> >> can anyone give details of
> >> a) howto open a particular s/s in the current directory
> >> and
> >> b) howto access particular cells in that particular s/s
> >> and
> >> c) howto close that s/s
> >>
> >> I have to admit that I'm hoping some kind soul will help before I'm
> >> consigned to grappling with what looks like grooling documentation!
> >> thanks alot matthew.
> >
> >


From wjdandreta at worldnet.att.net  Wed Jun 21 16:25:24 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 20:25:24 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com>
Message-ID: <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net>

Hi Anders,

>Do you REALLY need to use dBASE? Are you interacting with some other
>programs that uses the same database?


I don't really need to use Dbase but I need something that works with DOS
and is platform independent.

I found XBase which is an open source C++ database library that uses dbase
data structures and is platform independent. If Python does not already have
Dbase access, I guess I will have to figure out how to put a Python wrapper
around Xbase.

>Maybe you could use SWIG to create a Python module for using it.
>If you're not in a hurry I could maybe do it, but not until late
>august.

What is SWIG?

>// Anders
>




From loewis at informatik.hu-berlin.de  Tue Jun 27 04:44:37 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Jun 2000 10:44:37 +0200
Subject: Garbage collection versus reference counting
References: <lqI55.485$U3.22441@news.kpnqwest.fi> <kuituwupoy.fsf@lasipalatsi.fi>
Message-ID: <p6qaeg7zf62.fsf@informatik.hu-berlin.de>

Erno Kuusela <erno at iki.fi> writes:

> nope; unless i've misunderstood, a hybrid approach is on its way.
> read the python-dev archives for more information.

Somebody (Tim?) referred to that as "cleanup of cyclic objects",
rather than garbage collection. I like this characterization, as it
indicates that you get all (at least most) of the advantages of
garbage collection, with only few of the disadvantages.

Regards,
Martin



From dgoodger at bigfoot.com  Mon Jun 12 23:12:43 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Mon, 12 Jun 2000 23:12:43 -0400
Subject: Newbie: displaying binary numbers
In-Reply-To: <8i1q49$2nk3$1@news1.wire.net.au>
References: <8i1q49$2nk3$1@news1.wire.net.au>
Message-ID: <B56B1EEA.4C55%dgoodger@bigfoot.com>

on 2000-06-12 00:55, Steven Adams (adams_s at lab.eng.usyd.edu.au) wrote:
> I'm trying to display integers as binary numbers, I know there are the calls
> for making hexadecimal and octal strings, but couldn't find anything for
> binary?
> 
> are there any modules, or operations not in the docs?
> 
> all I need is to convert a number to a binary with n bits.

# enjoy!
# -- 
# David Goodger    dgoodger at bigfoot.com    Open-source projects:
#  - The Go Tools Project: http://gotools.sourceforge.net
#  (more to come!)

def bin(number, n=None,
        octal={'0':'000','1':'001','2':'010','3':'011', # fixed mapping,
               '4':'100','5':'101','6':'110','7':'111'}): # evaluated once
    """Converts an integer to a binary string, optionally fixed-width."""
    import string                                       # not needed in 1.6
    octstr = oct(number)
    if octstr[-1] == 'L':                               # for long integers
        octstr = octstr[:-1]
    binstr = string.join(map(octal.get, octstr), "")    # in 1.6: "".join()
    if n is None:   # strip off leading 0's
        # takes advantage of: if no match, string.find() returns -1
        return binstr[string.find(binstr, '1'):]
        # return binstr[binstr.find('1'):]              # 1.6
    else:
        return ("0" * n + binstr)[-n:]                  # populate with 0's

# some test code:
for i in range(16) + [33, 101, 255, 257, 1001,
                        2**16-2**15-2**13-2**11-2**9-2**7-2**5-2**3-3,
                        1010101010101010101010101L]:
    print "bin(%s): %s" % (i, bin(i))
    print " " * 30 + "bin(%s, 5): %s" % (i, bin(i, 5))




From moshez at math.huji.ac.il  Fri Jun  9 13:12:44 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 20:12:44 +0300 (IDT)
Subject: Regular expression matching performance
In-Reply-To: <Pine.GSO.3.96.1000609115142.4766B-100000@aplexus.jhuapl.edu>
Message-ID: <Pine.GSO.4.10.10006092009140.27416-100000@sundial>

On Fri, 9 Jun 2000, John H. Zouck wrote:

> I'm a relative newbie to python, and have been rewriting a fairly simple
> perl program in python, mainly because I like python much better for
> structural and engineering reasons. The pythonprogram runsmuch more slowly
> then the perl program, and I set out to find out why. The  python profiler
> revealed the program was spending most if its time in regular expression
> matching, so I decided to find out if this was very much slower than perl. 

Writing Perl in Python is a hazardous affair. Perl's IO and REs are heavily
optimized, because that's what the language is used for the most. Python's
performance in those areas is worse. One way to lessen the difference is 
thinking a bit about the problem <wink>.

In your case, I'd recommend starting with a string.search(line, 'Loopback')
before going into the "re" code: it's much faster in python. If that isn't
fast enough, try to use Marc Andre Lemburg's mxTextTools 
(http://www.lemburg.com).

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From mfletch at tpresence.com  Tue Jun 20 21:56:45 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 20 Jun 2000 21:56:45 -0400
Subject: Current date and time
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33D5@VRTMAIL>

http://www.python.org/doc/current/lib/module-time.html

>>> time.strftime( '%A %d %B %Y %H:%Mhr', time.gmtime( time.time()))
'Wednesday 21 June 2000 01:55hr'

or, for the local time (as opposed to the UTC time)...

>>> time.strftime( '%A %d %B %Y %H:%Mhr', time.localtime( time.time()))
'Tuesday 20 June 2000 21:56hr'

Documentation is your friend :) ,
Mike

-----Original Message-----
From: Aeneas [mailto:shan_mu <nospam]
Sent: Tuesday, June 20, 2000 9:19 PM
To: python-list at python.org
Subject: Re: Current date and time


How do you format the result as--
"Tuesday  20 June 2000  21:18hr" ?
Thanks.
...



From thomas at xs4all.net  Tue Jun 20 03:56:26 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 20 Jun 2000 09:56:26 +0200
Subject: nest class name problem
In-Reply-To: <394EB843.EEB3B68F@uniserve.com>; from bvdpoel@uniserve.com on Mon, Jun 19, 2000 at 05:18:11PM -0700
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net> <j8y35.1071$ul.67488@news1.rdc2.tx.home.com> <394EB843.EEB3B68F@uniserve.com>
Message-ID: <20000620095626.D26436@xs4all.nl>

On Mon, Jun 19, 2000 at 05:18:11PM -0700, Bob van der Poel wrote:

> Rainer Deyke wrote:
> > 
> > Jeff Massung <jmassung at magpiesystems.com> wrote in message
> > news:riu35.36$JM5.4708 at news.uswest.net...
> > > you can use Goption.__name instead of self.__name?

> > This would indeed work if it wasn't for name mangling.  Don't call it __name
> > if you want to access it from a nested class.

> Yes, but I'm using __name to keep the variable private/static to a
> particular instance of the Goption(). 
> 

Then you'll have to provide an access method:

class Goption:
	__name = ""
	def getname(self):
		return self.__name

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From cut_me_out at hotmail.com  Tue Jun  6 17:05:45 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 17:05:45 -0400
Subject: dictionary with complex keys
References: <393D54B6.EAF70234@magellan.umontreal.ca>
Message-ID: <etdbt1e4j12.fsf@m56-129-16.mit.edu>

> I'm trying to create a dictionary with the keys being lists.

Can you use tuples instead of lists?  That will work?

Hashing lists is kind of strange, because the list contents can change
arbitrarily, and you would have to have a hash value that was invariant
under those changes.  You could use the value of id (list) or something,
but then two lists with identical elements would not necessarily map to
the same hash value.

Alex.


From sabren at manifestation.com  Mon Jun 26 10:03:37 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 26 Jun 2000 10:03:37 -0400 (EDT)
Subject: nevermind
Message-ID: <Pine.BSF.4.10.10006261001210.23903-100000@chi.pair.com>

Okay, never mind about the floats. I read the 123.4+2 thread (although
I don't get the same error), and then found Tim's old FixedPoint module:

ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/FixedPoint.py

Sorry bout that,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------





From akuchlin at mems-exchange.org  Sun Jun 18 11:48:19 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 18 Jun 2000 11:48:19 -0400
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <3dbt0z3s8s.fsf@newcnri.cnri.reston.va.us>

"William Dandreta" <wjdandreta at worldnet.att.net> writes:
> I recently read a message that suggested that using regular expressions
> might be faster than the find function. I've heard the term regular
> expression used from time to time but I don't know what it means. Could
> someone give me a simple explanation?

Regular expressions are a mini-language for defining a certain set of
strings; the re module then takes a regular expression and tells you
if a given string matches the expression.  For example, [abc] means
'a', 'b', or 'c'; \bA.*?\b means a word beginning with 'A' (\b is a
special sequence meaning "match a word boundary").  As you can see,
regular expressions are concise but can easily be obscure.

Regular expressions will practically never be faster than a simple
string.find, so if you're searching for a fixed string, don't use
regular expressions; their extra expressive power costs you in
performance.  But they can do things that aren't possible with the
string module alone, and would require several lines of Python code 
to implement.

Consult the Library Reference, or the Regular Expression HOWTO for
more info.  (http://www.python.org/doc/current/lib/module-re.html,
http://www.python.org/doc/howto/regex/)

--amk


From thomas at cintra.no  Wed Jun  7 20:25:49 2000
From: thomas at cintra.no (thomas at cintra.no)
Date: Thu, 08 Jun 2000 00:25:49 GMT
Subject: What is Python3000??
References: <393e426f.77267284@news.online.no> <8hlkdf$nun$1@slb7.atl.mindspring.net>
Message-ID: <393ed59f.2598904@news.c2i.net>

Ops! 

Cought you on a bad day did I? ;->

I looked for a few seconds then posted away like crazy. Sorry.
Insomnia will make you do stupid stuff. 

Now get working on that FAQ!!

On 7 Jun 2000 13:58:07 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>In article <393e426f.77267284 at news.online.no>,
>Thomas Weholt <thomas at cintra.no> wrote:
>>
>>I want to know more about Python3000. How is it going to be different
>>from Python as we know it? More features, new syntax ?? 
>
>Nobody -- repeat, NOBODY -- has the answers to these questions.  Py3K is
>a codename for a project THAT HAS NOT YET STARTED.  Py3K is predicated
>to a certain extent -- to what extent, NOBODY KNOWS -- on the Fred
>Brooks maxim, "Build one to throw away."
>
>There are certain design decisions that were made for Python 1.x that
>Guido would like to revisit.  Some of these decisions *MAY* be changed
>in ways that break 1.x code.  Py3K is the name of the version that *MAY*
>have these incompatibilities.
>
>Should I make this a friggin' FAQ?
>
>(Thomas, if it seems like I'm pounding on you, it's because I know
>you've been reading comp.lang.python -- if you weren't aware of this,
>you haven't been paying attention.)
>--
>                      --- Aahz (Copyright 2000 by aahz at netcom.com)
>
>Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
>Hugs and backrubs -- I break Rule 6
>
>"You could make Eskimos emigrate to the Sahara by vigorously arguing --
>at hundreds of screens' length -- for the wonder, beauty, and utility of 
>snow."  -- PNH to rb in r.a.sf.f



From rushing at sPaM.nightmare.com  Mon Jun  5 01:43:47 2000
From: rushing at sPaM.nightmare.com (rushing at sPaM.nightmare.com)
Date: Sun, 04 Jun 2000 21:43:47 -0800
Subject: How to free a port that has been used by a python server?
References: <etdg0qvtkel.fsf@w20-575-109.mit.edu>
Message-ID: <8hfb5q$553$1@brokaw.wa.com>

In article <etdg0qvtkel.fsf at w20-575-109.mit.edu>, Alex
<cut_me_out at hotmail.com> wrote:
> 
> Is there some way to free up port 1729 when proxy gets destroyed?  
> 
> I'm running RedHat Linux, but I have also seen this problem with
> Solaris. 

If you wait a couple of minutes, the port will be freed.

If you don't like waiting, try something like this (from asyncore.py):

        def set_reuse_addr (self):
                # try to re-use a server port if possible
                try:
                        self.socket.setsockopt (
                                socket.SOL_SOCKET, socket.SO_REUSEADDR,
                                self.socket.getsockopt (socket.SOL_SOCKET, socket.SO_REUSEADDR) | 1
                                )
                except:
                        pass

-Sam



From tanzer at swing.co.at  Sat Jun 10 14:36:56 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sat, 10 Jun 2000 20:36:56 +0200
Subject: PATCH: Augmented assignment 
In-Reply-To: Your message of "Fri, 09 Jun 2000 17:45:10 GMT."
             <39442c45.54923673@news.telus.net> 
Message-ID: <m130q7t-000wcDC@swing.co.at>

invalid.address at 127.0.0.1 (David) wrote:

> On Fri, 9 Jun 2000 16:39:38 +0200, Thomas Wouters <thomas at xs4all.net>
> wrote:
> 
> >As promised, here is my patch to add augmented assignment (+=, -= and
> >family) to Python. It's still Work In Progress, and probably requires a
> >rewrite of some magnitude before Guido will accept it.
> 
> 
> Why would the Python community want this?  It obfuscates code, in as much
> that newbies aren't likely to intuite what the symbols mean.
>
[snip]
>
> And it's not like it adds new functionality nor improves code readibility.
> It just saves the lazy programmer a whopping one or two keystrokes.

It doesn't obfuscate code looking like:

   nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1

And it saves more than a few keystrokes.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From ted.drain at jpl.nasa.gov  Mon Jun 26 16:23:50 2000
From: ted.drain at jpl.nasa.gov (Ted Drain)
Date: Mon, 26 Jun 2000 13:23:50 -0700
Subject: Getting double* from NumPy array?
Message-ID: <ted.drain-2606001323500001@mjordan.jpl.nasa.gov>

I have some existing C/C++ code that can accept a double array ( int len,
and double* ptr) to do some processing.  I want to allow users to create
arrays in python using NumPy and then provide an interface to pass these
arrays to my C/C++ code.  

Assuming a 1-dimensional NumPy array, is there an easy way to extract a
pointer to a double and length from the python object?  The C/C++ isn't
going to change the object so I'd prefer not to make a copy of it.

Thanks
Ted

-- 
Ted Drain   Jet Propulsion Laboratory    Ted.Drain at jpl.nasa.gov


From jerome.quelin at insalien.org  Mon Jun  5 14:27:34 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Mon, 5 Jun 2000 20:27:34 +0200
Subject: Python's RE
References: <393B7A18.D0F1AA78@barak-online.net>
Message-ID: <960229654.1963595132@news.libertysurf.fr>

Evgeny Zemlerub <oblom at barak-online.net> wrote:
>How exectly do I something that reassembles perl's :
>$string = "qwerwe kkljlksdfg (first string) asdkjflkjg
>(secondstring)jlskjlfdkg(one more string)";
>@words = ($string =~ /\((.+?)\)/smg);
>foreach(@words){
>        print "$_\n";
>}
Python does not support non-greedy quantifiers (ie, *?, +? and {,}?)... In
fact, Perl is the only language that does, I think. You are to do it another
way in Python.

> print re.search(r'\((.+?)\)',
>'aaa(foo)bbb(bar)asdfsadf',re.S|re.M).groups()
Replace the (.+?) part of your regexp with ([^)]+) and it should work the way I
think you want it to work

Jerome
--
jerome.quelin at insalien.org


From dsavitsk at e-coli.net  Thu Jun 22 15:26:51 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Thu, 22 Jun 2000 19:26:51 GMT
Subject: Tutorials
References: <8ite7j$pv5$1@lola.ctv.es>
Message-ID: <%Lt45.13898$dF.574607@news1.rdc1.il.home.com>

look down about 20 entries:)
also www.python.org for listings

"JAZ" <jazudaire at yahoo.com> wrote in message
news:8ite7j$pv5$1 at lola.ctv.es...
> Where can I find tutorials about Python? Heard a lot about it and I'm
> interested.
>
> Regards
>
> Alberto
>
>




From rwallace at esatclear.ie  Sat Jun 17 17:26:02 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Sat, 17 Jun 2000 22:26:02 +0100
Subject: Advice on a Learning strategy
References: <39490B9E.950C2576@home.com>
Message-ID: <394BECEA.1D9C@esatclear.ie>

bo Vandenberg wrote:
> 
> I'm trying to put together a basic toolkit for general consultant
> programming in a win32 environment. (I know this is vague but I have to
> be flexible at this point.
> 
> My thoughts are to work in Python and Delphi.
> 
> Python is going to be for all the client side customization and Delphi
> will be my engine to do what doesn't need to be interpreted. I'm
> considerring C++ but I don't know if I want the learning curve.
> 
> VB vs Delphi    Delphi programs seem more self contained than VB. I also
> like the possibility of going to Kylix in the future. If I write it in
> Delphi I think the client is less likely to accidentally break it (no
> promises for Microsoft though).
> 
> I would like to use Tkinter but I think speed concerns will have me move
> to wxwindows and pythonwin.
> 
> Has anyone gone this route, or a compatible route, before who would be
> willing to share some advice???

I've used (separately) C++, VB, Delphi and Python.

If you don't want the learning curve of C++ - reasonable enough, it's
not the easiest language to learn - then I think I'd recommend doing the
whole lot in Python; I don't see any advantage to doing part of the
project in VB or Delphi.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From JeremyLynch at fsmail.net  Mon Jun 26 16:34:43 2000
From: JeremyLynch at fsmail.net (JeremyLynch at fsmail.net)
Date: Mon, 26 Jun 2000 20:34:43 GMT
Subject: learning-python question
Message-ID: <8j8ep3$tkn$1@nnrp1.deja.com>

hi,
I have just started learning python, although I am already confident in
c/c++, and I finding some of python's features confusing. This may seem
like a *very* stupid question so ignore it if you want to :)
My problem simplifies to:

count = 5

def count_func()
     print count
     count = count + 1

count_func()

Why can count_func() not see count (it throws up a 'name'
or 'attribute' error)

thanks if you can help,

----------------------
jeremylynch at fsmail.net
----------------------


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


From stevena at neosynapse.net  Fri Jun 30 16:23:01 2000
From: stevena at neosynapse.net (Steven D. Arnold)
Date: Fri, 30 Jun 2000 16:23:01 -0400
Subject: Python MP3 Player
In-Reply-To: <LC675.2355$iN5.540818@ptah.visi.com>
References: <395CE50A.3BA16777@bioeng.ucsd.edu>
 <4.3.1.2.20000630150810.01ff3df0@phear.dementian.com>
Message-ID: <4.3.1.2.20000630161838.02026c00@phear.dementian.com>

At 07:34 PM 6/30/2000 +0000, Grant Edwards wrote:

>That's not really doing it in Python, then, is it?

It's an interesting question.  To me the answer is "some yes, some 
no."  For example, xrange is written in C, yet it is a Python command.  You 
can call it from Python, use it in Python code, the whole nine yards.  I 
think it'd be great to have an MP3 library written in C with a Python 
interface that you could load and use just like any Python module. :-)  I 
do see why you'd object, though.


steve


--
Steven D. Arnold           Que quiero sera         stevena at neosynapse.net
"If you have  built castles in the air, your work  need not be lost; that
is where they should be.  Now put the foundations under them!" -- Thoreau




From davidw at linuxcare.com  Wed Jun 14 12:00:23 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 14 Jun 2000 18:00:23 +0200
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
Message-ID: <87hfaws160.fsf@eugene.prosa.it>

You could always try elastiC, at www.elasticworld.com :-)

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From kapil2000 at my-deja.com  Wed Jun 21 10:44:04 2000
From: kapil2000 at my-deja.com (kapil2000 at my-deja.com)
Date: Wed, 21 Jun 2000 14:44:04 GMT
Subject: Titlebar Icon
Message-ID: <8iqkbf$81g$1@nnrp1.deja.com>

Hi,
I'm new at Python and have been playing around with Tkinter.  Is there
any way to change the image in the title bar of a Tk window?  How would
I do it in windows and unix/linux?

Thanks for the help =)
Kapil


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


From m.faassen at vet.uu.nl  Thu Jun 15 10:58:56 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 14:58:56 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <1ec8zim.1jmu7c7qqgdkwN%bparsia@email.unc.edu> <3948C878.E97ED299@uab.edu>
Message-ID: <8iaqvg$dru$2@newshost.accu.uu.nl>

Shae Erisson <shapr at uab.edu> wrote:
> Bijan Parsia wrote:

>> I haven't *per se*, mostly because I'm not set up to really do XP of any
>> kind (being all on my lonesome :)). I have done a bit of *remote* pair
>> programming, and that works pretty well in Python (best if you can share
>> remote buffers).

> I'm looking for people to try VirtualPairProgramming (Python) with me.
> The best setup I've come up with so far is:
> SpeakFreely

What's this? Voice over IP or something?

> Emacs with make-frame-on-display or screen with -x

I do need to investigate these options for VirtualTutoring..

I'll take a look at that wiki.

Regards,

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


From claird at starbase.neosoft.com  Tue Jun  6 14:51:41 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 6 Jun 2000 13:51:41 -0500
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <8hjfed$n0c$1@newshost.accu.uu.nl>
Message-ID: <1A784309EB6B0EDA.2ABDA3FA86DAD60F.6D0681B90EC55AFC@lp.airnews.net>

In article <8hjfed$n0c$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
			.
		[lots of true and
		important stuff]
			.
			.
>and comparing apples with oranges. CGI pages (or other server generated
>pages, for instance those generated by Zope) are not inherently slower
>than JavaScript pages, and in fact, a JavaScript page can be generated
>on the server-side. It's just not a valid comparison at all, as you
Absolutely.  Now that I think of it, most
of my JavaScript is actually machine-generated.
			.
			.
			.
>> Plus, JavaScript can directly call Java Applet methods, enabling
>> it to do functions that are simply impossible with CGI scripts.
>
>Can it? I wasn't aware of this. Then again, I seldom encounter Java
>Applets. You may be navigating a different web? I generally encounter
No, he's right.  Here's what it's good for:
the only way browsers know to do network pro-
gramming is with Java.  Personally, rather
than write full-blown Java applets, I write
tiny little methods that do atomic network
operations, and count on JavaScript as a
display method.  Incidentally, that's my only
answer to the, "is there any point to applets
outside of toys?" question.  I've seen some
JavaScript programming I truly admire, includ-
ing some that would make my threshold of
"useful", but always in domains so far afield
from mine that I can't say I've verified them
myself.
			.
		[more reminders
		that Python is
		*hot*]
			.
			.
-- 

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


From karl-sve at dsv.su.se  Fri Jun 23 07:34:38 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Fri, 23 Jun 2000 13:34:38 +0200
Subject: what is a .pyo file?
In-Reply-To: <m3em5pyoqu.fsf@boss.ksu.edu>
References: <m3em5pyoqu.fsf@boss.ksu.edu>
Message-ID: <Pine.GSO.4.20.0006231328370.29130-100000@luna.dsv.su.se>

On 23 Jun 2000, Venkatesh Prasad Ranganath wrote:

> what are the contents of .pyo file?  When is it generated?  What is it used for?

It's optimized byte code, generated by importing a module or using
py_compile.compile() in a python process run with the -O command line
option. The advantages of the .pyo over the .pyc file is just that it's
optimized and thus runs faster. I guess... :)

HTH,
  Kalle Svensson



From shecter at darmstadt.gmd.de  Thu Jun 15 12:21:36 2000
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 15 Jun 2000 18:21:36 +0200
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151553280.8145-100000@fep132.fep.ru>
Message-ID: <3949028F.3B512AF9@darmstadt.gmd.de>

Oleg Broytmann wrote:

> On Thu, 15 Jun 2000, Robb Shecter wrote:
> > h = HTTP('my.host.name', '/loginscript', 8080)
> > (code, msg, headers) = h.post()
>
> import urllib
> filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript")
>

Thanks for the reply, but I don't think that does what I'm looking for.
(Although - it could very well be that I just don't know how to use the library!)

How do you add HTTP parameters?
How do you specify that the connection should be POST or GET?

I took a look at the latest CVS version of urllib, and saw how it implicitly
switches between GET and POST.  I don't like that too much, because apparently it
doesn't handle parameters for you, so you still have to explicitly code
differently for GET and POST, and hope that the implicit switch does its thing.
Not too clean, IMO.

- Robb



From android at micron.net  Sun Jun 11 09:48:47 2000
From: android at micron.net (Peter Immarco)
Date: Sun, 11 Jun 2000 09:48:47 -0400
Subject: Python and ADO 2.5?
References: <39430E84.2679DD45@micron.net> <8hvsr1$rbs$12$1@news.t-online.com>
Message-ID: <394398BF.5EC9581D@micron.net>

Eduard,

Thanks, good info.  The two books I have due to show up tomorrow are:

Python Essential Reference
Python Programming on Win 32

What language was your main staple before Python and how do you like Python
relative to it?

Peter.

Eduard Hiti wrote:

> I used ADO 2.5 in my last Python project, and had no problems at all. Most
> of the time, ADO access in VB translates almost directly to python syntax.
> Features of ADO I used with no problem include parameterized commands, batch
> updating and client side cursors.
>
> One thing to keep in mind is the way pythoncom handles out-parameters of
> method calls: These values are returned as part of the method result, e.g
> the Execute method of the Connection object returns a tuple consisting of an
> RecordSet object and the RecordsAffected value (see MSDN):
>
>     recordset, affected = db_connect.Execute("select * from ...")
>
> In VB you would only get the recordset returned and RecordsAffected would
> get stuffed in a reference parameter of the call.
>
> Another point is handling of date values: They are returned as objects with
> type pywintypes.TimeType. The documentation for this type is very terse, and
> I had to dig into the pythoncom demos to find their interface: One useful
> method is Format(<formatstring>), which will return a string from TimeType:
>
>     import pywintypes
>     if type(x) == pywintypes.TimeType:
>         return x.Format("%d.%m.%Y")
>
> For more information on COM and Python I recommend Mark Hammond's "Python
> Programming on Win32" (which covers almost the entire COM body, albeit
> sometimes not too deeply: "Advanced Python Programming..." ?), which has a
> chapter on database programming.
>
> Good luck,
>
>     Eduard



From hzhu at knowledgetrack.com  Wed Jun 14 16:53:57 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 14 Jun 2000 20:53:57 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de>
Message-ID: <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com>

On Wed, 14 Jun 2000 20:09:37 +0200, Thomas Thiele <thiele at muc.das-werk.de>
wrote: 
>Aahz Maruch wrote:
>
>> Suppose you needed a file object to connect to some chunk of memory.  In
>> other languages, you would need to inherit from the Python file object
>> in order to ensure that random functions that expect file objects would
>> still work.  In Python, that's completely unnecessary.  It's probably a
>> lot simpler to create your own class, complete with open(), close(),
>> read(), and write() methods -- and because you've maintained *interface*
>> consistency, you can use your new class *ANYWHERE* a file object would
>> go.
>> --
>
>But there is no mechanism to ensure that the behaviour of your memfileobjectsis
>like a fileobject. For instance if you don't note that you forgot to write the
>write()-function.
>In a worst case scenario it works for many months until a user calls a function
>in which
>the write() -function is needed.
>

Here's one example why not forcing complete interface compliance may be a
good thing:

In the matrix for python project I've just added a bunch of statistical
distribution classes, each of them supposedly provide cdf, pdf, rand and
other methods.  Each method is a simple wrapper around some library function
but to do them all takes time.  So I only added the easier ones and the ones
I may need shortly.  I can already use them now, and the project goes along.
If each one had to confirm to a final complete interface specification the
whole thing might have been put off for a long time. If a user wants
something more he can easily add the missing pieces or ask me to do so.  You
may say that when a project is finished everything should be there and
working, but my list of everything has an infinite length.

Now it could be nice if there were a tool that flags the not-implemented
parts and put them into a TODO list.  For example it could give todo=['B.b',
'B.c'] from the following example:

class A:
  def a(): raise NotImplementedError
  def b(): raise NotImplementedError
  def c(): raise NotImplementedError

class B(A):
  def a(): pass

I'm not sure if this is feasible in general, but even a half-hearted job
would still be helpful.

Huaiyu

PS. The stats classes are only in the latest CVS version as I added them
last night.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From rburnham at cri-inc.com  Fri Jun 30 13:02:53 2000
From: rburnham at cri-inc.com (Roger Burnham)
Date: Fri, 30 Jun 2000 17:02:53 GMT
Subject: Job Opportunity
Message-ID: <395cfc50.7969243@news.gte.net>

Dear c.l.p.

I appologize for posting this here, but our submission to the Python job board
has not shown up for over a month...


CRI (Cambridge Research & Instrumentation) manufactures and sells cutting-edge
photonics systems for process control, materials QC, flourescent assays, and
digital imaging.  We are seeking team players desiring visibility and
recognition; offering the excitement and rewards of a rapidly growing pre-IPO
company.

CRI is looking for a software engineer with two or more years of programming
experience in object-oriented languages, with strong analytical skills, and the
ability to work in a team as well as on independent projects.  Familiarity with
Python, the Python numerical and imaging libraries, XML/DOM, and wxPython is an
asset.

CRI offers competitive compensation and benefits packages and is an equal
opportunity employer.  Send resume to:

CRI, Inc.
80 Ashford Street
Boston, MA 02134
Fax: (617) 787-4488
email: "Cliff Hoyt" choyt at cri-inc.com
web: www.cri-inc.com

Thank you,

Roger Burnham
rburnham at cri-inc.com


From jparkey at nospamrapidbs.com  Mon Jun  5 15:55:59 2000
From: jparkey at nospamrapidbs.com (John Parkey)
Date: Mon, 5 Jun 2000 20:55:59 +0100
Subject: Memory leak: assigning function object to instance variable
Message-ID: <8hh0pf$i7$1@phys-ma.sol.co.uk>

I have registered this bug, with code fragment below, but wondered if
anybody else was falling foul of memory leaks of this type.  We use xmllib,
which relies on assigning methods to instance variables for later callbacks.
Our recent volume testing has shown serious leakage relating to this problem
(I use it in other places too, so it's pretty serious for us).  Environment
is 1.5.2 on NT, build 132.  To see the problem, run the example below and
watch memory usage in task manager.

(Quick plug for Mark Hammond - thanks for loads of help, and finding/fixing
COM related leaks). I hope my week of leak hell will soon be over!

class fred:
 def __init__(self):
  self.indirectFunc = self.theFunc

 def theFunc(self):
  return "blah"

def test():
 f = fred()
 del f

if __name__ == "__main__":
 for x in xrange(1000):
  test()





From ggallag at red-bean.com  Tue Jun 27 17:33:47 2000
From: ggallag at red-bean.com (Greg Gallagher)
Date: 27 Jun 2000 21:33:47 GMT
Subject: Complex datastructures in Python
Message-ID: <8jb6jr$2uov$1@news.enteract.com>

I'm a little disheartened here... I know how to do this in Perl, but I'm lost in Python and 
searching for 'complex data structures' is landing me nowhere.  Basically,  I want
a dictionary whosevalues are lists of dictionaries.  Simple.
I want to basically loop through some data and so something like this:

while loop:
   new_forward = {'ip' : ip_address,
                  'hostname' : hostname,
                  'mac_address' : mac_address}

   forward[domain].append(new_forward)


My code is simplified from what I want to do, but basically it loops through some data
in a file, picks out come fields (ip, hostname, etc.) and puts them in a dictionary,
new_forward.  I want a list of dictionaries key'd by the value 'domain', also
picked out of the fields in the file.  In perl ingo, this would be a hash of
hashes, basically.

Simply put, what kind of operations can or should I be doing, because it never likes
the append.  I've even tried something like this:

   forward[domain] = (forward[domain], new_forward)

And that almost made the structure I want but, alas, it was an empty dictionary.  

Any pointers?  Also, any good documentation for 'complex datastructures', ala Perl's
Cookbook ?

thanks!!

Greg Gallagher
greg at red-bean.com
http://www.red-bean.com/



From schneiker at jump.net  Sun Jun 18 07:21:50 2000
From: schneiker at jump.net (Conrad Schneiker)
Date: Sun, 18 Jun 2000 06:21:50 -0500
Subject: Python 3000 -> No, but there's 'R***'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net>
Message-ID: <8iib7d$73f$1@news.jump.net>

Hi,

"Aahz Maruch" writes:
>
> In article <394BB8DE.AE1D6A7A at webtv.com>, balrog  <balrog at webtv.com>
wrote:
> >
> >Half-seriously, Ruby is nicer looking & acting;  it's part Python,
> >part Eiffel / Sather, with the quick & easy string handling of Perl.
>
> This is mainly addressed to comp.lang.ruby; I'm just including
> comp.lang.python because that's where this started:

Likewise.

> Y'know, those of you in the Ruby world who make a point of going around
> trashing other languages are probably doing more to hurt the acceptance
> of Ruby over the long haul than anything else. It's *rude*.

"Half-seriously ..." == "... going around trashing other languages..."????
"Half-seriously ..." == "... rude"????

Well, thanks anyway for your feedback, although it seems a little bit
over-the-top to me.

I for one have been pretty scrupulous about pointing out that Perl and
Python are for various reasons often more appropriate than Ruby.
Nevertheless, on the single FYI post that I made after the ruby-lang mail
list mirror was finally running and all the major domestic and foreign news
servers were at last all propagating it--i.e. at the time comp.lang.ruby
finally became a viable active newsgroup--you would never guess from the
ballistic responses that the respondents had bothered to read the first
couple of paragraphs of that post.

The Perl, Python, and Ruby newsgroups all carry discussions by their users
of what they variously view as their language's limitations, and I don't
think referring to such things amounts to trashing any of these languages or
being rude. Encouraging people who are dissatisfied with various aspects of
Perl or Python to check out Ruby hardly seems to amount to trashing Perl or
Python.

If someone on comp.lang.ruby wants to do something for which no suitable
Ruby module exists, I don't think it would be trashing Ruby, nor do I think
it would be rude, for you or anyone else to point out that they should check
out Perl or Python to solve their problem. Likewise for issues of
programming style, interpreter reliability, better error diagnostics,
integrated IDEs, more books, more English documentation, and so on.

> I was
> almost interested enough in Ruby at one point to check it out, but I
> have to say that the Attitude I've seen so far is even worse than the
> Perlers.

Hmm. Even though I mentioned that I still use Perl from time to time, and
made it clear that I thought Ruby was of value to some but by no means all
people, and invited no response, one of the more rabid Perlers nevertheless
accused me of trying to start a language war. I'm not sure if any Ruby
poster has stooped comparably low--at least not yet. :-) But of course I
would still never hold the posting of 1 or a few Perl hotheads against Larry
Wall, or Perl, or the large number of otherwise civil Perlers.

If the attitude of 1 or 2 or 3 people keeps you from checking out Ruby, it
seems to me that that sort of thinking amounts to sinking to the level of
the people you loathe, plus it seems to overgeneralize quite a bit from a
couple of recent data points. Obviously I don't want to encourage bad
attitudes, but what I have seen doesn't seem comparable to Perl fanaticism.
I happened to learn about Ruby on comp.lang.python last year in a series of
what seemed to me at the time to be very civil discussions. None of the more
recent postings has seemed notably less civil to me.

> It'd be real nice if the level-headed Ruby people restrained the
> hotheads.

Well, I'm not sure we (or perhaps they, in your apparent opinion) would
generally agree about who the hotheads are in this case :-) ... but in any
case I still encourage anyone who thinks that Perl and Python have been
unduly slighted to feel free to post their dissent on comp.lang.ruby. I'm
sure at least _some_ people will be somewhat more careful about how they
discuss Perl, Python, and Ruby as a result.

FYI, the generally explicitly and implicitly expressed opinion on
comp.lang.ruby has been (more or less) that Python is a much more readable
and a much better OO programming language than Perl.

Conrad







From mlauer at trollinger-fe.rz.uni-frankfurt.de  Mon Jun  5 04:45:33 2000
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 5 Jun 2000 10:45:33 +0200
Subject: Python, Glade, & GTK
References: <8gb1gf$r0u$1@news.ihug.co.nz> <39295964.BD653227@bioreason.com> <0WQZ4.29$vh.14640@news>
Message-ID: <393b68ad@nntp.server.uni-frankfurt.de>

Robert (rhicks at nospam.rma.edu) wrote:
> Glade ExampleWhy don't the GLADE maintainers just incorporate the ability to
> save as Python code? It would be the thing that makes me swing from KDE to
> GNOME.

This feature is included in newer GLADE builds.

:M:

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




From scarblac-spamtrap at pino.selwerd.nl  Tue Jun 13 12:55:54 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 13 Jun 2000 16:55:54 GMT
Subject: Two newbie questions
References: <20000613165659.A7828@basajaun>
Message-ID: <slrn8kcpp7.2rs.scarblac-spamtrap@flits104-37.flits.rug.nl>

Arkaitz wrote in comp.lang.python:
> Hi all,
> 
> I'm learning Python, as it seems everybody around here :-), and I found two 
> problems that I can't find the solution to:
> 1. As the strings are inmutable, if I want to modify the order of the characters
> in the string(e.g. sort them), what I do is:
> 
> st = "asdfasfasdf"
> l = []
> for c in st:
>     l.append(c)
> 
> Then I process the list, and the problem is that I have no idea of how I can get
> the string back, any help?

import string

st = "asdgasfasdf"
l = list(st) # A simpler way to make the list
l.sort() # Do something, e.g. sort
st = string.join(l, "")

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  6:53pm  up 99 days,  7:07,  6 users,  load average: 0.01, 0.10, 0.13


From m.faassen at vet.uu.nl  Thu Jun 15 08:16:00 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 12:16:00 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de>
Message-ID: <8iahe0$ici$2@newshost.accu.uu.nl>

Thomas Thiele <thiele at muc.das-werk.de> wrote:
> Aahz Maruch wrote:
[you only need the right amount of interface and it works]

> But there is no mechanism to ensure that the behaviour of your
> memfileobjectsis like a fileobject.
[snip]

i.e., interfaces in Python are implicit and not explicit. And in addition,
there's no compile-time static checking of interfaces.

These are seperate issues. There's been some work on doing interfaces
in Python. Check the types-SIG archives for possible futures here,
linked to static typing. And there's Jim Fulton's proposal (also the result
of discussions), with some implementation, over here:

http://www.zope.org/Members/jim/PythonInterfaces/Summary

There's a chance a system like this will be adopted in Zope eventually.
Zope is one of the biggest Python code-bases out there, and may have the
most developers as well. It's really starting to show the need for some
explicit interface system. I'm not talking about static checking here;
though run-time checking would be a vast improvement already. With
explicit interfaces, the code would be much more self-documenting than
it is now.

Regards,

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


From m.faassen at vet.uu.nl  Fri Jun  2 09:59:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Jun 2000 13:59:04 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk>
Message-ID: <8h8ej8$7c0$2@newshost.accu.uu.nl>

Jacek Generowicz <jmg at ecs.soton.ac.uk> wrote:
> Martijn Faassen wrote:
[the Python-Dutch conspiracy exposed!]
> Might as well change the official language of comp.lang.python to Dutch.

> We beginnen meteen.

> Python en Nederlands: wereldtalen voor de eenentwintigste eeuw !

Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?

Groeten,

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


From MarkH at ActiveState.com  Fri Jun 30 10:59:05 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 30 Jun 2000 14:59:05 GMT
Subject: Python COM Server Type Library
References: <8jfvvc$iql$1@nnrp1.deja.com> <qb5pls0uvkrbjg6hrkua0flurop7p1mpsb@4ax.com> <8ji8kj$7n4$1@nnrp1.deja.com>
Message-ID: <ZA275.2296$Tb7.16829@news-server.bigpond.net.au>

"Charles Medcoff" <cmedcoff at my-deja.com> wrote in message
news:8ji8kj$7n4$1 at nnrp1.deja.com...

> That gives me an idea.  Wouldn't it be nice to have a utility that you
> could point at a type library and would generate a python script to
> implement those components defined in the the library (with stubbed
> methods of course).  I wonder if the pyhton win32 extensions expose
> enought of the win32 api to do this?

Well, this is very nearly makepy.  Indeed, makepy _does_ do exactly this
for "source" (ie, event) interfaces, so it wouldnt be hard to extend to
optionally do it for all interfaces.

Mark.




From jriehl at lucent.com  Thu Jun 29 17:47:05 2000
From: jriehl at lucent.com (Jon Riehl)
Date: Thu, 29 Jun 2000 16:47:05 -0500
Subject: OO Approach to file parsing with Python?
References: <3958294F.4B17E6F7@attglobal.net>
Message-ID: <395BC3D9.FE452234@lucent.com>

	I'm surprised nobody has mentioned John Aycock's SPARK library.  This
allows you to define your grammar tokens as a class, and the grammar
productions in the class doc strings.  It was designed specifically for
the development of custom/little languages.  You may find it at:
http://www.csr.uvic.ca/~aycock/python/
	Hope this helps.
-Jon


From moshez at math.huji.ac.il  Wed Jun 14 16:36:16 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 14 Jun 2000 23:36:16 +0300 (IDT)
Subject: Which linux distros have Python by default?
In-Reply-To: <1VO15.607$i06.193222@ptah.visi.com>
Message-ID: <Pine.GSO.4.10.10006142334560.3119-100000@sundial>

On Wed, 14 Jun 2000, Grant Edwards wrote:

> I'm going to distribute a utility written in Python/Tkinter as
> a configuration/administration utility that goes with a Linux
> device driver.  I was just wondering how many customers are
> likely to have Python/Tkinter available.  I doubt that anybody
> is going to install Python just to run this program.

Pretty soon, everyone will want to have Python to use ESR's CLM2, a
configuration utility for the the kernel written in Python/Curses. 
Tkinter is a different matter though
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From mhammond at skippinet.com.au  Wed Jun 14 15:28:03 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:28:03 GMT
Subject: Python and Windows Scripting Host
References: <9Ov15.1411$vM.56343@typhoon1.ba-dsg.net> <39y15.839$Zh1.63369@news.easynews.com>
Message-ID: <8i8mbb$c3f$1@nnrp1.deja.com>

In article <39y15.839$Zh1.63369 at news.easynews.com>,
  "Roger Upole" <rupole at compaq.net> wrote:
> The Wscript object is not referenced directly as a Python variable.
> You create it via win32com.client.Dispatch:

Im afraid that is not correct.  When Python is running under the
Windows Scripting Host, the WScript variable is automatic.


> "Jim Bower" <jbower at lims.com> wrote in message
> news:9Ov15.1411$vM.56343 at typhoon1.ba-dsg.net...
...

> > However, I am getting "NameError: wscript" anywhere the wscript

Python is a case sensitive language.  You need to use "WScript"

Mark.


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


From matt at null.net  Mon Jun 19 21:56:22 2000
From: matt at null.net (Matthew Schinckel)
Date: Mon, 19 Jun 2000 19:56:22 CST
Subject: standalone programs
References: <8iknh9$2a3t$1@walton.maths.tcd.ie>
Message-ID: <20000619195623.390892@210.9.22.17>

bazza at ireland.com (Barry Kilcline) wrote:
> Folks
> 
> I am a novice python programmer with some experience in C++. I'm
> trying to create a fully stand alone programme to put on systems where
> python is not installed. For my sins, I'm working on a windows
> platform. Can anyone give me an idiot's guide on how to do this,
> please? I have tried using freeze but without much success. (I have
> VC++ installed)
> 
> Any help would be much appreciated
> 
> Regards Bazza

One option is to use mxCGIPython (can't remember the URL, find it on
Parnassus <http://www.vex.net/parnassus>, and just have a compiled
single-file binary and the program.

Matt.


From DOUGS at oceanic.com  Fri Jun  9 15:43:51 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Fri, 9 Jun 2000 09:43:51 -1000 
Subject: readlines() question
Message-ID: <8457258D741DD411BD3D0050DA62365907A1FA@huina.oceanic.com>

> [About...]
> > >contents = open('file').readlines()
> > >
> > >is 'file' closed after readlines() finishes? If not, is it 
> possible to
> > >somehow get either the file descriptor or the file object 
> of the opened
> > >file? Thanks,
> 
> [Aahz] 
> > Yes, it's closed.  The anonymous object created by open() gets
> > deallocated after the assignment; part of the cleanup is to silently
> > call close().
> 
[Moshe]
> While what Aahz said can't be said to be untrue, it's hard to call it
> true, either. That is what happens in the *current* *CPython*
> implementation. In no way does the language require it, and 
> specifically,
> for JPython, it is not true.

An answer to the second part of your question which implies you want to
explicitly control this.

The built in 'open' returns a file object.  The fileno() method of file
objects returns the file descriptor if you really need it.  File objects
have a wealth of methods explained in section 2.1.7.9 of the Library
Reference.  One of them is 'close()' which flushes buffers and closes the
file.  It also deallocates the file object.

Your example would then have to be:

my_file = open('file')
contents = my_file.readlines()

# Use the file object in any other way you need here.

my_file.close()

-Doug-



From MarkH at ActiveState.com  Tue Jun 27 20:21:27 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 28 Jun 2000 00:21:27 GMT
Subject: Shutting down internet connection
References: <3958FCB1.A1B62061@hotmail.com>
Message-ID: <byb65.49$Tb7.332@news-server.bigpond.net.au>

The win32ras module can do this.  Look for the "win32\Demos\rastest.py"
script.  Specifically:

rastest.py -d "Connetion Name"

Will hangup the named connection.

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

"hulme" <kupisch at hotmail.com> wrote in message
news:3958FCB1.A1B62061 at hotmail.com...
> Hi everyone,
>
> I am looking for a way to shutdown an internet connection that was
> established by dial-on-demand on a remote computer. i.e. there are two
> computers given both running WIN98 on starting a browser the connction
> gets dialed on the other machine. So, I would like to have a script that
> shuts down the connection from the computer I am browsing with. Any way
> to do this with Python or Windws Extensions.
> Thanks for help!
>




From 12klat at sightreader.com  Wed Jun  7 01:06:29 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 22:06:29 -0700
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <Pine.SGI.4.05.10006061254220.22654-100000@lemur.magnet.com>
Message-ID: <393DD854.56032E87@sightreader.com>



John Mitchell wrote:

> <snip>
> For heavy string-formatting programs with loose specifications, Python
> will beat the tar out of Java every time.  Matter of fact, I cant really
> think of a niche for Java.  Hmm.  Maybe I should beat management with the
> cluestick again.
>
> - j

Java definitely has a niche: embeded software for household appliances.
What's it doing on the web anyway?

(Note: Java was initially designed for appliances, such as coffee machines,
hence the name.  No really! It's true!)

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/b38c59f6/attachment.vcf>

From echuck at mindspring.com  Tue Jun 13 21:06:57 2000
From: echuck at mindspring.com (echuck at mindspring.com)
Date: Wed, 14 Jun 2000 01:06:57 GMT
Subject: How do you FTP ASCII if ftplib strips CR?
References: <8i5jhd+h2lu@eGroups.com> <uog55mk6r.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8i6lr7$s2o$1@nnrp1.deja.com>

In article <uog55mk6r.fsf at ctwd0143.fitlinxx.com>,
  David Bolen <db3l at fitlinxx.com> wrote:
> > ourFTP.retrlines('RETR ' + filename, open(os.path.join(localdir,
> > filename), 'w').write)
>
> Your problem is using the "write" method on a file, which just writes
> a buffer of data without any consideration for text "lines".  As an
> aside, note that by doing the open inside the retrlines() call
> that you lost access to the open object, and thus can't close() it.
> It's possible that some data may not get flushed to the file until
> your process exits.  I'd probably open the file separately and close
> it after the retrieval.

By losing access to the file object returned by open(), it gets gc'ed
and therefore closed and flushed immediately. You only have to worry
about standalone open() calls if you use JPython or think you will
someday.

-Chuck
--
http://webware.sourceforge.net


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


From cut_me_out at hotmail.com  Mon Jun  5 11:43:39 2000
From: cut_me_out at hotmail.com (Alex)
Date: 05 Jun 2000 11:43:39 -0400
Subject: Regular expressions trouble
References: <8hgcn2$1lh$1@nnrp1.deja.com>
Message-ID: <etdr9accevo.fsf@w20-575-109.mit.edu>

> I'm fairly new to the use of regular expressions, and I've come upon a
> problem I can't seem to fix. I'm trying to extract a keyword from a
> HTML-page and color it red. This works fine, except that it also
> extracts the keywords within tags.

Although with some effort you might be able to get them to work in this
case, regular expressions by themselves don't really cut it for that
sort of thing, in my opinion.  You might want to check out htmllib, and
parse the html properly.  If you do, have a look at htmllib.py to get an
idea of how to use it.  I found the library reference suffered from lack
of examples.

Alex.


From alan.gauld at gssec.bt.co.uk  Tue Jun 13 09:48:23 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Tue, 13 Jun 2000 14:48:23 +0100
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com> <01bfd2ed$42b67880$29cf8ad1@daves-dell>
Message-ID: <39463BA7.43A3CD65@gssec.bt.co.uk>

>           Pisses me off too. Long filenames great, but allowing
> spaces is just silly, when spaces have been command-line
> delimiters all along. 

So what? Unix has supported both for a long long time. You 
just escape the spaces or quote the names.... The only 
programs that seem to get really confused are Win16 or 
old DOS programs.

Alan g.

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


From moshez at math.huji.ac.il  Wed Jun 14 07:36:08 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 14 Jun 2000 14:36:08 +0300 (IDT)
Subject: python is hard to learn without c
In-Reply-To: <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net>
Message-ID: <Pine.GSO.4.10.10006141435400.2269-100000@sundial>

On Wed, 14 Jun 2000, jin choung wrote:

> howdy,
> 
> ummm, is there any book that you reccommend for a rank beginner?  that
> learning python book is REALLY not for beginners.

Ivan's "Teach Yourself Python in 24 Hours"
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From herzog at online.de  Thu Jun 29 07:55:36 2000
From: herzog at online.de (Bernhard Herzog)
Date: 29 Jun 2000 13:55:36 +0200
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <m3k8f9cqcc.fsf@greebo.nodomain.de> <slrn8ll5vu.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3pup0k8g7.fsf@greebo.nodomain.de>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> Here's a retake:
> 
> $PYTHONPATH/
>    AAA/
>       __init__.py
>    BBB/
>       __init__.py
>       aaa.py
>       ccc.py
>    CCC/
>       ...
> 
> Within BBB/aaa.py the statement 'import AAA' imports from AAA/__init__.py on
> Unix but from BBB/aaa.py on Windows.
> 
> The reason is that within the BBB package python first looks for a module
> called AAA within the package (which it found as aaa.py on Windows) before
> it looks for a package named AAA.

Exactly. Since BBB was apparently written with importing AAA in mind,
simply don't name the modules in BBB so that they hide AAA. 

This is very similar to the way local variables hide global variables.

> Although this is often the desired behavior, the other way around is also
> fairly common.  In my case AAA and CCC are packages written by others and
> BBB is a wraper pacakge that imports the other packages to do the real work
> but provides different interfaces for convenience.  Using identical names
> seems very natural here.

Well, you could do something like this:

    BBB/
       __init__.py
       myaaa.py
       myccc.py

and have a BBB/__init__.py like this:

import myaaa
AAA = myaaa
del myaaa

import myccc
CCC = myccc
del myaaa

> Therefore I think it is desirable for python to provide a mechanism for
> programmers to describe absolute path of imports.  Like on Unix, writing
> /bin/sh in stead of sh in a script will guarantee its behavior no matter
> where it is run and what the current PATH the user has.

I don't think you need to be able to use absolute paths. What would that
mean exactly anyway? What would indeed be useful would be to explicitly
refer to a module in the global module name sapce, i.e. one directly in
sys.path.

> My previous proposal was to use
> import pythonroot.AAA

'import global.AAA' or 'from global import AAA' would look better, I
think.

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


From darcy at vex.net  Fri Jun  2 08:22:45 2000
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 2 Jun 2000 12:22:45 GMT
Subject: Using strings with ' in them in SQL-queries
References: <39342055.1301760@news.c2i.net>
Message-ID: <8h88ul$rch$1@news.tht.net>

thomas at cintra.no wrote:
> I need to insert strings with ' in them in query-strings. I`m using
> the PostgreSQL-database. 

> If I try stuff like this :

> db = _pg.connection(....)
> db.query('insert into test (id, name) value (1, 'fdsfds''fdsf') ')

I assume you are using PyGreSQL.  The suggested method to call it is
to use the Python wrapper so...

import pg
db = pg.DB(...)

Now, if you use the insert method instead of a raw query the issue
disappears.

test = db.insert('test', {
    'id': 1,
    'name': "fdsfds'fdsf",
	'active': 'Yes',
	'desc': None,
})

The pg class does the checking and conversion for you.  It even converts
'T', 'TRUE', 'Y', 'YES', 1, '1' and 'ON' (case insensitive) to 't' and
everything else to 'f' if the field is a boolean.  The new version also
turns None into "NULL".

And the return value is the dictionary with the new OID as well as any
changes to the record due to triggers, rules and defaults in the database.

You-learn-about-apostrophes-with-a-name-like-D''Arcy-ly yrs.

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


From pduffin at hursley.ibm.com  Thu Jun 29 07:17:06 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 29 Jun 2000 12:17:06 +0100
Subject: Controlling an external program with Python
References: <slrn8lj6pa.l52.thomas@centurion.xs4all.nl> <3958C648.1D76@rd.francetelecom.fr> <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <8jc28s$8eb$0@216.39.151.169> <8jdf8j$8tm4$1@nntp6.u.washington.edu>
Message-ID: <395B3032.7FA89EA1@hursley.ibm.com>

Donn Cave wrote:
> 
> The worst mess is that collection of hacks and tricks to open the
> slave terminal without security exposure and in a way that allows
> features like Berkeley job control to work correctly.  It's a mess
> because it's poorly documented and, at least in the case of security,
> it deals with gaps rather than features.  Someone will have to sort
> those out and incorporate them into a few platform-independent functions
> that cover the necessary ground without taking too much onto themselves -
> I mean like forkpty() takes too much on itself, in my opinion, to be
> something we would want at the C level.
> 
> The more we can steal, the better.  There's too much platform specific
> folklore in here to write a good C implementation from scratch with
> man pages.
> 

Which is exactly what Don Libes has done for Expect. If you are interested
maybe you should talk to him about a collaboration to extract the knowledge
out of Expect into a C library which both Tcl/Expect and Python can use.


From moshez at math.huji.ac.il  Thu Jun  8 03:28:24 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 8 Jun 2000 10:28:24 +0300 (IDT)
Subject: Capturing Extension output (newbie)
In-Reply-To: <8hmd2b$17e@news.or.intel.com>
Message-ID: <Pine.GSO.4.10.10006081027260.3146-100000@sundial>

On Wed, 7 Jun 2000, Jeff wrote:

> Hi, I am trying to use python to extend some C functions someone wrote long
> ago.  In all the examples I have seen on the web and in tutorials, none of
> them use anything like "printf" !!  How can I get python to capture stuff
> that the C function sends to STDOUT???

Dup file descriptor 1 into the writing end of a pipe, and read from the
reading end.

Read documentation about:
os.pipe
os.dup2

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From thomas at xs4all.net  Tue Jun 13 12:30:43 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 18:30:43 +0200
Subject: can python do this?
In-Reply-To: <39465166.6A7E261A@stud.ntnu.no>; from petersc@stud.ntnu.no on Tue, Jun 13, 2000 at 05:21:10PM +0200
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no> <3946232F.61B28793@sightreader.com> <39465166.6A7E261A@stud.ntnu.no>
Message-ID: <20000613183043.X20700@xs4all.nl>

On Tue, Jun 13, 2000 at 05:21:10PM +0200, Peter Schneider-Kamp wrote:

> but-then-I-use-linux-and-not-unix-ly y'rs

Well, almost noone uses UNIX nowadays. I'm not sure where the trademark lies
now, nor what the official requirements are to call your product 'UNIX', but
I can say from my own experience that Linux is more of a UNIX OS than
several of the 'certified' UNIX OSes, like, say, SCO UNIX ;-P

linux <> unix and linux in unix-ly yr's
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas at xs4all.net  Fri Jun 16 12:45:14 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Jun 2000 18:45:14 +0200
Subject: Augmented assignment, update
Message-ID: <20000616184514.Q26436@xs4all.nl>

I've more or less finished the augmented-assignment patch I posted last
week. It can be found, including a longish explanation, here:

http://www.xs4all.nl/~thomas/python/

(For those who missed it, the previous posting is here:
http://www.python.org/pipermail/python-list/2000-June/060680.html
The link in the posting, however, points to the new patch.)

I do not know if augmented assignment will every be incorporated into Python
(this patch or something entirely different) but at least noone can claim the
groundwork is missing :-)

Feel free to send questions, comments and verbal abuse my way ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From scherbi at bam.com  Wed Jun 21 08:34:17 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 21 Jun 2000 12:34:17 +0000
Subject: mailing list duplicates?
Message-ID: <3950B648.29F8FC2B@bam.com>

Is anyone else seeing a lot of duplicates on the list?

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From gmcm at hypernet.com  Sun Jun 18 21:49:27 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 19 Jun 2000 01:49:27 GMT
Subject: Pythonwin-wxPython
References: <394D3024.7E2B1A11@hotmail.com>
Message-ID: <8F57DAB28gmcmhypernetcom@199.171.54.154>

kupisch at hotmail.com (ratze) wrote:

>My  problem is that when I am developing a wxPython application
>everytime the the start of my app fails due to an exception I cannot
>start it again and get 'Only 1 wxApp per process!'. How do I get around
>this??

Run the wxPython from the command line. You can get around this if you 
really want to, but it's not worth the struggle.

- Gordon


From thiele at muc.das-werk.de  Thu Jun  8 10:50:01 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Thu, 08 Jun 2000 16:50:01 +0200
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
Message-ID: <393FB299.BED98501@muc.das-werk.de>

>

And my wish is:

Something like a (optional) preprocesor that scans the whole file for unknown
arguments and double defined functions (int the same class)

Example One:

def function_running_very_long(...):
        ......
def other_function(.....):
       variable = function_running_very_long(...)
        print variabl  #forgotten "e"

I really don't like it, to get a "NameError: variabl" after some hours.......

Example Two:

class X:
        .....
        def function(......):    #you can write here what you want
                ....
         .....
        def function(......):     #if you have a second function with the same
name
                .....                      #(forgotten or forgotten to delete or
type-mistake)





From hzhu at knowledgetrack.com  Wed Jun 28 19:03:14 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 28 Jun 2000 23:03:14 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com>
Message-ID: <slrn8ll17c.25h.hzhu@rocket.knowledgetrack.com>

On Wed, 28 Jun 2000 15:16:11 -0700, Sean Blakey <sblakey at freei.com> wrote:

>OK, I think I see what you mean.  You are objecting to the way that python
>searches first for modules in the local directory before searching the normal
>PYTHONPATH, correct?
>
>For example, If you have a script string.py and foo.py in your local directory
>and foo has an "import string" statement, you want foo to import string from the
>standard library instead of from your string.py script.

That is right. If this can't be remedied the package namespaces are
essentially flat.


>It seems to me that you are expecting python too read your mind about where you
>want the modules imported from.  I believe python's behavior makes sens for most
>cases.

Nothing like that.  I can tell python my mind.  I just need the syntax for
doing so.  If you read my explanation, you'll see that this situation is
quite common.  After all, providing heirarchical name space for module names
is the very reason we need packages. Right?

On Unix you can say /bin/sh and expect the correct sh to be used whatever
your PATH and whatever other sh sits in different directories, including
current directory.  My little proposal in the original post was intedend to
add this ability to Python as well.


>If you don't want python to search the current directory for imports, there is a
>very simple solution: put this snippet above all of your other imports:
>import sys
>sys.path.remove('.')
>

This might be fine for personal use.  But I'm developing a general package
that is being used by many people.  I don't think it's safe to fiddle with
that.  And as I said, the functionality of packages can not (and absolutely
should not) depend on the order of sys.path.  Otherwise packages expecting
different ordering of sys.path cannot be used together.


Huaiyu
-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From ge at nowhere.none  Wed Jun 14 17:34:51 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 14 Jun 2000 21:34:51 GMT
Subject: Which linux distros have Python by default?
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial>
Message-ID: <%TS15.655$i06.206156@ptah.visi.com>

In article <Pine.GSO.4.10.10006142334560.3119-100000 at sundial>, Moshe Zadka wrote:
>On Wed, 14 Jun 2000, Grant Edwards wrote:
>
>> I'm going to distribute a utility written in Python/Tkinter as
>> a configuration/administration utility that goes with a Linux
>> device driver.  I was just wondering how many customers are
>> likely to have Python/Tkinter available.  I doubt that anybody
>> is going to install Python just to run this program.
>
>Pretty soon, everyone will want to have Python to use ESR's CLM2, a
>configuration utility for the the kernel written in Python/Curses. 
>Tkinter is a different matter though

Well, people who compile kernels will, anyway.  There are
getting to be a lot of users who never do a kernel build...

-- 
Grant Edwards                   grante             Yow!  PUNK ROCK!! DISCO
                                  at               DUCK!! BIRTH CONTROL!!
                               visi.com            


From 12klat at sightreader.com  Tue Jun  6 09:35:54 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 06:35:54 -0700
Subject: Newbie question.
References: <393CE029.27B87C91@dulcesol.es>
Message-ID: <393CFE3A.DF7370AE@sightreader.com>

You're right.  By default assignments inside functions create local
variables within the function, even if there happens to be a variable by
the same name in the module scope.

Specify 'global' to make your function assign to the module scope.

Example:

>>> AYS = "hey"
>>> def setEnviroment():
...  global AYS
...  AYS = "NUMBER 9"
...
>>> def getEnviroment():
...    return AYS
...
>>> getEnviroment()
'hey'
>>> setEnviroment()
>>> getEnviroment()
'NUMBER 9'

Pablo Prieto wrote:

> Hi all of you there!
>
> #This is my question
>
> AYS = ""
>
> def setEnviroment():
>   AYS = "NUMBER 9"
>
> def getEnviroment():
>   return AYS
>
> #This is the end...
>
> The function getEnviroment always return "". I guess AYS is declared
> twice, One is module-scope and the other is local to setEnviroment, so
> I've got two differents variables.
>
> How can I make the variable AYS global to setEnviroment, so I'd be able
> to return "NUMBER 9" in getEnviroment?. (Don't say return "NUMBER9".
> I'll get annoyed :))
>
> Well, Have a nice summer all of you (or Winter if you are in the other
> side of the floor).
>
> Pablo.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/ee160dca/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/ee160dca/attachment.vcf>

From android at micron.net  Mon Jun  5 10:53:51 2000
From: android at micron.net (Peter Immarco)
Date: Mon, 05 Jun 2000 10:53:51 -0400
Subject: Windows 2000 Server compatibility?
References: <393B4343.E3A96525@micron.net> <5wN_4.384$qF1.312697@news.pacbell.net>
Message-ID: <393BBEFF.3455A226@micron.net>

Grant,

Thanks downloaded 1.5.2.  I thought 1.6a+ was alpha code right?

Peter.

Grant Munsey wrote:

> Python 1.5.2 and 1.6a2 both work as advertized on Win 2K pro
> and Win 2K server
>



From emile at fenx.com  Mon Jun  5 16:14:53 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 5 Jun 2000 13:14:53 -0700
Subject: operator overloading
References: <f3trf8.08.ln@kazan>
Message-ID: <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com>

Here's a quick example of using __cmp__

>>> class Test:
 def __init__(self,a):
  self.a = a
 def __cmp__(self,other):
  return self.a < other.a


>>> a = Test(3)
>>> b = Test(4)
>>> print a<b
0
>>> print a>b
1


HTH,

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: <kazan at kazan.fenx.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Monday, June 05, 2000 1:01 PM
Subject: operator overloading


> Hello
>
> How can I overload the less than operator?
> learning python did mention something about __cmp__ but
did
> not provide any examples.
>
>
> Thanks in advance
>
> Erling
> --
> http://www.python.org/mailman/listinfo/python-list
>




From erik_wilsher at my-deja.com  Fri Jun 16 09:16:52 2000
From: erik_wilsher at my-deja.com (erik_wilsher at my-deja.com)
Date: Fri, 16 Jun 2000 13:16:52 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <Pine.BSF.4.10.10006152355270.4382-100000@chi.pair.com>
Message-ID: <8id9bm$ko3$1@nnrp1.deja.com>

In article <Pine.BSF.4.10.10006152355270.4382-100000 at chi.pair.com>,
  Michal Wallace <sabren at manifestation.com> wrote:

> I pretty much write all of our code, so I've had little experience
> with pair programming.. But I would LOVE to see an editor wired up
> over the net so that two people could work on code at the same
> time. (If I knew anything about GUI, I'd probably try hacking IDLE
> myself)

Try VNC (http://www.uk.research.att.com/vnc)  The idea is that you
share desktop between two computers.  This way you can edit,
run/compile, and see program behaviour *on both screens* at the same
time.  Computers can be different platforms (Mac/X/MSWin).

VNC is free, GNU GPL, multi platform and can work even with modem
connections (although that can be a slow and painfull experience if
speed < 56k).




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


From akuchlin at mems-exchange.org  Tue Jun  6 17:38:54 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 06 Jun 2000 17:38:54 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <8hjfed$n0c$1@newshost.accu.uu.nl> <1A784309EB6B0EDA.2ABDA3FA86DAD60F.6D0681B90EC55AFC@lp.airnews.net>
Message-ID: <3dvgzmtrpt.fsf@amarok.cnri.reston.va.us>

claird at starbase.neosoft.com (Cameron Laird) writes:
> gramming is with Java.  Personally, rather
> than write full-blown Java applets, I write
> tiny little methods that do atomic network
> operations, and count on JavaScript as a
> display method.  Incidentally, that's my only

Indeed.  I'm responsible for the JPython applet for controlling the
remote microscope, and it's been dissatisfying enough that I'm looking
hopefully at Mozilla, with the idea of writing a microscope client as
a set of Mozilla chrome, triggering actions on the server through
XML-RPC or HTTP.  The major stumbling block isn't a language-related
one, but rather that there's no open standard for a streaming video
format.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
  "Imagine a world without a fixed point."
  "I can -- your mother's dead."
  -- Storey and Philip, in Peter Greenaway's _8 1/2 Women_ (1999)


From mikael at isy.liu.se  Wed Jun 14 06:56:25 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Wed, 14 Jun 2000 12:56:25 +0200 (MET DST)
Subject: When to import
In-Reply-To: <lieeks4gt73r3va4ad79j5v43lufufkqgu@4ax.com>
Message-ID: <XFMail.000614125625.mikael@isy.liu.se>

On 14-Jun-00 Anders M Eriksson wrote:
 >  When does the import take place?
 >  Will there be any difference in the two cases below?
 >  If I have a module that I very seldom use should I only import it when
 >  needed (using an if statement) or is it better to allways import it?

Import when needed. I've saved a lot of compile/execute time that way.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    14-Jun-00
Time:    12:55:12

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


From embed at NOSPAM.geocities.com  Wed Jun 14 16:23:24 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 14 Jun 2000 16:23:24 -0400
Subject: tuple question
References: <3947255E.7CF77923@softhome.net>
Message-ID: <dQR15.3687$227.85555@nnrp1.uunet.ca>

I first came across the term tuple in relation to Relational Algebra and in
particular, in the realm of SQL and Databases.

A row of data in a database is often called a tuple. Really, it's a way of
creating a Structure like you would do in C, very efficiently, which once
created, is immutable. This means you can create a hash on it, and then use
the hash to get the piece of data out of a database table.   This is exactly
how Tuples work, and they are hashed and then placed into an associative
array (Python calls associative arrays Dictionaries).

So whereas there is one native Array type in C, the List type in Python is a
mutable dynamically sized array, and a Tuple is a fixed immutable grouping
of items, and a Dictionary is like an array where the key doesn't have to be
a number.  So, three kinds of things which have something in common with a C
array is better than just having a return to the Bad Old Days where C arrays
were fixed in size, mutable to the point that you could easily crash your
system, and totally non-polymorphic.

Warren




From dale at out-think.NOSPAMco.uk  Mon Jun 26 08:02:24 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 26 Jun 2000 13:02:24 +0100
Subject: Driving Excel from Python
Message-ID: <8j7gp0$agt$1@supernews.com>

(Sorry if this appears twice. I got an error during first attempt).)

I'm driving Excel from Python and most of it's working fine except for:

>>> xlsh.UsedRange.Address(ReferenceStyle=1)
TypeError: call of non-function (type string)

xlsh is an Excel Sheet object and I'm trying to get the used range in RxCx
format. However, when I try and specify ANY arguments to the Address
property, I get this error message. I've tried both positional and keyword
arguments.

I have the Python Win32 book but this this subject is not quite covered in
sufficient depth.

Any help appreciated.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From wolf at one.net  Sun Jun 25 11:06:21 2000
From: wolf at one.net (wolf at one.net)
Date: Sun, 25 Jun 2000 11:06:21 -0400
Subject: Preserving object's original address when restoring from shelf
Message-ID: <gl6cls0n47m18ngrckdmrk187bmtovir96@4ax.com>

Hi All,

First, a quick background. I'm writing a text adventure writing system
in Python and I'm having some trouble with the save/restore system.

My problem is when I restore an object with this line of code:

   __GameModule__.__dict__[VariableName] = SavedGame[VariableName]

   where __GameModule__ is the top level module
         VariableName is the name of the object
         SavedGame is the open shelf

it does indeed restore the object--but it changes the object's address
(ID #). Unfortunately the object is referenced by *many* other objects
in the game. For example the object 'Hill' would be referenced by any
location that leads to Hill, and these references still point to the
old Hill.

Am I being clueless here? Is there some simple way to restore objects
that doesn't invoke this problem?

One final note, I don't know if it matters, but the system is set up
with layered modules, each module using an import from from the module
beneath. For example:

TQ imports * from PAWS, Universe, TQLib.
TQLib imports * from Universe, PAWS
Universe imports * from PAWS

This avoids the need to include a module reference each time an object
is used, this is important since the system is designed to let
non-programmers write games. (PAWS and Universe are unchanging
librarires, TQLib and TQ are written for each game)

Would it help to have each module save/restore its own dictionary and
then do another IMPORT * FROM statement? How would this impact memory
storage, would it cause memory leaks or would Python garbage collect
correctly?

I would really appreciate any help on this one!
Respectfully,

Wolf

"The world is my home, it's just that some rooms are draftier than
others". -- Wolf


From mwh21 at cam.ac.uk  Thu Jun 15 06:51:41 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 15 Jun 2000 11:51:41 +0100
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <8iab6a$idm$1@hunter.orchestream.com>
Message-ID: <m3ya47w72a.fsf@atrus.jesus.cam.ac.uk>

"Tobe" <tbradshaw at orchestream.com> writes:

> Lisp.. productive ??

Common Lisp, yes.

M.

-- 
  If a train station is a place where a train stops, what's a
  workstation?                            -- unknown (to me, at least)


From smnordby at yahoo.com  Wed Jun 14 20:11:14 2000
From: smnordby at yahoo.com (sn)
Date: Wed, 14 Jun 2000 17:11:14 -0700
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net>
Message-ID: <39481F21.648FF15B@yahoo.com>

I think Alan Gauld's "Learning To Program" online tutorial at
http://members.xoom.com/_XMCM/alan_gauld/tutor/tutindex.htm and GvR's tutorial
in the Python documentation are excellent.  After that, the Learning Python book
will make a lot more sense.  My first OO experience was Java.  Much of what
didn't make sense to me in the books I was using makes sense now after doing
Python for a couple of months.  Go figure :)

-Steve-

Jin wrote:

> thanks much for all the input.
>
> i have no doubt that C++ is far more complex and complicated.  you guys are
> probably right that Python is a much easier language but i picked a bad book
> (for me) to start with.
>
> will look for the 24hour book.
>
> but i think the reason the C++ book is working so well for me is that it
> assumes nothing about prior knowledge in an OO language because C++ is the
> base upon which everything else is built.  the book realizes this and so all
> such concepts are explained in detail from scratch.
>
> the learning python book however seems to assume a lot of things that i
> would only know if i came from a C++ or perl background!
>
> thanks again, and i will pursue both paths to enlightenment and hopefully
> one language will bolster skills in the other.
>
> jin



From thomas at xs4all.net  Tue Jun 13 09:51:31 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 15:51:31 +0200
Subject: Python Productivity over C++
In-Reply-To: <3945FBB1.9CE9B636@sightreader.com>; from kens@sightreader.com on Tue, Jun 13, 2000 at 02:15:29AM -0700
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3945FBB1.9CE9B636@sightreader.com>
Message-ID: <20000613155131.V20700@xs4all.nl>

On Tue, Jun 13, 2000 at 02:15:29AM -0700, Ken Seehof wrote:

> Memory leak tools - Python doesn't need them (except for circular refs which
> aren't that tough anyway).  Python wins here.  The next major release of
> python will probably have true garbage collection, which will remove orphan
> circular refs.

That's a bit extreme. The next major release would be Python 2, aka Py3k,
which has been repeatedly stated as being 'far away and undecided' at the
least ;) The next minor release, 1.6, *might* contain some form of garbage
collection to remove circular refs, but it certainly does not throw out
refcounting in favor of GC. The garbage-collection bit might even be a
compile-time option, off by default.

> Profilers - Never looked, so I don't know if python has any good ones.  If
> I'm concerned with saving a few milliseconds, I code an extension.

The Python profiler is pretty good, but it doesn't profile beyond the python
level (it can't, it's written in Python ;) so you have to combine the python
profiler and a C-level profiler to get proper stats on your extensions.
But, I have to admit, I haven't used it much either, myself.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From MarkH at ActiveState.com  Mon Jun 26 20:43:11 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 27 Jun 2000 00:43:11 GMT
Subject: Python COM does not seem to like me.
References: <8j0ddv$fqb$1@nnrp1.deja.com> <39541965.45DCE978@san.rr.com> <8j813b$if7$1@nnrp1.deja.com> <8j8cja$ruj$1@nnrp1.deja.com>
Message-ID: <zMS55.5325$Sn2.35710@news-server.bigpond.net.au>

<millja at my-deja.com> wrote in message news:8j8cja$ruj$1 at nnrp1.deja.com...

> Okay, Im sorry.  Its all working now.  I am not at all used to these
> stupid Windows computers where you have to restart after you do
> anything.

Its not Windows fault, it is Python's!  Windows has enough faults that we
dont need people inventing new ones!  The problem here is that the MS Web
Server is effectively "embedding" Python.  This means that the Python
process that you are running your scripts from is still running.

The problem is simply that _Python_ does not pick up these sorts of
changes while running.  The Python process (not the machine) must be
restarted for these changes to take effect.  In your specific example,
restarting the Python process means restarting IIS.

This can also be done from the IIS config app, or from the standard NT
services applet.

Mark.




From paulb at infercor.no  Tue Jun 27 06:34:20 2000
From: paulb at infercor.no (Paul Boddie)
Date: Tue, 27 Jun 2000 12:34:20 +0200
Subject: Pick or Universe Module
References: <3957612D.B631299D@laddservices.net>
Message-ID: <3958832C.27E72F14@infercor.no>

Brad Losh wrote:
> 
> has anyone seen a python way to talk to a Pick or Unverse DBMS.  if
> there is nothing I would like to start something, just not sure how.

I think that you would have two principal options:

  * Use an ODBC connectivity solution and mxODBC. UniData, which is a product
    supplied by the same company as UniVerse, has some solutions developed for
    it which allow access via ODBC. In my opinion, you have to jump through
    lots of hoops using the "interesting" tools available in order to "publish"
    the data to ODBC client applications, and this might seem like too much
    work. I did hear that UniVerse is a bit less "backward" in terms of
    connectivity, though.

  * Write something which uses the client libraries. Again, not being able to
    draw on any UniVerse experience, I can say that UniData has libraries which
    could be wrapped by something like SWIG [1] in order to access the database
    API.

I haven't bothered to do this kind of work myself, but it would obviously be a
lot nicer to develop applications in Python than the BASIC variants usually
associated with these database products.

Paul


From thomas at xs4all.net  Sat Jun 17 06:00:03 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 17 Jun 2000 12:00:03 +0200
Subject: How Fast Does Python Run?
In-Reply-To: <8if21t$4u5$1@slb1.atl.mindspring.net>; from aahz@netcom.com on Sat, Jun 17, 2000 at 05:24:13AM +0000
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net> <39499C91.3A1586BE@sightreader.com> <8if21t$4u5$1@slb1.atl.mindspring.net>
Message-ID: <20000617120003.S26436@xs4all.nl>

On Sat, Jun 17, 2000 at 05:24:13AM +0000, Aahz Maruch wrote:

> >And python will get faster.  There is no intrinsic reason for perl
> >to be faster than python; it's just been around longer and has been
> >optimized more.

> Your first sentence is true.  Your second sentence is false.  Perl has
> platform-specific hacks that make it faster; it's extremely unlikely
> that Python will ever go the non-portable route.

Also, Perl has syntax and semantics specifically targeted towards execution
speed, which creates a lot of room for optimization. This is most obvious in
platform-dependant areas, like the tricks used to get file-io as fast as
possible, but becomes painfully obvious if you start looking at optimizing
Python. Because Python does *not* have special exceptions intended to
execute the code faster, except for the docstrings disappearing when you use
double optimization (-OO). You can probably speed up Python quite a bit if
you concentrate on the common task and do not mind fawlty or flawed results
for the little used features.

Fortunately (in my opinion), that isn't the Python Way. There is still room
for improvement without losing accuracy and simplicity (of Python) but that
*will* cost, in simplicity-of-CPython: The current optimization (which
translates 'load_name' opcodes into 'load_global' or 'load_fast' depending
on wether they seem to be local or global) is a good example: The actual
added code is not that large or complicated, but when you start adding new
features, you have to keep those 20-odd lines of code in mind, or you'll
screw it up. Add 10 more of those 'performance hacks' and adding to CPython
can become a real pain.

Also, the load_fast-translation-cycle is skipped if the code block has a
'from module import *' or 'exec' statement in it, because the
compiler/optimizer does not know what names it imports, and hence not what
names will be actually local when the code executes.

Optimized-Python-contradicts-'simplicitly'-both-ways-ly y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From claird at starbase.neosoft.com  Thu Jun 29 15:26:03 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 29 Jun 2000 14:26:03 -0500
Subject: Expect commentary (was: Controlling an external program with Python)
References: <1542456BE0832BEE.8A959B400FAE5129.26A3D7DF9AA9B305@lp.airnews.net> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com> <8jg3hb$iic$1@nntp6.u.washington.edu>
Message-ID: <E7C835CEADB13A99.F8E4B0FF3F4ACA75.A254171C72C804CB@lp.airnews.net>

In article <8jg3hb$iic$1 at nntp6.u.washington.edu>,
Donn Cave  <donn at u.washington.edu> wrote:
			.
			.
			.
>And then I have looked at the source.  I'm sure it has no peer
>when it comes to comprehensive cross-platform pseudotty support,
>but a lot of barnacles come with that.  I would steal a younger
			.
			.
			.
All glory to Don.  Yes, the source has convinced me:
1.  This is one of those things, like writing
    a book or having a baby, where you spend
    part of the time wondering if you really
    should have gotten into this; and
2.  I know of no better way to do all he's
    done.
-- 

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


From jmace at ior.com  Fri Jun 16 13:55:52 2000
From: jmace at ior.com (Jim Mace)
Date: Fri, 16 Jun 2000 10:55:52 -0700
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com> <slrn8kdfhg.10b.warlock@troll.myth>
Message-ID: <8idpn3$122u$1@hardcore.ivn.net>

#circumference calculator

print "This calculates the circumference of a circle"
PI = 3.141592645
R = input ("Enter the radius of your circle here: ")
C = 2*PI*R
print
print "The circumference of your circle is: %0.11f" % (C)
print
print """You can use any number you need to for how ever any decimal places
you want to go out to.  For example, I used %0.11f to do out four
decimal places, but if you want to go out more just change the value
of n in %0.(n)f.  This is a little more simple for a newer programmer
that may not have any programming experience with built in functions
and definitions."""

Jim Richardson <warlock at eskimo.com> wrote in message
news:slrn8kdfhg.10b.warlock at troll.myth...
> On Mon, 12 Jun 2000 21:10:35 -0700,
>  jim roush, in the persona of <jar at mminternet.com>,
>  brought forth the following words...:
>
> >I'm just learning Python.  The following code does not work.  It's
> >supposed to calculate the circumference after the user enters the
> >radius.  The program displays the prompt, but when I enter the radius
> >and the press enter, nothing happens.
> >
> >FYI, I'm using Python 1.5.2 on WinNT.
> >
> >pi = 3.141592654
> >radius = input("Enter is the radius: ")
> >circ = 2 * pi * radius
> >print circ
> >
> >
> I am just a newbie but the following code works for me (on linux, hence
> the #! first line)
>
> #!/usr/bin/env  python
> import string,math
> #Import string for the atoi function, since raw_input returns a string
> #Import math so pi has many digits :)
>
> radius=string.atoi(raw_input('Radius? '))
> circ = 2*math.pi*radius
>
> print circ
>
> #Or to print circ to only 2 places
>
> print 'Circumference is %.2f'%circ
> --
> Jim Richardson
> Anarchist, pagan and proud of it
> WWW.eskimo.com/~warlock
> Linux, because life's too short for a buggy OS.
>




From rcm at bcnartdirecte.com  Sun Jun 11 13:08:45 2000
From: rcm at bcnartdirecte.com (Rafael Cordones Marcos)
Date: Sun, 11 Jun 2000 19:08:45 +0200
Subject: 8-bit cleanliness
In-Reply-To: <20000611185452.N20700@xs4all.nl>; from thomas@xs4all.net on Sun, Jun 11, 2000 at 06:54:53PM +0200
References: <39430E84.2679DD45@micron.net> <20000611175536.A3365@bcnartdirecte.com> <20000611185452.N20700@xs4all.nl>
Message-ID: <20000611190844.A565@bcnartdirecte.com>

On Sun, Jun 11, 2000 at 06:54:53PM +0200, Thomas Wouters wrote:
> On Sun, Jun 11, 2000 at 05:55:36PM +0200, Rafael Cordones Marcos wrote:
> 
> > I just started to use Python a few days ago because I got fed up of so much convoluted Perl code. ;)
> > Anyway, I have to read some text files and process the words appearing in them. I have discovered, to
> > my surprise, that accents like (?, ?, ...) get replaced by a 4 character code. Is there any class/module/option
> > available to read *text* files with non english characters in them?

Thanks for the quick response!! And sorry for the tone of my email but it's been 
several days working full time on this to finally find that Python was mangling 
accented characters... which it finally was not doing!!

Waht happened is that I put print statements to follow some data structures arround.
I had printed a hash and the keys in that hash appeared with \XXX codes. But if
I print the key itself everything is fine!

Back to code!

Rafa


> Those non-ASCII characters do not get replaced by that 4-character code,
> those non-ASCII characters *are* that 4-character code ;-) If you use repr()
> on strings, non-printable characters are expressed as an octal number, to be
> able to reliably reproduce them:
> 
> >>> s = "????rgh"
> >>> s			# which is the same as repr(s), in interactive mode
> '\341\341\341\341rgh'
> 
> >>> print s
> ????rgh
> 
> \341 is the 'accurate' representation of the '?' character, it'll always be
> converted in the same actual character, regardless of your font settings.
> How it is displayed depends on your font or your locale settings, depending
> on what you use to view it ;-)
> 
> Just treat your strings as data, as you should anyway, and all will end up
> fine. Just be sure not to use 'repr()' (or ``) when you really mean 'print'
> or 'str()'.
> 
> -- 
> Thomas Wouters <thomas at xs4all.net>
> 
> Hi! I'm a .signature virus! copy me into your .signature file to help me spread!
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
Linux! The Choice of a GNU Generation! -> http://www.debian.org
Unix IS user-friendly, it just chooses its friends very carefully.



From hinsen at cnrs-orleans.fr  Wed Jun  7 12:36:06 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 18:36:06 +0200
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
References: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD> <14647.58023.493846.225899@buffalo.fnal.gov> <m3wvk1hm2u.fsf@chinon.cnrs-orleans.fr> <vmebt1dms3s.fsf@buffalo.fnal.gov>
Message-ID: <m3ln0hh2ix.fsf@chinon.cnrs-orleans.fr>

Charles G Waldman <cgw at fnal.gov> writes:

> Konrad Hinsen <hinsen at cnrs-orleans.fr> writes:
> 
> > I'd prefer a portable solution to one that reflects the system math
> > library. I had the same problem under AIX, and I solved it by patching
> > Python. I just wonder if someone has the patience to work out a
> > patch that would create the same behaviour independently of what the
> > math library does. Not me, unfortunately...
> 
> What do you think the consistent behavior should be?  Should
> math.exp(-10000) return 0, or raise some exception?

I have yet to encounter a case where returning 0 would cause a problem,
so I'd vote for that solution. But raising an execption systematically
would be fine as well, one could then catch it and return 0 if that's
what is wanted.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From bashxxxford at nospam.scripps.edu  Tue Jun 20 17:00:39 2000
From: bashxxxford at nospam.scripps.edu (Don Bashford)
Date: 20 Jun 2000 14:00:39 -0700
Subject: Python complements C++ for productivity
Message-ID: <rok8fkm5jc.fsf@gage.i-did-not-set--mail-host-address--so-shoot-me>

I've been using C++ for about 10 years, and Python less than a year.
I recently had the experience of implementing an algorithm for
Delaunay triangulation in both Python and C++.  I was working from my
reading of someone's Ph.D. thesis that had parts of the algorithm in
pseudo-code, and part in English.  

I started out in Python.  The work was pleasant and went fairly
quickly.  It was pleasing that a lot of the Python I wrote ended up
looking very close to the pseudo-code in the thesis.  The interpreted
nature of Python was very helpful, since it made it easier to inspect
results as I went along.  I also benefited from a Python graphics
package written by a colleague that hepled me visualize what the
algorithm was doing.  I think in the early phases, development in
Python was faster than it would have been in C++.

But the resulting code was slow---about 100 times slower than the
thesis author's C code.  I started re-implementing in C++.  The
similarity of the basic "class" ideas in the two languages made the
re-write very easy and fast.  In some cases, I just dropped the Python
code into the C++ file and edited it.   In the translation process,
the stricter type rules of C++ helped me find some bugs that I might
never have found in the Python version, and it also resulted in
improved design.

The only real unpleasentness in the whole process was when I had my
implementation partly in Python (higher level) and partly in C++
(lower, more number crunching levels).  Then I was using SWIG to
"wrap" the C++ parts for use from Python, but getting proper handling
of translations like Python lists to C++ containers was quite a pain
(although I learned a lot about SWIG typemaps in the process).

In short, I found Python to be a great prototyping language, with
pretty easy movement to C++, which is a great language when you want
strict type checking, data hiding and performance.


From mvanier at endor.bbb.caltech.edu  Tue Jun 13 20:59:47 2000
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: 13 Jun 2000 17:59:47 -0700
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu> <8F51D9DB3gmcmhypernetcom@199.171.54.154>
Message-ID: <jl7lbtm60s.fsf@endor.bbb.caltech.edu>

gmcm at hypernet.com (Gordon McMillan) writes:

> Michael Vanier: 
> 
> >I wrote a little python script using the thread module the other day.  I
> >wanted to be able to set a flag in the parent thread that would be read
> >in the child thread.  So I set a flag variable (an integer in this case)
> >in the parent, created the child thread, and then checked the value of
> >the flag every time through a loop in the child thread to see if it had
> >changed.  It turns out that the child thread apparently made a copy of
> >the flag, because when it changed in the parent it didn't change in the
> >child. 
> 
> >To fix this, I changed the flag to be a list with one integer element. 
> >Now, when I change the list element in the parent, the child sees it
> >too.  
> 
> 
> Let me guess. You did something like make "flag" a module level global, and 
> then did "from ... import *" in the other.

Even dumber than this; it was all in the same file.  I set the global in the
parent thread, and expected that the new thread would see the new value when
it was changed.

> 
> If you had referenced "module.flag", all would have worked. But that 
> stinks, too. Your thread should have a flag member, and the parent thread 
> should set it (through a setter, if you really want to be virtuous <wink>). 
> That way the flag becomes part of the thread's API, instead of a bass-
> ackwards dependency.
> 

Well, the thread wasn't an object; I just called a function and expected that
it would see the same data (see below).

> 
> > I gather that what this means is that there is a fundamental
> >distinction between atomic types like integers and reference types like
> >lists or dictionaries.  However, since python treats everything as a
> >reference this seems inconsistent to me.  So what I want to know is:
> >what is the cause of this behavior, and what is the justification for
> >it?  Or could it be a bug? 
> 
> Aahz is correct (as usual). It has to do with assignment and immutability. 
> To see assignment to any immutable (not in your global or local namespace), 
> you have to see the immutable through some larger context. Say, a module 
> namespace, or (as you did) through a container object.
> 
> - Gordon

So what you're saying is that a thread's namespace is not the same as the
global namespace of the parent.  This is puzzling, since that's what I
thought a thread was: effectively a new process which shares the parent's
namespace.  Is this in a FAQ somewhere?  Or am I just not getting it?

To further complicate things, here is a stripped-down version of the code
that actually works!

#! /usr/bin/env python

import os, thread

def thread_func():
    while 1:
        print "running child thread..."
        if stop:  
            print "child thread is done"
            break


stop = 0  
thread.start_new_thread(thread_func, ())
os.system("sleep 2")
stop = 1
print "parent thread is done"
os.system("sleep 2") # Wait for child thread to finish.
# end of test program

The original program has a Tk interface, which may account for the
difference.  Nevertheless, given what you've said, why does this work?

Mike


--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.


From timmcd69 at my-deja.com  Tue Jun 27 21:33:30 2000
From: timmcd69 at my-deja.com (timmcd69 at my-deja.com)
Date: Wed, 28 Jun 2000 01:33:30 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jb6th$j4$1@nnrp1.deja.com>
Message-ID: <8jbklb$gen$1@nnrp2.deja.com>

> What is missing that I would like to see:
>  - templates (parameterized types)
>  - operator overloading (so I can make classes
on par with builtin ones)

The absence of templates is disappointing, but at
least the unified type system makes this somewhat
less annoying.

Read over the language spec again, it DOES have
operator overloading, much the same as C++.

Tim



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


From digitig at cix.co.uk  Mon Jun  5 18:04:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 5 Jun 2000 23:04 +0100 (BST)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <slrn8jlm1j.bkc.wtanksle@dolphin.openprojects.net>
Message-ID: <memo.20000605230414.25437A@digitig.compulink.co.uk>

In article <slrn8jlm1j.bkc.wtanksle at dolphin.openprojects.net>, 
wtanksle at dolphin.openprojects.net (William Tanksley) wrote:

> >Great. Ever maintain anybody else's code?
> 
> Are you trying to insult me?  Or are you trying to make some kind of
> point?  If the latter, you have no idea what my experience level is, so 
> to
> communicate your argument you'll have to explain yourself more 
> completely.

I'm saying that /I/ write code that's easily maintainable in whatever 
language, I expect /you/ write code that's easily maintainable in whatever 
language, but when we're confronted with code that third parties have 
written it more often than not is /not/ easily maintainable. I consider it 
a highly desirable feature of a language if it makes it harder to write 
unmaintainable code.

> *My* point is that Python's job is to make it possible to write really
> good-looking code.  It's my job, as the programmer or employer, to 
> prevent
> bad-looking code.

I'm not in the fortunate position of always being able to control the 
third-party code I have to work with.

> >> Take a lesson from the people complaining about Python's 
> > indentation: one
> >> of the leading false accusations is that it's inflexible, that it 
> > won't
> >> allow people to choose thier own indentation.
> 
> >Lesson taken. I won't give any credence to those who complain about 
> case >sensitivity on the same grounds.
> 
> But this mythical person complaining about case-sensitivity is 
> *correct*;
> Python truly *is* case-sensitive.

And Python truly *does* let one choose the indentation level.


From gregm at iname.com  Mon Jun 12 09:49:00 2000
From: gregm at iname.com (Greg McFarlane)
Date: Mon, 12 Jun 2000 23:49:00 +1000
Subject: Scrolled canvas confusion
In-Reply-To: <8hgjp8$5pt$1@cubacola.tninet.se>; from Andr? Dahlqvist on 05 Jun 2000 at 06:16:23PM
References: <8hg4a3$a09$1@zingo.tninet.se> <8hgd0r$1p8$1@nnrp1.deja.com> <8hgjp8$5pt$1@cubacola.tninet.se>
Message-ID: <20000612234900.05519@nms.otc.telstra.com.au>

I have already mentioned this to Eric, but I may as well say it here
as well :)

Rather than re-invent the wheel, why not just include Pmw in with your
code.  It's easy:  the Pmw documentation tells you how to do it, but
you have to look hard to find it :( Look in the "Freezing Pmw" section
of the "Dynamic loader" page, which has a link from the main Pmw.0.8.4
page.

In summary, you run a script, bundlepmw.py, which creates a single
file, Pmw.py, which you distribute with your application.  No external
dependencies.

Greg

On 5 Jun, Andr? Dahlqvist wrote:
> > If you don't want to use Pmw (which is only Tkinter code anyway), 
> > then look at PmwScrolledFrame.py and extract the bits you want and
> > manage resources directly.
> 
> I definately don't have anything against Pmw, but I just don't want to 
> add another requirement for using this program. The program I'm
> working on is CML2, which is the new Linux kernel configuration tool.
> People are pissed off enough as it is that Eric Raymond chose to use 
> Python for this, so I don't want to add Pmw to the requirements:-)
> 
> If anyone wants to help out and add scrolling to the Tk interface of
> CML2, or just help out in general, you can grab the source from:
> 
> http://www.tuxedo.org/~esr/kbuild
>  
> // Andr?
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From just at letterror.com  Sat Jun 10 05:56:38 2000
From: just at letterror.com (Just van Rossum)
Date: Sat, 10 Jun 2000 10:56:38 +0100
Subject: Stackless Python for the Mac?
In-Reply-To: <l03102800b567b5427839@[193.78.237.152]>
References: <20000609213814.01734E2673@oratrix.oratrix.nl> Message by Jeff
 Senn <senn@maya.com> ,	     09 Jun 2000 15:10:52 -0400 ,
 <zoouk6v7.fsf_-_@SNIPE.maya.com>
Message-ID: <l03102803b567c06b188d@[193.78.237.151]>

At 10:40 AM +0100 10-06-2000, Just van Rossum wrote:
>Erm, I don't think Stackless didn't involve any changes that affect what
>you describe above.

Ick, don't post before having coffee... This should of course read more like
...I don't think the Stackless changes affect what you describe above...

Just





From quinn at photo.ugcs.caltech.edu  Fri Jun 23 06:00:29 2000
From: quinn at photo.ugcs.caltech.edu (Quinn Dunkan)
Date: 23 Jun 2000 10:00:29 GMT
Subject: boolean true and false values.
References: <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl> <memo.20000622231736.45677A@digitig.compulink.co.uk> <slrn8l69kp.geg.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <slrn8l6dff.k96.quinn@photo.ugcs.caltech.edu>

On 23 Jun 2000 07:56:39 GMT, Albert Hofkamp <hat at se-46.wpa.wtb.tue.nl> wrote:
>Today, I think I may have a part of the answer.
>My work is language and tools development of a specification language aimed at
>users with almost no experience in programming.
>Unlike Python, our specification language is statically typed.
>For my work, I program in C++.
>Obviously, that makes me biased towards a statically typed language :-)
>
>My trouble with python is that there are types, and at the same time they are
>not there, which is confusing to me.
>
>As an example:
>
>a="a'slrk;jwhg"
>a=1
>
>works, since Python dynamically adapts the type of a (this may be not entirely
>semantically correct, but I hope you understand my point).
>
>a=1
>print "a"+a
>
>does not work, as Python keeps telling me every time I try.
>
>Maybe I make a reasoning mistake somewehere, but to me a sometimes adapts
>dynamically, and sometimes not.
>As a language aimed at 'the masses', I think this is hard to explain.

Just remember one thing:

In C, variable names have types.  So if you say
int i = 4;
the symbol "i" has the type int, the number 4 doesn't.  In my (non CS-trained)
eyes, C doesn't really have any types, since you can change them 'en passant'
with a cast... or it sorta does, but there's weird holes and tricks.  I guess
that's why they call it weakly typed.  Anyway, I find the situation
non-intuitive.

In python, *values* have types.  So 4 is an integer, '4' is a string, and i is
a symbol, which is just a name for whatever it refers to.  It doesn't have any
type, it's just a name, the actual data has the type.  Like in a filesystem:
the file *name* doesn't have a type, it's the contents (text, binary, jpeg,
etc.).

I don't know a lot of languages, but of the ones I do, C and its derivatives
are the only ones that have the 'variable name is typed' behaviour, and I find
the C way quite difficult to understand properly (and C was my first
language!).

So python never dynamically adapts types, except for integer -> float
promotion things.  It's just that names don't have types at all, values do.

>Then the lack of booleans.
>In my work I noticed that having clean and orthogonal concepts helps starting
>programmers. In our language design the language tries to force users into
>thinking cleanly and orderly, otherwise they end up with a messy program which
>the machine doesn't like.

Yeah, well I do think it's easier to grasp "functions that return true are
true and functions that return false are false, and everything else is an
error" than "well, this this this and that are false, everything else is
true".  But having implicit booleans is just plain easier to type.  And if I
want to write in a B&D language that knows better than I how I should write my
programs, then there are a few really interesting ones out there which are not
python :)

>Having a separate concept 'boolean' forces users to really write tests, rather
>than arbitrary expressions which happen to deliver the right result.

But python is all about arbitrary expressions that happen to deliver the
right result :)

But you're right, just as disallowing mutation forces users to really write
modular functions, rather arbitrary collections of procedures that happen to
give the right result when you test a few boundry conditions.

>Last but not least, I am not sure whether you'd really want the dynamic nature
>of current Python. [ what do you mean, a migration path ? :-) ]
>From programming theory, I learned that there are invariants which have to be
>preserved. Something like "the variable 'total' contains the total sum of
>elements dealt with upto now". Obviously, total has a single type here, which
>makes the entire dynamic typing concept with sequences like 'a="rg"; a=1'
>unwanted.
>
>I realize that introducing new variables on the fly is extremely important for
>ease of hacking^H^H^H^H^H^H^Hprogramming, but letting it mess up the type
>concept seems an unwanted side effect to me.

Sounds like you're making a case that all newbies should be taught haskell or
scheme.  After all, invariants should be preserved.  The best way to do that
is to make them not vary.

>>> Also, I am missing the data type 'set'. It can be faked with a 
>>> dictionary, but it
>>> is just not the same. Any one else got this problem ?
>>
>>It can be implemented very easily as a class wrapper around a list, that 
>>checks for presence before adding an element, and defines union, 
>
>Yes, I know, but why is it not in the language ?

It is in the language (as in 'a builtin type'), just not in the distribution.
I posted a link before.

>[ Before I my mailbox gets filled with flames, I do love Python, and I am
>  pushing it as the standard glue language in our group. The comments above are
>  solely arguments for improving Python further
>]

Gosh, maybe your perfect language is already out there.  haskell has some of
the great things about python like whitespace-denoted blocks, along with
static typing, strict usage of a Boolean type, no options to vary your
invariants, and yes, a Set type.  And there's no need for all that wordy
redundant declaration like in C.


From mjackson at wc.eso.mc.xerox.com  Wed Jun 14 12:23:30 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 14 Jun 2000 16:23:30 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
Message-ID: <8i8bi2$1s8$1@news.wrc.xerox.com>

"Warren Postma" <embed at NOSPAM.geocities.com> writes:

> Having something
> look correct on the screen, but not work because of misuse of tabs versus
> spaces, or incorrect indentation, is something that still continues to bug
> me.

Indulge my curiosity - how many times has this actually happened in
your six months of using Python?

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	You can never solve all difficulties at once.
				- Paul Dirac




From jcoruna at umd.es  Mon Jun  5 13:34:32 2000
From: jcoruna at umd.es (juan Carlos =?ISO-8859-1?Q?Coru=F1a?=)
Date: Mon, 05 Jun 2000 17:34:32 GMT
Subject: MimeWriter and Excel file
Message-ID: <20000605.17343286@jcoruna2.bio.umd.es>

Hello all,
I have the following problem:
I send an email with python using the MimeWriter module and an excel 
file attached. When I send the to myself inside our intranet the 
attchmente arrives well, but if I send the email to anyone outside our 
intranet and open the attachment with MS Excel I see the base64 
encoded characters. Any ideas.
Here is the source:
#!/usr/bin/python

import time, string, StringIO, MimeWriter, base64, os
from smtplib import SMTP

def enviarcorreo(cliente):
    msjhtml = '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 
Transitional//EN">\n' + \
	      '<HTML><HEAD>\n' + \
	      '<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; 
=charset=3Diso-8859-1">\n\n' + \
	      '<META content=3D"MSHTML 5.00.2014.210" 
name=3DGENERATOR></HEAD>\n' + \
	      '<BODY bgColor=3D#ffffff>\n' + \
	      '<DIV><FONT face=3DArial size=3D2><SPAN 
class=3D400082209-17022000>' + \
	      'This is the body' + '</SPAN></FONT></DIV>' + \
	      '<DIV><FONT face=3DArial size=3D2><SPAN 
class=3D400082209-17022000>' + \
	      '</SPAN></FONT></DIV></BODY></HTML>'
    msj = 'This is the body'
    
    # inicializar mensaje
    outputfp = StringIO.StringIO()
    w = MimeWriter.MimeWriter(outputfp)
    w.addheader('From', '<jcc at umd.es>') 
    w.addheader('To', '<jcc at umd.es>')
    w.addheader('Subject', 'test')
    w.addheader('MIME-Version', '1.0')
    w.flushheaders()
    h = w.startmultipartbody('mixed')
    h.write('This is a multi-part message in MIME format.\n')
    sw = w.nextpart()
    sw.startmultipartbody('alternative')
    
    ssw = sw.nextpart()
    ssw.addheader('Content-Transfer-Encoding', '7bit')
    ssw.flushheaders()
    f = ssw.startbody('text/plain', [('charset', 'iso-8859-1')])
    f.write(msj)
    
    ssw = sw.nextpart()
    ssw.addheader('Content-Transfer-Encoding', 'quoted-printable')
    ssw.flushheaders()
    f = ssw.startbody('text/html', [('charset', 'iso-8859-1')])
    f.write(msjhtml)
    sw.lastpart()
    
    
    sw = w.nextpart()
    sw.addheader('Content-Transfer-Encoding', 'base64')
    sw.addheader('Content-Disposition', 'attachment; filename="List de 
precios UMD.xls"')
    sw.flushheaders()
    f = sw.startbody('application/x-msexcel', [('name', 'Lista de 
precios UMD.xls')])
    try:
        base64.encode(open('filename.xls', 'rb'), f)
    except IOError, error:
	print 'Error: %s. No enviado %s a %s' % (error, 
lista[cliente.field('Columnas_precios')], 
cliente.field('Correo_elect___precios_'))
	return
    w.lastpart()
    correo = SMTP('server.bio.umd.es')
    correo.set_debuglevel(0)

    correo.sendmail(fromaddr, 'Juan Carlos Coru?a <jcc at umd.es>', 
outputfp.getvalue())
    correo.quit()






From 12klat at sightreader.com  Thu Jun  1 00:35:00 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Wed, 31 May 2000 21:35:00 -0700
Subject: Is Python really slow?
References: <392F8286.13371C4E@mit.edu> <m3vh00ia9j.fsf@atrus.jesus.cam.ac.uk> <393004FA.846A54BF@mit.edu> <8gtn29$s8u$1@ites.inria.fr>
Message-ID: <3935E7F4.6F0D871F@sightreader.com>

Cedric Adjih wrote:

> Alexander <alecdem at mit.edu> wrote:
> > Michael Hudson wrote:
> >> Alexander <alecdem at mit.edu> writes:
> >>
> >> > Hi
> >> >
> >> > I have implemented  loop in Python
> >> >
> >> > for k in xrange(1,100000000):
> >> >    pass
> >> >
> >> > and the same one in  C
> >> > for ( i = 0 ; i < 1e8; i++ ) ;
> >> >
> >> > and the second one seems to be working hundreds of times faster than the
> >> > first one
> >> >
> >> > Can I do anything with that.
> >>
> >> Well, it's an empty loop so it's pretty useless to all mankind.
> >>
> >> Is this /just/ a troll, or has Python's (relative) lack of speed
> >> bothered you?  Yes, C can spin empty loops faster (especially if the
> >> optimiser removes the loop entirely...), but for doing most
> >> significant things, the difference is much less noticeable.
> >
> > Actually I've created array and did something like this :
> >
> > for k in xrange(1,10000):
> >   for m in range(1,10000):
> >     x[m] =  .... #some calculation of the value of an array's element x[m]
> >
> > and this work's realy slow compared to C. I don't mind it working 3 times
> > slower, but ..
> >
> > Perhaps it is possible, to do something like  implementing loops in C inside
> > Python ??
> > (though it seems to me it will not help)
>
>   You're right, Python is slow. That is fast to program with, but slow
> to run. For many cases, it is fast enough.
> For other cases, its speed is an issue, but programmer time is deemed
> more important.
> For other cases, one can profile, and write a C module (with the
> extension API, SWIG, CXX or SCXX), for the bottlenecks, when possible.
>
> If you need more speed than Python has (but not top speed), and something
> more clean than C++, you can try something in between, such
> as Java or Eiffel.
>
> -- Cedric

In practice (at least in my experience), it is very unusual for python to actually
be a bottleneck.  The reason is that most time is usually spent in library calls
written in C.  In those cases where a profile indicates that python code is a
bottleneck, I write a C module.  In just about all major software projects, only a
very small amount of code (as little as 1%-2%) eats up almost all of the CPU time,
and that is the part you will want to write in C.

When all is said and done, my python projects typically run faster than would have
if they had been written in C because my code is cleaner and my algorithms are
more clever when I write in python.  I spend a smaller portion of my time
debugging, and I don't avoid re-writing code that I don't like (in C or C++ it is
often hazardous to make any changes at all).

For me, the best balance is to write in python and use C or C++ as a slave
language.

- Ken Seehof




From pinard at iro.umontreal.ca  Wed Jun  7 11:50:37 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 07 Jun 2000 11:50:37 -0400
Subject: __getattr__ and __setattr__ troubles
In-Reply-To: Michael Hudson's message of "07 Jun 2000 15:03:11 +0100"
References: <8gehq1$kq6$1@news1.xs4all.nl> <8ggo6m$2v6h$1@news.aros.net>
	<oq1z2ad4wi.fsf@titan.progiciels-bpi.ca>
	<m3u2f6wowf.fsf@atrus.jesus.cam.ac.uk>
	<oqem6abcpf.fsf@titan.progiciels-bpi.ca>
	<m3itvlwpuo.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <oq4s75lcc2.fsf@titan.progiciels-bpi.ca>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> [...] Then the code in ceval.c calls [...]
> Does that help?

It might help understanding the internals if I ever dive in them, yes.
So I'm saving your letter for later study.  And might reply in a few
years, maybe!  Not so long ago, I replied to a letter Guido sent maybe
ten years ago, and sleeping here since.  The conversation merely resumed! :-)

> > > Overriding __getattr__ is good for "oo, ain't that cool" value,
> > > but it is probably best to try to avoid it - if you need it, fine,
> > > but consider alternatives first.

> > [...] implementing __getattr__ is part of an intermediary cleanup, before
> > attempting to move further, so to simplify code all over and gain better
> > maintainability.  [...]

> Fair enough.

And since then, yet another cleanup got rid of __getattr__. :-)

> > P.S. - I hope that Guido will do his best so future Python stays legible.

> Out of curiosity, here do you mean Python source, or the C source of
> Python?  Both are almost uniquely readable, but which did you mean?

Python, the language.  If Python get modified so people may start randomly
casing a single identifier, for example, Python will become less legible.
If Python stops requiring the `:' within compound constructs to announce an
increase in indentation, Python might become less legible.  And such things.

Of course, I'm a proponent of legibility, everywhere.  My C code is
considered legible by those having worked with it, and I hope Guido tries
to write legible C as well.  I did try my best to be legible in Perl like
for everything else, but failed to satisfy myself for big Perl projects.

P.S. - Despite my poor knowledge of English, I also try to be legible while
writing messages.  You might be amused to know that I'm anal enough to
rewrite sentences parts and change words many times, until the paragraphs
fill more nicely.  You might observe that many of my paragraphs roughly
look like rectangles.  Two spaces after a sentence end, in a message,
increase legibility.  So is ragged right with fixed width fonts, rather
the simultaneous left-right justification.  I care a lot about legibility.

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





From dgoodger at bigfoot.com  Thu Jun 29 23:14:21 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Thu, 29 Jun 2000 23:14:21 -0400
Subject: How to convert a string to hex?
In-Reply-To: <slrn8lmiqt.7hd.scarblac-spamtrap@flits104-37.flits.rug.nl>
References: <3952EC0D.D5CF9066@esec.com.au> <wdD45.7813$_55.169378@news2-win.server.ntlworld.com> <395B4C56.19EFB075@wanadoo.dk><slrn8lmiqt.7hd.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <B58188CD.5995%dgoodger@bigfoot.com>

on 2000-06-29 09:16, Remco Gerlich (scarblac-spamtrap at pino.selwerd.nl)
wrote:

> Jens Arnfelt wrote in comp.lang.python:
>> Is there any easy way to go the other way, like convert hex number
>> "0x26100511FFFF" to a string with the the following value: "26100511FFFF"
>> ?
> 
> x = 0x26100     # (your example is too large for my computer)
> 
> y = "0x%x" % x  # String formatting "%x" formatter makes hex numbers
> 
> y is now "0x26100".

No whole number is too large for Python! Just attach an "L", which we all
know is the symbol of: Integer Repair Man! vis,

>>> x = 0x26100511FFFF
OverflowError: integer literal too large
>>> x = 0x26100511FFFFL   # Our hero!
>>> x
41850246397951L
>>> hex(x)        # hex() works where "%x"%x doesn't, with long integers
'0x26100511FFFFL'
>>> hex(x)[2:-1]  # to strip off the leading "0x" and trailing "L", if desired
'26100511FFFF'

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From rsginca at aol.com.NO.Spam  Mon Jun 12 12:33:44 2000
From: rsginca at aol.com.NO.Spam (RSGinCA)
Date: 12 Jun 2000 16:33:44 GMT
Subject: Testing for a file in python?
References: <39450724.231994953@nntp.interaccess.com>
Message-ID: <20000612123344.29950.00002371@ng-fu1.aol.com>

>How would you go about testing if a certain file exists in python?

Use os.path.isfile( path )

Rick
Rick G (in CA)
(To reply remove .NO.Spam from my e-mail address)


From donn at u.washington.edu  Mon Jun 12 13:32:45 2000
From: donn at u.washington.edu (Donn Cave)
Date: 12 Jun 2000 17:32:45 GMT
Subject: Client with select.select()
References: <#F2#YuI1$GA.106@net025s>
Message-ID: <8i36rt$42mq$1@nntp6.u.washington.edu>

Quoth "A[r]TA" <arta at NOSPAMx-stream.nl>:
| What's wrong?
| I have some simple server programmed for experiments. You can chat with it.
| blabla.
| But with the code below, I want to connect with it. In the beginning it
| prints a welcome message.
| Or, it's meant to do. But it only prints the first line from the
| welcome-message and not the
| following. What am I doing wrong? Because in the logfile from my server,
| there's  a line
| in which stands that a login-connection has been made. So that's good.
| But what am I doing wrong? I want the whole message printed, not the first
| line.

The file object you make with makefile() is an input buffer.  When
you invoke readline() on that buffer, it fills itself with whatever
data it can get from its source, the socket, and then gives you the
first line of that data.  Then you throw away the file object, and
the rest of the data from the socket goes with it.

You probably would be better off without the file object.  Use s.recv()
to read from the socket, and string.split(text, '\n') to separate the
lines.  If the last string isn't empty, save it and append the first
string next time around.

I think you also may have the wrong idea about the write-set parameter
to select().  The test with select() is ``will I block?''  That is,
if read-set returns a file descriptor, you may read from that descriptor
without blocking - either there's data, or end of file or whatever may
cause a non-blocking state on a blocking descriptor.  The same is true
of the write set, mutatis mutandi.  The reader on the other end will
probably ignore what you're writing, you just won't block.  Sockets
are usually in this state.  I guess it's a good thing to check anyway,
but by no means is there any implication that the other end of the socket
has requested data.  ("block" means "wait".)

While I'm at it - chr(13) + '\n' is more conveniently spelled '\r\n'.

	Donn Cave, donn at u.washington.edu
----------------------------------------
| from socket import *
| from select import *
|
| s = socket(AF_INET, SOCK_STREAM)
| s.connect('127.0.0.1', 23)
| text = ''
|
| while 1:
|     q = [s.fileno()]
|     rd, wr, er = select(q, q, [], None)
|
|     for n in rd:
|         if n == s.fileno():
|             print rd, wr, er, q
|             text = s.makefile().readline()
|             print text
|
|     for n in wr:
|         if n == s.fileno():
|             g = 'password' +chr(13)+ '\n'
|             s.send(g)


From just at letterror.com  Tue Jun 27 05:31:56 2000
From: just at letterror.com (Just van Rossum)
Date: Tue, 27 Jun 2000 10:31:56 +0100
Subject: Improving cStringIO API (Request For Comments)
In-Reply-To: <39529B56.D54681E0@CORP.ECIRCLES.COM>
Message-ID: <l03102801b57e21040bf2@[193.78.237.126]>

At 11:03 PM +0000 22-06-2000, Michael D. Marchionna wrote:
[ snip ]
>Comments and suggestions?

It seems you're using the wrong object. [c]StringIO is a stream, not an
array. If you need a hybrib between a stream and an array, the thing to do
is to write one. Modifying a standard module is usually a bad idea.
(Especially in this case, where the cStringIO and StringIO modules are two
implementations of the same interface: both need to be kept in sync.)

Check out the array module. It should be easy enough to whip up a class
that does what you want, using arrays underneath.


Just





From pinard at iro.umontreal.ca  Tue Jun  6 15:23:04 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 15:23:04 -0400
Subject: Spurious coercing? [was: Re: 'if foo' calls getattr?!?]
In-Reply-To: Moshe Zadka's message of "Tue, 6 Jun 2000 20:46:57 +0300 (IDT)"
References: <Pine.GSO.4.10.10006062045150.1650-100000@sundial>
Message-ID: <oqsnuqbomf.fsf@titan.progiciels-bpi.ca>

Moshe Zadka <moshez at math.huji.ac.il> ?crit:

> On 6 Jun 2000, [ISO-8859-1] Fran?ois Pinard wrote:

> > And I would presume that if __cmp__ happened to be found, a call to
> > __coerce__ would be tried before going any further?

> Yes, indeed. The reason is that Python tries very hard *not* to be
> smart: the opposite of smart, in this case, being predictable. If you
> know __coerce__ is always called before binary operators take place,
> your code can be that much simpler.

I saw many emails in which people were surprised by Python behaviour, and I
interpret the surprise as a sign that predictability suffers.  For example,
A.__cmp__(B) will not be called for an equality comparison when A "is" B.
Yet, given the meaning of __cmp__, this sounds absolutely correct to me.

Also given the meaning of __coerce__, I would consider just equally correct
to avoid calling A.__coerce__(B) when type(A) "is" type(B).  The spontaneous
predictable thing, here, is that __coerce__ should *not* be used when it is
evident it should not.  Python violates the principle of least surprise,
here, and predictability as well, in my opinion.  That probably depends
on what one gets used to predict! :-).

If it was a good idea to optimise the __cmp__ case, it is also a good
idea to optimise __coerce__.  Also given that not doing it for __coerce__
implies a supplementary (small) burden on the programmer, usually more
than for the __cmp__ case.  I do not see any advantage for not doing it.

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





From djc at itga.com.au  Tue Jun  6 20:56:42 2000
From: djc at itga.com.au (Dave Cole)
Date: 07 Jun 2000 10:56:42 +1000
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net> <87em6as5sp.fsf@heresy.itga.com.au>
Message-ID: <8766rms3zp.fsf@heresy.itga.com.au>

>>>>> "Dave" == Dave Cole <djc at itga.com.au> writes:

>>>>> "Drew" == Drew Fisher <drew at level3.nospam.net> writes:
Drew> Hello.  I am still learning Python, but I came across a need for
Drew> generating dynamic graphs (network load average, uptime, boring
Drew> stuff like this :).

Dave> I started to write a module that would be used for the basis of
Dave> a nice graphing module.  It uses libart, freetype, and libpng.
Dave> You can see where I even started documenting it here:

[snip]

Maybe I should read more carefully next time.  You wanted dynamic, not
static :-)

- Dave



From tim.lavoie at mts.net  Tue Jun 20 10:15:00 2000
From: tim.lavoie at mts.net (Tim Lavoie)
Date: Tue, 20 Jun 2000 14:15:00 GMT
Subject: SOCK_RAW with python socket library?
References: <slrn8ksreq.i2.tim.lavoie@goibniu.beyondtv.net>
Message-ID: <slrn8kuul3.2i1.tim.lavoie@goibniu.beyondtv.net>

In article <slrn8ksreq.i2.tim.lavoie at goibniu.beyondtv.net>, Tim Lavoie wrote:
>
>Hello,
>
>I've been tinkering with some socket programming lately, basically trying
>out ideas I get from browsing W. Richard Stevens' "UNIX Network
>Programming". Anyway, I'd like to play in Python too, so I'm using the
>socket library functions. When I try to use SOCK_RAW though, I get:
>
>   socket.error: (93, 'Protocol not supported')

Uuuuurrrrrrghh. Found it, looking at Jeremy's ping code, mentioned by
Jeff... If I add the third, protocol argument, all is well. 

Sorry, must RTFM. MUST....  GET.... COFFFEEEEE......


-- 
One doesn't have a sense of humor.  It has you.
                -- Larry Gelbart


From mfletch at tpresence.com  Sun Jun 18 19:55:41 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 18 Jun 2000 19:55:41 -0400
Subject: Why is Python so slow ?- revisited.
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL>

If I'm not mistaken, the most commonly executed bytecode in the interpreter
is set_lineno (or something similar).  Later versions of Python have a -O
flag that optimises out those byte codes (at the expense of debuggability).
The pattern encouraged seems to be to do all your debugging without the
flags, then generate -O versions when you want
faster-running-but-harder-to-debug versions.  Never actually _used_ the
pattern myself, of course, so I don't really know how much of a speed up it
gives.  Not particularly sure that comment lines require a set_lineno,
either, suppose I should let someone with a clue answer this question :) .

Also, don't discount the expense of an extra function call.  Function calls
were (and still are to a lesser extent) extremely heavy in Python five years
ago.

Enjoy yourself,
Mike

-----Original Message-----
From: William Dandreta [mailto:wjdandreta at worldnet.att.net]
Sent: Sunday, June 18, 2000 7:33 PM
To: python-list at python.org
Subject: Why is Python so slow ?- revisited.
...
The only thing that makes sense to me is that Python is spending 85% of the
time processing the comment lines and 15% of the time doing useful work.
There is a string.pyc so I would have assumed that it would not contain any
comments or descriptive strings but apparently that's not true.
...



From ge at nowhere.none  Fri Jun 30 15:34:03 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 30 Jun 2000 19:34:03 GMT
Subject: Python MP3 Player
References: <395CE50A.3BA16777@bioeng.ucsd.edu> <4.3.1.2.20000630150810.01ff3df0@phear.dementian.com>
Message-ID: <LC675.2355$iN5.540818@ptah.visi.com>

Steven D. Arnold wrote:
>At 06:24 PM 6/30/2000 +0000, Grant Edwards wrote:
>>In article <395CE50A.3BA16777 at bioeng.ucsd.edu>, Curtis Jensen wrote:
>>
>> >Does anyone know of a Python MP3 Player?
>>
>>Surely you're kidding?
>>
>>Do people really do real-time DSP type stuff in Python.
>
>Why not, provided the most time-sensitive stuff were written in
>C or C++?

That's not really doing it in Python, then, is it?

Heck, you can write device drivers in Python --- as long as
the actual driver is in C.  ;)

>Surely most of the user interface, except perhaps visuals (a
>graphics display that moves in time with the music), could be
>beneficially written in Python.

Sure, you could write the GUI in Python, but I guess when I
think of an MP3 player, I interpret that to mean the thing
that's decoding and playing the music.

-- 
Grant Edwards                   grante             Yow!  Let's climb to the
                                  at               TOP of that MOUNTAIN and
                               visi.com            think about STRIP MINING!!


From rwallace at esatclear.ie  Fri Jun 16 20:12:35 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Sat, 17 Jun 2000 01:12:35 +0100
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <8idie3$l4i$1@mail.pl.unisys.com>
Message-ID: <394AC273.24A9@esatclear.ie>

Clive Tooth wrote:
> 
> Russell Wallace wrote in message <394889C3.1ED5 at esatclear.ie>...
> >Correct me if I'm wrong, but I thought memory was the limiting factor?
> >i.e. that a good algorithm on typical hardware can fill all the
> >available memory with digits of pi in a few days or so?
> 
> You are wrong. The discussion was about methods of calculating digit N of pi
> (in some base) without computing digits 1 thru N-1 . This can be done in
> about O(log(N)) bits of memory.

Yes, that's what I mean.  The O(log N) memory consumption would, it
seems to me, make it possible to reach far higher values of N than if
you had to calculate digits 1 thru N-1, even if the amount of
computation required isn't much different.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From scarblac-spamtrap at pino.selwerd.nl  Sun Jun 18 05:40:21 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 18 Jun 2000 09:40:21 GMT
Subject: string.join question (was Re: urllib question)
References: <Pine.LNX.4.10.10006171424430.19831-100000@mail.fizzylab.com> <8igv24$5li$1@slb7.atl.mindspring.net> <394C4FF9.46B011DC@fizzylab.com>
Message-ID: <slrn8kp63p.b30.scarblac-spamtrap@flits104-37.flits.rug.nl>

paul at fizzylab.com wrote in comp.lang.python:
> Nothing. That's the problem. The item I want to append to the first never gets
> evaluated/appended. I get the first item -- the host name -- and I make a valid
> request for the root of the webserver, not what I want. It's as if I requested
> 'http://www.site.com/' rather than 'http://www.site.com/thispage.html'  which is
> not what I want.
> 
> See the code I submitted initially . . . . the first example works, the second
> does not.
> 
> #!/usr/bin/python
> import string, urllib
> 
> u = urllib.urlopen(string.join(["http://cnn.com/","index.html"]))
> print u.read()
> u =
> urllib.urlopen(string.join(["http://some.host.com/","monstrousQueryString?withParameters"]))
> 
> print u.read()

String.join() puts spaces in between the strings by default. The first one
works because it gets "http://cnn.com", which is incidentally the same page
as "http://cnn.com/index.html".

Maybe just use + instead of string.join? :)

urllib.urlopen("http://some.host.com/"+"monstrousetc") 


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"Early to rise, early to bed, makes a man healthy, wealthy and dead." -- TP


From olczyk at interaccess.com  Thu Jun  8 11:46:23 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Thu, 08 Jun 2000 15:46:23 GMT
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com> <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us> <20000608073337.A324967@vislab.epa.gov>
Message-ID: <3941bf74.166708937@nntp.interaccess.com>

On Thu, 08 Jun 2000 07:33:37 -0400, Randall Hopper <aa8vb at yahoo.com>
wrote:

>Wonder if Python could have something similar.  Folks could post URLs to a
>"getit.py" (or setup.py) script, and users could literally cut-and-paste
>this URL into a Python install tool that would just "do the install".  That
>is, don't bother them with where to FTP or surf, how to get the files,
>where to put the packages, what the name of the distdir in the tarball is,
>etc.  Just install the package and do what it takes to make it happen.
>That'd be cool.

You've got to be kidding! 
Haven't you heard of "melissa" or "ILoveYou"?
When I asked about CPAN, I wasn't concrened about installation.
I can do that myself.
What I was concerned about was a vast repository of available scripts.


From cgfandrich at my-deja.com  Thu Jun 15 13:25:03 2000
From: cgfandrich at my-deja.com (cgfandrich at my-deja.com)
Date: Thu, 15 Jun 2000 17:25:03 GMT
Subject: Embedded Python and Restricted Execution
References: <8i35cl$ch8$1@nnrp1.deja.com> <8F5198BEAgmcmhypernetcom@199.171.54.194>
Message-ID: <8ib3gn$4qv$1@nnrp1.deja.com>

In article <8F5198BEAgmcmhypernetcom at 199.171.54.194>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> cgfandrich at my-deja.com wrote:
>
> >I'm embedding Python in an application and I'd like to be able to run
> >Python code (including callable PyObjects) in a restricted
environment -
> > much like rexec - but I'd like to control the environment in C/C++.
> >I've tried the following and had some success:
> >
> >1. Create a new module.
> >2. Add a "__builtins__" dictionary to the module.
> >3. Copy references from existing "__builtins__" to new "__builtins__"
> >(except for "__import__" and "open" - use my functions for those 2)
> >4. Use the new "__builtins__" dictionary whenever I want to run code
in
> >restricted environment.
> >
> >When importing modules, I've also had to make sure that the module
> >being imported gets the correct "__builtins__" dictionary.  The only
> >problem is ... if I have 2 modules "test1" and "test2" that use the
> >same Python source code, I can't run them in the different
> >environments.  It seems like they both run with
> >whichever "__builtins__" was used to import the first module.
>
> I think you're saying that test1 and test2 both import some other
module?
> If that's the case, the import in test2 will find the module already
in
> sys.modules. So if test1 did an unrestricted import, test2 will find
the
> other module to have an unrestricted builtins.
>
> No easy answers for you. I assume you've checked out what rexec and
bastion
> do (in pure Python). You might consider having 2 separate
interpreters (as
> in Py_NewInterpreter, not 2 instances of python). Or you could
completely
> take over the management of namespaces (as in using the equivalent
of "mod
> = __import__(...)" which doesn't automatically put mod in
sys.modules).
>
> - Gordon
>


Thanks for the help.  That was the problem (test2 found test1's
imported modules in sys.modules).  Ended up writing a C++ version of
rexec (and ihooks) that has its own modules dictionary.

-chris


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


From al at servana.com  Mon Jun 26 16:05:43 2000
From: al at servana.com (al at servana.com)
Date: Mon, 26 Jun 2000 20:05:43 -0000
Subject: Script which only runs in Windows
Message-ID: <8j8d2n+qbe9@eGroups.com>

Hello.

We have an odd problem with Python which we cannot resolve.

Below, is a script we have been running in Win32 for months.

We moved our scripts from Win32 to Slackware Linux and IF statements 
which
are not in functions or classes will cause errors when you
try to run the script.

I have had retyped the script thinking it was the tab/space
issue, but nothing changes - in Linux we cannot use IF outside
a function or CLASS for some reason.

We are posting this here as it seems very serious and we were hoping 
someone else had seen the problem.

-----------------------------------------
import os
import sys
import time
import string
import sugdb
import gadfly

sBillDir = "dirname"
sSugLogin = "login"
sSugPWord = "password"
sSugSite = "171.1.1.1"

#
# The following Dirs MUST end in a path slash
#
SOutputDir = "/home/DATA/"
sDBLocation = "/usr/local/bin/db/"

#
# The script is called without any passin values.
#
# It starts by getting the date and extrapolating strings to use for 
file
# dates
#
# Next, the FTPASGET script is called and its return value is checked
#
# If all was well, then the DB processing script is called
#

#
# The following declarations are very important.
# This reads the date and creates the strings for the following
#

iTime = time.time()

# This is the format string for the date of the files we need to look 
for (like: 06/09)
sFileTime = time.strftime("%m/%d", time.localtime(iTime))
sFileName = time.strftime("%m%d", time.localtime(iTime))
sDirTime = time.strftime("%d%b", time.localtime(iTime))
sDateStart = time.strftime("%m-%d-%y", time.localtime(iTime))
sProcStart = time.strftime("%H:%M:%S", time.localtime(iTime))

#
# here is where we do some work
#
try:
    sResult = os.system('python ftpasget.py -q -g * -d ' + sFileTime 
+ ' -l ' + sSugLogin + ' -p ' + sSugPWord + ' ' + sSugSite + ' ' + 
sBillDir + ' ' + sOutputDir + sDirTime)
except:
    sGetMinutes = 0

iCurrTime = time.time()

#
# PROBLEM: The following will fail unless in a function 
# The failure error is:
#      Name Error: sResult
#
if sResult == 0:
    sGetMinutes = (iCurrTime - iTime) / 60
else:
    sGetMinutes = 0

iTime = iCurrTime

------------------------------------
If we run the code in Win32, or from a function in Linux, it works 
without error.

Any suggestions would be priceless.




From mfletch at tpresence.com  Tue Jun 13 23:42:49 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 13 Jun 2000 23:42:49 -0400
Subject: TCP packet size?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33BA@VRTMAIL>

Hmm, I'm very surprised you're needing to get down to this level when you're
using something as "high level" as XML as your encoding.  XML is so
inefficient as a network protocol that most people would just use standard
TCP streams (i.e. open a stream socket and let TCP take care of the
ordering/guaranteeing/decomposition/reassembly), instead of trying to send
TCP packets as raw packets (heck of a lot of work for no real benefit I can
see).  Or do you mean that you're doing a network sniff and are upset that
the packets are getting broken up and reassembled?
 
I do lots of work with UDP-based protocols where these issues come up, but
since UDP is non-guaranteeing, you wind up explicitly dealing with the
issues regardless (they're not easy, incidentally, lots of soft tradeoffs
dependent on current network conditions, required QOS, priority among
messages, etceteras).  I tend to favour breaking messages up more (I target
256 bytes, include 26 byte UDP header), rather than less, as small UDP
packets may get through a crowded network, while larger UDP messages will
get dropped when load is high (or at least, that's what certain articles
have me believe :) ).  To do it right, of course, you need a dynamically
adapting system that increases message size when load is low (and does a
thousand other optimisations, as well).
 
If I've missed the point, bear with me and explain again.  Since I've not
actually grokked what the "problem" being tackled is, it's quite likely I
have missed the point one way or another.  Enjoy yourself,
Mike
 
-----Original Message-----
From: chris [mailto:chris at rpgarchive.com]
Sent: Thursday, June 15, 2000 2:53 AM
To: Python List
Subject: TCP packet size?


Is there a way to ensure or predict packet size when using TCP?  I'm
developing a Multi-user application that functions much like a muli-player
game.  The applications are sharing xml data.  The xml data can get very big
3000-6000 bytes.  Unfortunately, when the size of the network message gets
over about 2000, its to get broken up into smaller packets.  So, when I send
a 3000 byte message it gets cut up into 2-3 packets.  Anther problem I get
is when I send two small messages back to back, they are sometimes received
together.  Is there something I'm missing here? 

I tried using headers and breaking up messages into smaller chunks, but the
size of the receiving packets doesn't always match what I sent.  I'd
appreciate any help or suggestions on how I can tackle this problem.  I'd be
happy to send someone my network code. 


Thanks 

-- 

chris davis

chris at rpgarchive.com

 

http://www.rpgarchive.com <http://www.rpgarchive.com> 

news and adventure database!

 

http://www.openrpg.com <http://www.openrpg.com> 

open source role playing!
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000613/5293ff7b/attachment.html>

From nospam at bitbucket.com  Tue Jun  6 01:28:51 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Mon, 5 Jun 2000 22:28:51 -0700
Subject: SSL and POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com> <etdog5fcx7a.fsf@w20-575-109.mit.edu> <8hhttv$7r6$1@nnrp1.deja.com>
Message-ID: <R_%_4.30$WI2.3203@newsfeed.avtel.net>

jsantaniello at my-deja.com wrote in message <8hhttv$7r6$1 at nnrp1.deja.com>...
>The cgi (which I have no control over) returns a nice html page that
>says it can't process my request because of a "format or context error".
>
>A simple test html page I made:
><html>
><body>
><form method=POST action=https://www.experian.com/cgi-bin/ccredit.cgi>
><input type=hidden name=0 value=1>
><input type=SUBMIT>
></form>
></body>
></html>
>
>works from a browser but:
>
>import httplib
>h = httplib.HTTPS('www.experian.com')
>h.putrequest('POST', '/cgi-bin/ccredit.cgi?0=1')
>h.putheader('Accept', 'text/html')
>h.putheader('Accept', 'text/plain')
>h.endheaders()
>errcode, errmsg, headers = h.getreply()
>print errcode # Should be 200
>f = h.getfile()
>print f.read()
>f.close()
>
>from a module returns a different page with the "context" error
>business. And it's not a cookie issue.
>
>Am I doing something wrong with specifiing POST?

You're sending it like a GET.  IIRC, the URL is /cgi-bin/ccredit.cgi
and the data is "0=1".  You'll also need a Content-Length header line
with a value of 3:
  h.putheader('Content-length', '%d' % len(data))
  h.endheaders()
  h.send(data + '\r\n')
(Maybe the \r\n is redundant.)
--
Phil Mayes    pmayes AT olivebr DOT com
Olive Branch Software - home of Arranger
http://www.olivebr.com/






From gmcm at hypernet.com  Thu Jun 22 09:45:58 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 22 Jun 2000 13:45:58 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <8F5B6C0A8gmcmhypernetcom@199.171.54.154>

William Dandreta wrote: 

[looking for dBase module]
>I am running uner DOS. Any version will be fine, even older ones as long
>as I can index it.

You're kind of SOL. You need a pure Python solution (because of DOS). 

There's a couple of DBF readers on the Vaults DB page: 
http://www.vex.net/parnassus/apyllo.py/973100124

but you'll have to hack your own writing.

- Gordon


From dworkin at ccs.neu.edu  Wed Jun 28 18:47:57 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 28 Jun 2000 18:47:57 -0400
Subject: smtplib
In-Reply-To: "Peter Timaratz"'s message of "Mon, 26 Jun 2000 18:58:09 -0500"
References: <000801bfdfca$61227820$b6cb45c6@timaratz>
Message-ID: <vndzoo5a0de.fsf@camelot-new.ccs.neu.edu>

"Peter Timaratz" <timaratz at lisco.com> writes:

> As a kindness to all the idiots like me perhaps this intro can be added to
> the smtplib module:
> 
> "This module implements SMTP and does not send e-mail. If you want to send
> e-mail with Python then learn SMTP. If you don't want to learn SMTP then you
> can use one of the many Perl modules available that don't require you to
> know SMTP."

Nice (but misguided) attempt at sarcasm.

The smtp module certainly does send email, and correctly, via smtp.
That is what it is for.

It doesn't write the email for you.

Anyone reading the docs for the module, particularly the example,
would have seen this written in very clear language:

**
Note that the headers to be included with the message must be included
in the message as entered; this example doesn't do any processing of
the RFC 822 headers. In particular, the `To' and `From' addresses must
be included in the message headers explicitly.
**

Perhaps you should read the documentation before you suggest changes to it.

-Justin

 



From jkraska1 at san.rr.com  Thu Jun  8 00:52:30 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 08 Jun 2000 04:52:30 GMT
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com> <fo6ujsofinc9n1ea2dtk96lrpuijje4j2t@4ax.com>
Message-ID: <393F2729.524BBA58@san.rr.com>

> Ok, I got it working now.  The complete lambdabuse.py file, so far, is:
> Functions to aid in grossly abusing lambdas. 

"Hi, everybuddy! My name is Ben Wolfson, and I'm a lambda-abuser"

*everyone at once*: "Hi, BEN!!!!!"




:^)



C


From cstromberger at earthlink.net.cut_here.com.org.gov.net.com  Thu Jun  1 18:54:14 2000
From: cstromberger at earthlink.net.cut_here.com.org.gov.net.com (Chris Stromberger)
Date: Thu, 01 Jun 2000 22:54:14 GMT
Subject: A Mountain of Perl Books + Python Advocacy
References: <20000405053920.20579.qmail@web2102.mail.yahoo.com> <slrn8hdj3i.47b.tsummerfelt1@co146453-a.kico1.on.wave.home.com> <ur9bc7i3p.fsf@ctwd0143.fitlinxx.com> <slrn8hggff.lv.tsummerfelt1@co146453-a.kico1.on.wave.home.com> <39184F02.E6AD745D@basho.fc.hp.com>
Message-ID: <c3qdjs8c1msmr6i8hjp7p5fo862r15nu7k@4ax.com>

On Tue, 09 May 2000 11:46:42 -0600, "John W. Stevens"
<jstevens at basho.fc.hp.com> wrote:

>I would invite everybody to read Eric Raymond's article: "Why Python?" 
>Honestly, I expected to see comments about it already in this news group
>. . . but maybe I was just late, and missed the thread.

Thank you for the invitation.  Where might one find this article?

Thanks,
Chris


From sekter at matavnet.hu  Wed Jun 14 04:35:15 2000
From: sekter at matavnet.hu (Arpad Kiss)
Date: Wed, 14 Jun 2000 10:35:15 +0200
Subject: connectable Python COM server
Message-ID: <01cd01bfd5db$783de6a0$3404a8c0@geometria.hu>

Hi,
I have a connectable COM server implemented in Python, but I am not able to
use IConnectionPoint interface  from VB.
VB needs type library information at design time - if I want to handle
events - and I don't know how to produce it.
Thanks in advance,
Arpad Kiss




From mwh21 at cam.ac.uk  Sat Jun  3 17:12:32 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Jun 2000 22:12:32 +0100
Subject: Automatic reload()?
References: <8hbnt1$62i$1@news2.isis.unc.edu>
Message-ID: <m3r9aexydb.fsf@atrus.jesus.cam.ac.uk>

"Daehyok Shin" <sdhyok at email.unc.edu> writes:

> Is there no program to reload all the changed modules automatically?

I have this in my ~/.pythonrc:

def rr(keepem=sys.modules.keys()):
    for i in sys.modules.keys():
        if type(sys.modules[i]) is types.ModuleType and i not in keepem:
            print "reloading",i
            reload(sys.modules[i])

I can't remember quite what the "keepem" bit's for, but I presume
there was some reason I put it in...

HTH,
Michael

-- 
  Ya, ya, ya, except ... if I were built out of KSR chips, I'd
  be running at 25 or 50 MHz, and would be wrong about ALMOST
  EVERYTHING almost ALL THE TIME just due to being a computer!
                                              -- Tim Peters, 30 Apr 97


From gansevle at cs.utwente.nl  Wed Jun 21 08:03:55 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Wed, 21 Jun 2000 12:03:55 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <8iq78m$u5m$1@nnrp1.deja.com>
Message-ID: <8iqav3$om$1@nnrp1.deja.com>

In article <8iq78m$u5m$1 at nnrp1.deja.com>,
  Fred Gansevles <gansevle at cs.utwente.nl> wrote:
>
> If you use enums only as a 'conveniant way to represent values', the
> next (quick-hack, untested) might be usefull:
>
> class Enum:
>     def __init__(self, **kw):
>         self.__dict__.update(kw)
>
> enum = Enum(Tag = 0, Date = 1, Ammount = 2)
>
> print "Date-field:", enum.Date
>

It just occured to me: If you'd like the enums to be 'global', you can
use the following (also-quick-but-even-dirtyer hack):

def enum(*kw):
    globals().update(kw)

enum(Tag = 0, Date = 1, Ammount = 2)

print "Date-field:", Date

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From chibaA at TinterlogD.Tcom  Fri Jun 23 20:52:18 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Sat, 24 Jun 2000 00:52:18 GMT
Subject: simple question...
References: <3953e13a.761314922@news1.on.sympatico.ca> <3954056E.E3C18DF@bioeng.ucsd.edu>
Message-ID: <395405c4.770668532@news1.on.sympatico.ca>

... alright, I'll try and be more clear.

At the end of a script A, I would like to initiate script B.  I don't
want to run script B as a function in A; rather, I want to quit A, and
start script B.  Is this possible?




On Fri, 23 Jun 2000 17:48:46 -0700, Curtis Jensen
<cjensen at bioeng.ucsd.edu> wrote:

>chibaA at TinterlogD.Tcom wrote:
>> 
>> Hello,
>> 
>> I'm sure this is a simple question...
>> 
>> how would I call another python script from within a script.  ie. I
>> want to basically stop the script that's running, and start running
>> another.  I don't have any particular functions in the target script,
>> and I don't want to continue running the script I'm currently in (ie.
>> I don't want to use import xxxxx).  I would also like to avoid a
>> system call if possible.
>> 
>> Any ideas?
>> 
>> TIA!
>
>I'm not sure what you're asking.  However, anytime you want to run a
>script, you have to import it.  If you run it from the command line,
>it's implicitely imported.
>
>
>
>
>
>
>
>
>
>
>-- 
>Curtis Jensen
>cjensen at bioeng.ucsd.edu
>http://www-bioeng.ucsd.edu/~cjensen/
>FAX (425) 740-1451



From DOUGS at oceanic.com  Sat Jun 17 17:10:10 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Sat, 17 Jun 2000 11:10:10 -1000
Subject: Python Grep (was: writing (Gnu)MAKE in Python)
Message-ID: <8457258D741DD411BD3D0050DA62365907A23A@huina.oceanic.com>

[John said:]
> > ...I would like to have Bash, sed, and grep as Python
> > programs rather than compiled C.
> 
[and Corageous asked:]
> Why?

Consider this an attempt at translation.

There have been posts in the past about having a Python shell.  Perhaps
thats what John wants.  Not that I'd like that.  It may be that John wants
the functionality of sed and grep more easily accessible in Python.  That is
something I care about (thus the self serving attempt at manipulating this
thread. ;-)

Grep functions in particular are something that I wonder about.  The
following is an experiment:

#!/usr/bin/python
#
#       pygrep.py
#
#
"""     An attempt to compare the use of the grep
        command with a 'Pythonic' method of finding
        all lines in a file that contain a string.
"""
import string, os

def pygrep(the_string,the_file):
    """ Search for and return all occurrances of lines
        in the_file that contain the_string.
        This is simplistic and unprotected but its
        purpose is only to learn how to make it fast. """

    find = string.find

    holder = open(the_file,'r')
    while 1:
        line = holder.readline()
        if not line:
            break
        if find(line,the_string) <> -1:
            print line

def mygrep(the_string,the_file):
    """ This is usually what I do when I need this. """

    command = 'grep %s %s' % (the_string,the_file)
    response = os.popen(command,'r')
    lines = response.read()
    print lines

if __name__ == '__main__':
    import time

    test_file = "/usr/local/devices/motor.10"
    test_string = "26086594"

    first = time.time()
    pygrep(test_string,test_file)
    second = time.time()
    mygrep(test_string,test_file)
    third = time.time()

    print "Python: %s" % ((second - first),)
    print "OS    : %s" % ((third - second),)

I run this and get:

$ ./pygrep.py
Punc,202,26086594,SG6,1,2,30days,-9,53,732970,358877,0,67,176761,39326,1,24.
94.89.251,

Punc,202,26086594,SG6,1,2,30days,-9,53,751771,371322,0,68,179118,40654,1,24.
94.89.251,

Punc,202,26086594,SG6,1,2,30days,-9,53,787679,393889,0,68,184387,42710,1,24.
94.89.251,

Punc,202,26086594,SG6,1,2,30days,-9,53,732970,358877,0,67,176761,39326,1,24.
94.89.251,
Punc,202,26086594,SG6,1,2,30days,-9,53,751771,371322,0,68,179118,40654,1,24.
94.89.251,
Punc,202,26086594,SG6,1,2,30days,-9,53,787679,393889,0,68,184387,42710,1,24.
94.89.251,

Python: 6.7979799509
OS    : 0.0741490125656

Am I missing a Pydiom that would narrow the gap or is using the OS function
the best way?

-Doug-



From stalacticus at my-deja.com  Sun Jun 18 23:31:00 2000
From: stalacticus at my-deja.com (stalacticus at my-deja.com)
Date: Mon, 19 Jun 2000 03:31:00 GMT
Subject: JOB: Python Programmer wanted. Sydney Australia
Message-ID: <8ik45k$rd0$1@nnrp1.deja.com>

If job postings are not part of this newsgroup's charter, sorry,
but I couldn't find a monthly posting.

Also, I tried to add this job to www.python.org, but the job
board doesn't seem to be getting updated.  Where do python
programmers looking for jobs hang out?


############################

Memetrics (Sydney, Australia)

Memetrics is a group of leading international engineers, scientists,
and business professionals who are working together to implement a
range of radically new presentation optimization systems using Lisp and
Python. We have already attracted substantial private and government
funding and have implemented working prototypes of our system.

Now we are conducting an international search for smart and creative
software engineers who would like to join our Sydney endeavor. Because
we?re a young organization, you will have an excellent opportunity to
have an impact on the evolution of the company. We offer a great work
environment full of interesting and challenging problems.

Required Experience and Skills

Ideally you will possess at least three years experience in an object
oriented development environment. Exposure to Lisp, Python or CORBA in
a commercial setting would be advantageous. Candidates should be
comfortable using Unix and/or NT.  Successful candidates also must have
excellent communication skills, and be willing to work closely within a
tight-knit team.

Salary

We want the best and are willing to pay accordingly. Compensation is
based on skills and previous experience in addition to a generous pre-
IPO stock option package. Australian and overseas applicants will be
considered. Come to the city of which Lonely Planet says ??you would
have to die and go to heaven before you see a more spectacular setting
for a city.? Please send r?sum?s to jobs at memetrics.com.




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


From mwh21 at cam.ac.uk  Tue Jun 13 12:52:04 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 2000 17:52:04 +0100
Subject: How many times inner functions are compiled ?
References: <000e01bfd555$ef3f4220$582d153f@tim>
Message-ID: <m3og5534nf.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Jerome Quelin]
> > def outer():
> >     def inner():
> >         print "spam"
> >     inner()
> > outer()
> > outer()
> >
> > In the snippet code above, I call outer() two times. I would like to
> > know how many times python will compile the inner function ?  One time
> > when compiling the script/module, or two times, ie at every invocation
> > of the outer function ?
> 
> Just van Rossum gave you the best answer, but since he was out-voted 2 to 1
> I thought I'd jump in to even the odds <wink>.  "inner" is compiled exactly
> once, independent of how often "outer" is called (even if you never call
> outer, "inner" is compiled once).  However, the "def" stmt is executable in
> Python, and at runtime, each time "outer" is called, upon hitting the "def
> inner():" block the (pre)compiled code object for "inner" is wrapped up in a
> function object and bound to "outer"'s local vrbl name "inner".  The
> "function object" adds stuff like values for default arguments to the code
> object (but you don't have any default arguments in this example, so you'll
> have to use your imagination).

To illustrate with code:

>>> def outer():
...     def inner():
...         print "bob"
...     return inner
... 
>>> outer() is outer()
0
>>> outer().func_code is outer().func_code
1

Cheers,
M.

-- 
  how am I expected to quit smoking if I have to deal with NT 
  every day                                                -- Ben Raia


From emile at fenx.com  Thu Jun 22 11:36:52 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Jun 2000 08:36:52 -0700
Subject: boolean true and false values.
References: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <0b1501bfdc5f$c2a8a7e0$1906a8c0@fc.fenx.com>

----- Original Message -----
From: "Tony J Ibbs (Tibs)" <tony at lsl.co.uk>
<snip>
> (This explanation does not, of course, work for my pet
hate in Python, which
> is that 0 at the start of a number means it is in octal
(does *anyone* care
> about octal anymore?)
<snip>

Only when I want to know about the bits.  It's so
much easier to recognize 122 octal as having
alternate bits set than the equivalent in decimal
or hex.


Emile van Sebille
emile at fenx.com
-------------------






From newsuser at linux2.johnmckown.net  Fri Jun  9 06:35:13 2000
From: newsuser at linux2.johnmckown.net (John McKown)
Date: Fri, 9 Jun 2000 05:35:13 -0500
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <slrn8k1i31.qtn.newsuser@linux2.johnmckown.net>

On Thu, 8 Jun 2000 23:18:26 -0700, Lawrence Kesteloot <lk at veriomail.com> wrote:
>
>If one of the goals of Python 3000 is to make it easy for new
>programmers to learn, how about removing the distinction
>between = and ==?  The only kinds of examples I can think of
>where this is useful is something like:
	[snip]

I totally disagree. I, personally, dislike having a single token mean more
than one thing. The concepts between assignment (=) and equality (==) are
totally different. The use of = and == might be confusing for a newbie.
But somebody who really wants to learn will learn. I know that I have problems
because I use multiple languages, and sometimes I try to put Perl expressions
in Python programs (doesn't work <grin>). In your example of A=B=5, you'd
have to figure out which of the two incompatable meanings is to be the correct
one. Is it A is assigned the value of true or false based on B being equal
to 5 or is it A is assigned the value of B which is assigned the value of 5 
(meaning A & B are both assigned the value of 5). Some would want it one
way, and some the other. If = is assignment and == is equality, then there
is no confusion. And I have different languages which intepret it differently,
so I make mistakes when I move between them. That's why many of my programs
don't do that sort of think. I use simplier, unambiguous statements and hope
that the compiler / interpreter is "smart".

Anyway, just my thoughts, remember how much they cost you.
John


From tormy at stud.ntnu.remove_this.no  Tue Jun  6 10:45:29 2000
From: tormy at stud.ntnu.remove_this.no (Tor Andre Myrvoll)
Date: 06 Jun 2000 10:45:29 -0400
Subject: Help, Image Filters with numpy
References: <393803A2.9AE23E9C@visionart.com> <8h9284$rc2$1@info3.fnal.gov> <39380EC8.3100E9F5@visionart.com>
Message-ID: <qcog5ezx4m.fsf@durga.research.bell-labs.com>

* Pete Shinners
| i'll be sticking with smaller kernels, no larger than
| 5x5 likely. nonetheless, i've seen a lot about FFT's
| but not much 'learning' or 'beginning' info. is there
| somewhere i can go to get a handle on the basics of FFT?

The FFT is nothing but an effective way of computing the discrete
Fourier transform (DFT) of a finite length signal. Any introductionary
text on digital signal processing explains these things without being
too mathematical about it (although you need to know your basic linear
algebra and analysis). I like the following book:

J. G. Proakis & D. G. Manolakis, 
"Digital Signal Processing, - Principles, Algorithms, and Applications"

Here's a link as well:

http://dir.yahoo.com/Science/Mathematics/Software/Fast_Fourier_Transform__FFT_/

-- 
- Tor Andre Myrvoll
"Winos don't march." - F. Zappa


From mo at nospam.com  Wed Jun 21 19:12:17 2000
From: mo at nospam.com (Mo)
Date: Wed, 21 Jun 2000 16:12:17 -0700
Subject: Regex engines
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org> <8igpab$7tf$1@slb2.atl.mindspring.net> <3950B2BE.AF756B9F@hursley.ibm.com> <3dd7laes4x.fsf_-_@mems-exchange.org>
Message-ID: <39514BD1.227B1131@nospam.com>

Andrew Kuchling wrote:
> 
> [Posted only to c.l.tcl and comp.lang.python; comp.lang.perl doesn't exist.]
> 
> Paul Duffin <pduffin at hursley.ibm.com> writes:
> > front end to a C++ program. Note that is Tcl, not Tk. And someone posted
> > don't use Tcl, use Python because it has a Tk binding and "Tcl is crap".
> 
> Ummm... If you mean the thread "C++ and Tcl", that person was writing
> about Perl, not Python.
> 
> > I would love to see (and this is not my idea) all of our communities
> > working together to produce really simple building blocks which can be
> > used by all of us. e.g.
> >       Regular expression engine.
> 
> Unfortunately all the regex engines seem inextricably tied in to their
> respective languages.
> 
> * Python 1.5 uses a hacked version of PCRE, which is a very
> good library, but it doesn't handle Unicode, though its author has said
> he's thinking about adding UTF-8 support.  I think PHP and some Scheme
> interpreters also have bindings for PCRE.
> 
> * Python 1.6 will use a new regex engine called SRE to handle both
> 8-bit and UTF-16 strings; the compiler is still written in Python at
> this point, though the engine itself is in C.
> 
> * Perl 5's regex engine is unreadable code, and disentangling it from
> the rest of Perl seems impossible.
> 
> * Tcl uses a Unicode-aware engine written by Henry Spencer, but it's
> not available as a standalone library.  At the Ottawa Linux conference
> last July I asked about releasing it, but have seen no sign of it
> almost a year later.
> 
> * Ruby seems to use a modified version of one of the GNU regex
> libraries; I can't tell if it's GNU rx or regex.
> 
> Unifying this situation seems impossible; maybe when SRE's compiler is
> translated to C, or if Spencer's library is released in a standalone
> version...


Did you know that there is a standalone version of henry's regexp
engine ported to Java? It is included in the Jacl interp (Tcl in Java),
but it is just a package, so you should be able to use it anywhere.
In fact, it is just two file Regexp.java and Regsub.java. You should
check it out.

Mo DeJong
Red Hat Inc


From gmunsey at adobe.com  Mon Jun  5 09:01:08 2000
From: gmunsey at adobe.com (Grant Munsey)
Date: Mon, 5 Jun 2000 06:01:08 -0700
Subject: Windows 2000 Server compatibility?
References: <393B4343.E3A96525@micron.net>
Message-ID: <5wN_4.384$qF1.312697@news.pacbell.net>

Python 1.5.2 and 1.6a2 both work as advertized on Win 2K pro
and Win 2K server

"Peter Immarco" <android at micron.net> wrote in message
news:393B4343.E3A96525 at micron.net...
> Is there a version of Python that runs well on Windows 2000 Server?  Any
> gotchas?
>
> Peter.
>
>




From mfletch at tpresence.com  Mon Jun 12 10:35:25 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 12 Jun 2000 10:35:25 -0400
Subject: where is PyOpenGL?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33AF@VRTMAIL>

Win32 binary installer is available from the PyOpenGL _project_ page (we
haven't put up an official homepage on sourceforge yet, should do that
soon).

See: http://sourceforge.net/projects/PyOpenGL/

Enjoy,
Mike

-----Original Message-----
From: Randall Hopper [mailto:aa8vb at yahoo.com]
Sent: Monday, June 12, 2000 10:09 AM
To: Gabor Papp
Cc: python-list at python.org; pyopengl at egroups.com
Subject: Re: where is PyOpenGL?


Gabor Papp:
 |Where can I download PyOpenGL for win32 from? I've tried the
 |following:
 |http://starship.python.net/~da		- the link seems to be
broken
 |http://PyOpenGL.sourceforge.net/	- it's empty

   http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=626358133

is David's last post on the topic.  On the PyOpenGL mailing list, Robin
<rjh at cita.utoronto.ca> mentions you can fetch the latest PyOpenGL from CVS
using the instructions at:

   http://www.pythonpros.com/cvs.html

See the "ViewCVS Page" link at the bottom as well.  Or if you'd like a
tarball of the latest release (PyOpenGL-1.5.5), let me know via and I'll
send you one.

-- 
Randall Hopper
aa8vb at yahoo.com

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



From pointal at lure.u-psud.fr  Mon Jun 19 07:54:54 2000
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Mon, 19 Jun 2000 11:54:54 GMT
Subject: how to pause a forked process
References: <3949019E.1D9B53B3@sec.noaa.gov>
Message-ID: <394e09d1.1044958891@news.u-psud.fr>

On Thu, 15 Jun 2000 10:17:40 -0600, j vickroy <jvickroy at sec.noaa.gov>
wrote:

>: Hello,
>: 
>: How can execution be paused in a child process?
>: 
>: This is what I tried:
>: 
>: import os
>: import time
>: 
>: pid = os .fork()
>: 
>: if pid == 0: # child
>:     # do some work
>:     time .sleep (5)
           ^ Is there really a space here?
>: 
>: else: # parent
>:     # do some work
>: 
>: The sleep() call is flagged as:
>:     AttributeError: sleep


---
Laurent POINTAL - CNRS/LURE - Service Informatique Experiences
Tel/fax: 01 64 46 82 80 / 01 64 46 41 48
email  : pointal at lure.u-psud.fr  ou  lpointal at planete.net 


From tanzer at swing.co.at  Mon Jun 19 01:47:50 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 19 Jun 2000 07:47:50 +0200
Subject: Circular import 
In-Reply-To: Your message of "Sun, 18 Jun 2000 15:16:59 EDT."
             <394D202B.1DF60864@colconsulting.com> 
Message-ID: <m133uPW-000wcEC@swing.co.at>

"Frank V. Castellucci" <frankc at colconsulting.com> wrote:

> With the assumption that class A and class B reference each other, how
> do you avoid the Python error with:
> 
> ---------- Composite.py -----------
> 
> from Leaf import Leaf
> 
> class Composite:
> 	def __init__( self ):
> 		self.__theLeaf = Leaf(self)
> 
> -------- Leaf.py --------------
> 
> from Composite import Composite
> 
> class Leaf:
> 	def __init__( self, aComposite ):
> 		self.__theComposite = aComposit

In your example, you don't actually reference Composite in Leaf.py --
therefore no import is necessary.

If you really need to reference Composite in Leaf.py, there are two
simple options :

- import inside the function(s) which reference Composite

- put the import statement at the end of Leaf.py (this only works if
  all references to Composite are inside functions, or more
  accurately, if Composite is not referenced bvy any code executed
  before the import statement)

Should you want to cross-reference at the module level you are out of luck.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From fiona at sitegnome.com  Thu Jun 22 08:09:32 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 22 Jun 2000 12:09:32 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 22nd, 2000
Message-ID: <20000622120932.24761.qmail@synop.com>

Greetings!

Below are the entries that made it into http://python.faqts.com tonight.

Fiona


## New Entries #################################################


-------------------------------------------------------------
Is there a Win32 module that I can use to set directory privileges, that has the functionality of cacls.exe using python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3914
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond, Albert Hopkins

Check the win32security module plus the pywintypes module for direct 
access to security objects.

The "problem" is that Python makes no attempt to make it easier than it 
is in C/C++.  There is a fair bit of magic to weave, involving a number 
of steps to successfully create the ACLs, ACEs and SDs!  Getting the 
code wrong can cause a few problems ;-)

There are no "nice" wrappers around this stuff simply because I 
personally dont have the experience with them.  The few times I need to 
do security related things I search MSDN for C sample code, and 
translate it to the relevant win32security/pywintypes calls.

As an example of the verbosity required, below is some code from the 
book examples <plug>Chapter 16 - Windows NT Administration</plug> that 
simply creates a security descriptor ready to be applied to the 
necessary object.
(In fact, this is probably a reasonable percentage of what you need)

# A utility function that creates an NT security object for a user.
def CreateUserSecurityDescriptor(userName):
    sidUser = win32security.LookupAccountName(serverName, userName)[0]
    sd = win32security.SECURITY_DESCRIPTOR()

    # Create the "well known" SID for the administrators group
    subAuths = ntsecuritycon.SECURITY_BUILTIN_DOMAIN_RID, \
               ntsecuritycon.DOMAIN_ALIAS_RID_ADMINS
    sidAdmins = win32security.SID(ntsecuritycon.SECURITY_NT_AUTHORITY,
subAuths)

    # Now set the ACL, giving user and admin full access.
    acl = win32security.ACL(128)
    acl.AddAccessAllowedAce(win32file.FILE_ALL_ACCESS, sidUser)
    acl.AddAccessAllowedAce(win32file.FILE_ALL_ACCESS, sidAdmins)

    sd.SetSecurityDescriptorDacl(1, acl, 0)
    return sd


-------------------------------------------------------------
Is there a help or man function, like in unix-like OS, that we can use to describe any function of python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3910
-------------------------------------------------------------
Fiona Czuczman
Richard Chamberlain, William Dandreta

You can use docstrings which are there for most things.

import sys
print sys.__doc__

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

There is a Windows help file called Pythlp.chm that I use. You can 
download from the python webside. It has an index and you can seach for 
keywords.


-------------------------------------------------------------
How do I get Tkinter to resize frames etc. when their parents (e.g. a Toplevel instance) are resized?
http://www.faqts.com/knowledge-base/view.phtml/aid/3913
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain, John Grayson

It depends on the kind of manager you use, but I'll use pack as an 
example.

from Tkinter import *
root=Tk()
Frame(root,borderwidth=3,relief=GROOVE).pack(side=LEFT,fill=BOTH,expand=
YES)
root.mainloop()

If you look at www.manning.com/grayson you can download two chapters of
Python and Tkinter programming one of which is screen layout. (of course
you'll be much better buying a copy ;)


You'll also find over 20,000 lines of example code at the same site...


-------------------------------------------------------------
Is there some documentation for the python interface to postgresql?
http://www.faqts.com/knowledge-base/view.phtml/aid/3911
-------------------------------------------------------------
Fiona Czuczman
Rod Haper,Erno Kuusela

The PyGreSQL package ftp://ftp.druid.net/pub/distrib/PyGreSQL.tgz
(Latest released version) has a tutorial directory.

The README that comes with the module documents the postgresql-specific 
api. the most recent versions support DB-API also.


-------------------------------------------------------------
Why does 123.4+2=125.40000000000001?
http://www.faqts.com/knowledge-base/view.phtml/aid/3909
-------------------------------------------------------------
Fiona Czuczman
Ruud de Rooij

>From http://www.python.org/1.6/

  The repr() of floating point numbers now gives full (17-digit)
  precision. This means that a number like 8.1 (which cannot be
  represented exactly in binary) will show up as 8.0999999999999996 when
  displayed using repr(). This shouldn't break any code but it may
  confuse users, especially since the interactive shell uses repr() to
  display all floating point results. Use str() or explicit precision in
  string format (e.g. "%.12g" % x) to see the the more familiar, rounded
  form.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I get the current date and time?
How do I print out the current time?
http://www.faqts.com/knowledge-base/view.phtml/aid/3600
-------------------------------------------------------------
Mike Hostetler, Fiona Czuczman
http://www.python.org/doc/current/lib/module-time.html,Jeff

I didn't think that the official documentation was very clear on this 
(but it is there), so I'm adding it here.

If you do

>> import time
>> print time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(time.time()))

You'll see something like:

2000-06-08 15:05:09

To dissect this:

time.time() gets the time in seconds since the epoch

time.localtime(secs) creates a tuple

time.strftime(string,tuple) creates a string of the time, using the 
formatting in "string" (that is what the "%"'s are doing there).
To see the full table of formatting options, see 
http://www.python.org/doc/current/lib/module-time.html

Also, the system man page, ('man strftime')? documents a closer set of 
the formatting options actually supported by the libc, which is what the 
python interpreter actually calls into. this will come into play mostly 
on 'older' machines, but python runs on them too!






From millja at wwc.edu  Wed Jun 14 10:35:59 2000
From: millja at wwc.edu (millja)
Date: Wed, 14 Jun 2000 07:35:59 -0700
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com> <Pine.GSO.4.20.0006140956480.27166-100000@luna.dsv.su.se>
Message-ID: <20c2f050.1f2b139f@usw-ex0109-069.remarq.com>

Hum, to tell you the truth, I'm pretty sure its installed
by default with Slackware 7.  If not by default, I know its
on the CD I have.  But you are correct when you say that
most slackware users wouldn't have a problem installing it
anyway.
jason


* Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautiful


From jeremy at beopen.com  Wed Jun 21 14:24:41 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 21 Jun 2000 14:24:41 -0400
Subject: List without duplicates?
References: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net> <3950892E.62BF3B@marconicomms.com>
Message-ID: <6CD53C35918DB5FE.AEBA8270A2610649.F703A3BCCDE5C9C7@lp.airnews.net>

Wolfgang Grafen <Wolfgang.Grafen at marconicomms.com> writes:

> > Is there a built in Python function that eliminates duplicates in a list?

> Try this ...
> 
> old_list = [1, 2, 3, 2, 1, 4, 3, 5, 2]
> 
> 
> unique_list = []
> 
> for member in old_list:
>     if not member in unique_list:
>         unique_list.append(member)
> 

This is a bad way to do it.  The "not member in unique_list" test
involves a linear search of unique_list.  This gives your uniquify
function O(n^2) behavior.  The dictionary version posted earlier uses
the dictionary's hash table and achieves constant time for the test.
It will perform much better.

Jeremy


From kens at sightreader.com  Tue Jun 20 18:29:41 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Jun 2000 15:29:41 -0700
Subject: Speed with Python
References: <QH645.94$QA1.13929@news.uswest.net>
Message-ID: <394FF055.47163332@sightreader.com>

In situations where I want speed, I prototype in python, then write extensions.
This is a convenient approach since the test suite you write for the prototype
can be used on the extension unchanged.

I do all GUI work in python (wxPython) and I have not noticed a difference in
speed as compared to C++, probably because most time is spent in library calls
anyway.  I suspect you will have similar results with database work, though I
haven't done high volume database work in python myself.  The benchmarks that
show 10-20 fold differences are doing number crunching or equivalent low-level
processing.  In real world situations, where your inner loop is calling into
libraries, the difference is negligable in my experience.

Jeff Massung wrote:

> Many threads with this kind of name have popped up, comparing the speed of
> Python to C, Java, etc. and trying to determine how fast some of the
> algorithms in Python are.
>
> My question however, is how many of you who are Python gurus/junkies (no
> offence, I myself am a C and Forth junkie) :) use Python for speed intensive
> processing (not graphics, but databases and such)? Does efficiency of
> developement outweight the eventual speed of use?
>
> Cheers,
> Jeff

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000620/47f5d59b/attachment.vcf>

From mfletch at tpresence.com  Fri Jun  9 02:02:07 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 9 Jun 2000 02:02:07 -0400 
Subject: Why is Python so slow?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F339D@VRTMAIL>

1) You're using an _ancient_ version of Python... 5 years old?  Something
like that.  There's been at least 1 factor-of-two increase in speed since
then.
2) You might possibly be using poor python idioms.  For instance, you may be
using file.readlines() or the like, which, on a 486 (likely with limited
memory) would wind up taking forever to read the file into memory, then
would thrash as it processed the file, and finally would thrash on reading.
3) Low-powered machines often require low-level tools.
4) Python will seldom be as fast as C save where you are making heavy use of
extensions (such as numeric python)

On a PIII 500 under (stackless) Python 1.5.2, the following gives these
results...

p:\>speedy
Data file creation: 16.6439999342 seconds
Data file processing (simple): 20.239000082 seconds

Creation generates 360,000 random numbers and writes them to disk.
Processing consists of reading in each line, stripping off the white space,
splitting into fields, converting to floats, rounding to 0 decimal places,
then writing back to disk.
Total memory usage is about 2.5MB on my machine.

By this (very rough) timing, it seems the 486 is about 1/300th the speed of
the PIII?  I don't recall them being that slow, but I don't have any lying
around on which to test.  Factor in 1.5.2 vs 1.2 and it'd still be 1:150 .
I'm guessing there's something else causing a slowdown, or something
significantly different about your processing.

Enjoy yourself,
Mike

8<_________ speedy.py ___________
import string, random

def createDataFile( file = 'z:\\temp\\test.dat' ):
	file = open( file, 'w')
	for line in xrange( 120000 ):
		file.write( '%s %s %s\n'% (random.random(),
random.random(),random.random()) )
	file.close()
def processDataFile( file = 'z:\\temp\\test.dat' ):
	outfile = open( file+'.out', 'w')
	file = open( file)
	data = file.readlines( 16000 )
	while data:
		for line in data:
			line = map( round, map( float, string.split(
string.strip(line) )))
			if line:
				outfile.write( '%s %s %s\n'% tuple(line) )
		data = file.readlines( 16000 )

if __name__ == '__main__':
	import time
	t = time.time()
	createDataFile()
	print 'Data file creation: %s seconds'% (time.time()-t)
	t = time.time()
	processDataFile()
	print 'Data file processing (simple): %s seconds'% (time.time()-t)


-----Original Message-----
From: William Dandreta [mailto:wjdandreta at worldnet.att.net]
Sent: Friday, June 09, 2000 1:19 AM
To: python-list at python.org
Subject: Why is Python so slow?


I am just learning Python and I wrote a little program to modify a text file
with 120,000 lines. It took 1hr and 30 mins. I wrote a C program that
created the text file from a .dbf and it only took 2 mins.

Both programs ran on the same machine (486-133mhz) under Concurrent DOS (a
multi-user mutitasking OS).

The Python version was 1.2.

Why does it take Python 50 times longer than C?

Bill Dandreta


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



From 12klat at sightreader.com  Mon Jun  5 09:02:04 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Mon, 05 Jun 2000 06:02:04 -0700
Subject: Starship ever gonna fly?
References: <f9oljso6hl7r7blv64an77h8lfj6dhr6f9@4ax.com> <20000605012821.V469@xs4all.nl>
Message-ID: <393BA4CB.C4DC60D7@sightreader.com>



Thomas Wouters wrote:

> On Sun, Jun 04, 2000 at 04:09:58PM -0700, Kirby Urner wrote:
>
> > The loss of Starship was a real tragedy.  Glad to
> > read it's recovering.  But is it?  The same not-
> > so-reassuring text has remained unchanged for some
> > time.
>
> > Part of what made the Python community seem so vital
> > was all these add-ons (like PyOpenGL).  But with
> > Starship down for the count, so many of the pointers
> > in Vaults of Parnassus are just dead, stone dead.
>
> If you search the list archive or browse the newsgroup for the last few days,
> you'll see that Starship is moving, or has been moved, this weekend. It'll
> take a few days for the DNS changes to settle, though. After the move, it
> should be more actively maintained. (By BeOpen, if I'm not mistaken.)
>
> > Whoever killed Starship did real and lasting damage
> > it seems.  Did anyone call the FBI?
>
> I was under the impression that the Big Crash was hardware-related, not a
> malicious attack.

The Perl people did it.

> --
> Thomas Wouters <thomas at xs4all.net>
>
> Hi! I'm a .signature virus! copy me into your .signature file to help me spread!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000605/41843ca3/attachment.vcf>

From akuchlin at mems-exchange.org  Fri Jun 16 11:08:53 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 16 Jun 2000 11:08:53 -0400
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu> <39498084.972E36D2@san.rr.com>
Message-ID: <3dr99xfyt6.fsf@amarok.cnri.reston.va.us>

Courageous <jkraska1 at san.rr.com> writes (about Common Lisp):
> Very high learning curve. Syntactically heavy weight. High
> current cost of maintenance. Continually deteriorating
> support and availability of technical expertise. ...

What about Dylan as an alternative?  I've only read a little bit about
it, and not actually used it for anything, but it doesn't seem like
too large a stretch for Python users.  The problem is, of course, that
Dylan isn't very widely used, and hence there aren't many fancy tools
for it, but it still seems like the closest existing thing to an
optionally statically typed Python.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Of course, for a lot of people, death was a welcome change. Grinding poverty
takes the edge off most things, including life.
  -- Tom Baker, in his autobiography




From rupole at compaq.net  Sun Jun 11 13:29:47 2000
From: rupole at compaq.net (Roger Upole)
Date: Sun, 11 Jun 2000 17:29:47 GMT
Subject: Pythonwin is Great!
References: <8hp2ku$a25@news.or.intel.com> <0ud7ks4t2b1s2an6es8q42lc5bd2rp1oeu@4ax.com>
Message-ID: <f0Q05.23905$wp.1821758@pouncer.easynews.com>

I agree entirely.  When I get ready to use a package or module
I'm not familiar with, the first thing I do is fire up Pythonwin and
play with it interactively.  The parameter list and autocomplete
save hours of looking through documentation (or source, if the
docs aren't up to snuff.).
       Roger Upole

<robin.escalation at ACM.org> wrote in message
news:0ud7ks4t2b1s2an6es8q42lc5bd2rp1oeu at 4ax.com...
> "Jeff" <no_spam at umich.edu> wrote:
>
> >Pythonwin is terrible!
>
> On the contrary I find it to be an essential part of development on
> the Windows platform. Build 129 has an incredible folding editor,
> multi-pane debugger, and configuration options. Mark Hammond has done
> us an amazing service. Did you happen to notice that it was free?!
>
> >If there are no answers to these questions, someone needs to work on
> >pythonwin's "development" capabilities.
>
> I believe the answer is that we all need to!
>
> My abilities are not up to this particular task, but I will soon make
> available some code I have been working on in another domain. Code
> that would not yet be ready without PythonWin!
>
> This is what open source is all about, no?
>
> Yes, we all have improvements we'd like to see. Is any software ever
> complete or perfect?
>
> -----
> robin                          robin at illusionsexeculink.com
> media artist /                 remove illusions to reply
> information architect          www.execulink.com/~robin/
>
> The form of e-coli bacteria that causes geomagnetic storms on earth.




From invalid.address at 127.0.0.1  Sat Jun 24 22:20:45 2000
From: invalid.address at 127.0.0.1 (David)
Date: Sun, 25 Jun 2000 02:20:45 GMT
Subject: PyType_Type?
References: <Ebo45.13372$Uw3.865124@bgtnsc04-news.ops.worldnet.att.net> <slrn8l9p1v.elg.thomas@centurion.xs4all.nl>
Message-ID: <39566a69.81770781@news.telus.net>

On 24 Jun 2000 16:39:38 GMT, thomas at xs4all.nl (Thomas Wouters) wrote:

>Probably Fawlty compiler/linker flags

Probably from Barcelona.



From krist at vanbesien.org  Wed Jun  7 08:59:11 2000
From: krist at vanbesien.org (Krist van Besien)
Date: Wed, 7 Jun 2000 12:59:11 GMT
Subject: (newbie) syntax problem compiling python scripts...
Message-ID: <393E471F.4AD05865@vanbesien.org>

I'm trying to compile a script as follows:

% python
>>> from py_compile import *
>>> compile ('htmlformat.py')
and then I get:
>>> compile ('htmlformat.py')
inconsistent dedent
  File "htmlformat.py", line 520
    f = open('includefile')
                           ^
SyntaxError: invalid token     

The piece of offending code looks like this (it's a part of the mailman
package btw):

def MailmanLogo():
    t = Table(border=0, width='100%')
    if mm_cfg.CUSTOM_FOOTER:
        includefile = os.path.join(mm_cfg.TEMPLATE_DIR,
mm_cfg.CUSTOM_FOOTER)
        f = open(includefile)
        text = f.read()
        f.close()
        t.AddRow([text])      

If I copy the codesnippet into a seperate file everything compiles
alricht, but in the context of the complete source python somehow
refuses to accept the "open" token, as if the definition was fracebly
forgotten, or overridden or such (I'm wildly speculating here).

What's going on? How do I debug this? 

TIA 
Krist



--
Ing. Krist van Besien    | At work:            | At home:
Software Engineer        | krist at west.nl       | krist at vanbesien.org
West Consulting B.V.     | http://www.west.nl  | 
Delft, the Netherlands   | +31 (0)15 2191604   | +31 (0)6 50235002
--
Why do programmers get Halloween and Christmas mixed up?
Because OCT(31) == DEC(25)


From pierre.pompanon at wanadoo.fr  Tue Jun  6 07:26:42 2000
From: pierre.pompanon at wanadoo.fr (pierre.pompanon at wanadoo.fr)
Date: Tue, 06 Jun 2000 13:26:42 +0200
Subject: - Sondage -
Message-ID: <E12zHVK-0000OV-00@aweb-pub.actionweb.fr>



Bonjour,

Voulez vous participer ? l?essor d?une Start up ?

Voulez vous devenir actionnaire sans d?bourser un centime ?

Rejoignez-nous sur le site 
http://lesondage.free.fr/html/sond02.php3

remplissez le questionnaire de participation et que l'aventure commence!!!

L'Equipe du sondage           





From moshez at math.huji.ac.il  Sat Jun 10 01:40:31 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 08:40:31 +0300 (IDT)
Subject: readlines() question
In-Reply-To: <8hrrr6$fj4$1@slb2.atl.mindspring.net>
Message-ID: <Pine.GSO.4.10.10006100839410.1511-100000@sundial>

On 9 Jun 2000, Aahz Maruch wrote:

> In article <Pine.GSO.4.10.10006092339580.29560-100000 at sundial>,
> Moshe Zadka  <moshez at math.huji.ac.il> wrote:
> >
> >Ummmmm...I'm known as a professional paranoid in many places, so I usually
> >code like:
> >
> >my_file = open("blah")
> >try:
> >	...my_file.read, my_file.readlines, etc.
> >finally:
> >	my_file.close()
> 
> And if you're *really* paranoid ;-)
> 
> try:
> 	my_file = open("blah")
> 	...my_file.read, my_file.readlines, etc.
> finally:
> 	my_file.close()

Not really. If 
my_file = open("blah")

fails there's

a) no need to close my_file
b) my_file has not been assigned yet

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From jar at mminternet.com  Tue Jun 13 14:47:27 2000
From: jar at mminternet.com (jim roush)
Date: Tue, 13 Jun 2000 11:47:27 -0700
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com> <20000613113018.S20700@xs4all.nl>
Message-ID: <c80dkssvpq3pqcsqjm9g8309p0o2fj124s@4ax.com>

>Your program should work fine, and it does, here. However, if you are using
>IDLE, the problem is a known bug in IDLE ;) You have to type the number and
>press enter in a different window than the one that shows the 'Enter is the
>radius:' text. I dont have the exact text of the bug at hand, sorry, but if
>you fiddle around, you'll find the right window.
>
>It's fixed in newer versions of IDLE, by the way, as well as a number of
>other things. If you use IDLE a lot, you might want to grab the newer
>version: http://www.python.org/idle
>
>> pi = 3.141592654
>> radius = input("Enter is the radius: ")
>> circ = 2 * pi * radius
>> print circ


That was it.  I upgraded to IDLE 0.5 and it works fine.  Thanks.


jar at mminternet.com
James A Roush


From thomas at cintra.no  Fri Jun  9 04:58:24 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 09 Jun 2000 08:58:24 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <3940af8b.236270458@news.online.no>

Yo,

PythonWin has some great stuff, like "partial" code-completion. I too
have som problems with code gone haywire, sometimes Pythonwin gets
stuck, but most of the time it`s a good alternative to IDLE.

I use Emacs under Linux and it works just fine. Just wish the
code-completion-thingy in Pythonwin could somehow be implemented in
Emacs. That`s probably never gonna happen, but an editor with these
capabilities would help the python grow bigger, and Windows-converts
like myself get over the habits gotten from programming in
Delphi/JBuilder. Code-completion, Code-Insight, parameter-lookups ...
It would be nice wouldn`t it??!

Ah well ... 

Thomas




From eduard.hiti at t-online.de  Wed Jun 14 15:33:01 2000
From: eduard.hiti at t-online.de (Eduard Hiti)
Date: Wed, 14 Jun 2000 21:33:01 +0200
Subject: WIN32API GetLastError()
References: <8i3460$ppm@news.or.intel.com> <8i5vjl$bh9$1@nnrp1.deja.com> <8i67nm$l37$12$1@news.t-online.com>
Message-ID: <8i8mqg$k0k$11$1@news.t-online.com>

Ok, I accidentally mailed my previous message directly to Mark Hammond via
email, and he replied by email. For anybody interested, I will break
netiquette and post his message here for clarification:

>The problem is that Python has printed a result between the CreateEvent()
>call, and the GetLastError() call.  The GetLastError() of zero almost
>certainly relates to the file IO used for either reading the next command,
>or printing the result.
>
>If I put the following code in a .py file:
>
>import win32event, win32api
>a = win32event.CreateEvent(None, 0, 0, "Test")
>b = win32event.CreateEvent(None, 0, 0, "Test")
>print win32api.GetLastError()
>
>When I run it, it yields:
>183
>
>Mark.

Tried it, worked!!

Bye,

    Eduard Hiti





From neurocrat at one.net.au  Sat Jun 10 09:24:49 2000
From: neurocrat at one.net.au (Neurocrat)
Date: 10 Jun 2000 23:24:49 +1000
Subject: ? Py 1.6 final
References: <Pine.OSF.4.10.10006091136370.28972-100000@prfdec.natur.cuni.cz> <3942121C.CDC581B2@stud.ntnu.no>
Message-ID: <853dmlk6se.fsf@one.net.au>

Peter Schneider-Kamp <petersc at stud.ntnu.no> writes:

> From Guido's announcement of moving Python development to BeOpen:
> 
> "There's one piece of bad news: Python 1.6 won't be released by June
>  1st.  There's simply too much left to be done.  We promise that we'll
>  get it out of the door as soon as possible.  By the way, Python 1.6
>  will be the last release from CNRI; after that, we'll issue Python
>  releases from BeOpen.com.Michal Bozon wrote:"
> 
> add to that: from all I know he might be on honeymoon right now ;-)

He'll be away until 17th June, then he'll be heading off to the
O'Reilly Open Source Conference shortly afterwards. Don't hold your
breath for 1.6.



From rwallace at esatclear.ie  Thu Jun 15 03:46:12 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Thu, 15 Jun 2000 08:46:12 +0100
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca>
Message-ID: <394889C3.1ED5@esatclear.ie>

Robert Israel wrote:
> Plouffe has another algorithm that can find a digit in an arbitrary
> base.  This takes C*n^3*log(n)^3 time for the n'th digit, so it is not
> faster than other methods that would compute the first n digits, but it
> does take very little memory.

Correct me if I'm wrong, but I thought memory was the limiting factor? 
i.e. that a good algorithm on typical hardware can fill all the
available memory with digits of pi in a few days or so?

> This one has "a more theoretical rather
> than practical interest".

Well, anything beyond the first 10-20 digits of pi is of purely
theoretical interest :)

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From kens at sightreader.com  Thu Jun 15 23:35:00 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:35:00 -0700
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net>
Message-ID: <3949A064.57412B43@sightreader.com>

Hey!  How did this turn into a Python vs. Lisp thread?

Peter Immarco wrote:

> Well I've been getting my feet wet with "Python programming in Win32", a
> great book.  I'm a long-time C/C++/VB/Assembler programmer and I must
> admit, I'm going to enjoy the "straightforwardness" of Python.  Seems to
> take a lot of the pain out of writing applications.
>
> However my C++ friends are badgering the crap out of me and they were
> already ribbing me about VB. :)
>
> Peter.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/fbb4accd/attachment.vcf>

From landauer at apple.com  Thu Jun 22 18:29:58 2000
From: landauer at apple.com (Doug Landauer)
Date: Thu, 22 Jun 2000 15:29:58 -0700
Subject: Python website
References: <3951fe9c_2@nnrp1.news.uk.psi.net>
Message-ID: <landauer-856CDE.15295822062000@news.apple.com>

> Has <http://www.python.org> disappeared, or is it just me?

It is just you:

       *plonk*,

and you have disappeared.

 :-)


From scarblac-spamtrap at pino.selwerd.nl  Mon Jun 26 04:02:58 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 26 Jun 2000 08:02:58 GMT
Subject: Countdown
References: <3952952D.88E3588D@asme.org> <slrn8lc5qg.rqc.scarblac-spamtrap@flits104-37.flits.rug.nl> <Fwq8s2.A8z@world.std.com>
Message-ID: <slrn8le3bq.t65.scarblac-spamtrap@flits104-37.flits.rug.nl>

Will Ware wrote in comp.lang.python:
> Remco Gerlich (scarblac-spamtrap at pino.selwerd.nl) wrote:
> > for i in range(len(thing)):
> >    print thing[-i]
> 
> If thing = [1, 2, 3, 4, 5], then this piece of code will print
> 1, 5, 4, 3, 2. The code below has probably the less surprising
> effect of printing 5, 4, 3, 2, 1.
> 
> for i in range(len(thing)):
>     print thing[-i-1]

Sigh. All my posts have small errors recently. Here I'm trusting the
behavior of indexing in Python too much (it avoids off by one errors,
right?...)

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "A family way":
   Population increases by 5 per cent each week in Universe III (Central
   Texas Computing)... This is possible only if every female is
   constantly pregnant with dodectuplets (a litter of 12).


From gmcm at hypernet.com  Fri Jun 16 14:32:29 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 16 Jun 2000 18:32:29 GMT
Subject: freeze, standalone?
References: <394A1416.ABBE428C@penguinpowered.com>
Message-ID: <8F559F951gmcmhypernetcom@199.171.54.154>

[posted and mailed]

Juerg Schulthess wrote:


>I'm trying to use the 'freeze' and 'standalone' utilty to build
>standalone executables under Win32. 

I know what you're talking about, but you might want to include a bit more 
context when asking for help with stuff randomly snarfed from the 'net.

>The executables produced with
>'standalone' work fine. Now I'd like to burn the executable on CD. This
>adds a 'readonly' attribute to the executable and this seems to change
>the behaviour of the exe completely - it exits immediately and does
>nothing and exits without any output. no errors, nothing! This efect can
>be easily simulated by adding the 'readonly' attribute in the file
>properties. Has anyone succeeded in writing standalone executables
>produced from Python code to CDs that still work from the CD???
>Any help is appreciated!

Freeze.py (and Simple.py) will produce something that runs from a CD. 
Standalone won't. I have a barely tested fix to the C code that does make 
it work, but no time right now.

- Gordon


From olczyk at interaccess.com  Wed Jun  7 10:29:18 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 07 Jun 2000 14:29:18 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au>
Message-ID: <39405bd1.75665484@nntp.interaccess.com>

On Tue, 06 Jun 2000 08:55:46 GMT, "Mark Hammond"
<mhammond at skippinet.com.au> wrote:

>
>Re MFC: Microsoft dont seem too interested in MFC any more.  Ive certainly
>heard nothing about it, and everything new I see has alternatives that
>compete directly with it.
>
>Unfortunately, my personal opinion is that MFC itself will die pretty
>soon, kept alive only by the weight of all the old code written for it.
>So with that in mind, I dont believe that MFC will undergo enough
>significant enhancements that keeping up will be a problem!


MS may not be interested in MFC anymore, because there is not much
more they can change in it. IE. it is a muture library and to add
functionality or to clean up design mistakes would break old code.

I don't see that that necessarily means that it will die. ( I would
like to see it die though. )


From neilh at scintilla.org  Fri Jun 16 22:41:42 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 17 Jun 2000 02:41:42 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <3948D25D.7C83CD6@micron.net> <39499FC1.8C31F7B5@sightreader.com>
Message-ID: <GzB25.685$qz4.5984@news-server.bigpond.net.au>

> LOC (lines of code) is also a measure of programmer output.  The idea is
> that you evaluate your programmers by the sheer quantity of code they
> crank out.  The one who cranks out the most lines of code per week is
> clearly the best programmer (yeah right!).  Anyway, it is sometimes used
> as a rough extimate of productivity.

   And this is where Python really shines. Python's high level data
structures make it perfect for generating reams of C++ code automatically.
Be the hero on your development team :)

   Neil




From phyllis at simplex.com  Tue Jun 20 13:01:36 2000
From: phyllis at simplex.com (Phyllis Metzger)
Date: Tue, 20 Jun 2000 10:01:36 -0700
Subject: Simplex needs Database Developer
Message-ID: <394FA36F.AF1B5C0A@simplex.com>

About the Company: Simplex Solutions, profitable pre-IPO start-up
developing tools for
EDA. Management team and Board of Directors comprised of prominent
leaders from
Cadence, Synopsys, and Avant! Located in Sunnyvale, CA. Please respond
to
Phyllis at simplex.com.

About the Position:
Database Developer
Software developer with a very strong C++ background to develop a
proprietary database for our VLSI analysis tools. Design data structures

involving billions of objects in a way that optimizes rapid information
retrieval, memory management, and file access. Requires expertise in
C++, Unix or NT, data manipulation, data paging, file management;
experience
with persistent data.  Experience with threads, Python, or wxWindows
a plus.





From db3l at fitlinxx.com  Tue Jun 13 16:11:01 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Jun 2000 16:11:01 -0400
Subject: How do you FTP ASCII if ftplib strips CR?
References: <8i5jhd+h2lu@eGroups.com> <uog55mk6r.fsf@ctwd0143.fitlinxx.com>
Message-ID: <uitvdmje2.fsf@ctwd0143.fitlinxx.com>

David Bolen <db3l at fitlinxx.com> writes:

Whoops - I let a cut 'n paste typo slip through in my last post - change:

>     class AsciiFile:
> 	import os
> 
>         def __init__(self,name=None,mode=''):

to

          def __init__(self,name=None,mode='w'):

since it's really intended for an output file class, and without some
default mode the 'b' addition later could cause an error.

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


From dbroadwell at mindspring.com  Sat Jun 24 03:45:35 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sat, 24 Jun 2000 03:45:35 -0400
Subject: need insight...
Message-ID: <3954671F.51E53AE6@mindspring.com>

The code to follow, to what i know, is behaving erratically. it is
erratic in that in the section after the print 'checking... statement
the  'for item in directorylist:' seems to be skipping over entries in
it's list for no apparent reason... (note: what is skipps changes if the
list it is given chages, but from what i can tell, it always skips about
half of the things on the list. i have snipped out as much of the code
as i can and still leave it psudo functional..

I am seeking help in some capacity, and if i have commited some grevious
programming error, enough information to avoid it in the future..

If it dosen't demonstate the same problem i will post the code entire
(it is about 60 lines)

thanks!

# Code follows..

import os
import sys

filetype = 'm3u'
directorylist = []

def doit():
    cwd = []
    print 'reading current directory'
    cwd = os.listdir('.')
    print 'building directorylist'
    for item in cwd:
        if (os.path.isdir(item)):
           directorylist.append(item)
           cwd.remove(item)
    del cwd
    print 'Checking for subdirectories to check.'
    print directorylist
    for item in directorylist:
        print 'subdirectory ' + item + ' found, scanning.'
        os.chdir(item)
        subdir = []
        subdir = os.listdir('.')
        print subdir
        for thing in subdir:
            print thing,
            if (thing[(len(thing)-3):] == filetype):
               print 'Yup.'
               # join item(path) to file
               # add to filelist
            else: 
               print 'Nope.'
               subdir.remove(thing)
        print subdir
        del subdir
        os.chdir('..')
      
doit()

-- 
        ~We're all fallen angels who've forgotten how to fly~
        Remember how: http://www.cyber-action.com/oed/dbroadwell.html
--


From dbroadwell at mindspring.com  Mon Jun 26 19:28:51 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Mon, 26 Jun 2000 19:28:51 -0400
Subject: erk, Bug? ( long post )
Message-ID: <3957E733.4237B55B@mindspring.com>

All of my text is at the bottom...

import os

filelist = []
subdirlist = []
filetype = 'm3u'
output = 'out.txt'

def main():
    stop = 2
    depth = 0
    path = '.'
    scan(path)
    file(path)
    print subdirlist
    if (subdirlist):
       print 'Crossing the bridge of death.'
       while (depth <= stop):
           print 'Depth now at:',
           print depth
           depth = depth + 1
           temp = []
           print 'Scan path for subdir loop',
           print subdirlist
           if (subdirlist):
              temp = subdirlist
              for dir in temp: 
                  scan(dir)
                  file(dir)
              print 'recycling while loop, wish me luck.'
       else: 
#           print 'Stopping that while loop variable readout: '
#           print 'This is where merge will be called once stable'
           print 'Filelist:',
           print filelist
#           print 'Subdirlist:',
#           print subdirlist
#           print 'Last depth:',
#           print depth
           print 'Stop set at:',
           print stop

def scan(scanpath):
    print 'Checking ( dir ):',
    print scanpath
    os.chdir(scanpath)
    dirtemp = os.listdir('.')
    for item in dirtemp:
        if (os.path.isdir(item)):
           if (scanpath == '.'): 
              subdirlist.append(item)
#              print 'Added: ' +item+ ' to subdirlist'
           else: 
              subdirlist.append(scanpath + '/' + item)
#              print 'Added: ' +(scanpath + '/' + item)+ ' to
subdirlist'
    if (not(scanpath == '.')): os.chdir('..')

def file(filepath):
    print 'Checking ( files ):',
    print filepath
    filetemp = os.listdir(filepath)
    for item in filetemp:
        for x in range(len(item)):
            if (item[x:(x+1)] == '.'):
               if (item[(x+1):] == filetype):
                  if (filepath == '.'):
                     filelist.append(item)
                     print 'Added: ' +item+ ' to filelist'
                  else: 
                     filelist.append(filepath + '/' + item)
                     print 'Added: ' +(filepath + '/' + item)+ ' to
filelist'

def merge(filelist):
    if (filelist):
       print 'Sory dave actually merging the functions is disabled'
#       outbuffer=open(output,'w')
#       for item in filelist:
#           inbuffer=open(item)
#           outbuffer.write(inbuffer.read())
#           inbuffer.close()
#       outbuffer.close()
       print 'But if you want to know, I would have:'
       print 'Merged files:',
       print filelist
       print 'With:',
       print './' + output
    else:
       print 'No ' + filetype + ' files to process, Exiting...'

print 'Good morning Dave.'
main()

ERROR:
Good morning Dave.
Checking ( dir ): .
Checking ( files ): .
['DLLs', 'Lib', 'libs', 'include', 'Tools', 'Doc', 'source']
Crossing the bridge of death (while loop that falls off the net).
Depth now at: 0 (how many directories deep am i? .. fails)
Scan path for subdir loop ['DLLs', 'Lib', 'libs', 'include', 'Tools',
'Doc', 'source']
Checking ( dir ): DLLs
Checking ( files ): DLLs
Checking ( dir ): Lib
Checking ( files ): Lib
Checking ( dir ): libs
Checking ( files ): libs
Checking ( dir ): include
Checking ( files ): include
Checking ( dir ): Tools
Checking ( files ): Tools
Checking ( dir ): Doc
Checking ( files ): Doc
Checking ( dir ): source
Checking ( files ): source
Checking ( dir ): Lib/Plat-Win
Checking ( files ): Lib/Plat-Win
Traceback (innermost last):
  File "C:\PROGRA~1\PYTHON\TOOLS\IDLE\ScriptBinding.py", line 131, in
run_module_event
    execfile(filename, mod.__dict__)
  File "C:\WINDOWS\Profiles\mongoose\Desktop\grope.py", line 112, in ?
    main()
  File "C:\WINDOWS\Profiles\mongoose\Desktop\grope.py", line 29, in main
    file(dir)
  File "C:\WINDOWS\Profiles\mongoose\Desktop\grope.py", line 63, in file
    filetemp = os.listdir(filepath)
OSError: [Errno 3] No such process

Is there a reason that that wrror message fills me with dread? Why is
the interpreter punishing me? What did i do wrong?

It always dies on the first use of the files that calls a directoy that
has a subpath. I tested that ability alone in the interpreter and it
worked perfectly well. I am on python 1.5.2 on a windows 98 box. I did
see aome comments on DejaNews about it, and it was also involving
os.listdir.. But there was only one instance, and i do not see precisely
how they relate...

-- 
        ~We're all fallen angels who've forgotten how to fly~
        Remember how: http://www.cyber-action.com/oed/dbroadwell.html
--


From sblakey at freei.com  Wed Jun 28 18:16:11 2000
From: sblakey at freei.com (Sean Blakey)
Date: Wed, 28 Jun 2000 15:16:11 -0700
Subject: (Serious?) package namespace problem (and a proposal)
In-Reply-To: <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com>; from hzhu@knowledgetrack.com on Wed, Jun 28, 2000 at 09:56:27PM +0000
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com>
Message-ID: <20000628151611.A2972@freei.com>

On Wed, Jun 28, 2000 at 09:56:27PM +0000, Huaiyu Zhu wrote:
> 
> On Wed, 28 Jun 2000 14:57:41 -0600, Bjorn Pettersen <bjorn at roguewave.com> wrote:
> >Note that all the problems you describe completely go away if you do:
> >
> >  import AAA
> >  import BBB.AAA
> >
> >IMHO you deserve any/all complications you get when you do "from foo
> >import *", and I strongly believe that we shouldn't do _anything_ to
> >encourage this use...
> >
> 
> Noop.  The issue is not what names are imported from a package, but from
> which package/module files they are imported.
> 
> Perhaps I shouldn't have "abstracted" my example so much as to sidetrack the
> reader's minds.  In fact I was doing
> 
> from AAA import aaa
> from BBB.AAA import aaa
> 
> but the first statement does the wrong thing in the directory python/BBB
> because it imports from the file python/BBB/AAA.py instead of the intended
> file python/AAA.py.
> 
> Huaiyu
> 

OK, I think I see what you mean.  You are objecting to the way that python
searches first for modules in the local directory before searching the normal
PYTHONPATH, correct?

For example, If you have a script string.py and foo.py in your local directory
and foo has an "import string" statement, you want foo to import string from the
standard library instead of from your string.py script.

It seems to me that you are expecting python too read your mind about where you
want the modules imported from.  I believe python's behavior makes sens for most
cases.

If you don't want python to search the current directory for imports, there is a
very simple solution: put this snippet above all of your other imports:
import sys
sys.path.remove('.')

Note that this won't work if you have a script named sys.py in your current
directory.
    -Sean

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
Disraeli was pretty close: actually, there are Lies, Damn lies, Statistics,
Benchmarks, and Delivery dates.



From shapr at uab.edu  Fri Jun  2 11:04:18 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 02 Jun 2000 15:04:18 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl>
Message-ID: <3937D345.CC0935D5@uab.edu>

Martijn Faassen wrote:
> 
> Jacek Generowicz <jmg at ecs.soton.ac.uk> wrote:
> > Martijn Faassen wrote:
> [the Python-Dutch conspiracy exposed!]
> > Might as well change the official language of comp.lang.python to Dutch.
> 
> > We beginnen meteen.
> 
> > Python en Nederlands: wereldtalen voor de eenentwintigste eeuw !
> 
> Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
> nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?
> 
> Groeten,
> 
> Martijn

ja, maar is het niet vanzelfsprekend dat alleen mensen die Nederlands
spreken zich voor Python interesseren?

or-at-least-lives-with-someone-who-speaks-dutch-ly y'rs

(you can add an american with a dutch and scandinavian sounding name,
posting from finland :)
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From pduffin at hursley.ibm.com  Wed Jun 28 07:12:32 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Wed, 28 Jun 2000 12:12:32 +0100
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <3958C648.1D76@rd.francetelecom.fr>
Message-ID: <3959DDA0.5414CF@hursley.ibm.com>

Alexandre Ferrieux wrote:
> 
> Paul Duffin wrote:
> > While ptys are not rocket science they are also not particularly portable,
> > Expect handles all of this for you.
> 
> In unix, AFAIK the only nonportable part is the *naming* of the pty
> devices. That kind of decision is fairly trivial to make in
> sh/awk/perl/python/tcl...
> 

I think that you will find that there is a lot more too it than that, there
are all sorts of different ways of controlling / configuring the ptys,
defaults are different ...

> > Windows is also a real problem as it
> > does not have ptys and needs to be approached in a completely
> > different way.
> 
> Yes but Expect for Windows is a totally different distrib too :-(
> 

But the interface is the same.

> Hence the suggestion to consider writing a specific Python module (if
> there isn't already - I've just arrived in town): for unix it just opens
> the proper devices and substitutes the popen2() pipes with the masters;
> for windows more trickery is needed (namely, behaving as a debugger to
> trap IO calls!), but should be doable in win32all, shouldn't it ?
> 

Apparently there is a port of Expect for Python so why not use that
rather than try and reimplement it all over again. If you wanted to
be really nice you could create a library containing a nice abstract 
representation of ttys which hides all of the nastiness and then create
a simple binding to that library.

Expect has over 7 years of development and centuries of testing on many
different platforms, do you really want to reproduce that. If so why,
what exactly is wrong with Expect.


From g2 at seebelow.org  Sun Jun  4 03:56:20 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 04 Jun 2000 08:56:20 +0100
Subject: ANNOUNCEMENT: PythonWorks 1.0 now available (may 31, 2000)
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com> <0CmZ4.30723$WS3.355848@typhoon.we.rr.com> <39362f52$0$10880@senator-bedfellow.mit.edu> <cosZ4.31355$WS3.360373@typhoon.we.rr.com>
Message-ID: <393A0BA4.41C7@seebelow.org>

Stephen Hansen wrote:
> 
>     Wellll.. Picture a RAD -- C++Builder, Visual Basic, etc.. You have your
> entire form, the menu bar, everything all together in one pretty little
> place where you can Rapidly Develop your environment.
>     You can do some 'visual' editing in Visual C++, but its not all together
> on one form.. you have your menu off in one place, your 'bars' off in
> another, the dialogs off in another.. not 'windows', really...
> 

Some of us think this is an advantage, BTW. ;-)

(i-just-hate-it-when-the-desktop-leaks-through the IDE-<wink>)
   -ly y'rs,

=g2
-- 
_____________________________________________________________________

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


From bvdpoel at uniserve.com  Sat Jun 17 19:31:25 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 17 Jun 2000 16:31:25 -0700
Subject: executing os commands
Message-ID: <394C0A4D.E2B9DBD8@uniserve.com>


Is getstatusoutput() the perfered method to run os commands from within
a python script or is there something more generic? This works fine, but
I'm just thinking that since this is part of a unix only module....

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



From jae at ilk.de  Sat Jun 17 22:26:14 2000
From: jae at ilk.de (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Sun, 18 Jun 2000 04:26:14 +0200
Subject: Python Productivity over C++ (Encapsulation)
In-Reply-To: <394BF19D.23C26098@colconsulting.com> (frankc@colconsulting.com)
References: <20000615183521.59984.qmail@hotmail.com> <394BF19D.23C26098@colconsulting.com>
Message-ID: <18062000.1@sanctum.local.jae.ddns.org>

>>>>> "Frank" == Frank V Castellucci <frankc at colconsulting.com> writes:

    Frank> Hung Jung Lu wrote:


[...]
    >> 
    >> Type strictness is fine. It has all to do with performance: small, fast,
    >> efficient machine code. Encapsulation is a totally different story.
    >> 
    >> One of the biggest lies in modern computer language theory is that
    >> encapsulation is there for a good technical reason.
    >> 
    >> And generations of students were lied to, and sadly, 99% believed in what
    >> the textbooks said. Some of them became programmers, and only after 15 or 20
    >> years in the field would they wake up one day to realize that, heck, they've
    >> been lied to all along. :) (Doesn't all this remind you of Nazis and
    >> Communism? Collective blindness is far more common that you think, ha!)
    >> [snip]

    Frank> Crap. Encapsulation allows me to change the storage hidden
    Frank> by a class from a Stack to a MultiMap without any effect
    Frank> througout the rest of the system. It does not require
    Frank> anyone to recompile, doesn't require any more maintenance.

    Frank> I also use data hidding in Python.

    Frank> What a stupid response you made. You must be guilty of
    Frank> exposing the data members in the interface. This is
    Frank> sickening.

I don't know what's the worst: Hung's original mail (ill-informed
indeed), Frank's response (certainly correct, but also completely
unnecessary) or me letting myself getting dragged into this.

Oh, and why any reply to this is non-existent (and Frank's mail
unnecessary): Look up Godwin's Law ;-) (and thus Hung also lost the
argument...)

Bye, J

PS: Encapsulation in most languages, BTW, doesn't go far enough
(IMPAO).  Like, in C++ you have to show all the private/protected
members in the interface... I'd much rather put all of that in the
implementation, so it doesn't show in a header file.  [EOTangent]

-- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
          My WebHome: http://members.tripod.com/Juergen_Erhard
             Internet Movie Database (http://www.imdb.com)
         I have a firm grip on reality, now I can strangle it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000618/6fed7939/attachment.sig>

From ahujache at cse.msu.edu  Wed Jun 14 15:29:21 2000
From: ahujache at cse.msu.edu (Chetan Ahuja)
Date: 14 Jun 2000 19:29:21 GMT
Subject: tcl/tk vs. python/Tkinter - Example attached
References: <3946B170.17A2630@visionart.com>
Message-ID: <8i8meh$cnc$1@msunews.cl.msu.edu>

Daniel Kramer <daniel at visionart.com>  spoke thusly:
<snip>

> Since I am new to this lang I was just wondering if there was a way to
> get better python/TK performance... if anyone could try out the examples
> and let me know.. for me the tcl/tk one feels much more responsive but I
> don't have any hard numbers to prove that.

> I have been reading that wxWindows is the much faster solution.. but I
> am very familiar with TK and it's plenty fast for what I need.. but I
> worry that the Tkinter ver won't be fast enough once I add alot more to
> it.


> -Daniel



  Just adding my 2 cents here. I think I see  a very slight difference
  on  my linux machine with all the latest versions of
  interpreters. It's a celeron 366 with 128 Megs of RAM.
  The python version seems just a tiny bit slower if I move the boxes
  around very fast. And oh, the difference almost disappears for me on
  commenting out the following line.

     self.canvas.itemconfig(Node.currentNode[0] + "_base",fill = "red")
   
     Could this change in color be causing the  perception of
     jumpiness....?? The boxes always remain grey in the tcl version.


      Chetan
      


From scherbi at bam.com  Wed Jun 28 07:16:00 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 28 Jun 2000 11:16:00 +0000
Subject: Need help: Python and Oracle on Linux x86
References: <B57EB243.44D%Dan@Grassi.org>
Message-ID: <3959DE70.43D296B@bam.com>

Dan -

The usual problem here is the line ORACLE_L in the makefile.  Here's mine:

ORACLE_L=-lnetv2 -lnttcp -lnetwork -lncr -lnetv2 -lnttcp -lnetwork -lclient
-lvsn -lcommon -lgeneric -lmm -lnls
rtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lnetv2 -lnttcp -lnetwork -lncr
-lnetv2 -lnttcp -lnetwork -lclient -lv
sn -lcommon -lgeneric -lepc -lnlsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3
-lclient -lvsn -lcommon -lgeneric -ln
lsrtl3 -lcore4 -lnlsrtl3 -lcore4 -lnlsrtl3 -lnsl -lgeneric -ldl -lelf -lposix
-lsql -lm -lm

(that's all one line)

This works on Linux with the 8.0.5 client.

Good luck,

Bill Scherer

Dan Grassi wrote:

> I am trying to use Python on a Linux system but need Oracle DB access.  I
> have tried compiling the DCOracle interface with no success, there are
> compile errors.
>
> I am trying to replace the Perl that is used with python but the DB access
> issue is a killer.  Can anyone help?
>
> Thanks,
>
> Dan
>
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From warlock at eskimo.com  Tue Jun 13 19:18:18 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Tue, 13 Jun 2000 16:18:18 -0700
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
Message-ID: <slrn8kdfhg.10b.warlock@troll.myth>

On Mon, 12 Jun 2000 21:10:35 -0700, 
 jim roush, in the persona of <jar at mminternet.com>,
 brought forth the following words...:

>I'm just learning Python.  The following code does not work.  It's
>supposed to calculate the circumference after the user enters the
>radius.  The program displays the prompt, but when I enter the radius
>and the press enter, nothing happens.
>
>FYI, I'm using Python 1.5.2 on WinNT.
>
>pi = 3.141592654
>radius = input("Enter is the radius: ")
>circ = 2 * pi * radius
>print circ
>
>
I am just a newbie but the following code works for me (on linux, hence 
the #! first line)

#!/usr/bin/env  python
import string,math
#Import string for the atoi function, since raw_input returns a string
#Import math so pi has many digits :)

radius=string.atoi(raw_input('Radius? '))
circ = 2*math.pi*radius

print circ

#Or to print circ to only 2 places

print 'Circumference is %.2f'%circ
-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From PA0215481 at ntu.edu.sg  Thu Jun  1 06:01:25 2000
From: PA0215481 at ntu.edu.sg (#VEDALA NARASIMHA KUMAR#)
Date: Thu, 1 Jun 2000 18:01:25 +0800 
Subject: jpython-archives
References: <kJVY4.1235$IO4.112187@news.flash.net>
Message-ID: <0CF260C495FED111A6610000F866308D0EDF8E28@mail3.ntu.edu.sg>

and one more thing, 

visit jpython-archives... very helpful. your questions are probably
already answered there.

http://www.python.org/pipermail/jpython-interest/

	-k

> ----------
> From: 	Scott Hathaway[SMTP:slhath at flash.net]
> Posted At: 	Wednesday, May 31, 2000 4:42 AM
> Posted To: 	comp.lang.python
> Conversation: 	jpython
> Subject: 	jpython
> 
> Does anyone know of a JPython tutorial?  I need something  basic that
> tells
> me how to create applets and then to get them into the browser.  I can
> make
> the demo app and compile it with jpythonc, but then I am lost as to
> how to
> view it in the browser.
> 
> Thanks,
> Scott
> 
> 



From scorder at incigna.com  Fri Jun 23 11:53:25 2000
From: scorder at incigna.com (Sam)
Date: Fri, 23 Jun 2000 15:53:25 GMT
Subject: python & excel and 1 remark
References: <39533572.AB90FD1F@tcomhp43.epfl.ch> <086ff63a.3e6f872e@usw-ex0105-035.remarq.com> <39533D22.B6C50A18@tcomhp43.epfl.ch>
Message-ID: <JP17LSOQ0389CL24CJE63626UHCAVIE2QJ@4AX.COM>

On Fri, 23 Jun 2000 12:34:10 +0200, Cedric Lefeuvre
<Cedric.Lefeuvre at tcomhp43.epfl.ch> wrote:

>Richard Chamberlain wrote:
>
>> Hi Cedric,
>>
>> It's possible to generate an excel file using win32com
>> extensions on a windows box. Get the extensions and a bit of
>> documentation at:
>>
>> http://starship.python.net/crew/mhammond/
>>
>> You might want to pick up a copy of Python Programming on Win 32
>> By Mark Hammond, Andy Robinson which has a chapter on this very
>> subject.
>>
>> Richard
>>
>
>Ok,Richard, but without using python on windows system????
If you don't care about formulas and formatting then you can generate
a csv (comma  seperated value) file pretty easily.  On windows csv's
are even associated with excel so all you have to do is double click
on it.  Otherwise you may look into KSpread from the KOffice suite.  I
believe they are using python as their scripting language.  I don't
know if the filters are in place yet though.

-Sam


From embed at NOSPAM.geocities.com  Wed Jun 28 13:23:34 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 13:23:34 -0400
Subject: Pickle for C++?
References: <395A1F7E.4112B2A3@muc.das-werk.de> <vJp65.11228$227.232032@nnrp1.uunet.ca> <395A3049.2E783D21@san.rr.com>
Message-ID: <Svq65.11254$227.232542@nnrp1.uunet.ca>

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:395A3049.2E783D21 at san.rr.com...
>
> > > Is there any solution out there or have I to write my own C++ - pickle
> > > algorithm?
> >
> > Embed Python into your C++ application, and don't bother trying to tie a
> > static language to a dynamic one by making a set of fixed C strucutres.
> > Believe me it's a waste of time.
> >
> > You can embed the Python DLL or shared library into your C++ application
in
> > a matter of minutes, and then unpacking your data will be a single line
call
> > to a Python function that will unpack your structure and a simple set of
> > functions to get and set the data in your python objects.
>
> This is good advice, but I think the question remains outstanding
> for those individuals in situations where, for whatever reason, doing
> what you recommend really isn't appropriate.
>

Suppose I implement Module X that can unpickle anything that Python can
pickle.  I believe it is reasonable to assert that Module X must therefore
be either (a) Python, (b) a fairly Large Subset or Reinvention of Python.

I suggest that in any case where Python is not appropriate, Pickle may also
not be appropriate.

XML-RPC/SOAP might become an intriguing possibility for those situations.

Warren




From donn at u.washington.edu  Mon Jun 19 19:25:07 2000
From: donn at u.washington.edu (Donn Cave)
Date: 19 Jun 2000 23:25:07 GMT
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHAECICEAA.jschmitt@vmlabs.com>
Message-ID: <8ima4j$1jsk$1@nntp6.u.washington.edu>

Quoth "John Schmitt" <jschmitt at vmlabs.com>:
[... quoting myself]

|> But there might a be a good chance for a radically different solution
|> to the basic problem.  OOP is an obvious fit.  The two step, configure
|> and make strategy that's common today is an obvious target for
|> replacement.
...
| Anyway, how is OOP an obvious fit to this problem in general?  Not obvious
| to me.  The general idea of make seems pretty good to me, but as I just
| learned, it's being applied to problems I don't understand very well.

When you write a makefile, you're defining attributes and methods for
your target objects.  There's inheritance, too - the target for
spud.o: spud.c inherits its method from a generic .o.c target.  It's
very limited in make itself, of course.

	Donn Cave, donn at u.washington.edu


From rurban at sbox.tu-graz.ac.at  Tue Jun 27 11:04:55 2000
From: rurban at sbox.tu-graz.ac.at (Reini Urban)
Date: Tue, 27 Jun 2000 15:04:55 GMT
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com>
Message-ID: <3958c185.103629691@news>

scoota at my-deja.com wrote:
>I'm writing an automation interface for a large system embedding Python
>as the script language.  To keep the scripts as easy to write as
>possible (for the end-user) I'm providing a set of extension objects
>written in C++.  What I would ideally like is for these objects to fire
>events which are then handled in the Python script.  So far I can see
>no platform independent way of doing this except using explicit
>callbacks (which is messier than I would like).

And you really believe that COM events are less messier than your
approach? come on. 
Explicit callbacks are always easier to write and understand than
registration with the COM eventsink manager object.
But on the other hand you loose portability and transparency for the
end-user. This is what COM is good for.

the german software ag has a DCOM port for linux. try that if you have
to.

--
Reini Urban
http://xarch.tu-graz.ac.at/autocad/news/faq/autolisp.html


From dsavitsk at e-coli.net  Sat Jun  3 14:39:47 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sat, 03 Jun 2000 18:39:47 GMT
Subject: Why is my Tkintering so bad?
References: <XL%Z4.2161$gr4.67106@news1.rdc1.il.home.com> <39391328.820786453@news1.on.sympatico.ca>
Message-ID: <Thc_4.2797$gr4.100715@news1.rdc1.il.home.com>

> So by adding these two lines, your code should work.
>
> That said, it is not a very pythonic way to do things.  Most python
> programmers will create a class that encapsulates the behaviour. See
> below for an example that demonstrates this.
>

thanks for the suggestions, i really do appreciate the time it takes to
answer these things.
i am digging myself out of a year of vb programming, and it seems to teach a
lot of habits that don't transfer well to other languages.

doug




From mhammond at skippinet.com.au  Sun Jun 18 00:22:08 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 18 Jun 2000 04:22:08 GMT
Subject: block comments
References: <4mP25.63531$WS3.629201@typhoon.we.rr.com> <20000617203650.U26436@xs4all.nl>
Message-ID: <Q7Y25.1543$qz4.13001@news-server.bigpond.net.au>

"Thomas Wouters" <thomas at xs4all.net> wrote in message
news:20000617203650.U26436 at xs4all.nl...


> I usually use normal comments: my editor makes it fairly easy to add '#'
(I
> usually use ' ## ' to make clear that it's intended as a
mass-comment-out)
> at the start of each line in a particular block.

Note that IDLE and Pythonwin also both have commands for quickly
commenting and uncommenting a large block.

> > should python have block comments? (py3k)

Personally, I would like to see them!

> Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!
Nah - but I will leave it here if it helps ;-)

Mark.





From pete at visionart.com  Fri Jun  2 14:57:38 2000
From: pete at visionart.com (Pete Shinners)
Date: Fri, 02 Jun 2000 11:57:38 -0700
Subject: Help, Image Filters with numpy
Message-ID: <393803A2.9AE23E9C@visionart.com>

i have some image data, greyscale, and i'd like to do
some simple filter work on it. i'm trying to think of
a "most optimized" way to get this done. i'm not sure
about all the tools that numerical python gives me, so
i thought i'd ask to see if parts of this are already
taken care of.

say i define a quick soften filter...

soften = array([[1,2,1], [2,4,2], [1,2,1]])

how can i quickly/easily apply this to my image
which is stored in another array? i've been hunting
examples that would do this for me, but no luck

i assume it would be something like this...

filterweight = sum_values_from_array(soften)
subimage = image.extract_subimage((x, y), soften.size())
result = (soften*subimage)/filterweight


ok, i admit i haven't learned much of numpy yet,
but that is in process. but doing something like
this to all the pixels should work, correct?

it'd be great if someone could solidify the code
to the correct numerical python equivalents! but
i'm more interested in hearing how i should best
be doing this.
thanks


From thomas at cintra.no  Mon Jun 26 03:39:53 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 26 Jun 2000 07:39:53 GMT
Subject: OT ? : Icon collection
Message-ID: <3958021a.1951516@news.online.no>

Hi,

I need a set of icons for a project and was wondering if anybody know
of any? It`s mostly web-based with a few desktop clients. Are there
any free, at least for non-commercial projects, icon-collections
available? ( That doesn`t suck )

Thomas


From hniksic at iskon.hr  Wed Jun 21 08:42:36 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 21 Jun 2000 14:42:36 +0200
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <8iq78m$u5m$1@nnrp1.deja.com>
Message-ID: <dpu2en9pdv.fsf@mraz.iskon.hr>

Fred Gansevles <gansevle at cs.utwente.nl> writes:

> If you use enums only as a 'conveniant way to represent values',

Is there any other way to use enums?  It certainly seems the way it's
used in C.

> class Enum:
>     def __init__(self, **kw):
>         self.__dict__.update(kw)
> 
> enum = Enum(Tag = 0, Date = 1, Ammount = 2)
> 
> print "Date-field:", enum.Date

Yummy!  Very elegant.


From shaleh at debian.org  Mon Jun  5 18:36:40 2000
From: shaleh at debian.org (shaleh at debian.org)
Date: Mon, 05 Jun 2000 22:36:40 GMT
Subject: indirect references
Message-ID: <8hha1n$phl$1@nnrp1.deja.com>

How do I do this in python:

foo = 'string'
bar = foo

somehow assign to bar and change the value of foo.



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


From akuchlin at mems-exchange.org  Wed Jun 28 17:33:32 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 28 Jun 2000 17:33:32 -0400
Subject: zope, python vs apache, perl
References: <Pine.SGI.4.05.10006271047340.13130-100000@lemur.magnet.com>
Message-ID: <3d66qtwkwj.fsf@kronos.cnri.reston.va.us>

John Mitchell <johnm at magnet.com> writes:
> cool stuff to steal:
> - easy persistent objects (no more stupid databases!  woo!)
> - call any object through the web
> - rich multilayered, if odd, template library.

- The HTTPRequest and HTTPResponse classes, which are convenient,
though a bit convoluted.  I hope to contribute back a test suite for
these classes some time.

We're following a similar approach with Quixote, keeping ZODB and the
above classes, but using our own, simpler, publisher and template
language.  

--amk


From leonhardt at informatik.tu-darmstadt.de  Mon Jun 26 11:01:31 2000
From: leonhardt at informatik.tu-darmstadt.de (Thomas Leonhardt)
Date: Mon, 26 Jun 2000 17:01:31 +0200
Subject: SuSE Linux, python and setgid
References: <8j72mh$7q0$1@sun27.hrz.tu-darmstadt.de>
Message-ID: <8j7r5b$q8o$1@sun27.hrz.tu-darmstadt.de>

Thanks,
I haven't got it yet, but I think I'll get there.
Somehow I hoped to get aroung compiling something.
But that's because of my NT background ;O)

Thomas



"Thomas Leonhardt" <leonhardt at informatik.tu-darmstadt.de> schrieb im
Newsbeitrag news:8j72mh$7q0$1 at sun27.hrz.tu-darmstadt.de...
> Hi,
> I have got the problem that I should install a script on our server
> which was written in python. The problem is that the script checks
> wether the group id was set as specified by the file permissions.
>
> But somehow the gid is not set when I execute the script. Is there
> something like suidperl for python or smoethingelse special with
> SuSE linux which causes the gid not beeing set??
>
> Thanks for help
> Thomas
>
>




From dsavitsk at e-coli.net  Thu Jun 29 14:52:59 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Thu, 29 Jun 2000 18:52:59 GMT
Subject: NEWBIE QUESTION
References: <395B11C0.4C619958@sicon.net>
Message-ID: <fWM65.27186$dF.1007563@news1.rdc1.il.home.com>

i am in the same boat as you.  i believe you must use calldll/windll (in the
dynwin.zip package) from sam rushing http://www.nightmare.com to work with
(functions in) dll's.
i tried it once without much luck, but sam gave me some pointers on using
it. i haven't retried yet as the hardware that the dll controlls is in a
different state.

"sergio" <smoriano at sicon.net> wrote in message
news:395B11C0.4C619958 at sicon.net...
> Hi all:
>     I`m working with oython a few days, and i want to now if i can work
> directly with a .dll file, or i must run makepy with the .dll,  or i
> must make a .pyd file, if it's like this, Can you tell me how can i make
> a pyd file
>
>




From conversy at lri.fr  Wed Jun 21 07:40:56 2000
From: conversy at lri.fr (Stephane Conversy)
Date: Wed, 21 Jun 2000 13:40:56 +0200
Subject: Tkinter and simulated transparency
Message-ID: <3950A9C7.59BA7E86@lri.fr>

Hi.

I have a top level window with an embeded canvas
To simulate transparency, I want to grab the image in the root window
at the place where the toplevel will be displayed, and put the
image in the canvas.

Is it possible to do that with Tkinter (grab a part of the root window)
?

BTW, what is the purpose of the non-refresh behavior of the toplevel
window ? If I put a canvas into it and then I pack() it, it's background

is displayed (as for the other widgets)..

Thanks.

--
St?phane Conversy
http://www-ihm.lri.fr/~conversy/
mailto:conversy at lri.fr





From fdrake at beopen.com  Mon Jun  5 11:59:52 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Mon, 5 Jun 2000 11:59:52 -0400 (EDT)
Subject: mixed solution: unicase (unique allowed case) (was: Re: Case-sensitivity: )
In-Reply-To: <8h92ko$5e1@news.Informatik.Uni-Oldenburg.DE>
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>
	<392456A9.993FB15C@concentric.net>
	<39245ED7.9C3A4FAB@prescod.net>
	<Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu>
	<vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA>
	<cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com>
	<8g58j4$11ei$2@thoth.cts.com>
	<8g6abe$s5m$17$1@news.t-online.com>
	<cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com>
	<20000522183552.A29224@mudge>
	<8gcoij$60n$1@news-int.gatech.edu>
	<8h92ko$5e1@news.Informatik.Uni-Oldenburg.DE>
Message-ID: <14651.52856.809392.435977@cj42289-a.reston1.va.home.com>

Johannes Nix writes:
 > - an assignment to the same name with different case in the same
 > module is a Case Error exception.

Johannes,
  I presume CaseError would be a subclass of NameError.  Perhaps it
should be called IdentifierCaseError?


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From jcm at bigskytel.com  Mon Jun 19 12:58:05 2000
From: jcm at bigskytel.com (David Porter)
Date: Mon, 19 Jun 2000 10:58:05 -0600
Subject: Question
In-Reply-To: <394E4B46.B7D6D424@darwin.cwru.edu>; from jmwarfe@darwin.cwru.edu on Mon, Jun 19, 2000 at 12:33:10PM -0400
References: <394E4B46.B7D6D424@darwin.cwru.edu>
Message-ID: <20000619105805.A1855@bigskytel.com>

* Mike Warfe <jmwarfe at darwin.cwru.edu>:
> Hello, does anyone know how to set the directory permissions in Win NT,
> using Python?  

os.chmod should work.

The permisions are set in octal; if you don't know the syntax, look up the
man page for chmod (on the net).

  david.

  



From digitig at cix.co.uk  Tue Jun  6 18:20:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Jun 2000 23:20 +0100 (BST)
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <hbX_4.6274$Hz.46116@news-server.bigpond.net.au>
Message-ID: <memo.20000606232048.11341A@digitig.compulink.co.uk>

In article <hbX_4.6274$Hz.46116 at news-server.bigpond.net.au>, 
neilh at scintilla.org (Neil Hodgson) wrote:

>    Its often hard to argue against freedom because we have a natural
> tendency to desire liberty. I see the big benefit in having programming
> language restrictions is that they facilitate reasoning about what code 
> does
> by eliminating the need to consider additional realms of variability.
> 
>    "OK, so sacrifice your freedom but leave mine alone!"
> 
>    By making the restrictions optional, the code is made harder to
> understand because of the additional factor that then goes into
> understanding it. Because of the optional strict mode of Visual Basic, 
> you
> have to remember when reading code whether it is in a strictly typed 
> module
> or not.

Ok. You say you've just switched to the "a_" convention. Come back when 
you've been using it for a year or so, and tell me whether you've actually 
find it useful or not. At the moment we have practical experience pitted 
against a theoretical problem. If the theoretical problem turns out to be 
real, then there's a problem to resolve; if not, why fix what's not 
broken?


From wjdandreta at worldnet.att.net  Fri Jun 16 15:52:13 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 16 Jun 2000 19:52:13 GMT
Subject: List of functions called in Python source files?
References: <3Xc25.3507$Uw3.210307@bgtnsc04-news.ops.worldnet.att.net> <20000616192134.375851@210.9.20.69>
Message-ID: <Nzv25.4945$Uw3.294689@bgtnsc04-news.ops.worldnet.att.net>

Hi Matt,

I solved the problem by taking every function in the .map file and searching
through the souce files to see if it was there. The Python script I wrote to
do it took 15 to 20 mins but it worked. Thanks for your suggestion.

Bill
Matthew Schinckel wrote in message <20000616192134.375851 at 210.9.20.69>...
>"William Dandreta" <wjdandreta at worldnet.att.net> wrote:
>> I need to get a list of the external (ones not in the same source
>> file) functions called in each Python source file. Does anyone know an
>> easy way to do this?
>>
>> Bill
>
>You can try using the Python profiler - but it will give you all of the
>functions, from this file and from any imports.
>
>I have only used to it compare speeds, but it seems flexible enough to
>be able to use it to list functions (and possibly where they came from).
>
>Matt.




From h__l at my-deja.com  Fri Jun 23 13:34:59 2000
From: h__l at my-deja.com (h__l at my-deja.com)
Date: Fri, 23 Jun 2000 17:34:59 GMT
Subject: python optimization
References: <8iu5dt$s6b$1@nnrp1.deja.com> <8iu7t0$5jg$1@slb7.atl.mindspring.net>
Message-ID: <8j0742$aoi$1@nnrp1.deja.com>

In article <8iu7t0$5jg$1 at slb7.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> >I know of -O0  <turns it off?>
> >and also -O2 <default it seems for the RPM I d/l for python 1.5>
>
> Nope, -00 should be the same as -02; they both strip docstrings from
> .pyc files for slightly faster loading (but not execution).  That's it
> for optimizations.  (Unfortunately, I can't find docs for this right
> now; I'm not even sure that -O2 is legal.)
>

Well what I  have done is recompiled the python interpreter it would
seem.  This SHOULD be valid with optimizations turned on in full but
only for the interpreter as it is done with gnu gcc.  Unfortunately
after installing the recompiled interpreter I seem to have lost zope
which I was running on and off.  Now I am trying to fix that by redoing
the interpreter in -O2 as it was before (with no luck I might add).
LOL, how's that for optimization?

hmmm so far I feel I am barking up the wrong tree.

ME


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


From JohnH at PHM.GOV.AU  Wed Jun 28 00:17:41 2000
From: JohnH at PHM.GOV.AU (Hirsch, John)
Date: Wed, 28 Jun 2000 14:17:41 +1000
Subject: Controlling Acrobat Reader with COM
Message-ID: <A90DD8D7B82FD311B4D00008C7DBD6EAB5EB75@phm-exch.phm.gov.au>

"Robin Becker" <robin at jessikat.demon.co.uk> wrote in message
news:Q38VarAdKTW5Ewcx at jessikat.demon.co.uk...
>> ...

>> I tried already with correctly escaped filenames and always got the same
>> error ie 'c:\\Python\\Installs\\test.pdf' got the same error as
>> 'gnarg.pdf' (and I don't have any docs called gnarg.pdf :)).

>Its likely that Acrobat needs a full-blown control container.  You may
>have to run it like the OCX demos in Pythonwin are run - check out
>pywin\demos\ocx.

>Mark.

If you can use wxPython it has an ActiveX control container in which you can
run Acrobat reader.
There is a demo called ActiveXWrapper_Acrobat.py which does exactly that.
It could be worth a look.

Jon




From PClaerhout at CREO.BE  Thu Jun 15 05:40:38 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 15 Jun 2000 11:40:38 +0200
Subject: Mapping a network drive
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03013011E2@msgeuro1.creo.be>

Hello to all,

anyone an idea how I can map a drive using pythonwin, and see
if it was successfull or not? I'm running python 1.5.2 on WinNT 4.

Pieter



From tbryan at python.net  Sun Jun  4 06:09:42 2000
From: tbryan at python.net (Thomas A. Bryan)
Date: Sun, 04 Jun 2000 06:09:42 -0400
Subject: Mathematica and python 1.5
References: <39391D55.2A8F1ED7@mail.danbbs.dk>
Message-ID: <393A2AE6.9A0F9EB3@python.net>

Poul-Erik Hansen wrote:
> 
> I need help in interfacing Mathematica 4 and python 
> 1.5 via pyml. When I try to compile the makefile for 
> pyml I get the following error message 
> "arrayobject.h: No such file or directory". 
> I have tried to play round
> with the makefile but without any succes.

arrayobject.h part of the source code for Python.
Developers of extensions sometimes forget to mention 
whether you'll need the Python source tree to compile 
their extensions.  Since you have Python and pyml is 
complaining, I would guess that you downloaded a 
binary version or Python or that Python came with your 
machine or OS.  If you're a Linux user, there are 
usually separate "package" and a "package-devel" packages 
or languages like Python.  The first lets you use Python, 
the second lets you develop/compile extensions that 
require various header files and such.  

The easiest solution may be to download the Python 
sources from www.python.org.  You could even get the 
new Python 1.6 sources and make sure that pyml still 
works with 1.6. :)

Good luck.

---Tom


From ray at commsecure.com.au  Thu Jun 15 08:21:39 2000
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Thu, 15 Jun 2000 12:21:39 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net>
Message-ID: <39487299.4F30790C@commsecure.com.au>

Jerome Chan wrote:
> 
> Has anyone done any extreme programming in python? Care to share your
> experiences?
> 
> <http://www.extremeprogramming.org/index.html>

My company is currently sponsoring prizes for a course at the
University of Sydney based upon extreme programming all using
Python as the programming language. All the assignments look
to be of high quality which is great ... except if you have
to choose a winner out of them.

We do most of our work using Extreme programming techniques
and all of it in Python.


From donn at oz.net  Sun Jun 18 12:55:41 2000
From: donn at oz.net (Donn Cave)
Date: 18 Jun 2000 16:55:41 GMT
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>
Message-ID: <8iiuud$572$0@216.39.151.169>

Quoth "John Schmitt" <jschmitt at vmlabs.com>:
| I'd love to take the diversion of writing Make using Python.  The things
| that are keeping me from doing this are, in order:
|
| - I'm not completely convinced that doing this is a good thing
| - my employer has not asked me to, and I doubt they would see the value in
| it that I do
| - Make is big and complicated and reproducing all the functionality that we
| need (complete with quirks and bugs) would be time-consuming
| - if I couldn't demonstrate its value immediately the work would not be
| appreciated, especially since we have many other projects worthy of my time
|
| The benefits as I see them (in no order):
|
| - provide a Make that really, really works
| - Make would work the same everywhere!  wouldn't that be great?!
| - not only would Make work the same way everywhere, it would take care of
| foreslash and backslash differences too!
| - Make bugs would be easier to fix
| - new features would be much easier to add
| - if I got it into our tool set, just maybe, Python will be accepted as a
| (preferred?:-) development tool
| - if I did it, it would be my opportunity to learn Python and to learn Make
| fairly well
| - did I mention portability?

You must have a more detailed notion of a ``Make that really, really
works''.  If it's just another implementation, I'd give you slim odds,
against the ``make is big and complicated'' point you allude to above.
But there might a be a good chance for a radically different solution
to the basic problem.  OOP is an obvious fit.  The two step, configure
and make strategy that's common today is an obvious target for replacement.

The sad thing is that the Python distribution itself would be the last
place we could use this system.

| In a smaller way, I would like to have Bash, sed, and grep as Python
| programs rather than compiled C.

I think you're alone on that one.

	Donn Cave, donn at oz.net


From mbel44 at dial.pipex.net  Fri Jun 30 08:46:27 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Fri, 30 Jun 2000 13:46:27 +0100
Subject: Python COM Server Type Library
References: <8jfvvc$iql$1@nnrp1.deja.com>
Message-ID: <qb5pls0uvkrbjg6hrkua0flurop7p1mpsb@4ax.com>

Charles Medcoff <cmedcoff at pyrasol.com> wrote:

>I think I already know the answer to this but I'll ask it anyway...
>
>Is there anyway to generate a type library for a COM server object
>written in Python?  I know this has implications as far as what type of
>binding is used.  That being the case I suspect the answer is no.

As Mark said it cant be done automatically, but you can always create
the type library by hand. 



Toby Dickenson
tdickenson at geminidataloggers.com


From aahz at netcom.com  Fri Jun 23 10:52:15 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jun 2000 14:52:15 GMT
Subject: SAP R/3 and Python ?
References: <3953212C.84E3127F@lemburg.com> <8ivqtr$oo8$1@nntp9.atl.mindspring.net> <8ivsoe$7gu$1@news1.wdf.sap-ag.de>
Message-ID: <8ivtiv$jo8$1@slb7.atl.mindspring.net>

In article <8ivsoe$7gu$1 at news1.wdf.sap-ag.de>,
Joerg Wegener <joerg.wegener at sap.com> wrote:
>"Aahz Maruch" <aahz at netcom.com> wrote in message
>news:8ivqtr$oo8$1 at nntp9.atl.mindspring.net...
>>
>> Disclaimer: I know little about R/3, too.
>>
>> IIRC, R/3 uses a database backend.  At worst, it should be possible for
>> you to dump data directly into R/3 as long as you're careful to follow
>> the virtual schema.
>
>By all means, do NOT attempt this! SAP introduced an additional layer of
>indirection between its applications and the database. The logical structure
>is stored an the ABAP Dictionary, a data dictionary. Applications access the
>database using the definitions found there.

<grin>  Yet another application of Aahz's Rule:

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

Do note, please, that I was careful to say "virtual schema"; I know that
SAP does not have a direct correspondence with the underlying DB.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Why don't they make poppasmarts or didjeridon'ts?  --Aahz


From claird at starbase.neosoft.com  Fri Jun 30 08:16:20 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Jun 2000 07:16:20 -0500
Subject: Following a file, or cloning tail
References: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org> <8F62AF73Egmcmhypernetcom@199.171.54.194> <36D5F4DFCA9BCA46.B5AE15F2F19F393C.28F9B766F10B5E23@lp.airnews.net> <8F62C20E2gmcmhypernetcom@199.171.54.154>
Message-ID: <C311C39CF376029F.651A9619D7C71B72.66E09700B142DAB0@lp.airnews.net>

In article <8F62C20E2gmcmhypernetcom at 199.171.54.154>,
Gordon McMillan <gmcm at hypernet.com> wrote:
			.
			.
			.
>The whole inode thing in that code is because it is a common Linux idiom to 
>rely on the fact that once a process has opened a file, said file can be 
>deleted by another process without the reading process noticing.
>
>This idiom is completely impossible on Windows, so there's no sense in 
>checking whether a new file with the same name has now been created.
>
>So what'll it be - a portable version that's useless on Linux, or one that 
>spends half it's time checking for something that can't occur on Windows 
><wink>?
>
>on-Windows-once-you've-grabbed-the-tail-
> you've-got-the-whole-damn-dog-ly y'rs
>
>- Gordon (who never uses popen on Windows anyway)

It's a good question.

Is tail(1) in POSIX?  I'm out of copies of POSIX documentation
again ...

It's never been a difficult question for me.  I've ended up
coding different tail functions (methods) for different pro-
jects.  Am I committing redundancy in my coding?  Barely,
and at a level with which I'm comfortable.  The general cat-
egory here is, "ambiguously defined stuff where the ideas
are so clear it's easier recoding them than trying to merge
conflicting requirements/definitions."

I most often need such a tail for log-scraping, and part of
my tail implementations generally is *not* to do what you
describe, but to stop reading the mv-ed i-node and pick up
with a newly-created file (a log, in my case).  This illu-
strates the ambiguity in tail's semantics.  I repeat:  I
resolve the ambiguity by my knowledge of requirements spe-
cific to a project.

You and I might have a slight vocabulary clash.  My code
often involves tests of os.name.  I think of the results
as "a portable version", without embarrassment.  Perhaps
your habit is to regard the same as *not* "pure" Python,
and correspondingly "importable".
-- 

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


From yawyi at gmx.de  Thu Jun 22 09:03:01 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Thu, 22 Jun 2000 15:03:01 +0200
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se>
Message-ID: <3953bf64$0$7689@personalnews.de.uu.net>

Niklas Frykholm <r2d2 at mao.acc.umu.se> wrote:
>Can someone explain this difference. (The processor on the Linux system is
>certainly not 40 times faster.)

This was discussed here a few months ago. The reason for the slow
startup is the winsound module (source: PC/winsound.c) which is
unnecessarily a builtin.  winsound causes Python to load the
Windows multimedia DLLs, which can be slow if your sound drivers
make it so.

Two solutions were discussed:
a) make winsound a .pyd
b) use some obscure "delayload" feature of the M$ linker. There
   is a ready to use python15.dll for Python 1.5.2 with this
   feature somewhere out there (I don't remember where, though,
   since I use Linux now)

IMHO solution a) should be standard in Python 1.6 (it isn't in
1.6a2).

Johannes



From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 28 10:22:51 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 28 Jun 2000 07:22:51 -0700
Subject: Speed with Python
References: <Pine.GSO.4.10.10006280840140.27161-100000@sundial> <395A12DF.9262B588@sightreader.com>
Message-ID: <066a40cd.1d8a48ab@usw-ex0105-038.remarq.com>

>> Given the same amount of development time, a python GUI is
likely to appear faster and smoother than the equivalent C++ GUI
because the python programmer has enough time to be clever.

Absolutely.

A while back I developed a gui in vb to an sql database. I
needed to make some changes which were significant in terms of
vb but trivial in python so I implemented a COM server and did
the database calls via odbc. I was actually very suprised with
the overall speed.

So I recently converted the whole application into Python with
Tkinter. It runs slightly faster than my original application,
mainly due to me having the time to be clever about it.

Richard

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From ben at co.and.co  Sat Jun 17 06:21:55 2000
From: ben at co.and.co (ben at co.and.co)
Date: Sat, 17 Jun 2000 10:21:55 GMT
Subject: TABs
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <394AD545.AC791A2B@san.rr.com> <394AF6C8.FFCBF182@houston.rr.com>
Message-ID: <7jI25.24589$CV4.61536@afrodite.telenet-ops.be>

Rod Haper <rhaper at houston.rr.com> wrote:
> Courageous wrote:
>> 
>> Well, speaking of tabs, in vim (.vimrc), I have:
>> 
>> set tabstop=4
>> set expandtab
>> 
>> This sets up vim to "on tab, take the cursor to the
>> next appropriate tabstop, but to insert spaces instead
>> of tabs when doing so". This is always just what I
>> want.
> 
> I used to set the tabstop option to 4 (my preference also) in vim but
> it could really be confusing when I'd edit a file that someone else
> had written using a mixture of tab characters and spaces to format
> their indentation levels.  Then I found a much better solution by
> using the softtabstop option.  Now I use
> 
> set expandtab 
> set softtabstop=4 
> set shiftwidth=4
> 
> This leaves the tabstop option set at it's default 8 characters, keeps
> the indentation spacing consistent (and sane) for mixtures of tabs and
> spaces, and let's me use my preferred indentation spacing of 4 spaces.

Unrelated, but maybe helpfull:
:inoremap :<return> :<return><tab>

Greetings,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From freytag at szs.ira.uka.de  Wed Jun 21 03:37:27 2000
From: freytag at szs.ira.uka.de (Thomas Freytag)
Date: Wed, 21 Jun 2000 09:37:27 +0200
Subject: PyThreadState_Get error with C extension module
Message-ID: <395070B7.5F245778@szs.ira.uka.de>

Hi all,

I still have a problem getting Python 1.6 alpha 2 to run on Windows NT,
in particular my own C extension modules. I have compiled the
python16.dll and the
python.exe by using the supplied Visual C++ project file from the
PCbuild directory
and all succeeds without a complaint. I also managed to compile my
extension module
which accesses a library of C functions implementing a Petri net
simulator engine.
But when I try to import my module, I receive the error message  "Fatal
Python error:
PyThreadState_Get: no current thread" and the interpreter crashes.
What's wrong ? Note that my module has been working for years under
Python 1.5.x and
doesn't use any threading. Its interface complies with the standard
interface given in the
example_nt directory. Under Linux it worked fine without changing a line
of code when
switching from Python 1.5  to 1.6.
Any hints ?

Thomas
______________________________________________________________________
Thomas Freytag                         Phone:         +49 721 608 4301
University of Karlsruhe - SZS          Fax:             +49 721 697377
D-76128 Karlsruhe / Germany            Email:   freytag at szs.ira.uka.de
WWW homepage:                  http://szswww.ira.uka.de/people/freytag
PGP private key:    http://szswww.ira.uka.de/people/freytag/pubkey.asc
______________________________________________________________________




From petersc at stud.ntnu.no  Thu Jun 15 03:17:09 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Thu, 15 Jun 2000 09:17:09 +0200
Subject: Python/COM/VfW
Message-ID: <394882F5.D94C1D20@stud.ntnu.no>

Can I use Python to get a COM object from Video for Windows?

short-question-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From JonesMiw at exchange.uk.ml.com  Tue Jun 13 08:21:30 2000
From: JonesMiw at exchange.uk.ml.com (Jones, Michael W (London))
Date: Tue, 13 Jun 2000 13:21:30 +0100
Subject: Problems installing win32com
Message-ID: <3103135DA599D111B12B00805FE63E7703EBD24B@lonim26407.uk.ml.com>

When attempting to install the Python Interpreter COM server I get the
following error??

I'm trying to install under NT4, Python 1.5.2

Registering COM server...
Traceback (innermost last):
  File "C:\Program Files\Python\Pythonwin\pywin\framework\scriptutils.py",
line 310, in RunScript
    exec codeObject in __main__.__dict__
  File "C:\PROGRA~1\Python\win32com\servers\interp.py", line 53, in ?
    Register()
  File "C:\PROGRA~1\Python\win32com\servers\interp.py", line 49, in Register
    return win32com.server.register.UseCommandLine(Interpreter)
  File "C:\Program Files\Python\win32com\server\register.py", line 403, in
UseCommandLine
    apply(RegisterClasses, classes, flags)
  File "C:\Program Files\Python\win32com\server\register.py", line 360, in
RegisterClasses
    addPyComCat, dispatcherSpec, clsctx, addnPath)
  File "C:\Program Files\Python\win32com\server\register.py", line 229, in
RegisterServer
    regCat = _cat_registrar()
  File "C:\Program Files\Python\win32com\server\register.py", line 85, in
_cat_registrar
    pythoncom.IID_ICatRegister
com_error: (-2147467262, 'No such interface supported', None, None)

Any help would be appreciated.

Thanks,
Mike Jones




From pecora at anvil.nrl.navy.mil  Thu Jun  1 14:59:57 2000
From: pecora at anvil.nrl.navy.mil (Lou Pecora)
Date: Thu, 01 Jun 2000 14:59:57 -0400
Subject: Bad Math Module Notation re:  atan2
Message-ID: <010620001459577031%pecora@anvil.nrl.navy.mil>

Maybe you've all kicked this around before, but I've just picked up
Python in the last month or so.  The notation for atan2(x,y) is 

 atan2(x,y)= arctangent( x/y)

Given the usual association of x with the abscissa and y with the
ordinate in mathematics, science and engineering it is easy to mistake
atan2(x,y) as arctangent (y/x) (x and y reversed!).  Or in another
view, according to the majority of mathematically trained people
atan2(x,y) as defined appears to deliver the arccotangent.  

Am I mising something?


From kens at sightreader.com  Wed Jun 21 09:28:43 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 21 Jun 2000 06:28:43 -0700
Subject: Manipulating specific bits
References: <Brl45.80022$WS3.716680@typhoon.we.rr.com>
Message-ID: <3950C30B.433E8F88@sightreader.com>

>>> byte = 0
>>> byte = byte | (1<<2)|(1<<5) # 2nd and 5th bit starting at the 0th bit
>>> byte
36

"|" = or operator
"&" = and operator
"<<" = left shift
">>" = right shift
"0x13" = hexadecimal representation of 19

Stephen Hansen wrote:

>     Okay, this is a Very Newbie-ish question.
>
>     What I need to do is manipulate specific bits in a byte, for
> transmission. I have absolutely *no* idea how to do this. :)
>     For example, say I have my 1 byte: 0 0 0 0 0 0 0 0. Say I want to set
> the 2nd and 5th bit to 1.. how the heck do I do that? :)
>     I've never actually even /blinked/ in the direction of bit-ish
> manipulation, so am utterly clueless. Next i'm gonna have to stick it into a
> struct I think, before transmission.. I think. dunno. Will look into that
> afterwards.
>
> Thanks. :)
>
> --S

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000621/5bd65a2e/attachment.vcf>

From db3l at fitlinxx.com  Fri Jun 16 17:16:50 2000
From: db3l at fitlinxx.com (David Bolen)
Date: Fri, 16 Jun 2000 17:16:50 -0400
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296DA8@mail.sauder.com>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> Oops!  You're right, as several people also pointed out to me in email.
> Actually I had this in .emacs and didn't realize it for years:
> 
> (setq-default tab-width 4)
> 
> I am not aware of serious adverse effect, though.

Eeek - that's evil.  Don't do that :-)

A TAB character has a perfectly good definition, and artifically
mucking with that definition just within the environment of your
editor is going to pretty much ensure that any other application that
expects the normal behavior will present or parse your files
differently.  Besides, Emacs makes it so easy to maintain indentation
at any level without changing the basic tab size.

>                                                    The good thing is that
> changing indentation is really easy.  You can use one keystroke backspace or
> C-d, for example.

Well, for a single line that's true, although of course if you set up
your language mode to your preferred indentation style, then just
hitting TAB anywhere on the line will re-indent that line according to
your preferences.

But for multiple lines you won't want to be explicitly deleting the
characters on each line, and once you have a region, adjusting the
indentation of the region (C-X TAB) is simple.

But to keep this posting slightly on topic for the group, I suppose
this is as good an indication as any that depending on at least the
presentation of TABs in code - perhaps except that used only locally -
is not the best choice.

> C-d, for example.  Presumably the evil will manifest itself when dealing
> with other programs, but I also set 'less -x4' so I don't even see it.

But as you say, you had to explicitly "reprogram" less for that behavior.

> Wondering what other people will see when reading my programs ...

They'll see a TAB character expand to get to the nearest 8-character
boundary.  At a minimum, if you must use a TAB this way within files
you work on, you should untabify them (or re-tabify them to standard
TAB widths) before any external distribution because assuming the raw
TAB character indents at multiples of 4 is definitely not going to be
the default assumption just about anywhere else, and if nothing else
it'll look different than you wrote it and perhaps not as nice (line
wrapping and all).

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



From jkraska1 at san.rr.com  Wed Jun 28 13:02:25 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 28 Jun 2000 17:02:25 GMT
Subject: Pickle for C++?
References: <395A1F7E.4112B2A3@muc.das-werk.de> <vJp65.11228$227.232032@nnrp1.uunet.ca>
Message-ID: <395A3049.2E783D21@san.rr.com>

> > Is there any solution out there or have I to write my own C++ - pickle
> > algorithm?
> 
> Embed Python into your C++ application, and don't bother trying to tie a
> static language to a dynamic one by making a set of fixed C strucutres.
> Believe me it's a waste of time.
> 
> You can embed the Python DLL or shared library into your C++ application in
> a matter of minutes, and then unpacking your data will be a single line call
> to a Python function that will unpack your structure and a simple set of
> functions to get and set the data in your python objects.

This is good advice, but I think the question remains outstanding
for those individuals in situations where, for whatever reason, doing
what you recommend really isn't appropriate.

Perhaps he should take a look at ILU, methinks.




C/


From del.vento at libero.it  Mon Jun 26 07:34:29 2000
From: del.vento at libero.it (Davide Del Vento)
Date: 26 Jun 2000 13:34:29 +0200
Subject: Shoud I use Java or Python?
Message-ID: <00062612550704.00594@dagomachine>

Hello world!
I am a physicist, with experience in "C" language (not too much C++) and no
previous web-based experiences.
Now I have to develop some applications based on a client-server queries on
databases and remote (but not only) big images manipulation. 
It should work indepentently of client platform and it should be simply ported
from windows to linux server.

This project will be used by scientist that will be interested in use some of
the ESA's SMART-1 images, see http://sci.esa.int/home/smart-1/index.cfm for
details.

I think that using java will be fine (I will call "exec" in order to run other
application that will do the eavy work on server), but what's about python? 
I use linux and I found it installed on my PC, so I visited the web-site (and
the newsgroup), but I don't understand if there are some advantage in writing
phyton application instead in writing Java ones or not (for my case, obviously).

Thank you very much for your help.
Bye,
;Davide
PS forwarding answer in mailbox will be appreciated.


--
Posted from [212.41.208.23] 
via Mailgate.ORG Server - http://www.Mailgate.ORG


From igorr at ifi.uio.no  Mon Jun 26 15:43:52 2000
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 26 Jun 2000 21:43:52 +0200
Subject: reload()'ing modules and how it affects references
Message-ID: <xjvn1k8grd3.fsf@gjallarhorn.ifi.uio.no>

[the article is a little bit long -- sorry]

Hi,


I'm trying to understand how reload() works. "Learning Python" and
Python's documentation warn against the dangers of combining 
from ... import ... and reload(), so that is fine, I use 
"import modulename".

What is rather pecular is that when I store a reference to a class (or
a class instance, for that matter), it does not get updated when I
reload() the entire module again:

# fooclass.Foo.Y is printed, runtime1.py import's fooclass
$ ./runtime1.py 
5 ; 5
			# here, I've changed fooclass.Foo.Y to 6 in emacs
reload( fooclass )
5 ; 5
5 ; 5
			# now re-assign the element explicitly
ml[0] = fooclass.Foo
6 ; 5
6 ; 5
			# and the instance
ml[1] = fooclass.Foo()
6 ; 6
6 ; 6


The Python code looks like this:

import sys
import fooclass

ml = [ fooclass.Foo, fooclass.Foo() ]
while 1:
    print ml[0].Y, ";", ml[1].Y
    next = sys.stdin.readline()
    if not next: break
    else:
        exec next
    print ml[0].Y, ";", ml[1].Y
# end while


(Apparently it does not matter whether the class reference/class
instance reference is stored in a variable or a list)

Now, for the questions:

1) What is the most elegant way of assigning all references to a
certain class to a new definition of this class after a reload? (I can
keep a dictionary/list of all the references, and assign them
explicitly right after the reload() call, but it just looks too ugly).

I suppose that it is impossible to make an instance of a class
suddenly "become" an instance of the new definition of the same class,
so that is fine.

2) When I call a method in class Foo that looks something like this:

def __repr__( self ):
    print <In Foo: Foo.Y ==", Foo.Y, "; self.Y == ", self.Y, ">"


after a reload the changes to Foo.Y affect Foo.Y but _not_ self.Y.
That is, the output from the test similar to the one above looks like
this (the function I call is __repr__):

$ ./runtime1.py 
6 ; 6 ; <Foo instance: self.x == 3, Foo.Y == 6, self.Y == 6>
reload( fooclass )    
	# now Foo.Y == 5 in the module file
6 ; 6 ; <Foo instance: self.x == 3, Foo.Y == 5, self.Y == 6>
6 ; 6 ; <Foo instance: self.x == 3, Foo.Y == 5, self.Y == 6>


Why does it happen -- I thought that Foo.Y and self.Y were supposed to
refer to the same object from within a class method? Am I missing
something essential?


tia and very willing to rtfm


ivr
-- 
Simula-konsulent? Simsulent? "I eat Java for breakfast!".
                                           -- thorkild


From eric at estinc.com  Thu Jun 29 17:20:57 2000
From: eric at estinc.com (Eric Lee Green)
Date: Thu, 29 Jun 2000 21:20:57 GMT
Subject: postgreSQL successes (Interbase?)
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com> <395B78AB.69DDC4F4@email.sps.mot.com>
Message-ID: <395BBDDB.C4C2161B@estinc.com>

Norman Shelley wrote:
> 
> Have you looked at Interbase (newly made OpenSource)? 

Interbase is not yet Open Source. An item posted on their web site seems to
imply it won't be until the fall that it is Open Sourced -- if it actually
happens. I'm not holding my breath.

I believe we looked at the (non-Open Source) version briefly and found that it
would not work on some of the embedded platforms where we required it (such
as, e.g., network appliances based upon non-Intel processors with BSD-based
operating systems). That ruled it out. 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From digitig at cix.co.uk  Thu Jun 15 15:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Jun 2000 20:28 +0100 (BST)
Subject: Where are the math functions?
References: <20000615134357.A1087@better.net>
Message-ID: <memo.20000615202830.52185A@digitig.compulink.co.uk>

In article <20000615134357.A1087 at better.net>, parkw at better.net (William 
Park) wrote:

> In fact, GNU math.h includes many higher functions like
>     j0, j1, j2, y0, y1, y2
>     lngamma, gamma, ...
>     erfc, erf, ...
>     cbrt,
> but Python 'math' module doesn't include them.  So, you have to create a
> new module of your own.  

Is Python based on portable C? If so (I hope it is), then it's right not 
to wrap non-standard libraries as standard. But as Gnu is open source, 
it's a pretty handy place to get them! I needed them for an academic 
project, and it's harder to cite gnu source, but the next ones I want for 
real work I think I'll get that way!


From m.faassen at vet.uu.nl  Wed Jun  7 08:18:40 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 12:18:40 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr>
Message-ID: <8hlej0$g84$1@newshost.accu.uu.nl>

Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
> m.faassen at vet.uu.nl (Martijn Faassen) writes:

>> Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
>> nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?

[knip een boel 'rationele' verklaringen]

> Er zijn dus merdere rationale verklaringen, dus geen behoefte om
> aan kosmische krachten te geloven!

Da's waar, misschien is er alleen een samenzwering, en geen cosmisch
effect. :)

Groeten,

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


From thor at localhost.localdomain  Wed Jun 28 14:33:21 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 28 Jun 2000 18:33:21 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <slrn8lkocq.70v.thor@localhost.localdomain>

On Wed, 28 Jun 2000 11:02:37 -0700, Steve Juranich  wrote:
>I am *brand* new to python (as of Sunday), and I just got to the section
>in the tutorial about classes.  I was wondering why there really isn't
>such an idea as a "private" member of classes? 
>security holes?  Also, what benefit is gained by allowing a user to
>add/delete members from existing classes? 

Well a kind of virtual methods or a Java interface. 

> Wouldn't it be more desirable
>to create his/her own derived class?
>
>Please understand, I am not attempting to start a flame war.  These are
>all questions asked in sincerity.
>
>Thanks in advance. 
>
>---------------------------------------------------------------------------
>Stephen W. Juranich				sjuranic at ee.washington.edu
>Electrical Engineering  	   http://students.washington.edu/sjuranic
>University of Washington		http://truk.ee.washington.edu/ssli
>
>


-- 



From andre at beta.telenordia.se  Mon Jun  5 08:52:18 2000
From: andre at beta.telenordia.se (André Dahlqvist)
Date: Mon, 05 Jun 2000 13:52:18 +0100
Subject: Scrolled canvas confusion
Message-ID: <8hg4a3$a09$1@zingo.tninet.se>

Hi

I have an application (using Tkinter) which currently stacks a bunch of 
buttons on top of each other in a Frame. The number of these buttons
change dynamically, which sometimes causes them to not fit on the 
screen. For this reason I'd like to use a scrollbar. Since the Frame class
doesn't have a yscrollcommand I think I need to pack this frame into a
Canvas, and then somehow use the create_window method with
the frame as an argument. What confuses me though are the various 
coordinates, like the scrollregion of the Canvas and the x and y 
coordinates to the create_window method. Also I don't know what 
sizes to give the Frame and Canvas to make them behave as I want
them to. If possible I'd like to do this with plain Tkinter, no Pmw.

I would greatly appreciate any help. A small example would be
great.

// Andr?


From jerome.quelin at insalien.org  Fri Jun 30 04:14:03 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 10:14:03 +0200
Subject: Class design: accessing "private" members
Message-ID: <962352843.1667731268@news.libertysurf.fr>

Hi all,

When designing a class, I wrote some accessors (get_var() and set_var()
methods) for my "private" members (private in the sense that I don't want
somebody to access them directly from outside the class.
I was wondering if it is good practice to acceed to those private members from
_inside_ my class, or should I use the accessors methods too?
Well, ok, it would be slower (function calls are always more expensive), but
isn't it more OO-well-behaviored?

-- File test.py
class Spam:
    def __init__(self, name="eggs"):
        self.set_name(name)  # Which is the
        self._name = name    # better/cleaner?
    def set_name(self, name):
        self._name = name
-- End of file

Thanks,
Jerome
--
jerome.quelin at insalien.org



From sholden at bellatlantic.net  Tue Jun 27 00:15:40 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Tue, 27 Jun 2000 04:15:40 GMT
Subject: Using MimeWriter
References: <Pine.LNX.4.21.0006261559480.23273-100000@lade.src.no>
Message-ID: <39582A60.37F30B6A@bellatlantic.net>

IMHO mimecntl is a more usable module than MimeWriter, though YMMV.
Worth finding ont he starship, though.

regards
Steve Holden

"Morten W. Petersen" wrote:
> 
> Does anyone have a (somewhat simple) script (or know of one) where
> MimeWriter is being used? Both for sending simple (body only) messages,
> and more complex multipart messages with files attached?
> 
> -------------------------------------------------------------
> How to reply to email:
>   http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
>   http://home.sol.no/~vidaandr/news/FAQquoting.html (english)
> 
> Who is the bigger fool? The fool or the fool who follows him?

-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From matthias.huening at univie.ac.at  Fri Jun 30 03:22:51 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Fri, 30 Jun 2000 09:22:51 +0200
Subject: ODBC
References: <8jf7nm$n48$1@www.univie.ac.at> <395B5E26.77666E49@sec.noaa.gov>
Message-ID: <8jhhv1$4a62$1@www.univie.ac.at>

j vickroy <jvickroy at sec.noaa.gov> schrieb in im Newsbeitrag:
395B5E26.77666E49 at sec.noaa.gov...
> I have not seen this behavior when repeatedly accessing an SQL Server
> database using Python 1.5.2 on a Win NT 4 platform.
>
> What happens if you repeatedly open and close the database **without
> executing any other statements** between successive "opens" ?
>

When I do that everything works fine.
The crashes are caused by odbcjt32.dll. Perhaps it's just mij configuration
(German Win98, Python 1.5.2, Pythonwin 125). Or the specific sql-statements
(first select something from one Access-database, then insert somethin into
another database). Are there any changes to odbc in later releases?
Anyway, I found a workaround which is acceptable for this project, but
which is not a real solution for my problem.

Matthias





From JamesL at Lugoj.Com  Thu Jun  8 22:03:33 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 08 Jun 2000 19:03:33 -0700
Subject: Testing -- not a new message in a while
References: <slrn8k0hv9.pq3.kc5tja@garnet.armored.net> <39405003.CFDA87FC@Lugoj.Com>
Message-ID: <39405075.AD056205@Lugoj.Com>

James Logajan wrote:
> 
> There are hundreds per day; check with your Usenet news provider.
> 
> "Samuel A. Falvo II" wrote:
> >
> > Just testing; I haven't seen any new posts in this newsgroup for quite a
> > while.  Thanks for your patience.
> >
> > --
> > KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
> > Samuel A. Falvo II          |
> > Oceanside, CA               |
Oops; meant to e-mail that!


From ddb at crystal.uwa.edu.au  Mon Jun 26 23:34:09 2000
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Tue, 27 Jun 2000 11:34:09 +0800
Subject: jpython casting to abstract java class
References: <39570C59.2CBCFAEB@crystal.uwa.edu.au>
Message-ID: <395820B1.417B380E@crystal.uwa.edu.au>

> 
> Hi,
> I have a small(ish) code fragment below which fails to
> cast a python-subclass of an abstract-java-base-class back to that
> java-base-class when it is passed as input to another
> java (constuctor) method.

Further testing suggests that the "abstract" word in the above is
irrelevant. It should say: 

any JPython scripted class which inherits from, or subclasses a 
java based class cannot be cast back to that base class when 
passed back to another java based method requiring that 
base class type as an argument.
(hope that makes sense)

I have no idea where to start looking to figure out where it is going
astray. Help appreciated.
Thanks again
Doug


From bryce.sellers at trw.com  Wed Jun 28 12:10:04 2000
From: bryce.sellers at trw.com (Bryce Sellers)
Date: Wed, 28 Jun 2000 09:10:04 -0700
Subject: newbie questions about imports, threads, and optimizations
Message-ID: <8jd7f1$kic$1@murrow.sp.trw.com>

I have been using python off and on (unfortunately, mostly off) for about a
year now, and have got a couple of questions for someone who knows more than
I.

First, is there a way to do something like this:

import string, mystring
m = 'string'
`m`.captalize('hello')
m='mystring'
`m`.capitalize('hello')

In other words, use a string as a module or function name and execute that
portion of code.

Second, are there any tutorials or even reference manuals out there on using
Python with threads?  I haven't so much as even seen any sample code,
although I admit I haven't looked very hard.

Finally, I've noticed that a lot of the discussions surrounding
optimizations center on the fact that even the standard library function
names can be overwritten.  Why can't we just have a command-line argument
(or something along those lines) that tells python that all variables
defined in modules are read-only?  Now the interpreter can optimize the heck
out of it, and throw an exception if anything tries to modify something.  Of
course, this idea is so simple that I'm sure the only reason it hasn't been
done is because its a bad idea.  My question is, why is it a bad idea?

Thanks for the help,
Bryce




From wware at world.std.com  Tue Jun 13 13:03:25 2000
From: wware at world.std.com (Will Ware)
Date: Tue, 13 Jun 2000 17:03:25 GMT
Subject: circular slide rule
Message-ID: <Fw3rDq.5qJ@world.std.com>

"""This generates a Postscript file, which when printed on
an 8.5x11 sheet of paper, gives you a circular slide rule.
Without the transparent slider, it's probably a lot less useful
than a log table, but there may be some nostalgic fun in here
for somebody."""

import sys, math

center = (72 * 4.25, 72 * 5.5)

def p2r(r, theta):
    t = math.pi * theta / 180.
    return (r * math.cos(t) + center[0],
            r * math.sin(t) + center[1])

print '%!PS'
print '/Times-Roman findfont 10 scalefont setfont'

def axis(theta):
    x1, y1 = p2r(0, theta)
    x2, y2 = p2r(72 * 4, theta)
    print '%g %g moveto %g %g lineto stroke' % (x1, y1, x2, y2)

def span(xlim, dx, n):
    global x
    i = 0
    while x + 1.e-8 < xlim:
        nextx = x + dx
        if i == 0:
            label = '%.3g' % x
            gap = 4
        else:
            label = ''
            gap = 2
        L1 = math.log10(x)
        L2 = math.log10(nextx)
        r1 = (72 * 3.5) - L1 * (72 * 1)
        r2 = (72 * 3.5) - L2 * (72 * 1)
        th1 = 720 * L1
        th2 = 720 * L2

        x1, y1 = p2r(r1, th1)
        x2, y2 = p2r(r2, th2)
        x3, y3 = p2r(r1 + gap, th1)
        x4, y4 = p2r(r1 - gap, th1)
        print '%g %g moveto gsave %g rotate' % (x3, y3, th1)
        print '( %s) show grestore' % (label)
        print '%g %g moveto %g %g lineto stroke' % (x1, y1, x2, y2)
        print '%g %g moveto %g %g lineto stroke' % (x3, y3, x4, y4)

        x = nextx
        i = (i + 1) % n

def magnitude():
    x0 = x
    span(1.5 * x0, .005 * x0, 4)
    span(2 * x0, .01 * x0, 5)
    span(4 * x0, .01 * x0, 10)
    span(7 * x0, .02 * x0, 5)
    span(10. * x0, .05 * x0, 5)

print '.7 setgray'
for i in range(36):
    axis(i * 10)

print '0 setgray'
x = 1.
magnitude()
magnitude()

span(x + 0.00001, .001, 1)

print 'showpage'
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From thomas at xs4all.net  Sun Jun 18 07:49:28 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 18 Jun 2000 13:49:28 +0200
Subject: Overriding 'select'
In-Reply-To: <8ihgad$2qj$1@bob.news.rcn.net>; from s2mdalle@titan.vcu.edu on Sat, Jun 17, 2000 at 10:37:34PM -0500
References: <8ihgad$2qj$1@bob.news.rcn.net>
Message-ID: <20000618134928.V26436@xs4all.nl>

On Sat, Jun 17, 2000 at 10:37:34PM -0500, David Allen wrote:

> The problem is this.  I've got a Connection object.  Each instance
> of a Connection object has a field that is a socket.  The server
> that uses these Connection objects wants to call select() on 
> Connection objects, not their corresponding sockets.  This is
> because the Connection object providesa bunch of methods 
> specific to each connection, and holds a lot of data for the
> server.  

What you want to do is adding a 'fileno()' method to the Connection object.
select.select() does not care about what kind of objects it gets passed, as
long as they have a fileno() method that returns the proper filedescriptor
for that object.

So, add this to your Connection object:

    def fileno(self):
        return self.socket.fileno()

and all should be well. The select.select() returns the objects passed in,
not their filedescriptor, so you do not need to do your own bookkeeping.

> Any ideas on speeding this up, or rewriting in a better
> way?  Is there any way to get around the read-only problem
> on sockets?

Yes, create a wrapper object. But then, you already had that ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas.heller at ion-tof.com  Wed Jun 14 16:05:06 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 14 Jun 2000 22:05:06 +0200
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com> <dCL15.4673$cQ.38267@news-server.bigpond.net.au> <8i8m11$bni$1@nnrp1.deja.com>
Message-ID: <01cf01bfd63b$d6ad0120$4500a8c0@thomasnb>

[Mark writes]
> 
> It appears the only real Unixism (apart from hard-coding "ls -l" etc)
> is the use of os.popen().
> 
> It appears very unlikely that os.popen() will be made portable in the
> 1.6 timeframe.
> 
> An option that _could_ make it would be to hack os.py, and attempt to
> use win32pipe.popen for os.popen.  There is a precedent -
> os.path.abspath() attempts to use win32api.GetFullPathName() if
> available.
I have some mixed feelings with these hidden dependencies in the core
distribution of optional extension modules.

What would happen if win32api is not installed? Fail silently?
Print a warning? Require the installation of win32all?

Thomas




From phd at phd.russ.ru  Tue Jun 27 07:14:19 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 27 Jun 2000 11:14:19 +0000 (GMT)
Subject: Using MimeWriter
In-Reply-To: <Pine.LNX.4.21.0006261559480.23273-100000@lade.src.no>
Message-ID: <Pine.LNX.4.21.0006271114070.2223-100000@fep132.fep.ru>

On Mon, 26 Jun 2000, Morten W. Petersen wrote:
> Does anyone have a (somewhat simple) script (or know of one) where
> MimeWriter is being used? Both for sending simple (body only) messages,
> and more complex multipart messages with files attached?

   http://www.faqts.com/knowledge-base/view.phtml/aid/1323/fid/236

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From dworkin at ccs.neu.edu  Mon Jun 12 21:15:57 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 12 Jun 2000 21:15:57 -0400
Subject: "import foo" and "from foo import bar"
In-Reply-To: jajs@NOSPAMretemail.es's message of "Tue, 13 Jun 2000 00:52:07 GMT"
References: <slrn8kaa4u.1r0.jajs@Pascual.es>
Message-ID: <vnd3dmiwfci.fsf@camelot-new.ccs.neu.edu>

jajs at NOSPAMretemail.es (Juanjo ?lvarez) writes:

> It's better to write the "from string import join" form or it's exactly
> the same (in memory use and performance) that "import string"?

There is a small performance benefit to the 'from' approach, but I
usually find that it is outweighed by the fact that it is much easier
to read and understand the code later if you don't do that.

The only time that I would recommend doing that sort of thing (other
than in modules that explicitly suggest that you do so) is if you are
going to reference that name _many_ times, such as inside a huge for-loop.

In that case, it may be worth it to get it into a local variable first.

-Justin

 



From pinard at iro.umontreal.ca  Tue Jun  6 10:55:46 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 10:55:46 -0400
Subject: Sequence-length - Missing the obvious ?
In-Reply-To: aahz@netcom.com's message of "2 Jun 2000 20:42:42 GMT"
References: <3934db4e@nntp.server.uni-frankfurt.de>
	<3938090C.9FB5D057@stud.ntnu.no>
	<8h9682$v4h$1@nntp9.atl.mindspring.net>
Message-ID: <oq1z2aeu4t.fsf@titan.progiciels-bpi.ca>

aahz at netcom.com (Aahz Maruch) writes:

> In article <3938090C.9FB5D057 at stud.ntnu.no>,
> Peter Schneider-Kamp  <peter at schneider-kamp.de> wrote:

> >As far as I understand the for loop goes on until an IndexError is
> >raised. Then the error is cleared and the for loop ends.

> Let's be technically precise here: an exception is raised and caught
> by the "for" handler; the for loop is then broken (precisely as if
> the break statement was executed).  Like any other caught exception,
> the exception is thrown away (not "error is cleared").  --

Are we precise?  If the `for' loop runs to completion, the `else:' clause
of the for-loop gets executed.  If a `break' statement is executed, the
`else:' clause is not.  So, the `for' loop is not broken "precisely".  :-)

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





From bawolk at pacbell.net  Sun Jun 25 17:32:10 2000
From: bawolk at pacbell.net (Bruce Wolk)
Date: Sun, 25 Jun 2000 14:32:10 -0700
Subject: Event Binding and Variable Passing
References: <8j0g88$htm$1@nnrp1.deja.com>
Message-ID: <39567A5A.5B692C5D@pacbell.net>

You need to use a lamda expression.

def fun(x):
	print x
b.bind('<Button-1>', lambda event, b=x: fun(b))

The event argument is ignored here.

Happy TKinter programming!

Bruce


kstar18 at hotmail.com wrote:
> 
> hi,
> 
> I'm relatively new to python and am experimenting with some Tk binding.
> I was wondering how I would pass a variable to a function that is called
> when a Tk event occurs.
> 
> For example,
> b = Button(root, text = 'hi')
> b.pack()
> x = 5
> b.bind('<Button-1>', self.fun(x))
> 
> def fun(self, event, x):
>    print x
> 
> This doesn't seem to work, and the compiler gives me complaints about
> 'self' and the variable being passed.
> 
> Any help would be greatly appreciated.
> 
> Thanks!
> 
> Kapil
> 
> This method doesn't seem to work.
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From cut_me_out at hotmail.com  Mon Jun  5 08:20:46 2000
From: cut_me_out at hotmail.com (Alex)
Date: 05 Jun 2000 08:20:46 -0400
Subject: How to free a port that has been used by a python server?
References: <etdg0qvtkel.fsf@w20-575-109.mit.edu> <8hfb5q$553$1@brokaw.wa.com>
Message-ID: <etdsnusco9t.fsf@w20-575-109.mit.edu>

Cool, I'll try it out.

Thanks, Sam.

Alex.


From hinsen at cnrs-orleans.fr  Wed Jun 21 09:12:12 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 21 Jun 2000 15:12:12 +0200
Subject: 123.4+2=125.40000000000001
References: <3950A30D.DEE4CF4D@brb.midat.de>
Message-ID: <m3em5rch5f.fsf@chinon.cnrs-orleans.fr>

Peter Koppatz <pkop at brb.midat.de> writes:

> >>> 123.4+2
> 125.40000000000001   # is this OK?
> >>> print 123.4+2
> 125.4
> >>>
> 
> On my calculator: 123.4+2 = 125.4 !?

Because your calculator works with a decimal representation of
numbers, whereas Python (and the vast majority of other programming
languages) uses a binary representation. In binary, the result of your
calculation has an infinite number of digits (like 1./3. has in
decimal) and cannot be represented exactly in a computer. That causes
the effect you observe.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From michael.drumheller at boeing.com  Wed Jun 14 15:32:24 2000
From: michael.drumheller at boeing.com (Michael Drumheller)
Date: Wed, 14 Jun 2000 19:32:24 GMT
Subject: Numeric python
References: <Pine.LNX.4.10.10006141123440.26946-100000@intel3.wag.caltech.edu>
Message-ID: <3947DDC7.C59F94A3@boeing.com>

The place to get it is http://numpy.sourceforge.net/.  FYI I downloaded
the rpm file onto my redhad 6.1 and it Just Worked.
Mike

Michael Todd Feldmann wrote:
> 
> Could anyone point me in the right direction to get Numeric python?  I
> have a linux box and tried

-- 
Michael Drumheller--Boeing Mathematics & Computing Technology
PO Box 3707, M/S 7L-40, Seattle 98124-2207, (425)865-3520 Fax:2965
michael.drumheller at boeing.com


From ejr at lotus.CS.Berkeley.EDU  Thu Jun 15 19:54:28 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 15 Jun 2000 23:54:28 GMT
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru>
Message-ID: <8ibqbk$o8j$1@agate.berkeley.edu>

And Oleg Broytmann writes:
 - 
 - GET:  filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript?%s" % urllib.urlencode({"param": value}))
 - POST: filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript", urllib.urlencode({"param": value}))

You only answered half of his request.  The other half is _not_
supported by urllib: adding HTTP headers.

I believe he's looking for a libwww-perl equivalent in Python.  
I am, too.  LWP is one major reason I still write Perl code on 
a daily basis.  The design is very nice.  You have an LWP::UserAgent 
instance to which you can attach a cookie jar, specialize, etc.  
You form HTTP::Request instances, modify the HTTP parameters, 
and pass them to the UserAgent.  You get back an HTTP::Response, 
which has all sorts of handy methods (is_error, is_success, etc.).  
You can also have the response sent to a callback as it arrives.

It's a very user-friendly design.  The same set of packages 
provides both urllib-like simplicity and advanced features.  
Unfortunately, LWP only supports HTTP/1.0, so it can't harness 
the power of pipelining and chunks.  The LWPng that was supposed 
to replace it seems to be on permanent hold.  It'd be a nice 
thing to see in Python...  See http://www.linpro.no/lwp/ for more.

Oh, and LWP code is somewhat more legible than the examples you've
provided, imho, but also can be more verbose.  The easy version:

  use LWP::Simple; use URI::Escape;
  $doc = get "http://my.host.name:8080/loginscript?param=" .
             uri_escape($value);

More complicated:

  use LWP::UserAgent; use URI;

  $ua = new LWP::UserAgent;
  $url = URI->new("http:://my.host.name:8080/loginscript");
  $url->query_form("param" => $value);

  $req = new HTTP::Request 'GET' => $url;
  $req->header('Accept' => 'text/html');
  $res = $ua->request($req);

  if ($res->is_success) {
  # ...

Perl borrowed from Python's OO model, so borrowing from a few 
Perl modules is fair...  ;)

Jason


From godzilla at netmeg.net  Fri Jun  2 11:39:33 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Fri, 02 Jun 2000 15:39:33 GMT
Subject: HTMLParser: getting one extra space around &gt;/&lt;/^M tags
Message-ID: <m21z2grsut.fsf@netmeg.net>

I am helping someone convert html archives from a mailing list back to
mbox format for forwarding to mail-archive.com.

i am using HTMLParser to pull out the body of the message into plain
text, and AbstractFormatter/DumbWriter ...

The message bodies are all inside <PRE> tags. so nofill is set to 1
which means self.formatter.add_literal_data(data) is used to add data
to output. This is what i want for preformatted messages.

There are lots of &gt; and &lt; tags in the body of messages, as
mhonarc/hypermail translate the '>' quote reply and the '<>' of
embedded email addresses using these html tags. (the bodies are
relativly HTML free except for these).

In addition, the end-of-line characters __in the body of the mail
message__ use '^M'.

The problem is this: the formatter is adding one extra space around
these three characters ( &gt &lt; and ^M).

so a back-in-text-mode mail message which looks like this:

=====
 Hello, why are you even reading this email.
 I have nothing important to say. 
 Stop it.
=====

the extra space on the left hand side which i assume comes from
translation of ^M

or this:

====
 and so then you said:

  >  i dont like you

 and i say, so what, and you said:

  >  thats what!


again, extra single space at beginning of line and on either side of
the '>'.

I cannot for the life of me (after 2 hours) find where in htmllib or
formatter these extra spaces are being inserted.

any one know?

many many thanks

les schaffer


From jah at mail.ru  Sun Jun 11 07:44:42 2000
From: jah at mail.ru (Igor E. Poteryaev)
Date: Sun, 11 Jun 2000 17:44:42 +0600
Subject: Will logging system be useful for python programmers ?
Message-ID: <39437BAA.2DC4CE90@mail.ru>

Hi, all

I want to write logging system in python. 

Main purposes:

- ability to send debug, warning, trace and alike messages to 
  various output streams like files, sockets, console, syslog ets.

- flexible format of log messages

- little overhead of logging calls

- ability to customize volume of logging by config files

- internationalization of message strings


I tried to find something already implemented in python, but without success.
Becouse I'm lazy enough, don't want to write it from scratch and want to
reuse some existing design, I searched again and found two implementation in Java: 
JLog:  http://www.alphaworks.ibm.com/tech/loggingtoolkit4j
log4j:  http://www.log4j.org

Both have nearly all I want to see in logging system.
I think that JLog is a little smaller and easier to reimplement, but
log4j is open sourced and clear from possible legal issues.

So, my questions to community are:

Will logging system be useful for python programmers ?
Which toolkit to use as staring point for python reimplementation ?
Which license will be most adequate ? (Puthon license, I suppose)
Maybe I missed something already existing in python ?

cheers,

Igor E. Poteryaev



From fgbrown at ionet.net  Wed Jun 14 21:48:11 2000
From: fgbrown at ionet.net (F. G. Brown)
Date: Wed, 14 Jun 2000 20:48:11 -0500
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <8i9cdf$t1c$1@ionews.ionet.net>

Jin,
I somewhat agree with you about the book, but have found that by using the
web page tutorials that can be found at the python.org site in conjunction
with the book they both work together in explaining things.  I've found the
following 2 to be helpful.  There are some others out there as well.

Look for the following:
Learning to Program
by Alan Gauld

Non-Programmers Tutorial For Python
Josh Cogliati

If all else fails Jin, the boys and girls here are always willing to help.
We all have to start somewhere.

Happy hunting,
A fellow newbie.
Fred
----------------------------------------------------------------------------
----

"jin choung" <jin.choung at gte.net> wrote in message
news:7QG15.2637$sI4.335806 at dfiatx1-snr1.gtei.net...
> howdy all,
>
> i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
> tried to pick up python before moving on to c++.  however, i was trying to
> work my way through the book LEARNING PYTHON and while it says that there
is
> no need to have prior experience in languages like c or c++, i really
found
> that that was not true.
>
> anyone else feel this way?
>
> i'm currently working my way through a 24 hour c++ book and i think this
is
> going to work for me.  oh, well, will be back to contribute here in a
couple
> of months... in the meantime, i'll just leech valuable nuggets of wisdom
> from you guys.
>
> thanks
>
> jin
>
>




From wjdandreta at worldnet.att.net  Mon Jun 19 22:15:32 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 20 Jun 2000 02:15:32 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com>
Message-ID: <8tA35.10459$C44.626417@bgtnsc05-news.ops.worldnet.att.net>

Hi chris,

Here is the fastest version of my program.
----------------------------------------------------------
#corrects Y2K leapyear error and
#strips off all the " marks

from strop import strip, splitfields,atoi,joinfields

lu4 = open('lu4.txt','r')
newlu4 = open('lu4.new','wb')

while 1:
#for i in range(6000):
  line = lu4.readline()
  if not line: break
  fields = splitfields(joinfields(splitfields(line,'"'),''),',')
  fields[2] = joinfields(splitfields(fields[2],' '),'0')
  fields = map(strip,fields)
#convert date to YYYYMMDD format
  if fields[2][-2:] == '00':
    fields[2] = '2000' + fields[2][-10:-8] + fields[2][-7:-5]
  else:
    fields[2] = '19' + fields[2][-2:] + fields[2][-10:-8] + fields[2][-7:-5]

#fix date for Y2K leapyear error
  if fields[2] == '20000301':
    fields[2] = '20000229'
  elif fields[2] == '20000401':
    fields[2] = '20000331'
  elif '20000301'<fields[2] <'20000501':
    fields[2] = '2000' + fields[2][4:6] + '%02d' % (atoi(fields[2][6:])-1)

  newlu4.write('%-16s,%8s,%8s,%10s,%10s,%-11s,%-11s,%-21s,%-51s\r\n' %
tuple(fields))
---------------------------------------------------------------------
Bill

Christian Tismer wrote in message
<394D6579.8B2FF840 at appliedbiometrics.com>...
>Also feel free to send me your source code for inspection.





From jbauer at rubic.com  Fri Jun  2 19:12:27 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 02 Jun 2000 18:12:27 -0500
Subject: planning ALSA PCM audio interface
References: <8h904j$5e1@news.Informatik.Uni-Oldenburg.DE>
Message-ID: <39383F5B.9FF1AA58@rubic.com>

> I think the simplest approach would be to make a 
> hierarchy of objects: soundcard, device, channel, 
> voice. But perhaps it is not smart to reproduce 
> the whole C API...

Johannes,

You didn't mention swig in your post, but I assume
you're familiar with it:

  http://www.swig.org

Swig makes it easy to wrap most of the C API, which
you should probably consider as a first pass -- i.e.
don't assume your interface will be smarter than
the original API -- because at the C level it probably
won't be.

After swigging the API, you can build nice clean Python
classes and play with different interface ideas until 
you're happy with the design.

first-swig-then-collect-the-swag-ly y'rs,

Jeff Bauer
Rubicon Research



From al_amerin at yahoo.com  Wed Jun 14 00:17:40 2000
From: al_amerin at yahoo.com (Al-Amerrho H. Amerin)
Date: Tue, 13 Jun 2000 23:17:40 -0500
Subject: importing module through variable
References: <8i5e9p$nm0$1@zingo.tninet.se>
Message-ID: <39470764.942ECADC@yahoo.com>

"Andr? Dahlqvist" wrote:

> Hi
>
> I have a variable which is assigned the name of a module. Is it possible
> to use this variable in an import statement? Someone told me I could use
> "exec("import %s" %var) and while this works I am not sure how I can
> access the module by it's name after that. Using something like
> var.modfunc() won't do it since 'var' is just a string containing the
> modules name.
>
> // Andr?

If you want access to the module instance that you've imported,  you can do
this:

exec('import %s' % var)
var  =  sys.modules[var]

You can now say:

var.modfunc()

Hope this works for you.


Al-



From hungjunglu at hotmail.com  Fri Jun  9 18:03:15 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Fri, 09 Jun 2000 15:03:15 PDT
Subject: Reloading on the fly: was: Re: Python Productivity over C++
Message-ID: <20000609220315.60036.qmail@hotmail.com>

--- In python-list at egroups.com, boud at r... (Boudewijn Rempt) wrote:
>Could you post an example of designing the main program to reload
>submodules? I'd like to get that working, too, and coming from
>languages where that's impossible, I've a bit of trouble coming up
>with a solution...

This is a very simple reloader, but it gives you an idea. Make two Python 
files names M1.py and M2.py:

M1.py
===============================================================
import sys
import thread
import time

def reloader():
    while 1:
        module_name = raw_input('module to reload: ')
        modules = sys.modules
        if modules.has_key(module_name):
            print 'reloading'
            reload(modules[module_name])
        else:
            print 'importing'
            exec 'import %s' % module_name
        time.sleep(1.0)

if __name__ == '__main__':
    thread.start_new_thread(reloader, ())
    while 1:
        time.sleep(1.0)

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

M2.py
=============================
print 'Hello World!'
=============================


Now, run the first file in a shell (or double click on it in Windows 
system), and then type in M2 at the prompt, then go to M2 and edit it by 
hand (change to something like print 'bye bye World'), and then type in M2 
again at the prompt. See the magic.

The thing is, if you modularize your huge python project into small modules, 
you can really achieve good reloading and that makes your debugging easier. 
This is especially true for large projects where the penalty of setting up 
the initial state is very time consuming (like when you are debugging a 
computer game module... there might be some error that happens only after 
you've played 27 dungeon rooms. Not that I have programmed games in Python, 
it's just an analogy.)

regards,

Hung Jung

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From shouldbe at message.com  Fri Jun 30 09:00:44 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 30 Jun 2000 13:00:44 GMT
Subject: Is this a leaker?
References: <GH075.17918$NP5.455672@newsread2.prod.itd.earthlink.net>
Message-ID: <0S075.17928$NP5.457944@newsread2.prod.itd.earthlink.net>

Let me expand this a little more
 PyObject* deqme(Py_BuildValue("i",0));
 PyObject* comm=PyObject_GetItem(args, deqme);
 Py_XDECREF(deqme);
comm isn't decref until later should I wait to decref deqme after I do
comm??

Also, here is a painful question Who is responsible for
return Py_BuildValue("i",0);
I skimmed thru python's source and it seems they Incref's on return values
so now it seems it is leaked.

Thanks
jamarijr at thomail.com
tho = hot.

"Arint?" <shouldbe at message.com> wrote in message
news:GH075.17918$NP5.455672 at newsread2.prod.itd.earthlink.net...
> comm =PyObject_GetItem(args, Py_BuildValue("i",2) );
> Is the Py_BuildValue leaking, because I don't save a pointer to the return
> value and decref it?  You should Decref comm too right?
>
> Thanks,
> Maybe I should change my name to major payne.
>
>
>




From englim at pc.jaring.my  Thu Jun  8 04:15:39 2000
From: englim at pc.jaring.my (Thomas)
Date: Thu, 08 Jun 2000 16:15:39 +0800
Subject: Pyrite
Message-ID: <393F562A.C0364C73@pc.jaring.my>

Has anyone used Pyrite?  Are there any sample programs?

Thomas.




From nhead at houbits.com  Wed Jun 14 08:23:07 2000
From: nhead at houbits.com (Nigel Head)
Date: 14 Jun 2000 12:23:07 GMT
Subject: TCP packet size?
References: <CC59868A0A76D311B1F50090277B321C1F33BA@VRTMAIL> <3948AC51.98E5EAEC@rpgarchive.com>
Message-ID: <8F5390E8Cabbrvfgtyloup@131.176.85.36>

I don't think this is reliably possible .. in Python or anything else.

My belief is that TCP is a "stream" prototcol - ie it just shovels a 
sequence of octets from here to there, grouped in some arbitrary way. Any 
concept of a "record" gets lost. Just 'cos you write the data in one lump 
gets you no control whatsoever over the pieces it gets received in. The 
only thing you know is that the octets will arrive, eventually, in the 
order you sent them in (or the link will drop).

But-then-if-beliefs-were-truths-we'd-all-be-Timbots'ly yours

Nigel.


chris at rpgarchive.com (chris) wrote in <3948AC51.98E5EAEC at rpgarchive.com>:
...
> A problem arises when I send a large message
>(3000-4000 bytes). Despite the fact that I give a large buffer size to
>socket.read() method, I often receive that message dissembled.  Also,
>sometimes I'll find small messages combined in the same read() call.
>The messages just aren't reassembled the way I thought TCP worked.  I'm
>hoping to find a way to received one message at a time in its entirety,
>or determine the start and end of my messages.
...


From PClaerhout at CREO.BE  Thu Jun 22 06:28:25 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 12:28:25 +0200
Subject: Versionstamp script included with PythonWin
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301210@msgeuro1.creo.be>

Hello all,

I found a script called verstamp.py under python\win32\scripts\VersionStamp,
which looks like a part of the win32 package from Mark Hammond. This script
makes it possible to give an exe or a dll a version stamp. Does anyone knows
if this works for reading the version stamp or did someone already made a
module for that?

Kind regards,


Pieter



From moshez at math.huji.ac.il  Tue Jun 13 14:25:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 21:25:05 +0300 (IDT)
Subject: Two newbie questions
In-Reply-To: <20000613165659.A7828@basajaun>
Message-ID: <Pine.GSO.4.10.10006132123420.21681-100000@sundial>

On Tue, 13 Jun 2000, Arkaitz wrote:

> Hi all,
> 
> I'm learning Python, as it seems everybody around here :-), and I found two 
> problems that I can't find the solution to:
> 1. As the strings are inmutable, if I want to modify the order of the characters
> in the string(e.g. sort them), what I do is:
> 
> st = "asdfasfasdf"
> l = []
> for c in st:
>     l.append(c)
> 
> Then I process the list, and the problem is that I have no idea of how I can get
> the string back, any help?

string.join(['a', 'b', 'c'], '') ==> "abc"

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From johnm at magnet.com  Mon Jun  5 14:53:25 2000
From: johnm at magnet.com (John Mitchell)
Date: Mon, 5 Jun 2000 14:53:25 -0400
Subject: Python's RE
In-Reply-To: <960229654.1963595132@news.libertysurf.fr>
Message-ID: <Pine.SGI.4.05.10006051448040.22654-100000@lemur.magnet.com>

On Mon, 5 Jun 2000, Jerome Quelin wrote:

> Python does not support non-greedy quantifiers (ie, *?, +? and {,}?)... In
> fact, Perl is the only language that does, I think. You are to do it another
> way in Python.
> 
> > print re.search(r'\((.+?)\)',
> >'aaa(foo)bbb(bar)asdfsadf',re.S|re.M).groups()

> Replace the (.+?) part of your regexp with ([^)]+) and it should work
> the way I think you want it to work

I'm pretty sure you're wrong, Python *does* support the non-greedy '+'
operator.  In Python 1.5.1:

>>> import re
>>> x=re.compile('a.+?b')
>>> p=x.match('axxxb awoob')
>>> p.group(0)
'axxxb'


The non-greedy ".+?" matches only up to the *first* "b", not the last.
Here's the greedy version:

>>> re.compile('a.+b').match('axxxb awoob').group(0)
'axxxb awoob'


I'd highly recommend switing to non-greedy operators, it's extremly
useful.  Regular expressions tend to be pretty hairy, and using this and
inline comments helps tremendously.


- j


"The opposite of bravery is not cowardice but conformity."
--Robert Anthony






From thomasm at legend.co.uk  Wed Jun 14 05:46:00 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Wed, 14 Jun 2000 10:46:00 +0100
Subject: cgi behaviouring with key=& url
Message-ID: <394753ef.0@katana.legend.co.uk>

Hi,

Using the module cgi, I send this url of this form :
http://host/cgi-bin/mycgi.cgi?status=enable&pattern=&folder=.

When I do a  cgi.FieldStorage()["pattern"].value this raise an exception as
= is not followed by
anything (I will probably have to add a "" after "pattern=" ).

Can so give me a good raison why the cgi module didn't define any pattern
Minifield with a None value ?

Thomas




From aahz at netcom.com  Mon Jun  5 18:22:47 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 5 Jun 2000 22:22:47 GMT
Subject: OSCON help?
Message-ID: <8hh97n$in3$1@slb0.atl.mindspring.net>

I'd like some help from one or two volunteers who are planning to be at
OSCON.  I'm giving a presentation at 3:30pm Thursday 7/20 about Python
threads (opposite the WorldPilot presentation).  Because I'm hearing
impaired, I expect to have some trouble hearing questions from the
audience.  What I'd like is for the volunteers to write down the
questions for me.  Any takers?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From paulb at infercor.no  Wed Jun 14 07:12:58 2000
From: paulb at infercor.no (Paul Boddie)
Date: Wed, 14 Jun 2000 13:12:58 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net> <960673850.532727042@news.pacbell.net>
Message-ID: <394768BA.1BCAD2B7@infercor.no>

Phylo Kramer wrote:
> 
> On Fri, 09 Jun 2000, Phil Fraering wrote:
> >Guido van Rossum <guido at python.org> writes:
> >
> >> Yet, here are some of the reasons why I am considering making Python
> >> case-insensitive:
> >>
> >> (1) Randy Pausch, a professor at CMU, found, when teaching Python to
> >> non-CS students in the context of Alice (www.alice.org), that the
> >> number one problem his students were having was to remember that case
> >> matters in Python.  (The number two problem was 1/2 == 0; there was no
> >> significalt number three problem.)

When you have a calculator as the first example of interactive Python (it's the
first thing people are encouraged to do in the tutorial), I would think that the
first problem one is likely to encounter is the 1/2 problem. I still find that
counter-intuitive.

> >> (2) I've met many people who are experienced and accomplished Python
> >> programmers but who still, when referring to Python modules or
> >> functions in an informal context (e.g. email), make up their own case
> >> conventions.  E.g. people will write me about the String module or
> >> about FTPLib.

I'm not a linguist or cognitive scientist but I think that unless people
remember such names letter-for-letter, they will obviously try and reconstruct
such names using the expectations they already have. Most people refer to the
File Transfer Protocol as "FTP" not "ftp", for example, because it is an
acronym.

In addition to that, many people use "case-forgiving" systems (those which are
either case-insensitive or case-preserving) which might distinguish between
case, but allow references to objects by disregarding the case "patterns" in an
identifier. Example: DOS/Windows filesystems.

> But when I reached into the library for certain modules I had to constantly
> refer to the docs because I could not remember if it was "selectColor" or
> "SelectColor","BaseHTTPServer" or baseHTTPserver".

Yes, some people use Java-style naming: NamesLikeThis for classes, and
namesLikeThis for methods and attributes. Others use what I might call C++-style
naming: names_like_this for everything. This won't cause nasty silent errors if
you just refer to such names incorrectly, though.

> Because Python has such a low learning curve, it seems that many developers
> see it as a macro language for their particular favorite. Java programmers use
> Java conventions, Perl programmers use Perl, TCL for TCL, C++ for C++, etc. In
> the long arduous path to learning any of these languages, a new programmer
> inherits the traditions of the teacher. But with a language that can be
> learned in a day, there is no time to study convention.

Very profound! But it is indeed hard to see a single convention even in the
standard modules.

> Consistency is all that I ask for.

What are the problems with case sensitivity? Remembering the right "pattern" of
upper and lower case letters in order to, amongst other things, avoid
introducing "rogue" variables into code.

What are the problems with case insensitivity? Being aware that some names which
might look very different actually refer to the same thing, because it is the
"pattern" of case in these names which is the only thing distinguishing them.

I would ask people to consider the wider issues. An interesting example is the
English spelling problem for certain names:

  initialize = 0
  ...
  if some_condition:
    initialise = 1
  ...
  if initialize == 1:
    do_something()

How can we solve this problem? Make "z" synonymous with "s"? ;-)

Paul


From scarblac-spamtrap at pino.selwerd.nl  Sun Jun  4 07:16:47 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 4 Jun 2000 11:16:47 GMT
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <slrn8jkej0.igk.scarblac-spamtrap@flits104-37.flits.rug.nl>

Bablos wrote in comp.lang.python:
> Now that I've got my FTP client doing some basic stuff, I now have a
> progress report which looks something like the following;
> 
> uploaded X bytes
> uploaded Y bytes
> uploaded Z bytes
> 
> Is there a way that I can get Python to print this all on the same
> line, so that it looks as if the line is being updated constantly, as
> opposed to having a ridiculously long list of 'uploaded...' lines?
> 
> Hope someone can help.

Print, sys.stdout.write and so on all just push bytes into stdout. Any
tricks you could use to do this are very platform dependent. However, maybe
printing a lot of ^Hs to clear a line will work; see what

for i in range(10000):
   print i,
   print "\x008"*8,
   
does.

Nice to see people from the Angband world here, btw :-)

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From jcm at bigskytel.com  Fri Jun  9 23:05:21 2000
From: jcm at bigskytel.com (David Porter)
Date: Fri, 9 Jun 2000 21:05:21 -0600
Subject: Programming Poll
In-Reply-To: <8hrim7$fjv$1@nnrp1.deja.com>; from epesh@my-deja.com on Fri, Jun 09, 2000 at 08:05:42PM +0000
References: <8hrim7$fjv$1@nnrp1.deja.com>
Message-ID: <20000609210521.A6218@bigskytel.com>

* epesh at my-deja.com <epesh at my-deja.com>:
> I have a programming skill poll back online, at
> http://cupid.suninternet.com/~joeo/poll.html, with the intent of trying
> to verify some suspicions regarding various programming languages. It's
> very informal; no tracking is done, so it's not scientific at all, but
> the more data I have...
> 
> Please check it out.

it returns 404.

-- 
Dress deftly your hair.  I will bring you every spring handfuls of your
hair. -- e.e.cummings in a blender 



From karl-sve at dsv.su.se  Thu Jun 15 17:45:45 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Thu, 15 Jun 2000 23:45:45 +0200
Subject: Which linux distros have Python by default?
In-Reply-To: <Pine.GSO.4.10.10006142334560.3119-100000@sundial>
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial>
Message-ID: <Pine.GSO.4.20.0006152332160.10224-100000@luna.dsv.su.se>

On Wed, 14 Jun 2000, Moshe Zadka wrote:

> On Wed, 14 Jun 2000, Grant Edwards wrote:
> 
> > I'm going to distribute a utility written in Python/Tkinter as
> > a configuration/administration utility that goes with a Linux
> > device driver.  I was just wondering how many customers are
> > likely to have Python/Tkinter available.  I doubt that anybody
> > is going to install Python just to run this program.
> 
> Pretty soon, everyone will want to have Python to use ESR's CLM2, a
> configuration utility for the the kernel written in Python/Curses. 
> Tkinter is a different matter though

But CLM2 isn't going to be standard for a while yet, not before 2.6, I
think. Not counting development kernels, that is. But they're not
installed by default, right? :)
OK, I'm not sure about this one, I admit. I don't even know how the
flamewars ended on the mailing list, but there was some resistance, and it
seems like a big change to a kernel that's supposed to be close to
release.

Another thing that probably will make Python default faster is that many
GNOME developers seem to like it (PyGNOME is included in Helix GNOME).
I don't know how it is on the KDE side, but with all the good things I've
heard about PyKDE, I wouldn't be surprised if there is some acceptance of
Python there too.

Regards,
  Kalle Svensson



From bmcd at es.co.nz  Fri Jun  2 19:11:08 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 2 Jun 2000 23:11:08 GMT
Subject: Bad Math Module Notation re:  atan2
References: <010620001459577031%pecora@anvil.nrl.navy.mil>
Message-ID: <slrn8jgfta.d4.bmcd@ripley.sulaco.net>

On Thu, 01 Jun 2000 14:59:57 -0400, Lou Pecora <pecora at anvil.nrl.navy.mil> 
wrote:
>Maybe you've all kicked this around before, but I've just picked up
>Python in the last month or so.  The notation for atan2(x,y) is 
> atan2(x,y)= arctangent( x/y)
>Given the usual association of x with the abscissa and y with the
>ordinate in mathematics, science and engineering it is easy to mistake
>atan2(x,y) as arctangent (y/x) (x and y reversed!).  Or in another
>view, according to the majority of mathematically trained people
>atan2(x,y) as defined appears to deliver the arccotangent.  
>Am I mising something?

The math module implementation gives:
1) Consistency with the C standard library (see K&R2 p251).
2) Consistency with the order you write it;
   I write the following

             y
       atan  -
             x

   by writing atan, then y, then x,
   which in Python is atan2(y,x) (for the four quadrant atan).

As other have pointed out, don't be confused by the misleading x<->y
notation change in the math module documentation. The math module does
The Right Thing(tm), particularly for the reason given in (1) above.

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From richard_chamberlain at my-deja.com  Mon Jun 26 10:25:51 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Mon, 26 Jun 2000 14:25:51 GMT
Subject: The automagic of Tkinter
References: <20000626135336.A8180@bork>
Message-ID: <8j7p5g$qib$1@nnrp2.deja.com>

Hi Egbert,

OK I'll let you get away with more than one
question per post, as long as they are not
difficult. ;-)

Your class Knop inherits from Button which in
turn inherits from Widget which inherits from
BaseWidget.

Button has the keyword argument master which
assigns None if no master is passed, which is
what you are doing. Widget passes this straight
to BaseWidget which does this:

if not master:
	if not _default_root:
		_default_root = Tk()
	master = _default_root

So Tkinter is kindly create a Tk() instance in
case you forgot, or were just being stubborn.

Here's mainloop:

def mainloop(n=0):
	_default_root.tk.mainloop(n)

so mainloop is gets called against _default_root
which just happens to be the Tk() instance that
Tkinter created for you.

When you create an a Tk() instance explicitly it
also calls it _default_root except this time you
get a nice convenient reference to it incase you
want to call methods against it.


Because most people are doing from Tkinter import
* you actually get the method mainloop() anyway.
So you needn't type root.mainloop() but just
mainloop() or call against any widget like you
did. It's probably not a good idea to do these
things though, unless no-one is ever going to
have to decipher your code.

and what does Knop mean?

Richard






In article <20000626135336.A8180 at bork>,
  Egbert Bouwman <egbert at bork.demon.nl> wrote:
> Good afternoon,
> The following small script does what I intended
it to do:
> it produces a button, and that button
dissappears when i click on it.
>
>     from Tkinter import *
>     class Knop(Button):
>        def __init__(self,tekst="KNOPJE"):
>            Button.__init__
(self,text=tekst,command=self.quit)
>            self.pack
(side=TOP,fill=BOTH,expand=YES)
>     root=Tk()
>     k=Knop()
>     root.mainloop()
>
> However when I replace the last three lines
with only:
>     Knop().mainloop()
> then the script still works.
>
> This leaves me with a number of questions
(sorry, richard):
> - if 'root=Tk()' is dispensable, then what is
its builtin replacement
>   and what does that replacement do
> - up till now I saw a class object as a
template, that has to be
>   instantiated in order to be of any use.
However here it seems to
>   lead its own life.
> - probably 'mainloop()' plays a crucial role
here, but what are all
>   its functions ?  What is the difference
between
>   'root.mainloop()' and 'Knop().mainloop()'
> I promised Richard to pose only one question
per post, but these
> questions really are one big question about the
Tkinter mechanism.
> egbert
>
> --
> Egbert Bouwman - Keizersgracht 197 II - 1016
DS  Amsterdam - 020 6257991
>
==================================================
======================
>
>



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


From senn at maya.com  Mon Jun 12 12:30:19 2000
From: senn at maya.com (Jeff Senn)
Date: 12 Jun 2000 12:30:19 -0400
Subject: Stackless Python for the Mac?
In-Reply-To: Just van Rossum's message of "Fri, 9 Jun 2000 22:04:30 +0100"
References: <l03102801b5670bc57ecc@[193.78.237.159]>
Message-ID: <8zwarhes.fsf@SNIPE.maya.com>

Just van Rossum <just at letterror.com> writes:

> At 3:10 PM -0400 09-06-2000, Jeff Senn wrote:
> >Has anyone contemplated (or started...or finished) a port of stackless
> >to the Mac?  Other than syncronizing the sources, there shouldn't be
> >any complications, right?
> 
> Almost none at all... It's only for stock 1.5.2 at the moment:
> 
>    http://www.petr.nl/just/PythonCoreStackless.sit.hqx
> 
> Instructions:
> - create a new folder inside the Python 1.5.2 folder.
> - drag the old PythonCore to it
> - place the new (stackless) PythonCore to the Python 1.5.2 folder
> - run ConfigurePython
> - done.
> 
> Good luck!
> 
> Just

Thanks, ...but... I tried this and get the "can't find PythonCore"
error when trying to run any Python stuff thereafter (until I put back
the original PythonCore).

> Almost none at all... It's only for stock 1.5.2 at the moment:
                                      ^^^^^^^^^^^
By this I assume you mean the distribution at:

ftp://ftp.cwi.nl/pub/jack/python/mac/*152.*

should work, no?

Thanks for any help,
-- 
-Jas




From steel12659 at mail.com  Fri Jun 16 11:47:32 2000
From: steel12659 at mail.com (Steel Building Promo)
Date: Fri, 16 Jun 2000 11:47:32
Subject: Steel Building Clearance
Message-ID: <20000616154759.F28191D03F@dinsdale.python.org>

Looking for an industrial strength steel building?

Look no further!

We have 8 new clearance buildings, one of which may fit your needs!


Simply send your name and a written request for 
more information to: 

mailto:steel987654 at mailcity.com?subject=BUILDING_DETAILS
(for removal instructions, please see below) 

And you will receive information via email within 24 to 48 hours!
You will also receive exclusive information about our canceled and clearance buildings, so
be sure to check your email box! 

IMPORTANT: For verification purposes, you MUST 
provide your name and a request for 
more information or no further information will be sent
to you. Your written request could be something as 
simple as writing "Please send more info." 


mailto:steel987654 at mailcity.com?subject=BUILDING_DETAILS
(for removal instructions, please see below) 
_______________________________________________________
To be removed simply hit the REPLY button and put the word 
REMOVE in the subject or click on this link:
mailto:steel12659 at mail.com?subject=Remove



From bwilk_97 at yahoo.com  Fri Jun  9 09:03:13 2000
From: bwilk_97 at yahoo.com (Bill Wilkinson)
Date: Fri, 09 Jun 2000 13:03:13 GMT
Subject: Pythonwin is terribly good.
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <lW505.42238$L3.321655@typhoon.midsouth.rr.com>

hmmm,  You said terrible, but I know that you meant terribly.
Here is what I mean.

Pythonwin is terribly cool!
Pythonwin is makes win32 programming terribly easy.
Pythonwin makes working with Excel/Access terribly enjoyable.
I am terribly productive with Pythonwin.
The Pythonwin debugger is terribly terrific.
I appreciate Marks work on Pythonwin terribly.
Each new release of Pythonwin is terribly better than the last.

additionally, without the Python com extensions, I would be terribly
screwed.

I have answers to both your problems... but I'm not tellin.... Ain't I
terrible. :)



"Jeff" <no_spam at umich.edu> wrote in message
news:8hp2ku$a25 at news.or.intel.com...
| Okay, 2 questions.  Seem like basic enough functionality to be in
pythonwin,
| but I dont see it.
| 1. How do you kill an infinite loop program without killing the pythonwin
| environment?
| 2. How can you 'unload' a module so that you can overwrite the .pyd
without
| leaving the pythonwin environment?
|
| If there are no answers to these questions, someone needs to work on
| pythonwin's "development" capabilities.
|
|
|
|




From embed at NOSPAM.geocities.com  Tue Jun 27 12:19:07 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 27 Jun 2000 12:19:07 -0400
Subject: more python evangelising...
References: <200006271406.AAA06218@mbuna.arbhome.com.au> <D9465.71358$7o1.1704294@news2.rdc1.on.home.com>
Message-ID: <qt465.9388$227.219932@nnrp1.uunet.ca>

> On the other hand, maybe it will be possible to add 'optional' features to
> Python 3000 that will make it suitable for that kind of development
without
> interfering with its simplicity and (development) speed.

Sometimes it's not so simple as "large system" versus "small system".  Let
me elaborate.

For example, some large systems in C that I have worked on were essentially
a large proprietary scripting language. Replacing that scripting language
with Python basically obsoleted the core of the large C system, and
basically left some C extensions, and Python.  That is a perfect example of
a good place to use a standard language instead of a complex C system that
wanted to be a language, but never quite arrived.  This is what in fact
happened to us with respect to two large commercial packages I used to have
to use: Fix DMACs and Wonderware InTouch, two industrial software packages
which are designed for SCADA and MMI applications in industrial control
systems.   What one really needs is a set of process control and monitoring
objects, and some nice drag-and-drop GUI tools, wrapped around a language
that is so easy to use, that "even an engineer can learn to use it". <smirk>

Warren





From jhe at webde-ag.de  Fri Jun 30 05:48:03 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 30 Jun 2000 11:48:03 +0200
Subject: MimeWriter (how to use/find boundaries)
References: <Pine.LNX.4.21.0006301138390.15363-100000@lade.src.no>
Message-ID: <8jhq9o$i7i$1@pollux.ip-plus.net>

"Morten W. Petersen" <morten at src.no> schrieb im Newsbeitrag
news:Pine.LNX.4.21.0006301138390.15363-100000 at lade.src.no...
> I've built a mail-sending function, which, when sending mail with
> attachments; does send the mail, with the encoded attachments; but it

We talked about that topic just recently, please check the mail archives and
the FAQTS to see if you find an answer there. Especially, see my mail on
adding a MIME-Version header.

Bye, J?rgen




From gherman at darwin.in-berlin.de  Fri Jun 30 08:16:39 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Fri, 30 Jun 2000 14:16:39 +0200
Subject: ANNOUNCE: Distutils 0.9
References: <20000630002620.A2329@beelzebub>
Message-ID: <395C8FA7.3F99809F@darwin.in-berlin.de>

Greg Ward wrote:
> 
>                      Python Distribution Utilities
>                               release 0.9
>                              June 29, 2000
> 
> [...]
> 
>     http://www.python.org/sigs/distutils-sig/
>     http://www.python.org/sigs/distutils-sig/download.html

Sorry, but all I see there are 0.8.x versions...

Dinu

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From gmcm at hypernet.com  Thu Jun  1 21:52:50 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 02 Jun 2000 01:52:50 GMT
Subject: Two questions on embedding :)
References: <XxCZ4.31420$WS3.364825@typhoon.we.rr.com>
Message-ID: <39371372$0$20961@wodc7nh1.news.uu.net>

Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:
[importing modules from C]
>I tried about a dozen different combinations of commands and so 
on and so
>forth until I just gave up and did PyRun_SimpleString("import 
sys");.
>
>    Does anyone have an example program anywhere that 
incorporates importing
>modules in a C++ level? Hrm. Actually, is there a need to bother? 
Is it any
>better then just doing a PyRun_SimpleString?

Not really. You're just mucking with the environment, not trying to 
return a result....

>    Later, I basicaly wanted to call: 
"languages.Translate('name','string')"
>and have the result returned to the C program for processing. 

Now you're better off diving in and getting your paws on the 
languages module, then the Translate function and calling it 
directly.

Check out
http://www.mcmillan-inc.com/embed.html 
and it's follow on page for examples (of both, actually).

- Gordon



From wtanksle at dolphin.openprojects.net  Sun Jun  4 18:29:08 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sun, 04 Jun 2000 22:29:08 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <memo.20000604004935.14417B@digitig.compulink.co.uk>
Message-ID: <slrn8jlm1j.bkc.wtanksle@dolphin.openprojects.net>

On Sun, 4 Jun 2000 00:49 +0100 (BST), Tim Rowe wrote:
>wtanksle at dolphin.openprojects.net (William Tanksley) wrote:

>> This is a classical example of false dichotomy ;-).  I liked Python's
>> indentation not because it enforced consistency, but rather because it
>> makes code look good.  

>But, of course, the programmer could make C look just as good by applying 
>consistent layout.

Good for C!  In fact, I do just that the majority of time at work.

>The nice thing about Python is that the programmer /will/ apply (locally)
>consistent layout.

That *is* one nice thing about Python.  However, it's only trivially nice;
were it to go away Python would still be a good language (imagine, for
example, that Guido invented a parser which could *tell* when a programmer
was using inconsistent spaces per tab and parse correctly in spite of
that).  It's *much* nicer that Python draws the program tree in two
dimensions than that it forces the programmer to be consistent about it.

>> In fact, it could be argued that Python does _not_ enforce consistency;
>> the programmer is free to choose any indentation for each block, so long
>> as each block can be parsed (which does require internal consistency).

>And, of course, one is free to choose any legal identifier, so the case 
>sensitivity doesn't enforce consistency either.

*Exactly*.  You can also choose program structures which make little sense
for your problem area (using 'try' where it doesn't belong), thus ruining
the pretty indentation, or choose two different abbreviations for the same
word as variable names (thus ruining your case-sensitive consistency).

>> So I don't really care about languages enforcing consistency; if I need
>> it, I'll enforce it myself.

>Great. Ever maintain anybody else's code?

Are you trying to insult me?  Or are you trying to make some kind of
point?  If the latter, you have no idea what my experience level is, so to
communicate your argument you'll have to explain yourself more completely.

*My* point is that Python's job is to make it possible to write really
good-looking code.  It's my job, as the programmer or employer, to prevent
bad-looking code.

>> Take a lesson from the people complaining about Python's indentation: one
>> of the leading false accusations is that it's inflexible, that it won't
>> allow people to choose thier own indentation.

>Lesson taken. I won't give any credence to those who complain about case 
>sensitivity on the same grounds.

But this mythical person complaining about case-sensitivity is *correct*;
Python truly *is* case-sensitive.

-- 
-William "Billy" Tanksley


From aahz at netcom.com  Fri Jun  9 18:35:37 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jun 2000 22:35:37 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch>
Message-ID: <8hrrfp$880$1@slb6.atl.mindspring.net>

In article <8hqbvr$fgr$1 at sunnews.cern.ch>,
Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote:
>
>First off, any claim of an order of magnitude increase in productivity
>is pure nonsense. See Brooks for an explanation or Boehms and McConnell
>for hard data.

That's half-true.  Lines of code per unit time is roughly constant per
programmer (though I think Python breaks even that rule).  Of more
interest is the extent to which one language has greater expressiveness
than another for a given domain.  So if Python programs are generally
one-tenth the size of C/C++ programs, there *is* an order of magnitude
increase in productivity.

Where I think Python really kicks butt is in maintenance, though.
Reading Python code is significantly easier than any other language I've
seen; rewriting Python code is also significantly easier.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From x_beelze at hotmail.com  Mon Jun 19 09:25:57 2000
From: x_beelze at hotmail.com (Chris Weichenberger)
Date: 19 Jun 2000 13:25:57 GMT
Subject: Tkinter integration of external Tk widget
Message-ID: <slrn8ks7j5.ohf.x_beelze@grizzly.cosy.sbg.ac.at>

 Hello everybody, 

I have the very special problem that some time ago I have written a Tk widget 
that displays molecules interactively which should now be ported to Python
using the Tkinter API. The widget is pure C code and this task seems to be 
everything else but straight forward. I had a look at the _tkinter.c and the
Tkinter.py sources and finally investigated the port of Tix to Python. 
Clearly, a wrapper based on the 'Widget' class has to be written but my main
concern is how to (dynamically) link the Tk widget .so library to the existing
Tkinter library? As far as I understood the Tix approach, one has to recompile
the package. Is there any way to circumvent this? Ideally, I am searching for
something likes Tk's 'load' function. I already unsuccessfully checked FAQs.
Is there anybody who can help me with that problem?

 Thanks in advance, 

    _Chris




From alan.gauld at gssec.bt.co.uk  Mon Jun 19 07:43:13 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 19 Jun 2000 12:43:13 +0100
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com> <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <394E0751.A6D6ACB8@gssec.bt.co.uk>

William Dandreta wrote:
> It is Python 1.2 because the later versions will not run 
> on my machine.

Can you say why? I had no problem getting python 1.5.2 running 
on my old 50MHz 486 laptop under DOS.

Just curious.

Alan g.

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


From jerome.quelin at insalien.org  Fri Jun 30 06:15:49 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 12:15:49 +0200
Subject: Are tk derived objects ever destroyed?
Message-ID: <962360149.1278823422@news.libertysurf.fr>

Hi all,

Look at the snippet code below:

-- File toplevel_creator.py
#!/usr/bin/python
from Tkinter import *
class Frm(Toplevel):
    def __init__(self):
        Toplevel.__init__(self)
        Label(self, text='spam').pack()
    def __del__(self):
        print 'Kai kai kai'
Button(text='create', command=Frm).pack()
mainloop()
-- End of file

When running this code and creating some toplevel windows with the
create button, the objects aren't destroyed (the __del__ function
doesn't get called) when I destroy the window.
I wonder why, since there aren't any other references to this toplevel
window. Unless Tkinter keeps tracks of all widgets created? Or
destroying the widgets does not clears all references to the widget?
Then what must I do to destroy the object when the window gets
destroyed?

Thanks,
Jerome
--
jerome.quelin at insalien.org



From phd at phd.russ.ru  Thu Jun 15 12:39:50 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 15 Jun 2000 16:39:50 +0000 (GMT)
Subject: Suggestion: "Completing" HTTP support in Python.
In-Reply-To: <3949028F.3B512AF9@darmstadt.gmd.de>
Message-ID: <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru>

On Thu, 15 Jun 2000, Robb Shecter wrote:
> Oleg Broytmann wrote:
> 
> > On Thu, 15 Jun 2000, Robb Shecter wrote:
> > > h = HTTP('my.host.name', '/loginscript', 8080)
> > > (code, msg, headers) = h.post()
> >
> > import urllib
> > filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript")
> >
> 
> Thanks for the reply, but I don't think that does what I'm looking for.
> (Although - it could very well be that I just don't know how to use the library!)
> 
> How do you add HTTP parameters?
> How do you specify that the connection should be POST or GET?

GET:  filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript?%s" % urllib.urlencode({"param": value}))
POST: filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript", urllib.urlencode({"param": value}))

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From MarkH at ActiveState.com  Fri Jun 30 21:33:55 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 01 Jul 2000 01:33:55 GMT
Subject: Reference Counting Woes in Python 1.5.2! Help!
References: <wr675.13560$227.259336@nnrp1.uunet.ca>
Message-ID: <7Ub75.2486$Tb7.18780@news-server.bigpond.net.au>

> I am using Visual C++ 6. It has debug libraries that are catching this
> problem printing an error message and quitting. When I go to release
version
> of code, I get access violations and then the code quits. Alas, what I'd
> like is to see the access violation location in source code, but the
darn
> debug library gets in the way!

It shouldnt be doing this!  An access violation when running under debug
build should definately break into the debugger, and definately does for
me.

You may be seeing Py_FatalError() being called.  This just prints a
message and calls "exit()", which does silently die.

Post Python 1.5, this routine as a block:

#ifdef _DEBUG
    DebugBreak()
#endif

You could try adding that to Python 1.5.2.  Then, in debug builds, a
Py_FatalError() will also break into the debugger.

> Has anyone else debugged pesky reference count problems in Visual C++
with
> Python 1.5.x and come out on the winning end? Your help or tips or
> suggestions on how to debug this sort of thing would be most welcome!

I find reference underflows easier to work with generally than leaks.  But
you definately need the exception in the VC debugger.  The symptoms are
generally that other code will die when trying to use the object with the
bad reference.  This can usually tell you at least the _type_ of the
object in question.

Another good hint - when re-running the same program multiple times under
VC, you almost always get identical addresses allocates between multiple
runs.  For example, if one run shows that object at address "0x123456" has
the ref count bug, restarting the program and rerunning under the debugger
will almost certainly give the same pointer the same address.  This makes
it easier to "back-step" through the offending code in multiple debugging
sessions. (eg, next step after determining the above may be to set a
break-point whenever "*(int *)0x123456" (ie, the objects reference-count)
changes.

Mark.




From dale at out-think.NOSPAMco.uk  Mon Jun 26 10:19:57 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 26 Jun 2000 15:19:57 +0100
Subject: Driving Excel from Python
Message-ID: <8j7oqn$e9q$1@supernews.com>

(Sorry if this appears more than once. I'm getting errors from news server).

I'm driving Excel from Python and most of it's working fine except for:

>>> xlsh.UsedRange.Address(ReferenceStyle=1)
TypeError: call of non-function (type string)

xlsh is an Excel Sheet object and I'm trying to get the used range in RxCx
format. However, when I try and specify ANY arguments to the Address
property, I get this error message. I've tried both positional and keyword
arguments.

I have the Python Win32 book but this this subject is not quite covered in
sufficient depth.

Any help appreciated.



--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants





From cut_me_out at hotmail.com  Sat Jun 10 10:35:04 2000
From: cut_me_out at hotmail.com (Alex)
Date: 10 Jun 2000 10:35:04 -0400
Subject: Reloading on the fly: was: Re: Python Productivity over C++
References: <20000609185450.91797.qmail@hotmail.com> <8hrjg3$dn2$2@news1.xs4all.nl>
Message-ID: <etdpuppeh9j.fsf@w20-575-109.mit.edu>

> Could you post an example of designing the main program to reload
> submodules? I'd like to get that working, too, and coming from
> languages where that's impossible, I've a bit of trouble coming up
> with a solution...

I posted something about that recently.  I've posted the code again
below.  I must admit, I haven't had much occasion to test it yet, as I
haven't been working on anything with complex module dependancies.
Also, if you had many modules, it would probably be a bit slow as it
stands.

import sys, string
from types import ModuleType

class Reloader:

    '''Here is my attempt at something that figures out an appropriate
    reloading order for you.  You create an instance of the class,
    call it 'rr', before you have imported any of the modules you'll
    be wanting to reload (the .pythonrc file would be a good place.)
    Then when you wish to reload the modules in a given namespace, you
    run the command

    exec (rr (globals ())) 

    In my case, the namespace would usually be __main__, so this would
    just go at the top of the script I'm currently working on in
    emacs.

    Alex.
    '''

    def __init__ (self):
        self.keepem = sys.modules.copy ()

    def get_new_modules (self):
        modules = {}
        for i, m in sys.modules.items():
            if (not self.keepem.has_key (i)) and \
               (type(m) == ModuleType):
                modules[i] = m
        return modules

    def clear_sys_modules (self):
        for i in self.modules.keys ():
            if sys.modules.has_key (i):
                del sys.modules[i]

    def get_dependancies (self, module_name):
        self.clear_sys_modules ()
        exec 'import ' + module_name
        new_modules = self.get_new_modules ()
        del new_modules[module_name]
        return new_modules

    def determine_reloads (self):
        self.return_commands = []
        while self.dependancies:
            for i, deps in self.dependancies.items ():
                if not deps:
                    break
            else:
                raise 'Cyclic dependancy(?)'
            exec 'import ' + i
            reload (eval (i))
            if self.namespace.has_key (i) and \
               (self.namespace[i] == self.modules[i]):
                self.return_commands.append (
                    'import %s; reload (%s)' % (i, i))
            del self.dependancies[i]
            for deps in self.dependancies.values ():
                while deps.count (i):
                    deps.remove (i)

    def __call__ (self, namespace):
        self.namespace = namespace
        self.modules = self.get_new_modules ()
        self.dependancies = {}
        for module_name in self.modules.keys ():
            deps = self.get_dependancies (module_name).keys ()
            self.dependancies[module_name] = deps
        self.determine_reloads ()
        return string.join (self.return_commands, ';')

rr = Reloader ()

if __name__ == '__main__':
    '''This is a test script, intended to be run with two modules, B.py
    and C.py somewhere in the directories in sys.path.  Their contents
    should be

    B.py:
    ######################################################################
    import C

    class B (C.C):
    
        def __init__ (self):
            C.C.__init__ (self)
        
    C.py:
    ######################################################################
    class C:

        def __init__ (self):
            pass
    '''
    import B, C

    bid = id (B.B)
    cid = id (C.C)
    exec (rr (globals ()))
    assert (bid != id (B.B)) and (cid != id (C.C))

    t = B.B ()


From mbel44 at dial.pipex.net  Thu Jun 29 07:29:23 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Thu, 29 Jun 2000 12:29:23 +0100
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194>
Message-ID: <ffcmlsc6p4lqohdd6tequdu918l0temns4@4ax.com>

gmcm at hypernet.com (Gordon McMillan) wrote:

>I think the "relative import" feature should be shot in the head (2 posts 
>today), but unfortunately most packages rely heavily on it.

A relative import is just as useful as a relative directory: it allows
intra-module imports, without each module needing to know where it is
packaged.

The real problem is that relative imports and absolute imports are
spelled the same way.

Toby Dickenson
tdickenson at geminidataloggers.com


From bill at libc.org  Sat Jun 17 16:53:39 2000
From: bill at libc.org (Bill Anderson)
Date: Sat, 17 Jun 2000 14:53:39 -0600
Subject: Which linux distros have Python by default?
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial> <%TS15.655$i06.206156@ptah.visi.com> <20000614175918.A5625@dophnic.yi.org> <RVW15.694$i06.216628@ptah.visi.com>
Message-ID: <394BE553.99FBC464@libc.org>

Grant Edwards wrote:
> 
> On Wed, 14 Jun 2000 17:59:18 -0500, Derek Moeller <moeller at dophnic.yi.org> wrote:
> >
> >> >Pretty soon, everyone will want to have Python to use ESR's CLM2, a
> >> >configuration utility for the the kernel written in Python/Curses.
> >> >Tkinter is a different matter though
> >>
> >> Well, people who compile kernels will, anyway.  There are
> >> getting to be a lot of users who never do a kernel build...
> >
> >Yeah, but the distribution maintainers will be forced to include it as
> >a default to be considered seriously.
> 
> Do most of the "consumer" distros even include kernel sources
> by default?  I'm not positive, but I don't think RH includes
> kernel sources in the standard "workstation" configuration.


In the case of Redhat (and I believe Mandrake), it is basically
irrelevant, since the installer and many admin tools require python,
making it a mandatory install. They do not install the devel files, but
python is available.


From android at micron.net  Mon Jun  5 02:05:55 2000
From: android at micron.net (Peter Immarco)
Date: Mon, 05 Jun 2000 02:05:55 -0400
Subject: Windows 2000 Server compatibility?
Message-ID: <393B4343.E3A96525@micron.net>

Is there a version of Python that runs well on Windows 2000 Server?  Any
gotchas?

Peter.



From rhicks at rma.edu  Wed Jun 14 09:07:34 2000
From: rhicks at rma.edu (Robert)
Date: Wed, 14 Jun 2000 13:07:34 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <qsL15.20$RB1.6366@news>

I don't agree. I am learning Python and have no C/C++ background. I am
learning it as my first language and I am really enjoying it. What I think
prior knowledge of any language gives you is a "mindset" on the overall
creation of a program (methodologies et al.). That has been the hardest part
for me.

Anyway...my 2 cents...

Robert

"jin choung" <jin.choung at gte.net> wrote in message
news:7QG15.2637$sI4.335806 at dfiatx1-snr1.gtei.net...
> howdy all,
>
> i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
> tried to pick up python before moving on to c++.  however, i was trying to
> work my way through the book LEARNING PYTHON and while it says that there
is
> no need to have prior experience in languages like c or c++, i really
found
> that that was not true.
>
> anyone else feel this way?
>
> i'm currently working my way through a 24 hour c++ book and i think this
is
> going to work for me.  oh, well, will be back to contribute here in a
couple
> of months... in the meantime, i'll just leech valuable nuggets of wisdom
> from you guys.
>
> thanks
>
> jin
>
>




From sabren at manifestation.com  Fri Jun 30 19:11:58 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 30 Jun 2000 19:11:58 -0400 (EDT)
Subject: MD5 - Passwords with python
In-Reply-To: <2vegj8.ia5.ln@odin.hmf.de>
Message-ID: <Pine.BSF.4.10.10006301910440.27864-100000@chi.pair.com>

On Thu, 29 Jun 2000 holger at fuessler-net.de wrote:

> Unfortunately, the standard MD5-module doesn't know about the
> structure of the password-string so I can't encode a password with
> it. I found a perl module which accomplishs my task and so I thought
> that there is also a python module. I thought redhat uses a lot of
> python stuff, so there should be a routine to generate suitable
> passwords. I'v already posted in comp.redhat.security. When I find
> the details, I'll post them here (if I can translate them to
> python).


http://www.sabren.com/code/python/crypt/


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From darrell at dorb.com  Tue Jun  6 00:35:53 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 5 Jun 2000 21:35:53 -0700
Subject: Executing a function based on user-input
References: <393C4BAC.568296DB@eudoramail.com>
Message-ID: <043901bfcf70$b3691b80$6401a8c0@home>

From: "Ramachandran Subramanian" <rsubramanian at eudoramail.com>
> I have a number of functions. Based on user input (a number) from a
> menu, I want to execute one of these functions.
> Is there a better way to do this?
> 

Don't know if this is better or not. 
If security is an issue http://www.python.org/doc/howto/rexec/

>>> glb={'go':go, 'stop':stop}
>>> while 1:
...     s=raw_input("-->")
...     exec(s, glb)
...
-->go()
go
-->for x in range(5): stop()
stop
stop
stop
stop
-->import os
-->os.listdir('.')
-->print os.listdir('.')
['reSubReturn.py', 't2.py~', 'tabIt.py']
-->




From richard_chamberlain at ntlworld.com  Sat Jun 17 12:00:31 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 17 Jun 2000 17:00:31 +0100
Subject: CodeWright anyone?
References: <NEBBIOGBDMDPEANLFKGHKEOPCDAA.jschmitt@vmlabs.com>
Message-ID: <C7N25.674$fw6.13719@news6-win.server.ntlworld.com>

John,

There is a bit about using CodeWright with Python here:

http://www.python.org/doc/howto/editor/node6.html

I've never used it so I can't comment (just remember seeing it)

Richard


John Schmitt <jschmitt at vmlabs.com> wrote in message
news:NEBBIOGBDMDPEANLFKGHKEOPCDAA.jschmitt at vmlabs.com...
>
> Does anyone use CodeWright to edit Python code?  I had a quick go at
setting
> up a lexer so that my project view works like it does in C, but I
obviously
> don't know enough about regular expressions and about what to do with
> matches in the CodeWright tool.
>
> John
>
>




From DOUGS at oceanic.com  Fri Jun 16 23:25:36 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Fri, 16 Jun 2000 17:25:36 -1000
Subject: Learning curves (was: Python so far)
Message-ID: <8457258D741DD411BD3D0050DA62365907A235@huina.oceanic.com>

[bhahn said:]

> X is time elapsed, Y is knowledge acquired.  So the common 
> usage of "steep"
> with regard to learning curves is in fact exactly backwards.  

Not exactly.  I'd say rather that the Y access is usually knowledge
_re_quired.

Thus:   (How do you HTML types do it <font courier, size 10>)

 Lots |            ZZZZZZZZ <- My experience of C++
      |           Z
      |          Z
      |         Z     
      |        Z    
      |       Z     
 Some |      Z     YYYWHYYY <- P$?#-hwt the other
      |     Z    Y             p language
      |    Z   Y
      |   Z  Y
 Not  |  Z Y       XCELLENT <- My experience of
 Much | ZY    X                Python ;-)
      |Z___________^_________________________

 ^                 ^
 |                 |
 \-- Knowledge     \-- Time required to attain
     required to       a working knowledge
     be productive

-Doug-



From embed at NOSPAM.geocities.com  Thu Jun 15 09:51:32 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Thu, 15 Jun 2000 09:51:32 -0400
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com>
Message-ID: <Ra525.3904$227.93795@nnrp1.uunet.ca>

> Indulge my curiosity - how many times has this actually happened in
> your six months of using Python?

I admit that using a proper editor helps. When using IDLE, or PythonWin, it
NEVER happens.

Nevertheless, I have been forced at various times to use Visual Studio, or
Some Editor That Was Around At The Time, for Python use, and it has bitten
me more than a few times.   I don't know an exact number, but let's say
about 25 times.

That's really not the point though, and I know it.

Really it's about the psychological damage done to me long ago by other less
humane languages and tools.

The lingering memory of COBOL and Fortran, as well as some horrible
incidents involving Makefiles, have given me the cold shivers with respect
to whitespace and indentation.

Accidental Complexity my ass. It's a conspiracy! ;-)  Perhaps COBOL and
Fortran and Makefiles were cruel experiments, foisted upon me by the Smoking
Man, and the other X-Files Illuminati.  There has been permanent damage done
to my brain by these three horrid tools.  All were forced upon me, a naieve
computer science student. The effects can never be repaired.

I still see the schizophrenia that resulted from this trauma in myself, and
in other programmers.  For example:

I have many developers in my office who would cry more loudly than I about
Python's "required indentation", and yet these same developers are always
telling me that all our company's programmers should tab and indent their
code the same way,
for "readability".   Few can see the contradiction there.

I'm with Python on this one, at the end of the day, but I'll still whine
about it when I get stuck with using a lame editor.

A strict indentation scheme is ideally intended to be used with editors that
can help a little bit.

Warren




From m.faassen at vet.uu.nl  Tue Jun  6 13:57:08 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 6 Jun 2000 17:57:08 GMT
Subject: Space Sensitive?
References: <8hj9vv$1kd@autodesk.autodesk.com>
Message-ID: <8hje1k$nfm$2@newshost.accu.uu.nl>

Akira Kiyomiya <akira.kiyomiya at autodesk.com> wrote:
> I hear Python is space sensitive, however, "Learning Python" from O'Reilly
> and "Python Essential Reference" from New Riders do not cover this topic.

> Could someone explain this one with a nice simple example why Python is
> space sensitive?

They don't? I don't have "Learning Python", but look at the section
'Line Structure and Indentation' in chapter 2 (page 13) of the Python
Essential Reference. 

Perhaps you're confused with 'case-sensitive'? :)

Example of Python's whitespace sensitivity (it's not just spaces. Tabs
too):

if x > 2:
    print "x is greater than 2"
    print "to be exact, x is", x
else:
    print "x is smaller than or equal to 2"
    print "Too bad, you don't win the prize"

As you see, the 'print' statements here are indented (by 4 spaces in this
case, which is common. Tabs are also frequently used. Just don't mix them)
The indentation determines the block structure. This is opposed lots of
other languages. Take for instance C, C++, Java, or Perl, where they use
'{' and '}' to determine block structure, or a language like Pascal where
'begin' and 'end' are used.

Case sensitivity is another topic. It means that to the interpreter, 'foo'
is not the same word (such as variable or function name) as 'Foo', and
also not the same word as 'FOO'. It matter whether you use uppercase or
lowercase.

Python's both whitespace-sensitive and case-sensitive.

Regards,

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


From erno at iki.fi  Tue Jun 20 16:15:36 2000
From: erno at iki.fi (Erno Kuusela)
Date: 20 Jun 2000 23:15:36 +0300
Subject: A few python questions
References: <slrn8ktmmj.2md.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <kupupcw1lj.fsf@lasipalatsi.fi>

>>>>> "Suchandra" == Suchandra Thapa <ssthapa at midway.uchicago.edu> writes:

    Suchandra> The first problem I've encountered is trying to
    Suchandra> determine the size of a file.

os.path.getsize(path)

    Suchandra> is there a way to get the file size of a file without
    Suchandra> doing this or using the stat function from the os
    Suchandra> module?

that works too, the indexing constants are in the stat module:
os.stat(path)[stat.ST_SIZE]

    Suchandra> Also is it possible to use PyKDE or PyQT without
    Suchandra> installing SIP?

(no idea)

    Suchandra> Finally, is there some documentation for the python
    Suchandra> interface to postgresql?  The rpm package that comes
    Suchandra> with redhat 6.2 doesn't have any documentation.

yes, the README that comes with the module documents the
postgresql-specific api. the most recent versions support DB-API
also. get it from <URL:ftp://ftp.druid.net/pub/distrib/>.

  -- erno


From ge at nowhere.none  Thu Jun 29 14:55:26 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 29 Jun 2000 18:55:26 GMT
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com>
Message-ID: <yYM65.2213$iN5.491623@ptah.visi.com>

In article <39588741.FB7EB5FE at hursley.ibm.com>, Paul Duffin wrote:
>Donn Cave wrote:
>> 
>> Quoth ge at nowhere.none (Grant Edwards):
>> | In article <kuzoobue0g.fsf at lasipalatsi.fi>, Erno Kuusela wrote:
>> |>>>>>> "Tony" == Tony Keating <keating at mech.uq.edu.au> writes:
>> |>
>> |>    Tony> start the program, then send various strings to the program,
>> |>    Tony> read some output from the program then send some more
>> |>    Tony> strings to the program.
>> |>
>> |> there are a couple of expect modules for python (so named because
>> |> there is an old tcl-based language called expect for doing just this).
>> |> parnassus lists 3 modules (<URL:http://www.vex.net/parnassus/
>> |> apyllo.py?find=expect>), i haven't
>> |> used any of them however...
>> |
>> | Or you can roll your own using the popen2 module.
>
>I think that you will find that Expect offers a lot more than a simple 
>pipe to the executable.

Yes, I know. That's where the "roll-your-own" bit comes in. If
all you _want_ is a simple pipe to the executable, then expect
isn't really the right choice.

-- 
Grant Edwards                   grante             Yow!  BARBARA STANWYCK
                                  at               makes me nervous!!
                               visi.com            


From yawyi at gmx.de  Fri Jun  2 17:21:07 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Fri, 2 Jun 2000 23:21:07 +0200
Subject: Help with SSL in 1.6
References: <Pine.LNX.4.21.0006021832560.13804-100000@lade.src.no>
Message-ID: <39382a75$0$25922@personalnews.de.uu.net>

Joseph Santaniello <joseph at src.no> wrote:
>I've just built OpenSSL and Python 1.6a for the express purpose of being
>able to use urlopen or similar to grab some ssl web-pages. But I can't get
>it to work. I get name errors when I try to use HTTPS or anything it seems
>because the "if hasattr(socket, "ssl"):" condition never seems to be
>true. And urlopen says https is an unknown protocol. What have I missed in
>setting this up so that I can grab ssl pages?

configure doesn't seem to test for OpenSSL.
You must edit Modules/Setup manually (copy from Setup.in,
if necessary):

- comment out:
#socket socketmodule.c  # socket(2); not on ancient System V

- uncomment and set proper path:
# Socket module compiled with SSL support; you must edit the SSL variable:
SSL=/usr
socket socketmodule.c \
        -DUSE_SSL -I$(SSL)/include -I$(SSL)/include/openssl \
                -L$(SSL)/lib -lssl -lcrypto

Johannes



From haugland at dl2sql9.princeton.edu  Wed Jun 21 11:56:30 2000
From: haugland at dl2sql9.princeton.edu (Jan Kristian Haugland)
Date: Wed, 21 Jun 2000 11:56:30 -0400
Subject: Nth digit of PI
In-Reply-To: <t7545.239$ZR1.2944@news.uswest.net>
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net>
Message-ID: <Pine.LNX.4.10.10006211154500.875-100000@dl2sql9.princeton.edu>

On Wed, 21 Jun 2000, Jeff Massung wrote:

> Please, this interests me -- I've had differention equations, but that's as
> far as I go. In theory, pi (and e) are infinite, but has that been proven
> that there is not end to their decimals?
> 
> Jeff

Someone else already answered your question, but what 
do you mean by "in theory, pi and e are infinite"? 




From mhammond at skippinet.com.au  Fri Jun  9 03:03:12 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 09 Jun 2000 07:03:12 GMT
Subject: Win32net.NetServerEnum Unicode problem
References: <8hlnd2$lrm$1@nnrp2.deja.com> <erz%4.2026$wp.166067@pouncer.easynews.com> <8hoglp$54l$1@nnrp1.deja.com>
Message-ID: <QE005.118$cQ.2216@news-server.bigpond.net.au>

<jcashman55 at my-deja.com> wrote in message
news:8hoglp$54l$1 at nnrp1.deja.com...
> I have Hammond's Python on Win32 book, and it doesn't list another
> parameter relating to domain name.

This is an error in the book.  Ive updated the book's page at starship.

Mark.





From mhammond at skippinet.com.au  Fri Jun  9 20:56:46 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 00:56:46 GMT
Subject: Debuggers.
References: <39483789.94409484@nntp.interaccess.com>
Message-ID: <ing05.709$cQ.9342@news-server.bigpond.net.au>

"Thaddeus L. Olczyk" <olczyk at interaccess.com> wrote in message
news:39483789.94409484 at nntp.interaccess.com...

> Everyone seems to say that pythonwin comes with one, but I don't see
> it. OTOH I have v 1.5.1

Pythonwin's debugger has improved enormously since 1.5.1 - it wasnt even
bundled with Pythonwin back then!

Get 1.5.2 and win32all-132.

Mark.




From m.faassen at vet.uu.nl  Wed Jun  7 09:51:25 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 13:51:25 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA>   <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net> <8hiq8d$ngm$1@newshost.accu.uu.nl> <slrn8jr8qm.b4q.wtanksle@dolphin.openprojects.net>
Message-ID: <8hlk0t$h4q$1@newshost.accu.uu.nl>

William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
> On 6 Jun 2000 12:19:25 GMT, Martijn Faassen wrote:
>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>> On 5 Jun 2000 12:01:04 GMT, Martijn Faassen wrote:
>>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
[snip]
>>>>> This is a classical example of false dichotomy ;-).

>>>>It's not.

>>> It actually is.  You're claiming that I hold one of two opinions; I hold
>>> neither one.

>>I'm claiming you haven't shown yet that consistency has nothing to do
>>with 'code looking good'. Anyway, you should've said something like that
>>in your first post -- now you explain what you meant. Now I'll explain
>>what I meant. :)

> You're claiming something that you've never written before -- I don't even
> understand what you're claiming that you claimed.  (Whew.)

Anyway, you pulled a false dichotomy out of your hat. I pulled the
"it's not out of my hat". It would've helped if you explicitized 
the dichotomy. :) Then I explicitized my 'no'.

> What I disagreed with was your false reduction of my argument to two
> views, neither of which I held.

All right; the views were consistency or decreased line count. Your view
appears to be that the advantage is lack of delimiters. Then we got into
an argument about consistency and I was continuing that argument there. :)

>>>>> I liked Python's
>>>>> indentation not because it enforced consistency, but rather because it
>>>>> makes code look good.  In fact, it could be argued that Python does _not_
>>>>> enforce consistency; the programmer is free to choose any indentation for
>>>>> each block, so long as each block can be parsed (which does require
>>>>> internal consistency).

>>>>"""
>>>>I liked Python's
>>>>case sensitivity not because it enforced consistency, but rather because it
>>>>makes code look good. In fact, it could be argued that Python does _not_
>>>>enforce consistency; the programmer is free to choose any case-spelling for
>>>>each variable name, as long as each variable name can be parsed.
>>>>"""

>>> But now you've converted my paragraph from a true statement to a false
>>> one.  The programmer is not in fact able to choose any case-spelling for
>>> each variable name; they must use the same one which was originally used,
>>> unless they wish to create or access another variable.

>>Perhaps that's unclear, but my reading of 'for each variable name' is for
>>each unique variable name -- I mean, a name can be referred to multiple
>>times, but it's not a different name each time.

> But again, this doesn't match with what I said about indentation.
> Python's indentation is flexible because you can choose your own
> indentation for each block, regardless of how other blocks are indented.
> This can be very handy!

Python's indentation is less flexible because you can't choose variable
indentation for a *single block*. You also can't choose not to indent.
Let alone that all options on where to place the delimiters are removed.
This means a definite cramping of indentation styles. I'm making the parallel 
with case sensitivity, where you can allow variable case-spelling for a 
single name or not. Not allowing variable case spelling for a single name is
less flexible and enforces consistency in case-spelling.

> Python's case sensitivity is not flexible: once you write a variable name
> one case, you're required to write it the same way.

Right. It's less flexible, just like Python's indentation strategy is
less flexible.

> I think I'm just stretching the metaphor, though.  I don't really think
> there's enough similarity between variables and indentation to really use
> effectively.

The argument is that reduced flexibility reduces variability in coding styles
(and decreases the chances of inconsistent style or absence of style).
I think there's a clear similarity there.

[snip]
>>So either you're saying you like Python's
>>indentation because it doesn't have braces (and that's the only reason),
>>or you're buying into the consistency argument after all. I believe I
>>gave you the braces option before, speaking of dichotomies.

> You did NOT give me the brace option before, as you can clearly see by
> reading the first paragraph.

You're right; I gave the reduced line count option; I checked. My
apologies.

> Now you give me it as though it were the
> ONLY reason; clearly, by my own discussion, it's not the ONLY reason;
> however, I will agree that it's a strong one.

The other reason (showing the tree structure) applies to all languages.
In those languages delimiters and not indentation is used to indicate
tree structure. For some reason those delimiters alone make code harder to
read for humans than the use of indentation.

[snip]
>>> Of course they should.  Lisp programmers get really mad if people post
>>> non-standard formatted code on c.l.lisp.  EMACS enforces it when writing,
>>> and angry programmers enforce it when reading ;-).

>>So in your case there's no reason to prefer Python's way of indenting,
>>*except* that there are no delimiters. Okay, thanks for making that 
>>clear. Or am I still missing something? :)

> You got it right.  Not *exactly* right, since you still seem to insist
> that I'm only allowed to like Python's indentation for exactly one reason.

I don't see how there can be another reason for you than a combination
of reduced line count and absence of delimiters. The tree structure argument
alone doesn't work, as delimiters show tree structure as well. But indentation
makes tree structure easier to grok for humans. Since other languages *can*
use indentation as well, the only reason that remains to prefer Python's
style would be that you like the absence of the delimiters/reduced line
count. Or more broadly speaking, an absence of redundancy.

[snip]
>>Yes, you were saying the same for case. Now, if I understand you right,
>>you like Python's indentation strategy more than the delimiter strategy.
>>The only two reasons I can come up with is:

>>  * less redundancy because there are no delimiters
>>  * less space taken by the delimiters

>>(because the two-dimensional structure argument makes sense for any language
>>if only you indent I'm not including that)

> You'd best include it anyhow -- because it's the reason.  It does NOT make
> sense for languages in which it's not, in fact, used.  There are not many
> languages in which indentation is used.

But indentation *can* be used in those languages. The *programmer* can
consistently enforce indentation. In most languages this is possible and
common practice. Therefore, the *only* reason you can like Python
better than those other languages is because of the lack of delimiters.
Because you don't buy the consistency argument, that's the only thing
left. Unless this is another false dichotomy and I'm missing something
essential. Coming up next, I see. :)

[snip]
>>Right, so that's the redundancy argument. Of course, if you used Lisp or C
>>without any indentation, you'd have that same feature too. Why do you
>>prefer the indentation approach to the delimiter approach? According to
>>your opinions on case-spelling one would think you'd prefer the former, as
>>it offers more freedom while the programmer still has the option just
>>fine to enforce any indentation strategy they want, if they want it.

> Excellent question, and this will hopefully clarify EXACTLY why you should
> include reason #3 in your list of reasons why I like Python's indentation.

> Almost all modern languages are tree-structured.  Therefore, they're most
> efficiently written and displayed as a tree.  Trying to write them as a
> delimited stream may be effective for computers (although it's still
> difficult), but it's terrible for humans.

Right. Agreed. Human interface reasons make indentation a better way to
display a tree than delimiters.

> Therefore, we write even languages which use only delimiters as though
> they were trees.

> Python breaks away from that.

So the only reason you prefer Python's approach would be the lack of 
delimiters. :) Reason #3 is no reason to *prefer* Python's approach --
you gain the same advantage in C as long as you indent properly, after all.

[snip]
>>> But in case-sensitive languages, words can be spelled *differently* in
>>> case: NO is not the same as No or no.  If the language were
>>> case-insensitive this would be impossible.

>>Hm, I'd say those would be different words, which vary only due to 
>>case-spelling. Not that the same word can be spelled differently in 
>>case.

> Notice how hard it is to communicate that those are different words,
> though?

That's because you confused the natural language 'English' (which is
tolerant of misflusner to vlibracious extent) with the programming
language Python (which isn't). It's harder to communicate the difference
between 'NO' and 'no' in Python than it is to communicate the difference
between 'x' and 'y' in Python because of this confusion, true.
But the difference between natural language words and words in
programming languages will have to be made clear to newbie programmers
pretty quickly anyway.

>  Even in typed communication we're not consistent, and in spoken
> communication we can't communicate that info at all.  Is it any wonder
> that newbies get confused?

It's no wonder newbies can get confused, but newbies will get confused
in any case about far worse things like immutability/mutability. It could
actually be good in a didactic sense to make newbies care about consistency
on all levels right away, because consistency is of extreme importance
if you want to produce good code. Also, newbies will have to read their
own code, and one can argue that the more readable this code is the
more easily they can understand it.

>>Ignoring that point, spelling the same word in different case-spellings
>>is only possible in case *insensitive* languages!
>>So I'm not quite sure what you mean here?

> Those words are pronounced and subvocalised the same -- unless, of course,
> the programmer was boorish enough to use any two of them in the same
> routine, in which case you'll probably have to read the routine again,
> this time subvocalising differently.  With case-insensitivity, you won't
> have to; your first reading was correct.

Except that now it's harder to read. Actually programmers use case
in many conventions. If something is spelled with a starting uppercase
in idiomatic Python code it's usually a class. If something is spelled
in all caps it's usually something intended to be a constant. So if I
see 'Foo', I think 'the class Foo', and if I see 'foo', I think 'the
instance foo'. If I see 'foo()' I think 'the function foo'. And if I see
'.foo()' I think 'the method foo'. It's almost like Perl; context changing
the meaning of the variable. As they're all objects. :)

> You know, for the first time it occurs to me that our learning styles may
> differ -- I'm somewhat auditory, and from the way you talk, I'm guessing
> you're primarily visual.

That's of course possible (though you do like indentation, which is of course
purely visual -- the auditory argument would be in favor of delimiters).

> The obvious solution is to compromise so that both styles of perceptivity
> are accomodated: the visual learners must be able to see things exactly
> the same (or they don't recognise them as being the same), and the
> auditory learners have to be able to vocalise and subvocalise things.

> Both of these are solved not by language fiat, but rather by convention:
> always use lower case.

Hm, but that will remove important visual cues that exist now. You'd have
to reintroduce those cues in some other sense that's both visual and
auditory. That's hard to do for names which have (MultipleWords or 
multiple_words), also hard for function calls. For classes you could
work with a prefix, perhaps. Obviously one has to engaged in a visual to
auditory translation when one reads code anyway. 'foo()' becomes 'here
foo is called', unless you vocalize the (). For me those are just 
*umph*. Hm. No real vocalization. :)

>>No, because I as a programmer happen to like the freedom different
>>case-spellings happen to offer. It's up to the *programmer* to
>>enforce the right conventions here. Throwing your arguments back at you
>>again. :)

> EXACTLY!  Now you're arguing 100% on my side -- only thing that's puzzling
> me is that you imagine you're disagreeing with me.  Evey single thing
> you're saying there is an argument ONLY for case-insensitive languages;
> because only with them can you choose case-spellings with freedom.

No, the argument is that I want *one single* case spelling for each
name in the system. I want to have freedom *there*. I don't know if that's
really a 'freedom', but I happen to like the restriction, which I'd
lose. I also like having the actual freedom of foo (instance) versus Foo
(class).

> And it's always up to the programmer to enforce conventions; I'm glad you
> finally agree on that.

No, I didn't say it *always* is up to the programmer to enforce conventions.
It's up to the programmer to enforce conventions when the system either
_can't_ enforce conventions or if the system enforcing certain conventions
makes that system significantly harder to use (and of course the system
needs to enforce the right conventions :).

I want my system to enforce the convention that 1 + "foo" makes no sense.
There are systems which don't enforce this convention and have another
one where it does make sense. But I like those less.

The system has to help enforcing conventions whenever possible, because
I want consistency, simplicity and clarity. Not only in my own code,
but elsewhere too. And a technological aid will help here in many cases.
The 'one best way to do it' principle (though of course in many cases
this best way can only be determined by social conventions).

>>>>> So I don't really care about languages enforcing consistency; if I need
>>>>> it, I'll enforce it myself.

>>>>A computer language would be worth less if it brought no consistency at
>>>>all.

>>> Granted.  But this doesn't seem to justify your slippery-slope treatment --
>>> we don't need to enforce your amount of consistency just because no
>>> consistency is bad.  We already enforce _some_.

>>In the case-sensitive version we have now, we already enforce some, but
>>not to the point the language becomes unwritable. It's also a simple
>>rule, which helps fit it in people's minds. The case-insensitive rule
>>would also be simple, but the code becomes less readable and not 
>>significantly more writable (in my opinion). The case-enforcing rule
>>makes code even more readable (possibly -- there are arguments about
>>this by the people who like different case-spellings), but it becomes
>>far less writable, and the rule has a bit of extra complexity.

> Code with case-insensitivity becomes vastly more writable,

I disagree. I don't like writing: foo = TFoo() or aFoo = Foo(), for
instance. I also like to be able to write: CONSTANT = 12 and know people
will always know it's that. Or for that matter, 'class Foo' and people
will always see instantly that 'Foo' is probably a class.

Perhaps classes should be in a separate namespace:

foo = classes.foo()

and constants too:

constants.foo

It's an idea.. :)

When I write code I think of the readers.

> and offers a
> danger (but not a certainty) of becoming less readable.

Consider that a certainty. Look at some Delphi code to see some variations
of *keyword* spelling, even.

> The writer must
> be instructed to prefer lower case -- but this is the same instruction you
> have to give to writers in a case-sensitive language.

Except that there, it's enforced by the libraries. If the beginner wants
to use a library, the beginner has to follow the library conventions. That's
a good technically enforced way to spread case idioms and consistency. 

[snip]
>>> And a lot of other
>>> things, all of which I hope to see.  I'm curious to see whether other
>>> people like them, and I'll defend them -- but unlike most of the people
>>> arguing about case sensitivity, an ENORMOUSLY minor issue,

>>Note that you're arguing about it as well. It's a minor issue with a lot
>>of effects on readability. And it's easy to argue about. So that's why
>>we do. :)

> Oh, of course.  I'm not annoyed by arguing about anything; I'm annoyed by
> the overstatement of the importance of case-sensitivity.

I think there's a disagreement on the importance of case-sensitivity.
While the issue is very simple, that doesn't mean the issue isn't
important. It's only minor in the technical sense. 

>>> I won't
>>> consider them prerequisites to using the new Python -- I'll use the new
>>> Python if it feels like an improvement.

>>Oh, I'll likely use the new Python if it's case-insensitive. But I'd
>>prefer it would not be so, and I'm just trying to spread a meme here;
>>the link with indentation. Your own meme-complex seems to reject it,
>>so I'll stop trying with you. :)

> It's not a terrible link, really; it just works the opposite of the way
> you'd like it to.

You haven't convinced me any more than I have convinced you. I won't
reiterate the arguments *yet again*, though. :)

[snip]
>>> And for case sensitivity it _just doesn't matter_.  The only people who
>>> are strongly helped or hurt by case sensitivity appear to be newbies, and
>>> they're being hurt.

>>I'd like some more evidence about how the newbies are being hurt,

> Guido cites some studies in CP4E, and the Alice people claim the same
> (although I've not read the studies).

Hm, haven't seen the studies Guido cites, except the Alice one, which 
is anecdotal and not focused on teaching programming but on teaching
a GUI with scriptability; in a GUI I'd probably prefer case insensitivity
at many points.

[wouldn't case insensitivity hurt the maintainer?]

> Honestly, it wouldn't hurt me; I wouldn't even care for more than a couple
> of seconds, as I decided which tradition I should follow (I'm good at
> adapting my coding style to that of the code I'm maintaining).

I'd rather have a single coding style (in that respect) so I don't have to do
any adapting. That's not only important when you go from application
to application, but also in a single application.

>>It's easy to say that it won't matter for communication and maintenance,
>>but could you try backing that up?

> It seems to be easy to say it'll hurt you -- do you mind backing that up?
> How much experience do you have with large projects (or small ones) in
> case insensitive languages?

I do have a year's worth of experience with Delphi, which is case
insensitive. Even though this was a single programmer project, I've
encountered code fragments that used uppercase for keywords, which I
adapted. I also worried about my own case-consistency; I wanted my code
to be case consistent but since the compiler didn't check I was never
sure, which did give me a 'messy' feeling.

I haven't waded through lots of Delphi source code written
by others though, except the Delphi Visual Component Library that
comes with it. Ask Tres Seaver, who is elsewhere in this thread; he
has more experience with this.

[snip]
>>> It may or may not be silly -- but it's _definitely_ false.  Python is
>>> _not_ inflexible; you can lay your code out in any way you want, so long
>>> as it's in the same tree shape as its semantic structure implies.

>>It's more inflexible. I can't do:
>>  start()
>>    do_a()
>>    do_b()
>>  commit()
>>And occasionally that would be nice.

> Granted.  But as I said, Python is still not "inflexible"; it's at worst
> less flexible (as you accurately said).  Case sensitivity _is_ accused of
> inflexibility, and in fact it IS inflexible.

Depends on the ways you look at it; just like indentation. It's
inflexible in the sense that if I write 'foo', you'll have to write
'foo' too if you want to refer to it. (I happen to like this inflexibility
because I like the enforced consistency)

It's *more* flexible than case-insensitivity in another way though. I can
use both 'foo' and 'Foo' for different names in the same namespace. 
This flexibility is exploited in common Python idioms.

> Furthermore, there's not
> even any linguistic reason for its existance; its only effect is to
> increase the number of possible symbols, and no programmer wants to
> maintain a program which actually uses case sensitivity to its full power.

Um, I use it to its full power because I like the enforced consistency.
I know you don't buy that argument, but that doesn't mean I don't. :)

What would 'using it to the full power' mean in your opinion?

> And for that matter, this example is easy enough to "solve" using a blank
> try-block:

> try:
>    start()
>    do_a()
>    do_b()
> finally: commit()

No, the semantics of this are entirely different and *wrong*. This
always commits, even if an exception was raised in my code, which is
*exactly* what I didn't want.

[snip]
>>> I don't think it's silly, either; I've complained enough about languages
>>> which don't do this right, such as Make.

>>Right -- Make is rather unwritable. There are no advantages there.
>>I still don't see how these people relate to our present case;
>>these people are complaining (without having actually tried it; anyone else
>>complaining is of course in his or her right to complain) about decreased
>>freedom you get with Python's indentation scheme.

> They're not complaining, by and large, about decreased freedom; if they
> were it would be a lot easier to ignore them (because they'd be correct).
> They're complaining about total lack of freedom.  They're wrong because
> Python has considerable freedom.

All right. Though there is a total lack of freedom in placing the
braces; you don't at all, really.

>>Now you say we should cater to people like that, offering them more
>>freedom (in case spelling), while nobody of *those* people appears to be
>>complaining about that, and while the reason many people (not you) like
>>Python's indentation *is* increased consistency and therefore increased
>>readability.

> Actually, I like the enforced consistency as well --

*faints* :)

> but I think it's far,
> drastically, more important that there be consistency between how the
> languages behaves and how it looks.

Ah. And but currently the language behaves in a case sensitive matter;
classes behave in a dictionary-like way, and string dictionary keys are
case sensitive unless you do something special. As a consequence,
getattr() and friends are case sensitive. It might be possible to
write a Python script that can determine if it's running in a case
sensitive version of Python or not.

> I really don't care if the programmer
> is required to dot his i's, keep case consistent, or use the same
> indentation for all 'if' clauses.  Those consistencies are _minor_
> compared to the ones we really need -- things like good variable names,
> clear syntactic structure, and so on.

Oh, agreed. Though again, the didactic argument might make sense that if
a newbie is taught consistency on the lower syntactic level, he is then
more likely to be also consistent on the other levels. Besides, 
consistency *is* important on the syntactic level anyway to get your
code through the compiler.

>>> There are two obviously terrible errors: inconsistency and inflexibility.
>>> There's no reason to assume that avoiding one will force everyone into the
>>> other, and I'm not making that mistake about you.

>>I'm not quite sure I understand the language. Right now we avoid
>>inconsistency and we also avoid inflexibility in the case-spelling
>>domain.

> Right now we are almost as inflexible as we could get.  We're not at a
> level I would call "bad" (I would imagine recieving errors when using
> alternate casings of words to be "bad").  On the other hand, loosening up
> a bit isn't going to bring us anywhere near to the realm of anarchy some
> Of yOu SeEm to PiCtuRe.

Yes, though perhaps a bit nearer than you suspect. Besides that, there's
the decreased flexibility (foo = Foo()) and definite increased inconsistency
*between* modules if not intra module.
 
>>There are minor disadvantages which apparently confuse newbies,
>>so we ought to think about how to improve matters.

> Good way of stating things.  Do you have any alternate ideas on this
> issue?  I don't -- but that's because I'm reasonably satisfied with this
> solution.

Better NameError and AttributeError reporting could help (there was a
patch posted to the newsgroup).

>>But case-insensitivity gives up consistency and case-enforcing gives up
>>flexibility; so it seems we shouldn't go anywhere. :)

> A false consistency is the hobgoblin of little minds.

Consistency-makes-it-easier-for-little-minds-to-read-code-ly yours,

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


From hzhu at rocket.knowledgetrack.com  Mon Jun  5 14:52:33 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Mon, 05 Jun 2000 18:52:33 GMT
Subject: ANN: MatPy - Matrix package for Python
Message-ID: <slrn8jnts8.tf.hzhu@rocket.knowledgetrack.com>

                MatPy - Matrix package for Python

This package allows numerical computation and plotting in Python with a
MatLab-like interface.  The implementation is stable and passes a small set
of tests.  I have started a SourceForge project on this.

If you are a Python user needing numerical computation, or an Octave user
looking for object-oriented programming and rich data structures, you might
want to give it a try.

It is currently implemented as wrappers around Numeric and Gnuplot packages.
Attractions for current Numeric python users include

- vectors and matrices behave as in linear algebra.
- default *, / are matrix-wise, not element-wise.
- both matrix- and element- functions (such as expm and exp, etc).
- provide interface to gnuplot for plotting.
- provides many convenience operators:  A.T().I() for inverse(transpose(A)).

Attractions for current Octave users include the availability of one of the
best OO language with data structures like lists and dictionaries of any
objects.

Where to get it?
    Home:     http://MatPy.sourceforge.net/
    Download: http://download.sourceforge.net/MatPy/MatPy-0.2.6.tar.gz
    Project:  https://sourceforge.net/project/?group_id=6490

What is there?
    This is an open source project hosted by SourceForge.  You'll find
    codes, tests and docs (html, ps, pdf), links to the project, CVS tree,
    mailing lists, etc.

Why another numerical package?
    To have the benefits of both Python and Octave, while utilizing the
    speed of the Numeric package.

How can I help in the development?
    You are welcome.  There are a lot to be done.  Please join SourceForge
    and let me know your SF username and your interest.  Or if you just 
    have comments, questions, patches, or anything else you want to say,
    mailto:hzhu at users.sourceforge.net.

Author and license:
    Huaiyu Zhu <hzhu at users.sourceforge.net>
    GNU Public License, version 2


Enjoy!


Huaiyu


From borlak at home.com  Wed Jun 14 08:15:10 2000
From: borlak at home.com (Michael Morrison)
Date: Wed, 14 Jun 2000 12:15:10 GMT
Subject: HTML Regular Expression
Message-ID: <iHK15.1036$OZ.185597@news1.rdc2.tx.home.com>

This is my first regular expression, and it's working okay, but I was
wondering if there was a better way to do this.

I read in an HTML document using urllib.urlopen, which returns the document
complete with tabs.  I don't want the tabs.  So I made this regular
expression:

   reobj = re.compile(r"(<.*?>|&#\d.{1,3})")
   text = reobj.sub('', text)
   text = string.replace(text, '\012', '')

How does that look?  The <.*?> is for the tabs, the &#/d.{1,3} is for the
&#183;  characters and the like.  I was told all the &# codes are 3 digits,
and that is the only way I could get the regular expression to erase 'em.

One last question, a pretty newbie one :)  I run a simple loop to keep my
program going...

while going:
    do_whatever
    time.sleep(2)

I do this in pythonwin, but it freezes up in windows, and I can't use ctrl-d
or ctrl-c or anything to stop it without killing it.  Is there a better
generic loop?

Thanks for any help and comments!





From delgado at olaf.de  Thu Jun 15 14:18:39 2000
From: delgado at olaf.de (Olaf Delgado)
Date: Thu, 15 Jun 2000 20:18:39 +0200
Subject: Extension types: conflicting use of "*"
Message-ID: <vl6bi8.jd.ln@olaf.de>

Hi Folks!

Let me first say that I do not write a lot to this group, but enjoy
reading it immensely. Thank you guys!

I am currently trying to put together an extension type for "small"
floating point matrices. Suddenly, after some supposedly innocent
changes, multiplication did not longer work. More precisely, whenever
I wrote a*b, the interpreter tried to do b*a. This is extremely
undesirable for matrix multiplication, as opposed to, say, integer
multiplication. :)

It turned out that this strange behaviour was toggled by supplying a
PySequenceMethods entry to the type object. Oops! I'm not quite sure
what exactly happens and do not feel like digging into the source code
either (well, at least not at the very moment). Perhaps someone else
knows and is willing to explain?

Right now, I have to write A[lo:hi:] instead of A[lo:hi], which is not
too bad. I need a special method to retrieve the list of rows in case
I want to loop over them, which is not too much of a pain, either.

Anyway, all this makes me look forward to Python 3k, the interpreter
of milk and honey. :)

Thanks for your attention!

Olaf

-- 
  ////
  Olaf  Delgado Friedrichs, Bielefeld, Germany
  `='   --- http://www.mathematik.uni-bielefeld.de/~delgado ---


From jajs at NOSPAMretemail.es  Thu Jun 15 23:27:01 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Fri, 16 Jun 2000 05:27:01 +0200
Subject: Which linux distros have Python by default?
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial> <Pine.GSO.4.20.0006152332160.10224-100000@luna.dsv.su.se>
Message-ID: <slrn8kj7k5.7fg.jajs@Pascual.es>

Kalle Svensson dijo sobre Re: Which linux distros have Python by default?,
en fecha: Thu, 15 Jun 2000 23:45:45 +0200

|On Wed, 14 Jun 2000, Moshe Zadka wrote:
|> Pretty soon, everyone will want to have Python to use ESR's CLM2, a
|> configuration utility for the the kernel written in Python/Curses. 
|> Tkinter is a different matter though
|
|But CLM2 isn't going to be standard for a while yet, not before 2.6, I
|think. Not counting development kernels, that is. But they're not
|installed by default, right? :)

|OK, I'm not sure about this one, I admit. I don't even know how the
|flamewars ended on the mailing list, but there was some resistance, and it
|seems like a big change to a kernel that's supposed to be close to
|release.

Yes, but my bet is that 2.4 kernel is still 2-3 months away, and on the
last post on the subject ESR says CML2 is nearly as functional as CML1.

Considering that the kernel configuration tool is not as critical as the
other parts (it won't wipe out your filesystem), I wouldn't be surprised to
see it in 2.4.



From cgw at fnal.gov  Wed Jun  7 11:24:39 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: 07 Jun 2000 10:24:39 -0500
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
References: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD> <14647.58023.493846.225899@buffalo.fnal.gov> <m3wvk1hm2u.fsf@chinon.cnrs-orleans.fr>
Message-ID: <vmebt1dms3s.fsf@buffalo.fnal.gov>

Konrad Hinsen <hinsen at cnrs-orleans.fr> writes:

> I'd prefer a portable solution to one that reflects the system math
> library. I had the same problem under AIX, and I solved it by patching
> Python. I just wonder if someone has the patience to work out a
> patch that would create the same behaviour independently of what the
> math library does. Not me, unfortunately...

What do you think the consistent behavior should be?  Should
math.exp(-10000) return 0, or raise some exception?



From quinn at pyro.ugcs.caltech.edu  Mon Jun 19 00:36:27 2000
From: quinn at pyro.ugcs.caltech.edu (Quinn Dunkan)
Date: 19 Jun 2000 04:36:27 GMT
Subject: Function that knows the names of its actual parameters
References: <394a6395.7217524@news.btx.dtag.de> <jwbnews-9D73CE.13205916062000@news.olympus.net> <394d399b.4366388@news.btx.dtag.de>
Message-ID: <slrn8kr8vq.j8g.quinn@pyro.ugcs.caltech.edu>

On Sun, 18 Jun 2000 21:22:36 GMT, Stefan Franke <spamfranke at bigfoot.de> wrote:
>Huh? Are we talking about the same thing? Keyword arguments either overwrite
>existing ones or supply new ones along with the call. What I am looking for is
>more or less (and arguably sensible) syntactic sugar for building a subdict
>of locals() given non-quoted variable names:
>
>magic(a, b, ...) <-> {"a": locals("a"), "b": locals("b"), ...}
>
>This is obviously not meaningful for expressions as arguments. I don't making
>this look like a function call is a good thing, but I do think the functionality is
>quite useful sometimes (Python, HTML, SQL, etc. generation...).

Well, you asked for it.

This is a slightly modified function I posted a while back.  It gets confused
by functions across multiple lines, and by expressions which involve
variables.  And a bunch of other things, I'm sure.

import sys

def arghack(*_):
    try:
        1/0
    except:
        frame = sys.exc_info()[2].tb_frame.f_back
    func_name = frame.f_code.co_name
    co = frame.f_back.f_code
    c = map(ord, co.co_code)
    c.reverse()
    def getop(st=c):
        try:
            op = st.pop()
        except IndexError:
            return None, None
        oparg = None
        if op >= 90:
            oparg = st.pop() + st.pop() * 256
        return op, oparg

    while 1:
        op, oparg = getop()
        if op is None: break
        if op == 127 and oparg == frame.f_back.f_lineno:
            break

    while 1:
        op, oparg = getop()
        if op is None:
            raise "I'm lost"
        if (op == 101 or op == 116) and (
                co.co_names[oparg] == frame.f_code.co_name):
            break
    d = {}
    argno = 0
    op, oparg = getop()
    while 1:
        if op == 101:
            d[co.co_names[oparg]] = _[argno]
        elif op == 124:
            d[co.co_varnames[oparg]] = _[argno]
        elif op == 131:
            break
        else:
            d[argno] = _[argno]
            while op != 131 and op != 101 and op != 124:
                op, oparg = getop()
            continue
        argno = argno + 1
        op, oparg = getop()
    return d

foo, bar = 1, 2
def magic(a, b):
    print arghack(a, b)

magic(foo, bar)


From jschmitt at vmlabs.com  Sat Jun 17 01:33:46 2000
From: jschmitt at vmlabs.com (John Schmitt)
Date: Fri, 16 Jun 2000 22:33:46 -0700
Subject: CodeWright anyone?
Message-ID: <NEBBIOGBDMDPEANLFKGHKEOPCDAA.jschmitt@vmlabs.com>

Does anyone use CodeWright to edit Python code?  I had a quick go at setting
up a lexer so that my project view works like it does in C, but I obviously
don't know enough about regular expressions and about what to do with
matches in the CodeWright tool.

John




From cut_me_out at hotmail.com  Sun Jun 25 19:06:28 2000
From: cut_me_out at hotmail.com (Alex)
Date: 25 Jun 2000 19:06:28 -0400
Subject: http redirect line in Python dynamic html script.?
References: <395671e6.929362672@news1.on.sympatico.ca>
Message-ID: <etdr99lbbt7.fsf@w20-575-101.mit.edu>

If it were me, I would factor the stuff that you need from
next_script.py to another module, and import that module in both
script.py and next_script.py.

Alex.



From hinsen at cnrs-orleans.fr  Wed Jun 14 05:17:40 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 14 Jun 2000 11:17:40 +0200
Subject: When to import
References: <lieeks4gt73r3va4ad79j5v43lufufkqgu@4ax.com>
Message-ID: <m34s6w1v0r.fsf@chinon.cnrs-orleans.fr>

Anders M Eriksson <anders.eriksson at morateknikutveckling.se> writes:

> When does the import take place?

When the import statement is executed.

> Will there be any difference in the two cases below?

If the import of module2 has no side effects, then the result of the
two scripts will be the same.

> If I have a module that I very seldom use should I only import it when
> needed (using an if statement) or is it better to allways import it?

There is no reason in principle to import everything always. Importing
is just another way to assign an object to a name. If something is used
rarely, and more importantly only in a well-defined region of the code,
putting the import statement close to the "consumer" is a very reasonable
approach in my opinion (and I do it myself).
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From jkraska1 at san.rr.com  Wed Jun 21 03:20:48 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 21 Jun 2000 07:20:48 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <39506D51.B179ED34@san.rr.com>

> I am using dictionaries somewhat but sometimes is it hard to know which data
> structure is best.

You use a dictionary when you want to quickly look up something
based on its identity or something you associate with it's identity.

You use lists for, well, just about everything else (in particular
for things which you want to keep in order, or which you might
wish to access by a fixed position in the list, i.e., the nth
item).

C/


From ge at nowhere.none  Mon Jun 12 12:45:01 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 12 Jun 2000 16:45:01 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <hs815.297$i06.89565@ptah.visi.com>

In article <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf at 4ax.com>, mkx at excite.com wrote:

>So my question is this: Is Python a more "Windows-friendly" language?

IMO, Python is more "person-friendly" than Perl.  Since many
Windows users are persons (though I sometimes wonder about the
its designers), Python could be considered more "Windows-friendly".

>Is it's base structure less related to any one specific operating
>system?

Perl isn't so much related to an OS, as it is to some other
utilities (awk,sed,grep) and languages (bourne shell).

>Is learning Python an exercise in learning the language, and
>not a Unix tutorial? 

Definitely.  Python is very OS-neutral.  The first non-trivial
application I wrote was created mostly under Linux, but runs
exclusively under Win32.  (By using some test-stub classes, it
will run under Linux, but it doesn't really do anything
useful.)

>Can I be successful even if I do not use Unix ;)

Yes, for some definitions of successful.  My definition of
success includes using Windows as infrequently as possible, but
your's may differ.  ;)

As far as Python goes, it works almost as well under Windows as
it does under Unix, and there are some handy libraries (like COM
objects) that exist for Windows and not for Unix.

-- 
Grant Edwards                   grante             Yow!  The entire CHINESE
                                  at               WOMEN'S VOLLEYBALL TEAM all
                               visi.com            share ONE personality --
                                                   and have since BIRTH!!


From strad at scheol.hm  Tue Jun 20 22:49:55 2000
From: strad at scheol.hm (Coltrey Mather)
Date: Tue, 20 Jun 2000 20:49:55 -0600
Subject: User input in curses
Message-ID: <Pine.LNX.4.21.0006202020140.2681-100000@Abarrach.scheol.hm>

What is the best (or even a fairly good) way to get multi-line user input
in a python/curses application?  Is there a way to pipe input from an
external editor? perhaps a simple python+curses implementation of ed?

+---------------------------------------------------------+
| Coltrey Mather                             Scheol, Ltd. |
| Lead Engineer                           strad at scheol.hm |
| http://www.scheol.hm  finger://strad at Abarrach.scheol.hm |
+---------------------------------------------------------+





From millja at my-deja.com  Tue Jun 13 23:27:33 2000
From: millja at my-deja.com (millja at my-deja.com)
Date: Wed, 14 Jun 2000 03:27:33 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com>
Message-ID: <8i6u37$re2$1@nnrp2.deja.com>

Note: I'm reposting because those messed up
commas are ugly.

I am having a bit of an issue.  My current
situation is that I am able to use Python with
ASP and I am able to use VBScript with ASP.
However, I am still trying to figure out how to
use the two together.

What is occurring is that I am working on a
website, most of which is already written with
VBScript.  I have a rather large function that I
would like to integrate.  Is there some way to
call this function from within the VBScript?
 Or, it would work if I can get numbers,
generated by the VBScript, into the Python
function and the results of this function back to
the VBScript.

Thanks for you time.

jason milliron



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


From MarkH at ActiveState.com  Tue Jun 20 20:19:08 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 21 Jun 2000 00:19:08 GMT
Subject: Changing directory ACLs
References: <394F96B3.34E31C68@darwin.cwru.edu> <slrn8kv94e.44t.ahopkins@ahopkins.dynacare.com>
Message-ID: <0ST35.3926$qz4.30371@news-server.bigpond.net.au>

"Albert Hopkins" <ahopkins at ahopkins.dynacare.com> wrote in message
news:slrn8kv94e.44t.ahopkins at ahopkins.dynacare.com...

> Check the win32security module.  I've never used it, but it appears to
> have what you're looking for.

Yes, Im confident that win32security, plus the pywintypes module for
direct access to security objects will do what you want.

The "problem" is that Python makes no attempt to make it easier than it is
in C/C++.  There is a fair bit of magic to weave, involving a number of
steps to successfully create the ACLs, ACEs and SDs!  Getting the code
wrong can cause a few problems ;-)

There are no "nice" wrappers around this stuff simply because I personally
dont have the experience with them.  The few times I need to do security
related things I search MSDN for C sample code, and translate it to the
relevant win32security/pywintypes calls.

As an example of the verbosity required, below is some code from the book
examples <plug>Chapter 16 - Windows NT Administration</plug> that simply
creates a security descriptor ready to be applied to the necessary object.
(In fact, this is probably a reasonable percentage of what you need)

Mark.

# A utility function that creates an NT security object for a user.
def CreateUserSecurityDescriptor(userName):
    sidUser = win32security.LookupAccountName(serverName, userName)[0]
    sd = win32security.SECURITY_DESCRIPTOR()

    # Create the "well known" SID for the administrators group
    subAuths = ntsecuritycon.SECURITY_BUILTIN_DOMAIN_RID, \
               ntsecuritycon.DOMAIN_ALIAS_RID_ADMINS
    sidAdmins = win32security.SID(ntsecuritycon.SECURITY_NT_AUTHORITY,
subAuths)

    # Now set the ACL, giving user and admin full access.
    acl = win32security.ACL(128)
    acl.AddAccessAllowedAce(win32file.FILE_ALL_ACCESS, sidUser)
    acl.AddAccessAllowedAce(win32file.FILE_ALL_ACCESS, sidAdmins)

    sd.SetSecurityDescriptorDacl(1, acl, 0)
    return sd





From tismer at tismer.com  Sun Jun  4 14:34:30 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 04 Jun 2000 20:34:30 +0200
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com> <39398EEE.C18EC987@tismer.com> <3939903D.79BD8D2D@san.rr.com>
Message-ID: <393AA136.36C34915@tismer.com>


Courageous wrote:
> 
> > As per unusual, I will incorporate this into Stackless
> > Python immediately if it is written as clear as you described
> > it.
> 
> It actually *is* pretty clear, I discovered that I needed to be
> very careful with head inserts. I ended up with 3 seperate critical
> conditions (and am not sure if there might be some more clever way
> of expressing them):

God are you quick. Do you have a copy of G-man's time machine? :-)

>     if ((dlist->vfront < 1) && (dlist->len < (2*dlist->alloclen/3)))
>     {
>         DlistReposition(dlist);
>         dlist->vfront--;
>         dlist->items[dlist->vfront]=object;
>     }
>     else if (dlist->vfront < 1)
>     {
>         DlistGrow(dlist);
>         dlist->vfront--;
>         dlist->items[dlist->vfront]=object;
>     }
>     else
>     {
>         dlist->vfront--;
>         dlist->items[dlist->vfront]=object;
>     }
> 
> Condition number 1 occurs when the virtual index is underflowing
> the actual allocated memory, but there in fact is plenty of memory
> to hold the current array. Therefore it is reposition using the
> "1/3 rule".
> 
> Condition number 2 occurs when the virtual index underflows the
> allocated memory, but the memory space is too small. The allocated
> region is doubled and then repositioned using the 1/3rd rule.
> 
> Condition number 3 occurs normally. The virtual index is
> decremented and the item is inserted, cost = 1.
> 
> I'm not certain exactly, but I believe that the first two conditions
> amortize away over all inserts at the head to O(1)+k.
> 
> The rule of 2/3 availabile memory being "enough" was an entirely
> arbitrary pick. There has to be some rule, however, otherwise
> condition number two will occur on every underflow, resulting in
> growth at every insert at the head. Ran into that one by mistake.
> 
> Oopsie. :)-

Tja. You touched hardcore stuff, so problems come for free :-)

But I'm quite confident that your "thirds" approachis fine.

ciao - chris

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



From 12klat at sightreader.com  Tue Jun  6 10:20:21 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 07:20:21 -0700
Subject: Newbie question.
References: <393CE029.27B87C91@dulcesol.es>
Message-ID: <393D08A4.355C220@sightreader.com>

Hey 4 out of 4, not bad!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/6065ca0e/attachment.vcf>

From jean-louis.replacebylastname at ixiasoft.com  Tue Jun 13 18:06:53 2000
From: jean-louis.replacebylastname at ixiasoft.com (Jean-Louis Couturier)
Date: Tue, 13 Jun 2000 18:06:53 -0400
Subject: Problems with COM
Message-ID: <Wey15.3428$UM2.51434@weber.videotron.net>

Problem : I need to access an object created by an OUT variable in a
function call to a COM object.
Code example :

import win32com.client
import pythoncom

_testQueryString="""<?xml version="1.0" encoding="UTF-8"?><query
VERSION="1.0" RESULTSPACE="R1"><andkey><key
NAME="Date"><interval><start><date><year>1996</year><month>January</month><d
ay>4</day></date></start><end><date><year>1997</year><month>December</month>
<day>30</day></date></end></interval></key><key NAME="Full
Text"><elem>commission</elem></key></andkey></query>"""

def debug(msg):
    print "[DEBUG]",msg

def get_docbase_service(servername="jeanlouix", docbasename="TEXTML Demo"):
    debug("Connecting to COM server")
    clientService=win32com.client.Dispatch("TextmlServer.ClientServices")
    debug("Connecting to server "+servername)
    serverService=clientService.ConnectServer(servername)
    debug("Connecting to DocBase "+docbasename)
    docBaseService=serverService.ConnectDocBase(docbasename)
    return docBaseService

def query_server(docBaseService, queryString):
    searchService=docBaseService.SearchServices
    resultSpace=win32com.client.Dispatch
    debug("Querying... ")
    searchService.SearchDocuments(queryString,resultSpace)
    return resultSpace.Count

def main():
    docBaseService=get_docbase_service()
    print query_server(docBaseService,_testQueryString)

if __name__=="__main__":
    print "Testing pyTEXTMLServer"
    main()

My example above gives the following error on the line "return
resultSpace.Count":
AttributeError: Count
The resultSpace object is a COM object that cannot be created beforehand.
It has to be placed in a VARIANT that is fed to the function
SearchDocuments.  How do I do this?  I tried creating a Dispatch object
without specifying a type but it doesn't seem to work.

Thank you in advance for any solution or direction to relevant
documentation.
I will be happy to provide more details if I have not been precise enough.

Jean-Louis Couturier
Developer
Ixiasoft






From ge at nowhere.none  Tue Jun 27 11:58:49 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 27 Jun 2000 15:58:49 GMT
Subject: Python COM for Linux?
References: <8ja7gv$687$1@nnrp1.deja.com> <3958c185.103629691@news>
Message-ID: <Za465.1871$iN5.400266@ptah.visi.com>

"Warren Postma" <embed at NOSPAM.geocities.com> wrote:

> 2. COM on Linux is a fudge, and not a very stable one at that.
> There is a COM implementation for Linux but it's not open
> source, so if you have a problem, you can't fix it, you're
> stuck.

In article <3958c185.103629691 at news>, Reini Urban wrote:

>the german software ag has a DCOM port for linux. try that if
>you have to.

I have a Python program that fetches mail from MS Exchange
Server using MAPI/COM.  I'd dearly love to run it under Linux.
Does anybody have an experience with accessing MS Exchange
Server via COM from Linux?

Open source COM for Linux would be nice, but closed source COM
under Linux can't be much worse than running under Win32.

-- 
Grant Edwards                   grante             Yow!  NANCY!! Why is
                                  at               everything RED?!
                               visi.com            


From janne at nnets.fi  Fri Jun 16 12:59:43 2000
From: janne at nnets.fi (Janne Sinkkonen)
Date: 16 Jun 2000 19:59:43 +0300
Subject: Learning curves (was: Python so far)
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <c6xsnuefq2b.fsf@rac6.wam.umd.edu> <39498084.972E36D2@san.rr.com> <1ecb5va.1nch8nwz2t5siN%bparsia@email.unc.edu> <bhahn-ya02408000R1606000950220001@news.corp.hp.com>
Message-ID: <m3em5xr280.fsf@kinos.nnets.fi>

bhahn at spam-spam.g0-away.com (Brendan Hahn) writes:

> X is time elapsed, Y is knowledge acquired.  So the common usage of
> "steep" with regard to learning curves is in fact exactly backwards.
> Too late now, though.

Or X is the skill acquired, Y is the effort. Then the current use of
steep is meaningful: it means you have to work hard to get anything
done at all (and possibly that later you may get a lot done with
little effort).

-- 
Janne


From cut_me_out at hotmail.com  Mon Jun 26 21:12:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Jun 2000 21:12:59 -0400
Subject: Getting double* from NumPy array?
References: <ted.drain-2606001323500001@mjordan.jpl.nasa.gov>
Message-ID: <etdem5khqp0.fsf@w20-575-109.mit.edu>

> Assuming a 1-dimensional NumPy array, is there an easy way to extract a
> pointer to a double and length from the python object?  The C/C++ isn't
> going to change the object so I'd prefer not to make a copy of it.

Yeah, there is.  You probably want the numpy_array->data member.  See 

http://starship.python.net/crew/hinsen/NumPyExtensions.html

for a great overview of this sort of stuff.

Alex.


From sabren at manifestation.com  Wed Jun 21 15:39:45 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Wed, 21 Jun 2000 15:39:45 -0400 (EDT)
Subject: apply for objects
In-Reply-To: <8ib6j4$dq5@cs.vu.nl>
Message-ID: <Pine.BSF.4.10.10006211536090.1727-100000@chi.pair.com>

On Thu, 15 Jun 2000, Hahn S wrote:

> class TestHandler:
> 	def start(self, **attrs):
> 		print "%s %s" % (name, attrs)
> 
> 	def dispatch(self, style, attrs):
> 		if style == 'begin':
> 			apply(self.start, attrs)
> 
> Now this won't work, because the self.start doesn't have a __call__
> property. Any ideas how to solve this?
> 
> The **attrs in start is necessary, because there is other code calling
> this function.


I don't think it has anything to do with __call__, since start is
a method. The problem is with your use of apply. The syntax is:

apply(somefunction, arglist, [kwargs])

... so you probably mean this:

       def dispatch(self, style, **attrs):
           if style == 'begin':
               apply(self.start, (), attrs)




Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------





From aahz at netcom.com  Tue Jun 20 09:27:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jun 2000 13:27:34 GMT
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk> <8imc1f$181$1@slb6.atl.mindspring.net> <394f2d55.0@katana.legend.co.uk>
Message-ID: <8inrg6$mm2$1@slb7.atl.mindspring.net>

[BTW, please maintain attributions for quoted text; it makes it much
easier to keep track when multiple people are writing.]

In article <394f2d55.0 at katana.legend.co.uk>,
Thomas Mangin <thomasm at legend.co.uk> wrote:
>Attribution restored: Aahz
>>Attribution restored: Thomas Mangin
>>>
>>>I notice that if a directory is root owned and chmoded 700 (let say /test)
>>>and if it contain a directory called test2
>>>os.path.isdir("/test/test2") return 0 if called by a "normal user"
>>>I was expecting an exception.
>>
>> Why were you expecting an exception?
>
>I created few directories for my application as "myself" but I am running
>the script as a cgi.
>
>drwx------ thomas thomas  [somwehere in the world]/test
>drwx------ thomas thomas  [somwehere in the world]/test/test2
>
>my web server and I are different users,  as "nobody" have no read/go
>through access on the parent (test) directory I am checking, it can not even
>try to perform any action on the child one (test2).

But it's not performing any action on test or test2; it's performing an
action on the path /test/test2.  I believe it actually uses the OS call
stat(), but that doesn't really matter; the point is that whatever
internal function gets called on the *entire* path returns an error,
therefore the path cannot be a directory.

>Nothing with the current return code indicate that the test failed.
>I was just guessing : return 0 means no directory present, not enable to
>perform the requested action ..

It means that for whatever reason (permission denied or non-existence or
because it's a file), the path you gave isdir() is not a directory.  If
you want more information than that, you need to get it yourself.  More
than that, if you want to test individual path components, you have to
do that yourself, too.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From notin at myisp.com  Sat Jun  3 10:49:54 2000
From: notin at myisp.com (notin at myisp.com)
Date: 3 Jun 2000 14:49:54 GMT
Subject: Find The Real G Spot with a Naughty Tonic!
Message-ID: <VRFJ6559.G80942KO@myisp.com>

Subject: Find The Real G Spot with a little Naughty Tonic!

 
DO NOT CONSIDER PARTYING OR HAVING 
SEX UNTIL YOU HAVE VISITED THIS SITE!

Do you burn out at the party or worse 
later when your partner still has the 
energy for sex?

Want stallion like stamina coupled with 
the ability to give your partner the 
orgasm they crave and deserve?

If so follow this link NOW!

http://www.big-penis.ic24.net/partydrugs.htm







From fdrake at beopen.com  Mon Jun  5 12:09:55 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Mon, 5 Jun 2000 12:09:55 -0400 (EDT)
Subject: Help with SSL in 1.6
In-Reply-To: <39382a75$0$25922@personalnews.de.uu.net>
References: <Pine.LNX.4.21.0006021832560.13804-100000@lade.src.no>
	<39382a75$0$25922@personalnews.de.uu.net>
Message-ID: <14651.53459.943389.828213@cj42289-a.reston1.va.home.com>

Johannes Stezenbach writes:
 > configure doesn't seem to test for OpenSSL.
 > You must edit Modules/Setup manually (copy from Setup.in,
 > if necessary):

  And the SSL support doesn't appear to be documented anywhere.  If
anyone would like to document the socket.ssl() function and the SSL
objects, I'd love to see a patch to the documentation, or just plain
text for the new material.
  If anyone's interested, please drop me a note!


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From senn at maya.com  Fri Jun 16 09:01:59 2000
From: senn at maya.com (Jeff Senn)
Date: 16 Jun 2000 09:01:59 -0400
Subject: How to prevent illegal definition of a variable in objects?
In-Reply-To: Roland Schlenker's message of "Thu, 15 Jun 2000 12:05:48 +0000"
References: <8i8ujt$hop$1@news2.isis.unc.edu>
	<3948C69C.7D37E1BF@attglobal.net>
Message-ID: <1z1x93ug.fsf@SNIPE.maya.com>

Roland Schlenker <rol9999 at attglobal.net> writes:
> How about this;
> 
> >>> class A:
> ...     def __init__(self):
> ...         self.x = 1
> ...     def __setattr__(self, name, value):
> ...         if name == "x":
> ...             self.__dict__[name] = value
> ...         else:
> ...             raise AttributeError

Or if (as I suspect) the goal is to "protect" the object from getting
new, different attributes after it has been initialized (e.g. from
outside the "class") you could do something like this:

class A:
     def __init__(self):
         self.x = 1
         self.foo = 'bar'
         ...other initialization...
         self._protect = 1           # <-- always last line of __init__
     def __setattr__(self, n, v):
         if not self.__dict__.has_key('_protect') or self.__dict__.has_key(n):
             self.__dict__[n] = v
         else:
             raise error("Protected Object") #or whatever

...a way to prevent someone from deleting attributes left as an
exercise to the reader... :-)

-- 
-Jas




From olczyk at interaccess.com  Tue Jun 20 19:32:34 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Tue, 20 Jun 2000 23:32:34 GMT
Subject: Removing items from a list in a loop
Message-ID: <3952fdf6.950540640@nntp.interaccess.com>

I would like to remove a bunch of items from a list *efficiently*.
Essentially  I would like to do this:

   for x in list:
        if Function(x):
	list.remove(x)

This does not work. Presumably for this reason:

list before removal
x1 x2 x3 x4 x5 

asume that both x2 and x3 need to be removed.

during removal
x1 x2 x3 x4 x5
     ^==x in loop

when removal done
x1 x3 x4 x5
     ^
next loop
x1 x3 x4 x5
          ^ skipping x3

Is there a safe way to iterate through a list and remove some items?



From mfein at aplcomm.jhuapl.edu  Tue Jun 20 15:47:09 2000
From: mfein at aplcomm.jhuapl.edu (Matt Feinstein)
Date: Tue, 20 Jun 2000 19:47:09 GMT
Subject: Computer Algebra System
References: <8inpa4$8iv$1@vishnu.jussieu.fr> <394FC515.972FAE0F@nih.gov>
Message-ID: <394fc892.31377969@aplnews>

On Tue, 20 Jun 2000 15:25:09 -0400, Jonathan Epstein
<Jonathan_Epstein at nih.gov> wrote:

>Matlab has an interface to Maple (their "Symbolic Toolbox", I think).
>
>It might be useful to study their API, and emulate it in a reasonable
>(and hopefully legal) fashion in Python.
>
>-Jonathan
>

It's pretty simple, at least from the user's point of view. You
declare your symbolic variables a la

sym x y z;

and then just go ahead and use the symbolic variables in the
interpreter in algebraic or arithmetic expressions the same way you
would use numerical expressions-- all the arithmetic and algebraic
operators used by the interpreter are appropriately overloaded. There
is an accompanying suite of functions and state variables that control
how the symbolic stuff is done. There is also an 'Extended' symbolic
toolbox that gives the user direct access to the Maple kernel.

--
Matt Feinstein
mfein at aplcomm.jhuapl.edu
Organizational Department of Repeated
and Unnecessary Redundancy


From m.faassen at vet.uu.nl  Thu Jun 15 10:00:17 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 14:00:17 GMT
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <8hs4d8$33b$1@animus.fel.iae.nl> <20000610120811.H20700@xs4all.nl>
Message-ID: <8ianhh$dro$3@newshost.accu.uu.nl>

Thomas Wouters <thomas at xs4all.net> wrote:
[discussion on naming]

Since there's a discussion about naming anyway, I'd like to ask what
the 'ad' stands for in __add_ad__? This may be another datapoint
in the discussion. :)

Regards,

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


From db3l at fitlinxx.com  Fri Jun 16 17:16:50 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 16 Jun 2000 17:16:50 -0400
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com>
Message-ID: <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> Oops!  You're right, as several people also pointed out to me in email.
> Actually I had this in .emacs and didn't realize it for years:
> 
> (setq-default tab-width 4)
> 
> I am not aware of serious adverse effect, though.

Eeek - that's evil.  Don't do that :-)

A TAB character has a perfectly good definition, and artifically
mucking with that definition just within the environment of your
editor is going to pretty much ensure that any other application that
expects the normal behavior will present or parse your files
differently.  Besides, Emacs makes it so easy to maintain indentation
at any level without changing the basic tab size.

>                                                    The good thing is that
> changing indentation is really easy.  You can use one keystroke backspace or
> C-d, for example.

Well, for a single line that's true, although of course if you set up
your language mode to your preferred indentation style, then just
hitting TAB anywhere on the line will re-indent that line according to
your preferences.

But for multiple lines you won't want to be explicitly deleting the
characters on each line, and once you have a region, adjusting the
indentation of the region (C-X TAB) is simple.

But to keep this posting slightly on topic for the group, I suppose
this is as good an indication as any that depending on at least the
presentation of TABs in code - perhaps except that used only locally -
is not the best choice.

> C-d, for example.  Presumably the evil will manifest itself when dealing
> with other programs, but I also set 'less -x4' so I don't even see it.

But as you say, you had to explicitly "reprogram" less for that behavior.

> Wondering what other people will see when reading my programs ...

They'll see a TAB character expand to get to the nearest 8-character
boundary.  At a minimum, if you must use a TAB this way within files
you work on, you should untabify them (or re-tabify them to standard
TAB widths) before any external distribution because assuming the raw
TAB character indents at multiples of 4 is definitely not going to be
the default assumption just about anywhere else, and if nothing else
it'll look different than you wrote it and perhaps not as nice (line
wrapping and all).

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


From thomas at xs4all.nl  Mon Jun 26 16:06:54 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 20:06:54 GMT
Subject: socket-module
References: <eOU#4853$GA.336@net003s>
Message-ID: <slrn8lfduh.k73.thomas@centurion.xs4all.nl>

On Mon, 26 Jun 2000 18:27:06 +0200, A[r]TA <arta at NOSPAMx-stream.nl> wrote:

>I'm having a question. Maybe a strange one, but don't blame me. Just
>a newbie or something like that :-)

>If you're looking at the source from SOCKET.PY, you see a
>two classes. One for the socket-object and one for the file-descriptor.
>But, this module (in Windows) doesn't import anything.
>And par example:

>def accept(self):
>    sock, addr = self._sock.accept()
>    return _socketobject(sock), addr

>This is the accept method. Where can I see the lowest-level-Python-code?
>I can't see the real code from HOW it's accepting the connection.
>Is this possible and how? It's basically for me to see the TCP/IP-packages
>Python is creating....(very low level)

I believe socket.py is only used for operating systems that have a slightly
different way of doing networking, like Windows and BeOS. If you look at the
top of socket.py, you'll see something like this:

try:
        from _socket import *
except ImportError:
        from socket import *

_realsocketcall = socket

Meaning that the real low-level API is pulled from either the _socket or the
socket module, which is likely a loadable library (a DLL, under windows) or
compiled-in module.

If you want to see the real gory details about the socket library, you'll
have to delve into the C source ;-P I think the socketmodule for Windows is
the normal socketmodule, with slightly different #defines. You can find
socketmodule.c in the Modules subdirectory of the source tree.

Good-luck-ly y'rs,
	Thomas.



From msoulier at nortelnetworks.com  Wed Jun  7 11:52:28 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 7 Jun 2000 15:52:28 GMT
Subject: popen in real time?
References: <8hjdn4$6kf$1@bmerhc5e.ca.nortel.com> <jtln0iy7x2.fsf@wazor.biostat.wisc.edu>
Message-ID: <8hlr3s$cfv$1@bmerhc5e.ca.nortel.com>

In article <jtln0iy7x2.fsf at wazor.biostat.wisc.edu>, William Annis wrote:
>
>        I've found using the Popen3 class in popen2 to be very helpful
>in these cases... turn off the buffering by setting the bufsize option
>to zero.

	Thanks, I'll try it out.

	Mike

-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From aahz at netcom.com  Thu Jun 15 16:58:50 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 20:58:50 GMT
Subject: Regular Expressions and Threads?
References: <3947D2F7.264BDC6E@uswest.net> <39481F1C.419364F5@uswest.net> <8i979n$4ss$1@slb6.atl.mindspring.net> <3dd7li696x.fsf@amarok.cnri.reston.va.us>
Message-ID: <8ibg2a$t1q$1@slb2.atl.mindspring.net>

In article <3dd7li696x.fsf at amarok.cnri.reston.va.us>,
Andrew M. Kuchling <akuchlin at mems-exchange.org> wrote:
>aahz at netcom.com (Aahz Maruch) writes:
>>
>> checking for the existence of a match, that could cause problems.  (As
>> with anything else, the re module should be thread-safe, but individual
>> re objects probably are not.)
>
>No, individual re objects should be.  The MatchObject you get back
>from a matching operation contains references to the string, groups,
>and original regex object, but a RegexObject shouldn't ever change
>after it's been created.  Tim Peters put some effort into making the
>cache thread-safe.  Brian, can you post the relevant bits of code so
>we can see what's going on?

Oh, right, for some reason I thought the regex object contained the
match object (Perl-based brainfart).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From pete at visionart.com  Tue Jun 20 19:44:22 2000
From: pete at visionart.com (Pete Shinners)
Date: Tue, 20 Jun 2000 16:44:22 -0700
Subject: speedup numpy ufuncs
Message-ID: <395001D6.BD6D9728@visionart.com>

i've been doing some simplish image processing
stuff with numpy and a thought occured to me about
a possible speedup.

if i have 4 or 5 arrays i want to add together,
the fastest way i've found to do this is by
calling the add ufunc many times, specifying an
output argument.
(this gets rid of extra temporary arrays)

array1 = array2 = array3 = array4 = arange(100)
add(array1, array2, array1)
add(array1, array3, array1)
add(array1, array4, array1)

i was wondering if there was a way to add these all
in one pass?
addmany(array1, array2, array3, array4, array1)

would this even make a difference? it just seems
internally numpy could loop over all the input arrays
in one shot, instead of looping for each addition pair.

(this all assumes that
 array1 = array1+array2+array3+array4 is creating
 several many temporaries. after my intial testing
 it seems this is the case)

anyways, it seems this would be a simple addition to
make to numpy (i believe i could crank it out myself)
but i wonder what the insight is on this. would there
really be much benefit by stripping of the excess 
looping?)


along these lines... what is the absolute fastest
quickest way to compute the average of many arrays?
i know (array1+array2+array3)/3 will not be it, but
is there a common "turbo" method for this?


From ben at co.and.co  Wed Jun  7 06:38:39 2000
From: ben at co.and.co (ben at co.and.co)
Date: Wed, 07 Jun 2000 10:38:39 GMT
Subject: OT: nederlands (Re: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr>
Message-ID: <PCp%4.109780$Qp2.291904@afrodite.telenet-ops.be>

Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
 
> Er zijn dus merdere rationale verklaringen, dus geen behoefte om
              ^^^^^^^
  Toch nog een beetje een franstalige tongval :-).

> aan kosmische krachten te geloven!

Groeten,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From ge at nowhere.none  Wed Jun 14 13:02:53 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 14 Jun 2000 17:02:53 GMT
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com> <8i86sf$e7c$0@216.39.141.36>
Message-ID: <1VO15.607$i06.193222@ptah.visi.com>

In article <8i86sf$e7c$0 at 216.39.141.36>, Dana Booth wrote:

>GE: I've got a Tkinter-based GUI interface used to do
>GE: administrative stuff for a device driver under Linux.  I know
>GE: RedHat installs Python and Tkinter by default.  Is there a
>GE: handy list somewhere of which Linux distributions can be
>GE: expected to have Tkinter installed?
>
>It's best to choose not to put on Python at install time, and then just
>retrieve the latest version from the Internet. Uncomment the lines in the
>Modules/Setup file before you build it pertaining to Tkinter, then you're
>assured of having the latest version.

I've got Python and Tkinter installed, that's not the problem.

I'm going to distribute a utility written in Python/Tkinter as
a configuration/administration utility that goes with a Linux
device driver.  I was just wondering how many customers are
likely to have Python/Tkinter available.  I doubt that anybody
is going to install Python just to run this program.

-- 
Grant Edwards                   grante             Yow!  ALFRED JARRY! Say
                                  at               something about th' DEATH
                               visi.com            of DISCO!!


From pduffin at hursley.ibm.com  Fri Jun 30 06:15:01 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 30 Jun 2000 11:15:01 +0100
Subject: Controlling an external program with Python
References: <1542456BE0832BEE.8A959B400FAE5129.26A3D7DF9AA9B305@lp.airnews.net> <nz855.1583$iN5.265679@ptah.visi.com> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com> <8jg3hb$iic$1@nntp6.u.washington.edu>
Message-ID: <395C7325.20A26E19@hursley.ibm.com>

Donn Cave wrote:
> 
> Quoth claird at starbase.neosoft.com (Cameron Laird):
> | In article <395B32D1.B926FDB3 at hursley.ibm.com>,
> | Paul Duffin  <pduffin at hursley.ibm.com> wrote:
> ...
> |> This sort of arrangement is only possible when people who are interested in
> |> Expect functionality can transcend any language *issues*, by accepting that
> |> the language is basically irrelevant apart from the effect in may have on
> |> the required functionality and work together.
> ...
> | I believe I've heard Don express surprise at Expect's
> | career.  He thought people would steal the ideas for
> | other (languages') libraries.  What he least expected
> | (sic), if I understand it correctly, was that the whole
> | world would be content to invoke Expect as the standalone
> | executable which we know utterly dominates its use.
> 
> Well, I'm surprised he's surprised!  Expect-Tcl works out
> so well that it more or less cancels out the ``wrong language''
> problem.  Mainly because Tcl (as I suppose Lisp/Scheme) can
> support an "expect" statement in a way that doesn't fit in
> Python (that I can think of - this is not a comment on the
> existing Python expects, with which I am not familiar.)
> 
> And then I have looked at the source.  I'm sure it has no peer
> when it comes to comprehensive cross-platform pseudotty support,
> but a lot of barnacles come with that.  I would steal a younger
> horse than this.
> 

You do like mixing metaphors, first it is nautical, then it is criminal,
anyone would think you were a pirate ;-).

The trouble is that any abstraction which does not deal with all those
barnacles will have limited use. There are many people out there using
Expect on older platforms who would not be pleased to lose support.


From thomas at cintra.no  Wed Jun  7 20:25:51 2000
From: thomas at cintra.no (thomas at cintra.no)
Date: Thu, 08 Jun 2000 00:25:51 GMT
Subject: Why not a "Deluxe" Python-distribution
References: <Pine.LNX.4.21.0006071336230.23668-100000@fep132.fep.ru> <sjt0l9h42t519@corp.supernews.com>
Message-ID: <393ed754.3036265@news.c2i.net>

Hmmm ... paying money. That`ll throw some people off. BUT, one
solution could be to put stuff like PostgreSQL, Apache, Zope, lots of
docs, code, examples, modules AND ( best of all ) a Linux-distribution
on the CD. That way it would be a complete development solution, and
the added value would probably make people more willing to spend a
couple of bucks on it. 

http://www.thekompany.com/home/ has a package in development called
PowerPlant which is more along the lines of what I described above. A
solution more aimed at Python alone would be even better.

The bottom line is that a solution for which user has to pay has to
something more that just what they could download. Most people can
afford a 5-10MB download, and that would cover lots, LOTS of
python-modules. Putting it on a cd wouldn`t be necessary. 

Am I making sense? I just think this would really benefit the
acceptance of python. If anybody has pointers on how we could do this
we could get started. What is needed? Would we have to compile a
version of the package for each and every Linux-distro,
Windows-version, BeOS-version (?) etc. A significant part of it is
platform-independant, so that shouldn`t be a problem. The rest, like
database-drivers, modules available in source etc. that`s another
matter. Ideas? 

The first thing could at least be to gather all platform-independant
stuff and compile a package of that, and take the rest from there. 

Thomas


On Wed, 7 Jun 2000 13:17:09 -0400, "Jason Cunliffe"
<jasonic at nomadicsltd.com> wrote:

>Yes, thanks for http://www.pythonware.com/downloads.htm#py152
>
>A Deluxe pre-compiled Python is an excellent idea.
>In a pragmatic sense,  I honestly think this is what step#1 of CP4E must
>be..
>
>Just give people one large super useful download without headaches which
>will let them do lots of stuff right from the start. Many people who would
>rather have a big initial download and then be able explore the abundance
>and power without confusion. Of course full docs and examples are a good
>idea too.
>For those with decent bandwidth [DSL etc], the size is not an issue.
>For those who do not have bandwidth they other options - for example I would
>be quite happy to make DeluxePython CD's for people as long as they cover
>the modest cost of materials, postage, and doingit.
>
>-Jason
>
>________________________________________________________________
>Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director
>
>Oleg Broytmann <phd at phd.russ.ru> wrote in message
>news:Pine.LNX.4.21.0006071336230.23668-100000 at fep132.fep.ru...
>> On Wed, 7 Jun 2000, Thomas Weholt wrote:
>> > Wouldn`t it be nice to have a deluxe edition of Python, with lots of
>> > the adopted modules not in the standard distro? Like PIL, PST, code
>> > from the sigs, database drivers etc.
>>
>>    You are absolutely right! To some extent, it is here:
>> http://www.pythonware.com/downloads.htm#py152
>>
>>    But it's certainly incomplete. Well, in the world of free / opensource
>> software it is simple: just do it yourself!
>>    Compile whatever you want, put it on your site, relax... and wait for
>> complains :)
>>
>> Oleg.            (All opinions are mine and not of my employer)
>> ----
>>     Oleg Broytmann      Foundation for Effective Policies
>phd at phd.russ.ru
>>            Programmers don't die, they just GOSUB without RETURN.
>>
>>
>
>



From md9tim at mdstud.chalmers.se  Mon Jun  5 09:23:06 2000
From: md9tim at mdstud.chalmers.se (=?ISO-8859-1?Q?Tim_Gahnstr=F6m?=)
Date: Mon, 5 Jun 2000 15:23:06 +0200
Subject: the windows interpreter
Message-ID: <Pine.SOL.4.21.0006051507510.9521-100000@scooter.mdstud.chalmers.se>

I downloaded and installed the python 1.5 for windows interpreter.
I can use some pyton comands at the prompt like

>>> print "Hi"
will causet the output 
Hi
I can also create simple files with idle and run them with F5.
But when I try to use an example program like

----
#! /usr/local/bin/python
import sys
if '-h' in sys.argv or '--help' in 
sys.argv or '--help' in sys.argv: print '''
help.py--does nothing useful (yet)
options: -h, -help, or --help-display this help
Copyright (c) Jacek Artymiak, 2000 '''
    sys.exit(0)
else:
    print 'I don't recognize this option' sys.exit(0)
---

Then it will not work at al, not even if I remove 
the first "linux inspired" line
I get the errormessage

----
Traceback (innermost last):
  File "F:\pyton\Python\Tools\idle\ScriptBinding.py", line 131, in
run_module_event
    execfile(filename, mod.__dict__)
  File "F:\pyton\Python\test.py", line 3
     if '-h' in sys.argv or '--help' in
                                        ^
 SyntaxError: invalid syntax
---


I would apreciate if anyone had a suggestion.

Tim Gahnstr?m



From timr at probo.com  Sat Jun 24 01:22:03 2000
From: timr at probo.com (Tim Roberts)
Date: Fri, 23 Jun 2000 22:22:03 -0700
Subject: Carrage Returns and Tabs in Windows Applications
References: <8ir53n$l5n$1@nnrp1.deja.com> <paf45.4867$qz4.36531@news-server.bigpond.net.au> <slrn8l36j8.b3d.quinn@pyro.ugcs.caltech.edu> <8it6p6$3b6$1@nnrp1.deja.com>
Message-ID: <j9h8ls8njenmrp0ca8jbjtcrb6l5n2aric@4ax.com>

jelizondo at my-deja.com wrote:
>
>Actually I may have worded my original message a little bit wrong.
>Basicly what I'm wondering is if there is a way to automate a series of
>commands in a windows application using python. For example.
>
>os.system(<program> <file>)
>
>then using python to automate a series of commands, sort of like batch
>and/or script file.

If you are trying to automate Windows applications, you have a hard row to
hoe.  However, it can be done.  There is enough of the Windows API in
win32all that you could use SendMessage to send WM_CHAR messages to an
application.  It's a pain, but it's possible.

Alternatively, if it is a popular commercial application, it may even have
a COM interface.  COM interfaces can be controlled from Python without too
much trouble.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From jkraska1 at san.rr.com  Sun Jun  4 15:18:03 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 19:18:03 GMT
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com> <39398EEE.C18EC987@tismer.com> <3939903D.79BD8D2D@san.rr.com> <393AA136.36C34915@tismer.com>
Message-ID: <393AAC0F.FFB4B65@san.rr.com>

> >     if ((dlist->vfront < 1) && (dlist->len < (2*dlist->alloclen/3)))
> >     {
> >         DlistReposition(dlist);
> >         dlist->vfront--;
> >         dlist->items[dlist->vfront]=object;
> >     }
> >     else if (dlist->vfront < 1)
> >     {
> >         DlistGrow(dlist);
> >         dlist->vfront--;
> >         dlist->items[dlist->vfront]=object;
> >     }
> >     else
> >     {
> >         dlist->vfront--;
> >         dlist->items[dlist->vfront]=object;
> >     }

Klaus Baldermann correctly pointed out that the following code is
both more clear/succinct and slightly more efficient:

    if (dlist->vfront < 1)
    {
        if (dlist->len < 2*dlist->alloclen/3) DlistReposition(dlist);
        else                                  DlistGrow(dlist);
    }
    dlist->vfront--;
    dlist->items[dlist->vfront]=object;
    dlist->len++;

I made substantial progress last night. Left to do is slice assignment,
sorting, creation of the immutable type to prevent modifaction during
sort, and slightly reimplementing the functions to match the same
available profiles of listobject.c methods.

I implemented concatenate to handle either dlist/dlist combinations
or dlist/list combinations, which is a nice feature, because it
allows you to fold a python list produced by third party software
into an empty dlist; it would also be possible to allow my deque
implementation to be concatenated in as well, but I didn't do it,
because I didn't want to force the module dependency.

Is there any regression testing software for python lists anywhere?
It's getting a bit hairy to verify everything now, and I'm
writing more and more of my own regression testing code.




C/


From fplancke at my-deja.com  Mon Jun  5 14:25:50 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Mon, 05 Jun 2000 18:25:50 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com> <oqvgzo9npk.fsf@titan.progiciels-bpi.ca>
Message-ID: <8hgrao$dkd$1@nnrp1.deja.com>

In article <oqvgzo9npk.fsf at titan.progiciels-bpi.ca>,
  =?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> wrote:
> Courageous <jkraska1 at san.rr.com> writes:
>
> > Well, yeah, but [0:3] as "element zero through three, inclusive"
>
> Read 0 up to 3, but not including it.
> That notation is very, very handsome
> and useful.  It is one of the jewels of Python.
> You did not see it yet?
> I fear you still write too much C, and too little Python :-) :-).

Too much C ? I don't think so, if you consider the usual C idiom:

for (i = 0; i < 3; ++i)
   add_to_my_slice(a[i])

I think C itself may have inspired the design decision (but I agree
with other people's arguments in favour of it).
Just my 0.02e worth...

--
Frederic van der Plancke
e-mail address: reverse and add dots etc. to:
      <com writeme vdplancke frederic>


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


From tom-main at REMOVEME.home.com  Fri Jun 16 11:14:59 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Fri, 16 Jun 2000 15:14:59 GMT
Subject: Can't find win32all-132
Message-ID: <Tvr25.13782$7o1.355124@news2.rdc1.on.home.com>

A week or two ago Mark Hammond announced that win32all-132 (based on 1.5.2)
was now available to all.  But I can't find it.  Anywhere.

Anyone else?

Tom.




From kens at sightreader.com  Thu Jun 15 22:56:33 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 19:56:33 -0700
Subject: more on python and GUI speed
References: <jGl25.2323$3E6.42734@news1.alsv1.occa.home.com>
Message-ID: <39499761.26C9403@sightreader.com>

Daniel Kramer wrote:

> I know I am beating this subject to death but I moved over to python from
> tcl to improve my programming skills (which are minimal), learn oop, and
> still write gui programs with ease.
>
> I have found the gui solutions to be a bit slow compared to tcl/tk.  Today I
> found something interesting.. I put my computer under heavy load and ran a
> simple tcl/tk app, python/tkinter app, and a wxPython app.. all were simple
> drawing a line on a canvas type apps.
>
>     The tkinter and wxPython apps were very choppy when the cpu was under
> load from another process.. where the tcl/tk app saw almost no performance
> hit at all.. it was still just as fluid as when my computer's cpu was idle.
>
> Anyway.. I am leaning toward going back to tcl/tk and just learning [Incr
> Tcl] to fill the oop gap.
>
> -Daniel

Can you post the wxPython code in question?  There are ways to inadvertently
make a wxPython app run slow.  Your results are surprising to me because I would
expect the line draw operation to be the bottleneck rather than the python
interpreter.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/b576cae6/attachment.vcf>

From mwelch at incigna.com  Thu Jun 29 12:18:08 2000
From: mwelch at incigna.com (Mike Welch)
Date: Thu, 29 Jun 2000 12:18:08 -0400
Subject: Text in Tables
Message-ID: <IBEALDELHCKACLINDGEKCECLCAAA.mwelch@incigna.com>

Hello, I'm a software developer trying to learn Python to generate
PDF files from ASP pages. One problem I'm having with tables in my
PDF files are that cells do not word wrap. This would not be a
problem if I knew how to put a line break (or new line) in my text.
In the docs it says \n works for new lines, but this does not work in
tables. Can someone please tell me how this is accomplished in
tables?

Thanks!

_________________________________________________
Michael B. Welch
Software Developer
Incigna, Inc. <http://www.incigna.com/>
424 East 4th Street
Cincinnati, Ohio 45202
E-mail: mwelch at incigna.com
Office: (513) 333-0313 Ext. 14
Fax: (513) 333-0413

"Your eSolutions Company"







From andres at corrada.com  Wed Jun 28 10:51:08 2000
From: andres at corrada.com (andres at corrada.com)
Date: Wed, 28 Jun 2000 10:51:08 -0400
Subject: Lista de correo sobre Python
In-Reply-To: <2B1262E83448D211AE4B00A0C9D61B0301301248@msgeuro1.creo.be>; from Pieter Claerhout on Tue, Jun 27, 2000 at 04:16:09PM +0200
References: <2B1262E83448D211AE4B00A0C9D61B0301301248@msgeuro1.creo.be>
Message-ID: <20000628105108.B29827@corrada.com>

On Tue, Jun 27, 2000 at 04:16:09PM +0200, Pieter Claerhout wrote:
> Can somebody help? I forgot my Spanish glasses...
> 
> 
> Pieter
> 
> -----Original Message-----
> From: andres at corrada.com [mailto:andres at corrada.com]
> Sent: Tuesday, June 27, 2000 1:44 PM
> To: JD
> Cc: python-list at python.org
> Subject: Re: Lista de correo sobre Python
> 
> 
> On Mon, Jun 26, 2000 at 09:47:13AM +0000, JD wrote:
> > Un foro sobre Python en:
> > python_dev-subscribe at egroups.com
> > 
> > Muchos usuarios dispuestos a resolver tus dudas, a ayudarte y deseando que
> > compartas tus conocimintos.
> > 
> 
> ?Cual es la justificacion para comenzar una lista nueva en espagnol cuando
> ya existe una?
> 

I was asking what is the justification for creating a new Spanish mailing list
in egroups when an older and already established Spanish list already exists
hosted by Pythoneers in Argentina.

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From db3l at fitlinxx.com  Tue Jun 13 15:53:48 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Jun 2000 15:53:48 -0400
Subject: How do you FTP ASCII if ftplib strips CR?
References: <8i5jhd+h2lu@eGroups.com>
Message-ID: <uog55mk6r.fsf@ctwd0143.fitlinxx.com>

al at servana.com writes:

> If you look at the FTPLIB docs, it states that RETRLINES will strip 
> CR when passing the text to the RETRLINES callback.

Actually, it strips the end of line entirely (which is CRLF in ASCII
FTP transfers).  Presumably that's to simplify the processing of the
lines, since you already know that each entry in the list (or each
string supplied to the callback) is a "line" from the transfer.

Within the FTP stream, every line of an ASCII transfer will always end
with CRLF (by specification - see below).

> So, how do you put the CR back in if you want them?

You could just add it whenever you need.  Note that if you want to
store the ASCII file locally using the proper line ending for your
platform, the most portable approach is probably to use os.linesep as
the separator (although I think using "\n" in a string and letting the
underlying C library perform the platform translation would probably
work just about as portably).

> When you FTP ascii from a non-similar platform, the loss of the CR 
> causes the text to FTP as 'one line'. (like UNIX-to-UNIX and NT-to-NT 
> work, UNIX-to-NT does not)

This has to depend on what you are doing with the retrieved lines.
The retrlines() method will return a list of individual lines, so they
can't be considered 'one line' unless you dump them in binary format
to some file or location that expected CRLF line endings.

(I wrote this before getting to the "write" call you use below, which
is why you get the raw data in the file)

> Yet ALL FTP tools do not do this, so only the Python scripts tend to 
> munge ASCII files.

Well, in reality, all FTP tools do really perform some end of line
processing for ASCII transfers - although you normally don't see it
because it's beneath the covers.

By definition, ASCII transfer in FTP is performed using a particular
NVT-ASCII representation (defined in the Telnet specification).  The
FTP RFC (RFC959) actually requires that all end of lines within an
ASCII transfer must be represented by CRLF.  That means that if you
are retrieving a file in ASCII mode from a Unix platform, the FTP
server on that platform actually _adds_ a CR to the ending of every
line before transferring it over the network.

Of course with the other tools you mention, you are probably just
retrieving a file to a local copy, so all you see is the final local
copy written to disk, at which point the FTP tool, knowing it's an
ASCII transfer, will have already applied the right end of line
meaning for the local platform.  But any FTP tool has to understand
how to potentially translate a local line ending format to the network
format required for the transfer.

> I was using the following notation, but cannot come up with another 
> where I can control the append of a CR:
> 
> ourFTP.retrlines('RETR ' + filename, open(os.path.join(localdir, 
> filename), 'w').write)

Your problem is using the "write" method on a file, which just writes
a buffer of data without any consideration for text "lines".  As an
aside, note that by doing the open inside the retrlines() call
that you lost access to the open object, and thus can't close() it.
It's possible that some data may not get flushed to the file until
your process exits.  I'd probably open the file separately and close
it after the retrieval.

What I would suggest instead is defining a small class to handle
access to your output file that takes each line string and writes it
to your file as a textual line, including line ending - something
like:

    class AsciiFile:
	import os

        def __init__(self,name=None,mode=''):
            if name:
                self.open(name,mode+'b')
            else:
                self.file = None

        def open(self,name,mode):
            self.file = open(name,mode+'b')

        def writeline(self,line):
            self.file.write(line)
            self.file.write(os.linesep)

        def close(self):
            self.file.close()
            self.file = None

and then code like:

    output = AsciiFile().open(os.path.join(localdir,filename),'w')
    outFTP.retrlines('RETR ' + filename,output.writeline)
    output.close()

should do what you want.  Note that I've chosen to have the class
actually process the file in binary and add the full line separator.
This should be portable to all platforms.  I believe you could also
just open the file in text mode and always append the string '\n' to
each line and it'll translate to the appropriate text line ending on
each platform, but I'm only positive that's going to work on Unix and
Windows.

Of course, none of this would be necessary if somewhere along the
lines each of the various systems had agreed on just how a textual
line should be terminated, but this is what we're stuck with.

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


From tom-main at REMOVEME.home.com  Sat Jun 17 09:41:11 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Sat, 17 Jun 2000 09:41:11 -0400
Subject: Can't find win32all-132
References: <Tvr25.13782$7o1.355124@news2.rdc1.on.home.com> <8ifann$pte$1@phys-ma.sol.co.uk>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E74@mail.sauder.com>

Thanks.

Tom.

"John Parkey" <jparkey at nospamrapidbs.com> wrote in message
news:8ifann$pte$1 at phys-ma.sol.co.uk...
> Tom,
>
> go to
>
> http://starship.python.net/crew/mhammond/downloads/win32all-132.exe
>
> jp
>
> "Tom" <tom-main at REMOVEME.home.com> wrote in message
> news:Tvr25.13782$7o1.355124 at news2.rdc1.on.home.com...
> > A week or two ago Mark Hammond announced that win32all-132 (based on
> 1.5.2)
> > was now available to all.  But I can't find it.  Anywhere.
> >
> > Anyone else?
> >
> > Tom.
> >
> >
>
>




From rbill at digisprings.com  Thu Jun  8 17:23:52 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Thu, 8 Jun 2000 15:23:52 -0600 (CST)
Subject: setblocking in Win32
In-Reply-To: <8homp6$a47$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.10.10006081332270.11627-100000@localhost.localdomain>

HI,

I have a suspicion on this, but please forgive the guess if I am wrong.

I think non-blocking means that an exception is raised immediately if the
socket cannot act on the call at that very moment.  In other words, if you 
say sock.accept(), and there is not an immediate connection, the
EWOULDBLOCK exception is raised so that the socket does not block
anything.  I think the EWOULDBLOCK is error # 10035 on WinNT- the error
code you are getting.  This means that what you have done so far is good
news- the socket is doing what it is supposed to do.

The trick is to pass through the exception, possibly like thus:

import socket
from errno import EALREADY, EINPROGRESS, EWOULDBLOCK, ECONNRESET, ENOTCONN

try: 
    conn, addr = sock.accept()
    print "connection received"
    handle_connect(conn, addr) # some method you've made
except socket.error, e:
    if why[0] == EWOULDBLOCK:
        print "socket would have blocked and I couldn't wait"
    else:  # in case it is a different error situation
        raise socket.error, e
        sock.close()

The next catch is that you now must loop through this to catch a
connection, so it is good to add a select to be nicer to the cpu
i.e.:

import socket
import select
from errno import EWOULDBLOCK, etc....

sock = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
sock.setblocking(0)
sock.bind("", 8002)
 
while 1:
    select.select([sock],[sock],[sock],2) # 2 is the arbitrary timeout
    try: 
        conn, addr = sock.accept()
        print "connection received", conn
        handle_connect(conn, addr)
    except socket.error, e:
        if why[0] == EWOULDBLOCK:
            print "would block, so proceeding after interupting socket"
        else:
            raise socket.error, e
            sock.close()

I hope this helps.

Robert

On Thu, 8 Jun 2000 jkingry at my-deja.com wrote:

> Hi, I think I'm going to be told what I don't what to hear, but from
> the error below, I think I have surmised that trying to set non-
> blocking on a socket in a Win32 environment doesn't work.
> 
> Are there any work arounds?
> 
> Joe
> 
> Traceback (innermost last):
>   File "D:\Prog\Python\Lib\threading.py", line 376, in __bootstrap
>     self.run()
>   File "D:\Prog\Python\Lib\threading.py", line 364, in run
>     apply(self.__target, self.__args, self.__kwargs)
>   File "io.py", line 80, in run
>     sock,(address,port) = server.accept()
>   File "D:\Prog\Python\Lib\plat-win\socket.py", line 36, in accept
>     sock, addr = self._sock.accept()
> error: (10035, 'winsock error')
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> 




From jwbnews at scandaroon.com  Fri Jun  9 20:32:49 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 09 Jun 2000 17:32:49 -0700
Subject: WIDESPREAD INCOMPETENCE AT BELL ATLANTIC
References: <B60FEC7EB.pduegowihyk@ooobpqvldjd.com> <394188BF.B39F1255@san.rr.com>
Message-ID: <jwbnews-F0008B.17324909062000@news.olympus.net>

In article <394188BF.B39F1255 at san.rr.com>, Courageous 
<jkraska1 at san.rr.com> wrote:

> > Perhaps the US Government should take a look at Bell Atlantic in
> > much the same way it's currently looking at Microsoft, with a similar
> > remedy in mind.
> 
> Bell Atlantic just Acquired GTE and doubled in size. It appears
> the FTC disagrees....

Indeed...now the customers will have to find a replacement for the "Hell 
Atlantic" term of endearment with which many refer to Bell Atlantic.

  --John (writing from "US Worst" land)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From tismer at tismer.com  Sun Jun  4 14:45:58 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 04 Jun 2000 20:45:58 +0200
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com>
Message-ID: <393AA3E6.253C88D1@tismer.com>


David Broadwell wrote:
> 
> > Not that this exactly belongs in a python ng but I'm told the
> > only usable onboard programming environment is quartus, an ansi
> > forth system by Neal Bridges.
> 
> Actually forth is the bios level language of the chip, before palm os
> loads. Just like a sparc, only smaller. Nope it dosen't really belong
> here, but is an option for the palm people. Still given the option of
> python of forth, i'll take python. (i used to know forth)

I'm supporting the Python port to Palm, of course.
In the long term, and if I can find the time, I'd
really love to rewrite Python completely in Forth.
That would really fit the machine, and it would
introduce Python in completely new areas.
One idea is to adopt the OTA standard which
you can find at http://www.europay.com .

I admit that this is a huge project. But I'm in
the MI5T (Mission Impossible 5oftware Team) and they
can do the impossible :-)

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



From 12klat at sightreader.com  Thu Jun  1 20:12:12 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Thu, 01 Jun 2000 17:12:12 -0700
Subject: Getting the directory size
References: <2B1262E83448D211AE4B00A0C9D61B030130118F@msgeuro1.creo.be>
Message-ID: <3936FBDC.4284A8BF@sightreader.com>

It seems unlikely that python is a bottleneck here, since the algorithm is I/O
bound (i.e. most of the time is spent accessing your hard drive).  Steve's
suggestion will help since you don't want to run getDirSize() twice, but I
don't think you'll be able to do any better than that (even if you wrote an
extension module).

Note that if you right-click on a large directory (such as C:\Windows) from
Explorer, and click "properties", NT will take a little while to get the Size
value (which is the same as your algorithm will produce).  You're algorithm
(with Steve's enhancement) should have the same performance as the "Windows
Properties" dialog (but be careful not to be fooled by caching effects - the
second time you get a directory size will automatically be much faster than the
first).

- Ken Seehof
kens at sightreader.com
www.sightreader.com/kens

Pieter Claerhout wrote:

> Hi all,
>
> I want to be able to find the total size of a directory on my NT machine,
> and I started with writing it using the os.path.walk function, and it looks
> something like this:
>
> <code>
> import os
> import sys
>
> def calcDirSize(arg, dir, files):
>         for file in files:
>                 stats = os.stat(os.path.join(dir, file))
>                 size = stats[6]
>                 arg.append(size)
>
> def getDirSize(dir):
>
>         sizes = []
>         os.path.walk(dir, calcDirSize, sizes)
>         total = 0
>         for size in sizes:
>                 total = total + size
>         if total > 1073741824:
>                 return (round(total/1073741824.0, 2), 'GB')
>         if total > 1048576:
>                 return (round(total/1048576.0, 2), 'MB')
>         if total > 1024:
>                 return (round(total/1024.0, 2), 'KB')
>         return (total, 'bytes')
>
> def main():
>
>         dir = sys.argv[1]
>
>         print "Testing directorySize..."
>         print "Directory: %s" %(dir)
>         print "Size: %s %s" %(getDirSize(dir)[0], getDirSize(dir)[1])
>
> if __name__ == '__main__':
>         main()
> </code>
>
> Is there any way to fasten up this beauty, because depending on the
> number of directories it has to walk through, it takes a long time. Is
> there maybe a function in the win32 modules who does this for me,
> but lots faster?
>
> Kind regards,
>
> Pieter Claerhout



From olczyk at interaccess.com  Sat Jun 17 03:49:49 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 17 Jun 2000 07:49:49 GMT
Subject: Python 3000
Message-ID: <394c2b15.634411296@nntp.interaccess.com>

Is there a link where I can look up what's being considered for Python
3000?
TIA


From timr at probo.com  Thu Jun  8 00:35:16 2000
From: timr at probo.com (Tim Roberts)
Date: Thr, 8 Jun 2000 04:35:16 GMT
Subject: SQL Variable Substitution Question
References: <8hmacu$ia9$1@news-int.gatech.edu>
Message-ID: <UF8UJS0SIUG4CFSACQCUVRL37GIMB82BEL@4AX.COM>

liang at cad.gatech.edu (Shengquan Liang) wrote:
>
>say i have yyyy = 1000
>
>in the documentation of DC Oracle, it says that
>i shoul d use 'select * from attt'
>		'where bbbb = :yyyy'
>
>to make it work.

No, SQL doesn't have any idea what your Python variables are.  To use the
":" feature, you have to have set an _SQL_ variable called "yyyy".  How you
do that varies with the backend.

In your case, it's probably easier to build the complete string in Python
and pass that to the database:

  'select * from attt where bbbb = %d' % yyyy

>BUT, when i tried it this way, the
>iDC oracle reponded by:
>
>  oci.err:(1008,'ORA-1008: not all variables bound\012')
>
>what's the problem with that?

You haven't bound the SQL variable yyyy to any value.  As I said, the
method for doing that varies with the backend.

--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From qpgmr at wsnews.worldshare.net  Mon Jun  5 11:59:52 2000
From: qpgmr at wsnews.worldshare.net (qpgmr at wsnews.worldshare.net)
Date: Mon, 5 Jun 2000 15:59:52 GMT
Subject: mixed solution: unicase (unique allowed case) (was: Re: Case-sensitivity: )
In-Reply-To: <8h92ko$5e1@news.Informatik.Uni-Oldenburg.DE>
	<392456A9.993FB15C@concentric.net>
	<39245ED7.9C3A4FAB@prescod.net>
	<Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu>
	<vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA>
	<cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com>
	<8g58j4$11ei$2@thoth.cts.com>
	<8g6abe$s5m$17$1@news.t-online.com>
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>
Message-ID: <14651.52856.809392.435977@CJ42289-A.RESTON1.VA.HOME.COM>

Johannes Nix writes:
 > - an assignment to the same name with different case in the same
 > module is a Case Error exception.

Johannes,
  I presume CaseError would be a subclass of NameError.  Perhaps it
should be called IdentifierCaseError?


  -Fred

--
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From thomas at xs4all.net  Thu Jun 15 09:25:41 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 15 Jun 2000 15:25:41 +0200
Subject: Best link for Python add-ins?
In-Reply-To: <3948D2EB.424B5E42@micron.net>; from android@micron.net on Thu, Jun 15, 2000 at 08:58:20AM -0400
References: <3948D2EB.424B5E42@micron.net>
Message-ID: <20000615152541.G26436@xs4all.nl>

On Thu, Jun 15, 2000 at 08:58:20AM -0400, Peter Immarco wrote:

> Just read about another wonderful library for Python, NumPy.  (Is that
> pronounced like "lumpy" or "lum-pie":)?  What's the best link that has a
> list of the libraries available for Python, windows 32 version?

http://www.vex.net/parnassus

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jeffm at centurionsw.com  Tue Jun 20 10:20:20 2000
From: jeffm at centurionsw.com (Jeff Mitchell)
Date: Tue, 20 Jun 2000 14:20:20 GMT
Subject: Designing an Application
References: <uyQj6IKt$GA.233@cpmsnbbsa03>
Message-ID: <394F7D67.98265754@centurionsw.com>

Disclaimer: I've not done this in python and I don't known anything
about Gadfly.

Most databases can be queried for metadata about a table (column names
and types) and a data structure can be built from that on the fly. The
database won't (at least in my experience) know enough to for you to
determine if the data is a reference to another object, so you'll have
to create a naming convention for columns that supply that, for example:

	create table students (
		student_id	integer,	## a field in the student object
		ref_animal	integer,	## reference to the animal table
		);

So, in this senario any field name that begins with 'ref_' is a
reference to data stored in table named after the underscore.

I've done this (in C, no less) and it works well. The biggest problem is
constructing the user interface on the fly , which is not terribly
difficult, unless you want it to look good ;-)

Hope this helps -- Jeff

lexberezhny wrote:
> 
> Hi,
>   I am not sure if this is appropriate to this list but I though I would try
> anyway. I am writing a Grade book program using a gadfly database, and a
> PyGTK front end. I split everything up into a database library (all the SQL
> stuff, and database initialization), grade book library (class definitions
> for a student, assignment, etc), and an interface.py GTK file. Here is an
> stripped down student class:
> 
> class Student:
>     def __init__(self, studentID):
>         student = database.getStudent(studentID) # runs the SQL query and
> returns a student list
>         self.Name = student[0]
>         self.Grade = student[1]
>         etc...
> 
> My future goal would be to allow the teacher to add his or her own
> properties to a student, and maybe even add independent object (ie a teacher
> wants to keep track of the favorite animals of each student, so there would
> be an object for an animal maybe, and a property in the student table to
> relate to the animal table, something to that effect, etc..) Well I can not
> think of a good way to implement this without putting a few hacks together.
> I guess what I'm trying to do is create dynamic classes? Or is there some
> other way to do this? Any suggestions or experience tips would be greatly
> appreciated. On a side note I am doing this for a school project, so I'm
> learning as I'm going along and willing to read any documentation, etc.
> 
>  - Lex Berezhny


From kuncej at mail.conservation.state.mo.us  Tue Jun 13 11:02:38 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Tue, 13 Jun 2000 10:02:38 -0500
Subject: Where should I put my HTMLgen directory?
References: <7BD10B680501D411B9DF009027E06F32011ACA@EXCHANGE>
Message-ID: <c5s15.52$K4.422@news.more.net>

> I have unpacked HTMLGen into:
> C:\Program Files\Python\Lib\HTMLgen
>
> But when I try to run:
> import HTMLgen
>
> I get the traceback:
> ...
> ImportError: No module named HTMLgen

when you do an "import" python looks for
the module in all the directories in sys.path

Below are four ways to let python know where
you have installed HTMLgen. I use #4 myself.

1) Add the directory to sys.path in your program
before you import the module:
    import sys
    sys.path.append('C:\\Program Files\\Python\\Lib\\HTMLgen')

2) Add the directory to your systems "PYTHONPATH" environment
variable before you start your program. For windows:
    SET PYTHONPATH=C:\Program Files\Python\Lib\HTMLgen
or if PYTHONPATH is already defined:
    SET PYTHONPATH=C:\Program Files\Python\Lib\HTMLgen;%PYTHONPATH%

3) Create a file 'C:\\Program Files\\Python\\HTMLgen.pth' containing the
single line:
Lib/HTMLgen

4) Import HTMLgen as a package
    a) create an empty file C:\\Program
Files\\Python\\Lib\\HTMLgen\__init__.py
    b) in your program, import HTMlgen like this:
        from HTMLgen import HTMLgen

All these procedures are explained in the documentation, and in
the various python books. None is specific to HTMLgen.

And, personally, I would not put HTMLgen in the 'Lib' directory - I think
of that as modules that are distributed with python, and may get overwritten
with a new release. I'd suggest installing it as
   C:\Program Files\Python\local\HTMLgen
or just
   C:\Program Files\Python\HTMLgen


  --Jeff




From it at onestopent.com.au  Tue Jun 13 03:44:52 2000
From: it at onestopent.com.au (Jason)
Date: Tue, 13 Jun 2000 17:44:52 +1000
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no>
Message-ID: <3945e37b$0$489$7f31c96c@news01.syd.optusnet.com.au>

hi thomas
thanks for your help again...
here's the error message i get when i run your script in IDLE
the listing is the first in the txt file
can you help pls....
does this script make the output file a tab file..i can't see any tab in the
script at all...can you let me know pls...

thanks
jason

FORM: ['Popular Recording']
CMTYPESWITCH: ['Popular']
RELOCORIGINAL: ['8F2C014FEFE7A61DCA2568D600267EBD']
CMCOUNTRY_ORIGIN: ['']
CMGENRE_CODE: ['']
CPTITLE: ['LIQUID']
CPARTIST: ['RECOIL']
CPOARTISTS: ['']
CMCATNUMBER: ['CDSTUMM173']
CMAPN: ['5016025611737']
CMARIAPRICECODE: ['C1145']
CMARIAMEDIACODE: ['21']
CMARIADISTRIBUTORCODE: ['MDS']
CMARIAPACKAGECODE: ['']
DMRELEASE: ['19/06/2000']
cMDateFormat: ['19/06/2000 12']
DMDELETE: ['']
cMDeleteStatus: ['No']
CMLISTPTRACKS: ['']
CMLISTPARTISTS: ['']
RELOCMEDIA: ['0FE68A46C3E72687CA2568D6002C9FAD']
CURMPRICE: ['21.71']
CMTYPE: ['Compact Disc']
CMRECORDCOMPANY: ['MUSHROOM DISTRIBUTION SERVICES']
CMPACKAGE: ['']
CMARIADISTRIBUTORHOUSECODE: ['BFM']
cDistributorHouse: ['BFM']
CMLOCALE: ['Y']
$UpdatedBy: ['CN=PPT/OU=AEROS/O=JUKEBOX']
$Revisions: ['05/05/2000 06']
Traceback (innermost last):
  File "C:\PROGRA~1\PYTHON\TOOLS\IDLE\ScriptBinding.py", line 131, in
run_module_event
    execfile(filename, mod.__dict__)
  File "C:\WINDOWS\Desktop\aeros1.py", line 6, in ?
    print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
IndexError: list index out of range





From rupole at compaq.net  Sun Jun  4 11:18:26 2000
From: rupole at compaq.net (Roger Upole)
Date: Sun, 04 Jun 2000 15:18:26 GMT
Subject: instantiate multiple automation objects
References: <3939d663.1392772@nntp.pipeline.com>
Message-ID: <6ru_4.14017$Rq.1311147@pouncer.easynews.com>

win32com.client.Dispatch always tries to connect to a running
instance first, so if you've already done a Dispatch for that
object, any subsequent call will return the same instance.
However, you could use DispatchEx to get a second
instance, as it apparently doesn't try to connect to a running
instance first.
           HTH
                  Roger Upole


"Yoon S. Choi" <ychoi at pipeline.com> wrote in message
news:3939d663.1392772 at nntp.pipeline.com...
> I am trying (without much luck) to create multiple instances of an
> automation object.  I can create 1 instance correctly using the
> function call win32com.client.Dispatch().  However, when I use the
> same function to try and create another automation object, it just
> refers to the first instance.
>
> Here is an example using Microsoft Word as the automation object:
>
> import win32com.client
> w1 = win32com.client.Dispatch( "Word.Application")
> w2 = win32com.client.Dispatch( "Word.Application")
> w1.Visible = 1  # Microsoft Word becomes visible
> w2.Visible = 1  # Nothing happens because w2 is referring to the same
> object as w1
>
> How can I for example create another instance of Microsoft Word?  In
> other words, w1 and w2 would refer to two separate automation objects
> of the same type.
> Can I do this without having to resort directly to the pythoncom
> module and its associated CoCreateInstance and CoGetClassObject
> functions?




From jeremy at beopen.com  Fri Jun  9 11:52:57 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 09 Jun 2000 11:52:57 -0400
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
Message-ID: <D3C4C6248A1F2644.8269264D36C33639.4C75B18E6ECC1D9B@lp.airnews.net>

hzhu at rocket.knowledgetrack.com (Huaiyu Zhu) writes:

> My wishlist for Python3K (This is not to start a flame war):
> 
> 3. The := operator for deepcopy as oposed to reference assignment.  Therefore
>    - the assignment (a = b) guarantees (a is b)
>    - the assignment (a := b) guarantees (a == b and a is not b).
>    

There are many kinds of objects where copying is not defined.  It's
unlikely we'll define copies for all of them.  It is also quite
difficult to guarantee that equality is defined for every kind of
object other than object identity.  In the case of an object that
contains references to other objects, a deepcopy that creates a
complete clone of every object referenced from the root is expensive
and probably unnecessary. Thus, I don't think we'll add a := operator.

I'd also observe that I seldom write programs where I need to make
copies (and almost never a full deepcopy).  If this programming idiom
is unusual, I don't see any advantage to having language syntax to
support it.

-- Jeremy Hylton <http://www.python.org/~jeremy/>




From jriehl at lucent.com  Wed Jun 28 20:32:01 2000
From: jriehl at lucent.com (Jon Riehl)
Date: Wed, 28 Jun 2000 19:32:01 -0500
Subject: Bindings for game library
References: <bOj65.1079$MZ3.2888@newsfeed.slurp.net>
Message-ID: <395A9901.CEB7CB42@lucent.com>

	Earlier this year I took an initial stab at the CDX game library. 
Unfortunately, my stuff got nuked with the Starship Python.  I'll go
ahead and take an action to update the following URL real soon now:
http://starship.python.net/crew/jriehl/pycdx/ (smirk)
	CDX is at: http://www.cdx.sk/
	If you see stuff you want in PyCDX, send many death threats to
jriehl at mindspring.com.  (DISCLAIMER: This death threat invitation only
extends to CDX/game development enthusiasts you sick Internet types.)
-Jon


From mktsite at centexchange.com  Thu Jun  8 01:31:53 2000
From: mktsite at centexchange.com (mktsite at centexchange.com)
Date: Thu, 8 Jun 2000 08:31:53 +0300
Subject: CentExchange - The best traffic builder
Message-ID: <MO329O6U.3788KFAK@centexchange.com>

Dear Sir or Madam,

Drive targeted traffic to your site with http://www.centexchange.com - 
The most effective way to draw customers to your site and keep them there, 
due to FOCUSED ADVERTISING. Our unique method can boost your sales and 
increase dramatically your revenues. This is a method by which advertisers 
pay us according to the number of surfer entries into their particular site. 
Also the money you invest in these customers will come back to you as payment 
for your products. Our Method guaranties that you pay only for surfers who 
stay in your site - you can see that its much more effective than banners, 
and it's completely risk-free!
We are expanding rapidly - come to our site http://www.centexchange.com 
(click on JOIN US - and then ADVERTISER) and see for yourself our unique 
advantages and the huge benefits you can gain by joining us. 

Prepare yourself for a flood of customers / MONEY !

For more information please feel free to contact me - mktsite at centexchange.com.

Have a great day,
Dan Marom
Sites marketing 
CentExchange.com




If you don't want to receive mail from us in the future, simply 
Reply with "REMOVE" in the subject. 




From richard_chamberlain at my-deja.com  Thu Jun 29 05:59:15 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Thu, 29 Jun 2000 09:59:15 GMT
Subject: How to concatenate external files ..?
References: <slm1odapjev11@corp.supernews.com>
Message-ID: <8jf6lf$v2j$1@nnrp1.deja.com>

Hi Randy,

import sys, string, glob, os

#First I create a list:
fls = glob.glob('C:\\wxpython\\wxPython-2.1.16\\docs\\wx\\wx*.htm')

#Then I guess I should open an output file for writing
outfile = open('temp.txt,'w')

for x in fls:
               file=open(x,'r')
               data=file.read()
               file.close()
               outfile.write(data)

outfile.close()

On your outfile you had 'r' which is read obviously you need to be able
to write to it so 'w' is what you need.

We then iterate over fls to give us each file name and open each one.
We then call the read method against it, which loads in the entire file
and assign that to data. we close the file and write the data to our
outfile. And then on to the next file. Finally we close the outfile.

You have to use read() with some caution because it (trys) to load the
whole file into memory which maybe an issue if it was sizable. In your
case it isn't an issue because they are simple html files.

Richard






In article <slm1odapjev11 at corp.supernews.com>,
  "Randolph MacKenzie" <rmack at eznet.net> wrote:
> I want to combine all the little *.htm files in a directory into one
big
> file.
>
> import sys, string, glob, os
>
> #First I create a list:
> fls = glob.glob('C:\\wxpython\\wxPython-2.1.16\\docs\\wx\\wx*.htm')
>
> #Then I guess I should open an output file for writing
> outfile = open('temp.txt,'r')
>
> Now what  .. how do I iterate over the list and specify that each
file is to
> be read and appended to "outfile" ?
>
> for x in fls:
>     ???
>
> -Randy
>
>


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


From drew at level3.nospam.net  Tue Jun  6 10:30:58 2000
From: drew at level3.nospam.net (Drew Fisher)
Date: Tue, 6 Jun 2000 08:30:58 -0600
Subject: drawing graphs in python
Message-ID: <393d0c07$0$17334@news.denver1.Level3.net>

Hello.  I am still learning Python, but I came across a need for generating
dynamic graphs (network load average, uptime, boring stuff like this :).

Is the a module in Python that allows this?  If so, could someone point me
to it?

Thanks in advance,
Andrew Fisher




From rajarshi at home.com  Tue Jun  6 20:40:36 2000
From: rajarshi at home.com (Rajarshi Ray)
Date: Wed, 07 Jun 2000 00:40:36 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <kk5rjski2pc4g4tor99tc2h5ursinprg80@4ax.com>
Message-ID: <393D999B.E1150BCB@home.com>

Kirby Urner wrote:
> 
> Rajarshi Ray <rajarshi at home.com> wrote:
> 
> >Besides the gee-whiz aspect of such algorithms, do they have any value?
> >After all, what makes pi pi is ALL of its digits in a particular
> >sequence.
> 
> Supposedly there were some deep points about number
> theory in this vicinity -- according to the web pages
> anyway.  I'm no expert.

Ah, that's what I wanted to hear. The importance of this discovery is
then what makes pi amenable to such a calculation, and also if other
transcendental numbers can be treated like this e.g. 'e'.

Don't you just hate presumptuous bastards like Ullrich who will jump
your gun without even considering if you actually have a point?? Perhaps
you should be glad you're a non-expert; the more expert you get, it
seems the more of a pompous ass you become too.

 
> Anyway, the "gee whiz" factor isn't to be underestimated,
> either.
> 
> Kirby

-- 
Sir, I have found you an argument. I am not obliged to find you
an understanding.

                                               - Samuel Johnson


From robinporter72 at yahoo.ca  Thu Jun 15 09:38:26 2000
From: robinporter72 at yahoo.ca (Robin Porter)
Date: Thu, 15 Jun 2000 09:38:26 -0400 (EDT)
Subject: TEXT MANIPULATION (in CSV format)
Message-ID: <20000615133826.11516.qmail@web1502.mail.yahoo.com>


I am wondering how difficult it would be to use PYTHON to go through a CSV (Comma separated value) formatted file, take two of the fields and merge them into one field.

An example would be:

Field 1             Field 2

"K"           ,       "1234"

I want to change to:

Field 1

K1234

Thanks in advance for your assistance.

Robin

 



Mr. Robin Porter
St. Lawrence College
Computer Engineering

Database Coordinator - Raiser's Edge

Alumni Development

 



---------------------------------
Do You Yahoo!?
Get your free @yahoo.ca address at Yahoo! Mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/8da43abd/attachment.html>

From digitig at cix.co.uk  Tue Jun  6 18:20:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Jun 2000 23:20 +0100 (BST)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.10.10006060938200.4711-100000@sundial>
Message-ID: <memo.20000606232050.11341B@digitig.compulink.co.uk>

In article <Pine.GSO.4.10.10006060938200.4711-100000 at sundial>, 
moshez at math.huji.ac.il (Moshe Zadka) wrote:

> On Mon, 5 Jun 2000, James Logajan wrote:
> 
> > If God had meant for languages to be case insensitive, He would not 
> > have
> > invented both upper AND lower case letters. If it is good enough for 
> > Him, it
> > is good enough for me!
> 
> GOD CREATED UPPERCASE LETTERS. everything else is the work of man.

god doesn't need to shout...


From dave at rotwang.freeserve.co.uk  Sun Jun 18 08:03:10 2000
From: dave at rotwang.freeserve.co.uk (Dave Berkeley)
Date: Sun, 18 Jun 2000 13:03:10 +0100
Subject: mpeg encoding
Message-ID: <8iie3s$epp$1@newsg1.svr.pol.co.uk>

I have written a python extension which allows me to control a Kodak DC215
digital camera for use in time lapse photography using the Kodak SDK. I can
now grab jpg images every 20 or so seconds, and leave it running all day.

Does anyone know how to convert a set of jpg images into mpeg using PIL?

Dave Berkeley




From shapr at uab.edu  Mon Jun 19 04:41:12 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 19 Jun 2000 08:41:12 GMT
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C8@VRTMAIL>
Message-ID: <394DDD4C.26269099@uab.edu>

Mike Fletcher wrote:
> 
> The byte code compilation makes the doc string an attribute of the function
> object, as opposed to a statement that needs to be executed on each
> iteration of the function.

> ...
> 
> The """ syntax would seem to be slower than putting # at the begining of
> everyline because with # it only needs to test the first character of the
> line, if it's # it can skip the rest of the line. But with """, it needs to
> check every character after until it finds the matching """.
> 
> ...

And as Thomas Wouters (I think) pointed out a few days ago, python -OO
gets rid of docstrings in the bytecode, along with other optimizations.
I have no idea if it works in Python 1.2 though.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From rol9999 at attglobal.net  Tue Jun 20 06:48:57 2000
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Tue, 20 Jun 2000 10:48:57 +0000
Subject: converting hex number contained in a string to an integer
References: <8ing7d$o6r$1@nnrp2.deja.com>
Message-ID: <394F4C19.185E4B64@attglobal.net>

ejfc at my-deja.com wrote:

> I have a string containing a number. This is an
> hex number. How can I convert this to an integer?
>
> >>> hexnumber
> '32'
> >>> 0x32
> 50
> >>> int(0x32)
> 50  # this is the result I want to get
> >>> hexnumber = '0x' + hexnumber
> >>> hexnumber
> '0x32'
> >>> int(hexnumber)
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
> ValueError: invalid literal for int(): 0x32

>>> import string
>>> string.atoi(hexnumber, 16)
50

Roland Schlenker



From thomas at xs4all.nl  Fri Jun 30 08:56:21 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 30 Jun 2000 12:56:21 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com> <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com> <geT65.2262$iN5.511399@ptah.visi.com> <m2lmznzk2t.fsf@mycroft.actrix.gen.nz>
Message-ID: <slrn8lp673.u8u.thomas@centurion.xs4all.nl>

On 30 Jun 2000 19:47:38 +1200, Paul Foley <see at below> wrote:
>On Fri, 30 Jun 2000 02:03:56 GMT, Grant Edwards wrote:

>> C predates the PDP-11 so we'll chalk it up to convergent
>> evolution.

>Oh no it doesn't.

Oh yes it does ;) The first bits of C, including, for instance, the
auto-increment operators, were designed and used on the PDP-7, not the
PDP-11.

> C was written for the Unix port to PDP-11.

More like 'during' than 'for'. UNIX and C both, together, evolved on for
quite a number of years (witness the new C & UNIX 'standards')

>(Perhaps you're thinking of Unix!?)

Well, UNIX has it's roots on the same PDP-7 that saw birth of the 'new B'
language, which eventually turned into 'C'. 'new B' was (not suprisingly)
based on the language 'B', and both were interpreted rather than compiled
(because the PDP-7 had too little resources to support a compiler.)

Of course, like the early bicycles, very few of us would actually recognize
the OS, the language or the hardware if they didn't have museumtags attached
;-)

(I saw an old PDP-11 used as a telco-switch in a museum a few months back;)


From gmcm at hypernet.com  Mon Jun 26 19:32:50 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 26 Jun 2000 23:32:50 GMT
Subject: unexpected behaviour in class variables
References: <8j83ml$2s7$1@slb3.atl.mindspring.net>
Message-ID: <8F5FCA994gmcmhypernetcom@199.171.54.194>

Andrew Dalke talks himself into a circle: 

[skipping much]

>import math
>def eggs(x):
> pass
>
>class Spam:
> builtin_func = math.cos
> def_func = eggs
> def p(self):
>  print self.builtin_func
>  print self.def_func
>  print math.cos

OK - builtin_func and def_func live in Spam.__dict__. Remember that <wink>.

>import math
>
>def eggs(x):
>    return math.sin(x)
>
>class Spam:
>    def __init__(self):
>        self.sin = eggs
>        self.cos = math.cos

Here, sin and cos (like wanton and lust) live in the instance dict.

An attribute found in the instance dict will never get wrapped. Only stuff 
found in the class __dict__ (or in a base class __dict__) qualifies for 
wrapping.

Why the wrapping? To hide away "self" - so it's implicit to the method 
object but explicit to the function object.

Make sense now?

- Gordon


From eq3pvl at eq.uc.pt  Thu Jun  8 07:39:54 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Thu, 08 Jun 2000 12:39:54 +0100
Subject: ANNOUNCE: early alpha Quixote Web development tool
References: <200006080135.VAA04781@mira.erols.com>
Message-ID: <393F860A.EAC32B61@eq.uc.pt>

"A.M. Kuchling" wrote:

> The basic idea is that a Web application is written as a Python
> package, and URLs are then treated as paths walking down into the
> package.  Objects, functions, and methods are accessible through URLs
> if they're listed in an __exports__ attribute of the object.
> (Obviously this is inspired by Zope's object publishing.)

If i get it right this will be a dtmless, zodbless, aquisitionless zope.
Just a Zpublisher-like ORB with a ASP-like template language.
Why not (re)use ZORB?

I'm one that thinks Zope is not for everything. I like when new web
projects
appear on pythonland. One of those nice projects is modpython. Maybe the
these two projects will work be complementary in the future.

regards

pedro lima



From xxx at yellow13.mimuw.edu.pl  Thu Jun 22 09:18:51 2000
From: xxx at yellow13.mimuw.edu.pl (Robert Szefler)
Date: Thr, 22 Jun 2000 13:18:51 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net>
Message-ID: <SLRN8L1NP4.1JM.XXX@YELLOW13.MIMUW.EDU.PL>

In article <t7545.239$ZR1.2944 at news.uswest.net>, Jeff Massung wrote:
>Please, this interests me -- I've had differention equations, but that's as
>far as I go. In theory, pi (and e) are infinite, but has that been proven
>that there is not end to their decimals?
>
>Jeff

oh, come on, pi and e are irrational, so their decimal representations are
infinite and non-periodic. It's quite easy to prove e's irrationality. And
it's a bit harder for pi.
Moreover, they're transcendental numbers (do I spell it right?) ie. not a
root of any integer-coefficient-polynomial. This is substantially harder
to prove than the above ones.

steve


From jkraska1 at san.rr.com  Wed Jun 21 20:25:14 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 22 Jun 2000 00:25:14 GMT
Subject: Python on FreeBSD
References: <TLc45.12472$dF.523323@news1.rdc1.il.home.com>
Message-ID: <39515D16.C4C1DA06@san.rr.com>

dsavitsk wrote:
> 
> i've been doing the python thing for a while on win32, but i finally got my
> bsd box up.  i did a make install, and python from the terminal window
> works.  i am wondering how to start idle,...

If you install python in /usr/local, you might create
a csh script ala:

#!/bin/csh
/usr/local/python/python /usr/local/python/tools/idle/idle.py





C/


From tseaver at starbase.neosoft.com  Tue Jun  6 23:27:59 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 6 Jun 2000 22:27:59 -0500
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newsh <slrn8jrb37.b4q.wtanksle@dolphin.openprojects.net>
Message-ID: <6E26036C81F31B9A.BB33FE25988B7171.E4D3580AA05779C6@lp.airnews.net>

In article <slrn8jrb37.b4q.wtanksle at dolphin.openprojects.net>,
William Tanksley <wtanksley at bigfoot.com> wrote:
>On 06 Jun 2000 17:34:20 -0400, Andrew M. Kuchling wrote:
>>wtanksle at dolphin.openprojects.net (William Tanksley) writes:
>>> I'm not the one that's claiming that removing case-sensitivity will ruin
>>> Python.  My claim is very simple: first, it's a minor change for us to
>>> make, and second, it helps newbies.
>
>>Like Aahz, I don't think the change is that minor.
>
>I suppose we'll just have to agree to disagree, then.  Because I can't see
>the change causing me to care -- my coding style will hardly change.

*Sigh* It's not *your* style I'm worried about -- I've seen stuff you
post, and *you* wrtie cleanly.  I have significant experience, however,
in working in *large* (100K+ LOC) bodies of code written in a case-
insensitive language (ObjectPascal) by case-negligent coders.  The
readability of case-mangled code, like that of inconsistently indented
code, is *horrible*.  "Coding standards" get ignored by this grade of
"programmer";  compiler errors don't.

>
>Now, the change WILL cause old code to not work -- but in case you missed
>it, it's been established that Py3K is going to do that.  Adding or
>removing this feature will not fix the code.
>
>>I'm also doubtful of the "helps newbies" claim; unless there have been
>>some other citations in this thread that I missed, we have exactly one
>>study using exactly one programming environment that reported CS
>>(case-sensitivity) was a problem, and even that report says "most [users]
>>continued to type case-incorrect tokens in their programs for a short
>>period", which implies that they did eventually stop.  The evidence is
>>hardly overwhelming, and basically purely anecdotal.
>
>Now here we can make some progress, I think.  Once CP4E is funded, I would
>hope that this is something they would do some work in.  Hopefully before
>the change is finalised in Py3K.
>
>However, from my experience teaching people programming and typing, people
>don't normally think or communicate in a case sensitive way, and a large
>portion of people have a really hard time doing so.  I've hypothesised
>that this is an artifact of visual versus auditory learning.

Readability is an important element of both learnability and usability
of a language;  case-insensitivity reduces readability in order to enhance
writability.  Ought we make it easier to write sloppy Python than
is presently possible?  I don't see how that can be a win, except for the
greenest of 4E's.

>>IMHO, if GvR was inventing a completely new language for teaching,
>>then it would be worthwhile to debate whether it should be CS or CI.
>>But he's not; Python is an implemented language with a sizable and
>>growing user base that's used to CS, and a number of published books
>>and other materials that will be invalidated by switching to CI. 
>
>And the fact is that they're going to be invalidated anyhow.  Nothing we
>do to CS will change that.  Part of the reason for Py3K is to allow Guido
>to make large, backwards-incompatible changes.

To what end?  I still have seen nothing except the anecdotal evidence from
the Alice project for the claim that CS reduces the overall usability, or
even learnability, of Python.

Gratuitous-incompatibility-buys-us-what?'ly

Tres.
-- 
===============================================================
Tres Seaver                                tseaver at digicool.com
Digital Creations     "Zope Dealers"       http://www.zope.org


From aahz at netcom.com  Wed Jun 14 21:58:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 01:58:57 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <394833BD.6A322985@san.rr.com>
Message-ID: <8i9d91$60t$1@slb0.atl.mindspring.net>

In article <394833BD.6A322985 at san.rr.com>,
Courageous  <jkraska1 at san.rr.com> wrote:
>
>In vi/vim, you can globally replace all tabs with:
>
><esc>:1,$s/<tab>/    /g

And if you want to save a couple of characters:

^[:%s/\t/    /g
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From peter at verecomm.com  Wed Jun  7 14:43:50 2000
From: peter at verecomm.com (Peter Kropf)
Date: Wed, 07 Jun 2000 11:43:50 -0700
Subject: Python under VMware/NT?
References: <r3h%4.23$cT1.480@news>
Message-ID: <393E97E6.30D7DAFE@verecomm.com>

I use VMWare 2.0 under RedHat 6.2 on a regular basis. One
of the tasks is to perform cross platform checks for some
GUI code written using Python and wxWindows. I've had no
problems and performance looks to be ok. Not great, but ok.


Chris Ryland wrote:
> 
> Has anyone tried running Python in a virtual machine under VMware/NT,
> whether in Linux or NT/2K/9x?
> 
> Odd question, I realize. ;-)
> --
> Cheers!
> / Chris Ryland, President / Em Software, Inc. / www.emsoftware.com

-- 
Peter Kropf
Codesmith
VEReCOMM Corp.
2475 Augustine Drive, Suite 102
Santa Clara, CA 95054
408-919-3107; Fax 408-969-0688
peter at verecomm.com


From mfletch at tpresence.com  Mon Jun  5 23:05:56 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 5 Jun 2000 23:05:56 -0400 
Subject: Network byte ordering question ...
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3394@VRTMAIL>

( val1, val2, val3, val4, val5 ) = unpack( ">16sBBBi", header )

Should do it, no?  That is, the header is a string of 16 bytes followed by 3
unsigned single-byte integers, followed by a (signed) 32bit integer, all in
Network Byte Order.  You pack/unpack from/to Python native values, which are
always in the machine's byte order.

Packing would use the same pattern, and takes care of translating to/from
network byte order.

Yay! another Waterloovian Pythonista!
Mike

-----Original Message-----
From: istevens at calum.csclub.uwaterloo.ca
[mailto:istevens at calum.csclub.uwaterloo.ca]
Sent: Monday, June 05, 2000 9:22 PM
To: python-list at python.org
Subject: Network byte ordering question ...


...
FYI, here's what I've tried:

    ( val1, val2, val3, val4, val5 ) = unpack( "@16s1s1s1s4s", header )

This works somewhat but I would really like val1 to read as a string
containing numeric values (to preserve leading zeroes) and the other
values to read as numbers.  This is also specific to little-endian
machines and I'd like to generalize it.
...



From 12klat at sightreader.com  Wed Jun  7 01:34:33 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 22:34:33 -0700
Subject: Question about Jpegs
References: <mKi%4.8088$r7.1635010@news1.rdc2.tx.home.com>
Message-ID: <393DDEE9.F0F3C4F5@sightreader.com>

Take a look at PIL (Python Imaging Library).

Michael Morrison wrote:

> This may not be the appropriate place for this question, but it's for a
> Python program and I know there are a lot of experienced programmers here,
> so....
>
> Does anyone know how to "resize" a .jpg or similar file?  My script is
> dealing with uploaded pictures, and I want it to automatically create a
> thumbnail for said pictures.  Does python by chance have a module that can
> do this?
>
> Thanks for any input

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/df93e56c/attachment.vcf>

From pauldubois at home.com  Mon Jun 26 12:22:01 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Mon, 26 Jun 2000 16:22:01 GMT
Subject: Array Memory Allocation Size
References: <39540628.B284C15F@bioeng.ucsd.edu> <m3bt0o4wa2.fsf@chinon.cnrs-orleans.fr>
Message-ID: <JqL55.4838$i5.75676@news1.frmt1.sfba.home.com>

print len(ravel(x)) * x.itemsize()

Also be sure to check out the new spacesaver features.

Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote in message
news:m3bt0o4wa2.fsf at chinon.cnrs-orleans.fr...
> Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:
>
> > I'm using NumPy to allocate heaps of arrays.  Is there a command to find
> > out how much space each array is using in memory?
>
> By C code, yes. From Python, no, because you cannot find out
> how much space each element takes up, and that is compiler dependent.
> --
> --------------------------------------------------------------------------
-----
> Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
> Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
> Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
> 45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
> France                                   | Nederlands/Francais
> --------------------------------------------------------------------------
-----




From kaimar at ut.ee  Fri Jun  2 19:10:28 2000
From: kaimar at ut.ee (Kaimar Karu)
Date: Sat, 03 Jun 2000 02:10:28 +0300
Subject: zope example at /url/
Message-ID: <39383EE4.C87460EE@ut.ee>

Hi,

When I try to follow the example of working with xml in zope
(http://www.xml.com/pub/1999/12/zope/index.html)
i get an error, when i'm in the part where i have to configure a
ZCatalog:
**
Error Type: NameError
Error Value: Eval
**
What could be wrong?

tia,
--
Kaimar Karu | kaimar at ut.ee | 251 13 468
       = What a wonderful world =




From mwh21 at cam.ac.uk  Fri Jun  9 10:55:57 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 15:55:57 +0100
Subject: Python Productivity over C++
References: <CC59868A0A76D311B1F50090277B321C1F339E@VRTMAIL>
Message-ID: <m3k8fyvr7m.fsf@atrus.jesus.cam.ac.uk>

Mike Fletcher <mfletch at tpresence.com> writes:

> Just a minor note, having spent all of last night (unsuccessfully)
> trying to track down a memory leak.  Yes, memory leak tools would be
> nice to have in Python.  Hidden memory leaks (cycles) are rare, but
> when they happen in a complex project they are often a nightmare to
> track down.  An optional garbage collector with a debug mode saying
> "hey, I found a cycle that wasn't cleaned up, the objects were:
> object types, object cycle description" would be really cool.  Yes,
> I'd probably remove it for the final proggy, but as a development
> tool it would be very, very, useful, and is currently (to my
> memory), not available.

Have you tried Neil Schemenauer's patch?

http://www.enme.ucalgary.ca/~nascheme/python/gc.html

It can help with some of these things.

There's also cyclops, which I think is somewhere on ftp.python.org.

Cheers,
M.

-- 
  Academic politics is the most vicious and bitter form of politics,
  because the stakes are so low.                      -- Wallace Sayre


From jmwarfe at darwin.cwru.edu  Tue Jun 20 12:03:06 2000
From: jmwarfe at darwin.cwru.edu (Mike Warfe)
Date: Tue, 20 Jun 2000 12:03:06 -0400
Subject: Directory ACL's in Python
Message-ID: <394F95BA.F3850E6A@darwin.cwru.edu>

Hello, I have been using Python to do administrative tasks on my NT PDC.
Is there a Win32 module that I can use to set directory privileges, that
has the functionality of cacls.exe using python? I am currently using
cacls to set the privileges by doing a os.system command in my script.
Thanks for the help in advance!


Mike Warfe



From timr at probo.com  Sat Jun 24 01:20:11 2000
From: timr at probo.com (Tim Roberts)
Date: Sat, 24 Jun 2000 05:20:11 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <8F5B6C0A8gmcmhypernetcom@199.171.54.154> <slrn8l7bqu.cmq.andyl@azaal.dircon.co.uk>
Message-ID: <A4H8LSO56CJL0004363KFJ0GMHFLT8BMQ6@4AX.COM>

andyl at azaal.dircon.co.uk (Andy Leighton) wrote:
>
>--
>Andy Leighton => andyl at azaal.dircon.co.uk
>"... January is your third most common month for madness" - _Sarah Canary_

  "Sarah Canary?"

  "No, semi-carnally!"

(This is on-topic for this newsgroup.  Take it away, Eric the orchestra
leader!)
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From emile at fenx.com  Thu Jun 15 23:00:18 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 15 Jun 2000 20:00:18 -0700
Subject: identifying operating system
References: <39499924.3BC84686@engineering.uiowa.edu>
Message-ID: <00dd01bfd73f$076f2820$1906a8c0@fc.fenx.com>

Look for Marc Lemburg's platform.py, or use
the standard sys.platform.

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: Juerg Tschirren <jtschirr at engineering.uiowa.edu>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Thursday, June 15, 2000 7:50 PM
Subject: identifying operating system


>
> Is there a way how I can identify the operating system my
Python script
> is running on (e.g., Linux, Windows, Mac, ...)?
>
> Juerg
> --
> http://www.python.org/mailman/listinfo/python-list
>




From bparsia at email.unc.edu  Thu Jun  8 01:37:08 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 8 Jun 2000 01:37:08 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <Pine.SGI.4.05.10006061254220.22654-100000@lemur.magnet.com> <393DD854.56032E87@sightreader.com> <393DF80A.BB53465C@prescod.net>
Message-ID: <1ebvw05.1qkeu0rre2fx7N%bparsia@email.unc.edu>

Well, since I'm here...

Paul Prescod <paul at prescod.net> wrote:

> Ken Seehof wrote:
> > 
> > Java definitely has a niche: embeded software for household appliances.
> 
> I really don't see it having particular virtues there. Forth is more
> efficient and more runtime flexible. Python is easier to code. Scheme is
> smaller to implement. C has better real-time characteristics. Any of
> them can speak TCP/IP.
[snip]

Paul paul paul *Paul*! <sniff/> You forgot Smalltalk! :)

(And Scheme is only "smaller to implement" or *easier* to implement if
you do it in Scheme, or Common Lisp, or...After all, writing a GC, and
doing tail recursion, and continuations, *and* macros...in *C*...whew!
:))

(Er...actually, you forgot common lisp. CL-HTTP on Lisp Machines run the
White House documentation server. Franz just released "AllegroServe"
which is a much easier to understand HTTP server for Common Lisp which
they intend to be portable and Free, I believe. Several inititives to
marry CL and HTML/XML for page generation. Etc.

I really recommend that folks take a look at AllegroServe. I'm no Common
Lisper, but the architecture looked very straightforward and, indeed,
quite neat.)

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From sabren at manifestation.com  Wed Jun 28 09:04:56 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Wed, 28 Jun 2000 09:04:56 -0400 (EDT)
Subject: Comments: full-text indexing of RDBMS
In-Reply-To: <395b7627.97196711@news.online.no>
Message-ID: <Pine.BSF.4.10.10006280856040.17347-100000@chi.pair.com>

On Tue, 27 Jun 2000, Thomas Weholt wrote:

> I get a list of all the tables in a specified database from the
> databasemodule, remove all database-system tables, goes thru each
> table, select * from table. The result ends up in a dictionary-object.
> I check all the values for plain text. Integers and numbers in general
> are ignored. A function take the plain text and returns a list of
> words found in that record. I put each word in a new table, gives it a
> unique id. A record in PostgreSQL has a OID, an uniqe
> object-identifier.  The OID is mapped against the ids of the words
> that occurred in that record. 


Hey Thomas,

  You emailed me a while back about ransacker... I've actually
improved the speed quite a bit since then, provided you use the bsd
module for the hash stuff, since performance of gdbm seemed to be a
large part of the problem... It's not perfect yet, but you might
want to try it out.

  http://ransacker.sourceforge.net/

  Your algorithm would still be what you have above, but ransacker
does a lot of it for you. If nothing else, you might consider it in
the prototyping stage.

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From gee308 at mediaone.net  Thu Jun  8 17:56:35 2000
From: gee308 at mediaone.net (Toy)
Date: Thu, 08 Jun 2000 21:56:35 GMT
Subject: quick newbie question on variables. please answer
Message-ID: <39401B5F.122C6DB4@mediaone.net>

Im a programming newbie, so most of these concepts are basic to me.  I'm
trying to write really basic  CGI programs just to get started.  How
would I pass a text variable that a user entered to find a man page for
a tool?  This is the code I wrote (startx is just a command I wrote in,
but how would I write it so any tool could be inputed) the 3rd line od
code is written wrong for sure:

#!/usr/bin/python
import commands
b = commands.getstatusoutput('man %x!') % startx
print = b

  I know that this is a real basic question, but I am 100% new to
programming and I am self teaching.  Also, are ther large security risks
if this program is on a web server(if so, how could I fix it?)?  Thanks
for your time in advance.





From morpheus at despair.rpglink.com  Fri Jun  2 12:34:35 2000
From: morpheus at despair.rpglink.com (Steve Lamb)
Date: Fri, 02 Jun 2000 16:34:35 GMT
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h7blm$j1b$1@slb2.atl.mindspring.net> <slrn8jeqfv.n4g.morpheus@teleute.rpglink.com> <8h8meo$a57$1@slb7.atl.mindspring.net>
Message-ID: <slrn8jfof4.o6c.morpheus@teleute.rpglink.com>

On Fri, 2 Jun 2000 10:15:59 -0600, Andrew Dalke <dalke at acm.org> wrote:
>So not only am I not a lawyer, but I'm not a wordsmith.

    But you play one one usenet.  ;)

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From darrell at dorb.com  Tue Jun 27 00:39:53 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 26 Jun 2000 21:39:53 -0700
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand>
Message-ID: <03ee01bfdff1$bd0b25c0$6401a8c0@home>

I find that often re.findall is easier to get right than re.split

>>> s="uid=joe, ou=org\\, unit, o=some org"
>>> re.findall("([^,]+)=([^,]+)", s)
[('uid', 'joe'), (' ou', 'org\\'), (' o', 'some org')]

--Darrell

----- Original Message ----- 
From: "Steffen Ries" <steffen.ries at sympatico.ca>


> Hi,
> 
> I'm trying use re.split to separate an LDAP Distinguished Name into
> its naming components. The naming components are of the form
> "attr=value" and they are separated by ",".
> 
> The part, which I can't solve is, that "value" may contain the
> sequence "\,", which obviously does not separate a naming component.
> 
> e.g.
> "uid=joe, ou=orgunit, o=some org"
> should be split into ["uid=joe", "ou=orgunit", "o=some org"]
> 
> "uid=joe, ou=org\\, unit, o=some org"
> -> ["uid=joe", "ou=org\\, unit", "o=some org"]
> 
> Is there a way to use re.split to achieve this? (I solved it searching
> [^\\], and splitting the string in a loop, but I was wondering if it
> can be done easier.)
> 





From richard_chamberlain at ntlworld.com  Sat Jun 24 10:26:47 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 24 Jun 2000 15:26:47 +0100
Subject: verifying date values
References: <PcQ45.16280$dF.645882@news1.rdc1.il.home.com>
Message-ID: <My355.514$F01.10888@news6-win.server.ntlworld.com>

Hi,

Marc Lemburg's mxDateTime module (do a search at www.vex.net/parnassus)
returns a RangeError if you create a date that doesn't exist. I've created a
checkDate method below which returns 0 or 1.

mxDateTime is an excellent module if you're going to be doing anything
complicated with dates or times.

Richard

import DateTime

def checkDate(y,m,d):
    "year,month,day"
    try:
        checkDate=DateTime.DateTime(y,m,d)
    except DateTime.RangeError:
        return 0
    else:
        return 1
def testDates(y,m,d):
    print '%d/%d/%d' %(y,m,d)
    if checkDate(y,m,d):
        print 'is ok'
    else:
        print 'is not valid'

def main():
    testDates(2000,3,10)
    testDates(2000,13,3)

if __name__=='__main__':
    main()


dsavitsk <dsavitsk at e-coli.net> wrote in message
news:PcQ45.16280$dF.645882 at news1.rdc1.il.home.com...
> is there a quick way to see if a string is a valid date? something like
>
> >>> isDate('5/22/1999')
> 1
>
> thanks
>
>
>




From xiarcexml at yahoo.com  Fri Jun 30 17:10:51 2000
From: xiarcexml at yahoo.com (Dave Williams)
Date: Fri, 30 Jun 2000 14:10:51 -0700 (PDT)
Subject: [Tkinter]: Where is a good place to learn how to use this...?
Message-ID: <20000630211051.11572.qmail@web5301.mail.yahoo.com>

I am fairly new to Python, and I have been reading a
book, and I have checked online for resources.  There
is a general "intro to" type off of the python.org
page, but it does not really cover things like the
equivalent of Java's my_textfield.getText() methods,
etc...

Where is a good (free maybe) online resource for
this?? And my "book" doesn't really seem to get into
Tkinter either, so is there one I should be reading
that does?

~Xiarcel


__________________________________________________
Do You Yahoo!?
Get Yahoo! Mail - Free email you can access from anywhere!
http://mail.yahoo.com/



From matt at null.net  Wed Jun 28 14:42:00 2000
From: matt at null.net (Matthew Schinckel)
Date: Wed, 28 Jun 2000 12:42:00 CST
Subject: Can anyone solve this?
References: <DD768E0BEA95D211B27A0008C74C86250A6B8E@BEKEXCH1>
Message-ID: <20000628124201.184982@210.9.21.253>

<rdudfield at my-deja.com> wrote in message
news:<8j737m$t5a$1 at nnrp1.deja.com>...
> Hi,
> 
> I've got a problem here which I just can't figure out.
> 
> a = [1,2,3] b = ["c",32] c = [1,2]
> 
> inputlist = [a,b,c]
> 
> Inputlist can have any number of lists in it. The lists in it can have
> any number of elements in them.
> 
> I need the code to give every possible combination of the lists in
> inputlist.
> 
> This is what the answer should be for the inputlist above.
> 
> outputlist = [[1, 'c', 1], [1, 'c', 2], [1, 32, 1], [1, 32, 2], [2,
> 'c', 1], [2, 'c', 2], [2, 32, 1], [2, 32, 2], [3, 'c', 1], [3, 'c',
> 2], [3, 32, 1], [3, 32, 2]] []

I did find something about combinatorics on starship yesterday, but I
can't remember where it was.





From wjdandreta at worldnet.att.net  Thu Jun 22 13:35:21 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 22 Jun 2000 17:35:21 GMT
Subject: Dictionary vs if-elif?
References: <UlS35.15101$Xx5.781176@bgtnsc06-news.ops.worldnet.att.net> <8itbu7$opv$1@slb6.atl.mindspring.net>
Message-ID: <t7s45.15493$C44.886800@bgtnsc05-news.ops.worldnet.att.net>

Hi Aahz,

Those are interesting alternatives. Thanks for the code samples.

Bill





From petersc at stud.ntnu.no  Sat Jun  3 17:42:25 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 03 Jun 2000 23:42:25 +0200
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it>
Message-ID: <39397BC1.570AF5C8@stud.ntnu.no>

Paolo Redaelli wrote:
> 
> but if I run a program that does x=float(x) where x = "0.000000E+00 "
> (being a string) here's the output.

Can you give some code? I tried the following:

=== BEGIN dumm.py ===
x = "0.000000E+00 "
x = float(x)
=== END dumm.py ===

it worked just fine. What version of Python are you using?

bye
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From bwarsaw at python.org  Fri Jun  2 09:46:14 2000
From: bwarsaw at python.org (bwarsaw at python.org)
Date: Fri, 2 Jun 2000 09:46:14 -0400 (EDT)
Subject: Status of List
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com>
	<3937B59C.3AFE081E@bam.com>
	<20000602153318.N469@xs4all.nl>
Message-ID: <14647.47782.21512.501187@anthem.python.org>

>>>>> "TW" == Thomas Wouters <thomas at xs4all.net> writes:

    >> Things have been very quite since yesterday from my point of
    >> view.  I'm just glad I'm not the only one.

    TW> Nope, it isn't you. I think it's Mailman. There was a sudden
    TW> burst of activity on the mailman-devel list, with a dozen or
    TW> so changes being checked in, and i think one of them broke the
    TW> usenet->mailman gateway. (The other way 'round seems to be
    TW> working fine.) I guess we'll have to wait for Barry to wake up
    TW> and spot it.

None of the recent spate of checkins have been propagated up to
python.org -- it could be a lurking bug in the semi 2.0b3 that it's
currently running.  But I had planned to update python.org sometime
today so if there's a problem with the current CVS snapshot, I want to
know about it.

Thomas, are you aware of a specific recent change that may have broken
Usenet->mail?  Email me directly if you want to discuss further.

    TW> [CC: Barry just in case he doesn't read the list via Mailman
    TW> ;-]

Naw, it's just that I don't have time to track python-list at all
these days :(

-Barry



From gmcm at hypernet.com  Sat Jun  3 17:35:20 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 03 Jun 2000 21:35:20 GMT
Subject: String.join revisited (URGENT for 1.6)
References: <28052000.5@sanctum.jae.ddns.org> <39346FAC.F2797EA5@cosc.canterbury.ac.nz> <8h320n$6fo$1@newshost.accu.uu.nl> <03062000.1@sanctum.jae.ddns.org> <8hb67j$71h$1@slb6.atl.mindspring.net> <03062000.2@sanctum.jae.ddns.org>
Message-ID: <39397a18$0$23541@wodc7nh7.news.uu.net>

Juergen A. Erhard <jae at ilk.de> wrote:

>The other objection was that [].join would be 'lists and tuples are
>*heterogeneous* containers. you can put *any* kind of Python 
object in
>them, not just strings' (as quoted from a mail by the effbot).
>
>Which is a non-argument... the exact same thing can be said 
against
>string.join(list, sep).  So why should list.join(sep) not simply do
>the same thing for non-strings that string.join(list, sep) does and
>raise an TypeError exception?[1]

Because it's a "internal state" error, not a "bad argument" error.

This whole friggin' thread is a *bad argument* error....

- Gordon



From quinn at riyal.ugcs.caltech.edu  Sun Jun 25 15:26:09 2000
From: quinn at riyal.ugcs.caltech.edu (Quinn Dunkan)
Date: 25 Jun 2000 19:26:09 GMT
Subject: Reloading modules
References: <8j5atk$ijs$1@newsg4.svr.pol.co.uk>
Message-ID: <slrn8lcnbp.drb.quinn@riyal.ugcs.caltech.edu>

On Sun, 25 Jun 2000 17:09:36 +0100, Duncan Smith
<buzzard at urubu.freeserve.co.uk> wrote:
>>>> from Graph import *
>.
>.
>.
>>>> reload(Graph)
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>TypeError: reload() argument must be module
>
>What do I need to do to successfully reload a module?  I'm currently wasting
>a lot of time re-starting PythonWin each time I want to test a changed
>module and I can't find the answer.  Thanks

'from x import ...' doesn't actually bring the module into the
namespace, just its contents.  You can get mostly what you want
with 

del sys.modules['Graph']
from Graph import *'

If it's just interactive testing, then feel free to use nasty tricks like
that.  But for real code, you should generally stay away from them.


From jkraska1 at san.rr.com  Wed Jun 21 01:34:09 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 21 Jun 2000 05:34:09 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au> <20000619112509.A20245@fangorn.csse.monash.edu.au>
Message-ID: <39505457.5223ADC5@san.rr.com>

> I saw a screenshot once of a vim session that used the ? (ascii 0273)
> character to display tabs. It was actually very visually appealing, and I was
> a little annoyed when I failed to get emacs to do the same thing.

I just spent the last 45 minutes vainly searching the web,
usenet, comp.editors, and the vim documentation for the
command to do this. I must not know the magic word. Oh
please, great Oz, enlighten me.


:)

C/


From scarblac-spamtrap at pino.selwerd.nl  Tue Jun  6 09:01:32 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 6 Jun 2000 13:01:32 GMT
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it>
Message-ID: <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl>

Paolo Redaelli wrote in comp.lang.python:
> Look at the this short program:
> 
> =======
> #!/usr/bin/env python
> 
> s = "0.000000E+00"
> print "Converting",s," into: ", float(s)
> print "Importing gtk"
> 
> import GDK
> from gtk import *
> from gnome.ui import *
> import GdkImlib
> print "Converting",s," into: ", float(s)
> =======
> Here's its output:
> =======
> [paolo at aldebaran paolo]$ python float.py
> Converting 0.000000E+00  into:  0.0
> Importing gtk
> Converting 0.000000E+00  into: 
> Traceback (innermost last):
>   File "float.py", line 11, in ?
>     print "Converting",s," into: ", float(s)
> ValueError: invalid literal for float(): 0.000000E+00
> =======
> 
> The only lesson I can extract from it is that GTK breaks float!

It still looks to me like gtk or gnome.ui is overwriting float, because of
your from module import * statements. What does "print repr(float)" say?

(I don't have gtk, can't test this myself).

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
This space intentionally left blank.


From jhe at webde-ag.de  Mon Jun 26 05:02:46 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Mon, 26 Jun 2000 11:02:46 +0200
Subject: How to change class instances behavior via reload?
References: <961962342.297785615@news.libertysurf.fr>
Message-ID: <8j764v$q5g$1@pollux.ip-plus.net>

"Jerome Quelin" <jerome.quelin at insalien.org> schrieb im Newsbeitrag
news:961962342.297785615 at news.libertysurf.fr...
> I would like to change the behavior of all the existing instances of a
> class when using the reload statement. I wonder if this is possible?

Laurent showed you how to do this automatically. Another technique I used to
DETECT whether the implementation of a class has changed is this:

class spam:
    def __init__(self):
        pass

    def checkReload(self):
        return self.__class__ != spam

Bye, J?rgen




From jkraska1 at san.rr.com  Sun Jun 25 00:46:44 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 25 Jun 2000 04:46:44 GMT
Subject: Python with braces.
References: <395683ae.299281703@nntp.interaccess.com>
Message-ID: <39558EFC.A2ECFA23@san.rr.com>

> Someone told me that there is an interpreter that can hanld Python
> using braces instead of indentation for defining blocks. Can anyone
> tell me where to find it?

Perhaps someone will, but why? The language you'd be using
wouldn't be Python.



C/


From dgoodger at bigfoot.com  Wed Jun  7 23:01:39 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Wed, 07 Jun 2000 23:01:39 -0400
Subject: (newbie) syntax problem compiling python scripts...
In-Reply-To: <393E471F.4AD05865@vanbesien.org>
References: <393E471F.4AD05865@vanbesien.org>
Message-ID: <B56484D3.4808%dgoodger@bigfoot.com>

on 2000-06-07 08:59, Krist van Besien (krist at vanbesien.org) wrote:
> What's going on? How do I debug this?

Maybe check for mixed tabs & spaces? Could be the indentation looks ok, but
really isn't... Just a guess.



From emile at fenx.com  Wed Jun 21 20:43:45 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 21 Jun 2000 17:43:45 -0700
Subject: Speed with Python
References: <QH645.94$QA1.13929@news.uswest.net>
Message-ID: <09e101bfdbe3$0255cda0$1906a8c0@fc.fenx.com>

I deal mostly with business applications, and when we got
started the systems could read/change/write about 100 times
a minute.  I still support a lot of that software, such
that whatever speed issues existed have been more than
adequately addressed by advances in hardware.  Speed
from that perspective has always been a function of
customer budget.

want-faster-buy-bigger-better-ly y'rs,

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: "Jeff Massung" <jmassung at magpiesystems.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, June 21, 2000 10:07 AM
Subject: Speed with Python


> Many threads with this kind of name have popped up,
comparing the speed of
> Python to C, Java, etc. and trying to determine how fast
some of the
> algorithms in Python are.
>
> My question however, is how many of you who are Python
gurus/junkies (no
> offence, I myself am a C and Forth junkie) :) use Python
for speed intensive
> processing (not graphics, but databases and such)? Does
efficiency of
> developement outweight the eventual speed of use?
>
> Cheers,
> Jeff
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From paul at prescod.net  Wed Jun 28 17:19:34 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 28 Jun 2000 14:19:34 -0700
Subject: Pickle for C++?
References: <395A1F7E.4112B2A3@muc.das-werk.de>
Message-ID: <395A6BE6.7E350345@prescod.net>

> I have a Python class that holds a lot of data.
> And I now want to send these datas over a TCP - connection via
> pickle and sockets. It works fine. But now I want
> that an application written in C++ is able to unpickle the data-class
> too and write it's content into an equivalent C++ - class/struct.

XML-RPC, SOAP and CORBA are designed specifically for these situations.
Throughput will not be as good (especially with the first two!) as with
Pickle.

Also consider "repr" rather than pickle. Again, throughput is not as
good but parsing is much easier.

-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski



From jkraska1 at san.rr.com  Mon Jun  5 01:54:50 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 05 Jun 2000 05:54:50 GMT
Subject: "dlist" -- beta is ready
Message-ID: <393B4144.4C1DC04C@san.rr.com>

I have a functioning implementation of my "dlist":
an ADT which is a vector similar to python's list,
with the addition of a buffer zone on the front as
well as the back. This gives the vector O(1)+k
performance on add/remove from both the head and
the tail, while maintaining constant time for random
access (with a small additional constant cost of
tracking a virtual index).

The implementation seems stable, and is passing
a small test suite I've written, so I think it works.

There are three items which I will finish soon
(perhaps sooner if someone asks for them):

1. sorting; probably shouldn't be too hard to add
   in Guido's/Tim's binarysort/sample sort code.

2. reversing.

3. invariant structures, in conjunction with 1.

This is a VC98 project, intended to compile/run
on windows platforms, although it could with some
small mods work on unix platforms with equal ease.

Simply request the source, and I'll send you a
copy of the whole project dir, including the small
test suite I have.

C/


From donn at u.washington.edu  Thu Jun  8 15:13:57 2000
From: donn at u.washington.edu (Donn Cave)
Date: 8 Jun 2000 19:13:57 GMT
Subject: Why can't I restore my redirected stdout?
References: <8hoo05$fc$1@news5.jaring.my>
Message-ID: <8hor9l$34qa$1@nntp6.u.washington.edu>

Quoth "Raymond Ng Tong Leng" <think2 at pd.jaring.my>:
| I am trying to implement a verbose and quiet mode in my Python script. So I
| wrote the following class:
|
| class Quiet:
|     def write(self, data):
|         pass
|
| Then I set sys.stdout to Quiet by:
|
| >>> sys.stdout = Quiet()
| >>> print 'hello'
| >>>
|
| 'hello' doesn't come out which is what I want. But, when I tried to restore
| sys.stdout's original settings:
|
| >>> sys.stdout = sys.__stdout__
| >>> print 'hello'
| >>>
|
| 'hello' is not printed either! How do you restore sys.stdout to its original
| settings? Also, I was wondering, are there other ways to suppress
| sys.stdout? I'm just curious whether I could suppress sys.stdout without
| redirecting to a class. It seems to me that I should be able to redirect
| sys.stdout to a null device or something. Thanks!

I'm not familiar with sys.__stdout__, maybe someone else knows that one.

If you're on UNIX, you can dup() the file descriptors around underneath
stdout, for this effect.  I append an example.

	Donn Cave, donn at u.washington.edu
----------------------------------------
import os
import sys

class Outs:
	def __init__(self):
		self.streams = []
	def setdevice(self, filename):
		#  Open an existing file, like "/dev/null"
		fd = os.open(filename, os.O_WRONLY)
		self.setfd(fd)
	def setfile(self, filename):
		#  Open a new file.
		fd = os.open(filename, os.O_WRONLY|os.O_CREAT, 0660);
		self.setfd(fd)
	def setfd(self, fd):
		ofd = os.dup(1)	      #  Save old stream on new unit.
		self.streams.append(ofd)
		sys.stdout.flush()    #  Buffered data goes to old stream.
		os.dup2(fd, 1)        #  Open unit 1 on new stream.
		os.close(fd)          #  Close other unit (look out, caller.)
	def unset(self):
		#  Restore previous output stream.
		if self.streams:
			sys.stdout.flush()
			os.dup2(self.streams[-1], 1)
			os.close(self.streams[-1])
			del self.streams[-1]


From egbert at bork.demon.nl  Mon Jun 26 07:53:37 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Mon, 26 Jun 2000 13:53:37 +0200
Subject: The automagic of Tkinter
Message-ID: <20000626135336.A8180@bork>

Good afternoon,
The following small script does what I intended it to do:
it produces a button, and that button dissappears when i click on it.

    from Tkinter import *
    class Knop(Button):
       def __init__(self,tekst="KNOPJE"):
           Button.__init__(self,text=tekst,command=self.quit)
           self.pack(side=TOP,fill=BOTH,expand=YES)
    root=Tk()
    k=Knop()
    root.mainloop()

However when I replace the last three lines with only:
    Knop().mainloop()
then the script still works.

This leaves me with a number of questions (sorry, richard):
- if 'root=Tk()' is dispensable, then what is its builtin replacement
  and what does that replacement do
- up till now I saw a class object as a template, that has to be 
  instantiated in order to be of any use. However here it seems to 
  lead its own life.
- probably 'mainloop()' plays a crucial role here, but what are all
  its functions ?  What is the difference between 
  'root.mainloop()' and 'Knop().mainloop()'
I promised Richard to pose only one question per post, but these 
questions really are one big question about the Tkinter mechanism.
egbert

-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From ksks at sprynet.com  Sat Jun 10 17:26:15 2000
From: ksks at sprynet.com (Kent)
Date: Sat, 10 Jun 2000 16:26:15 -0500
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <8hqhhi$396$1@mail.pl.unisys.com>
Message-ID: <8huc4t$gu7$1@slb2.atl.mindspring.net>

Clive Tooth <clive at pisquaredoversix.force9.co.uk> wrote in message
news:8hqhhi$396$1 at mail.pl.unisys.com...
> Robert Israel wrote in message <8houlu$j9s$1 at nntp.itservices.ubc.ca>...
>
> >[...]
> >There are two different algorithms here, for two different problems.
> >
> >The BBP algorithm can find a hexadecimal (or more generally, base
> >2^k) digit of pi, faster than any known algorithm that would find
> >all the digits up to this one.  This is the exciting one.
>
> Although BBP do mention that "This algorithm is, by a factor of
> log(log(log(n))), asymptotically slower than the fastest known algorithms
> for generating the n-th digit by generating all of the first n digits of
> log(2) or pi."

I'm not getting this. It's "faster than any known algorithm that would find
all the digits up to this one", but it's "asymptotically slower than the
fastest known algorithms
 for generating the n-th digit by generating all of the first n digits of
 log(2) or pi". How can it be both faster and asymptotically slower? Does
this mean "faster for small enough values of n, but slower for large enough
values"? If so, about how big are "small enough" and "large enough"?

too lazy to follow the links and try to figure it out for himself,
Kent.





From mwh21 at cam.ac.uk  Wed Jun  7 19:27:31 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 08 Jun 2000 00:27:31 +0100
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl> <393D1155.D067A821@decrc.abb.de> <m33dmqygdj.fsf@atrus.jesus.cam.ac.uk> <slrn8jt5ck.779.jepler.lnk@potty.housenet>
Message-ID: <m3bt1dvzq4.fsf@atrus.jesus.cam.ac.uk>

jepler.lnk at lnk.ispi.net (jepler epler) writes:

> On 06 Jun 2000 16:32:40 +0100, Michael Hudson
>  <mwh21 at cam.ac.uk> wrote:
> >My money's on so bizzarre interaction with libc, myself.
> 
> I'm guessing that 'import gnome.ui' sets the locale, including the
> numeric format:
> 	$ env LC_NUMERIC=en_US python -c 'import gnome.ui; print float("1.1")'
> 	1.1
> 	$ env LC_NUMERIC=de_DE python -c 'import gnome.ui; print float("1.1")'
> 	Traceback (innermost last):
> 	  File "<string>", line 1, in ?
> 	Value Error: invalid literal for float(): 1.1

$ env LC_NUMERIC=de_DE python -c 'import gnome.ui; print float("1.1")'

Gdk-WARNING **: shmget failed!
1.1

Versions?

Cheers,
M.

-- 
  Programming languages should be designed not by piling feature on
  top of  feature, but by removing the  weaknesses and restrictions
  that make the additional features appear necessary.
               -- Revised(5) Report on the Algorithmic Language Scheme


From parkw at better.net  Thu Jun 15 13:43:57 2000
From: parkw at better.net (William Park)
Date: Thu, 15 Jun 2000 13:43:57 -0400
Subject: Where are the math functions?
In-Reply-To: <memo.20000614223200.31901D@digitig.compulink.co.uk>; from digitig@cix.co.uk on Wed, Jun 14, 2000 at 10:32:00PM +0100
References: <owen-97F32C.10015712062000@news.u.washington.edu> <memo.20000614223200.31901D@digitig.compulink.co.uk>
Message-ID: <20000615134357.A1087@better.net>

On Wed, Jun 14, 2000 at 10:32:00PM +0100, Tim Rowe wrote:
> In article <owen-97F32C.10015712062000 at news.u.washington.edu>, 
> owen at astro.washington.edu (Russell E. Owen) wrote:
> 
> > In article <slrn8k8dfv.tu.hzhu at mars.localdomain>, 
> > hzhu at users.sourceforge.net wrote:
> > 
> > >I'm quite puzzled by not being able to find most of the common special
> > >functions, such as (incomplete) gamma, beta and bessel and erf.
> > >
> > >This arrises from effort to provide statistical functions for the MatPy
> > >package.  Many stat functions require incomplete gamma/beta and some
> > >variants.
> > >...
> > 
> > The random module (which is available on all platforms) offers beta and 
> > gamma functions. I can't comment on incomplete gamma, bessel and erf. I 
> > found the information about the random module in "Python Essential 
> > Reference" by David Beazley; the index lists beta and gamma, but not 
> > the others.
> 
> No it doesn't. It includes the beta and gamma /distributions/. These are 
> related to, but not the same as, the beta and gamma /functions/.
> 
> When I wanted the gamma function I just ported code from Numerical Recipes 
> in C, which only took a couple of minutes (longer to test, but I'd have 
> spent that time if I'd just typed in the C). Probably still be a copyright 
> violation to post it, though :-(

In fact, GNU math.h includes many higher functions like
    j0, j1, j2, y0, y1, y2
    lngamma, gamma, ...
    erfc, erf, ...
    cbrt,
but Python 'math' module doesn't include them.  So, you have to create a
new module of your own.  

If you need even more higher functions, there are Cephes and other
libraries around.  Try searching <freshmeat.net>.  Or, port the relevant
routine from "Numerical Recipes in C".

--William



From ben at co.and.co  Sun Jun 11 09:26:03 2000
From: ben at co.and.co (ben at co.and.co)
Date: Sun, 11 Jun 2000 13:26:03 GMT
Subject: spaces in filenames
References: <000201bfd350$0d818660$f6a2143f@tim>
Message-ID: <LrM05.166678$Qp2.427323@afrodite.telenet-ops.be>

Tim Peters <tim_one at email.msn.com> wrote:
> box.  And if you stick-- as I suspect the overwhelming majority of Windows
> users do --purely to GUI apps, spaces in paths simply don't bite you.

Try opening a "some bitmap with a space.bmp" in Paint, by
double-clicking (Win95/98). And try explaining people why this doesn't
work, when you're teaching an introductory course to Win95/98.

Greetings,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From daniel at visionart.com  Fri Jun 16 04:36:31 2000
From: daniel at visionart.com (Daniel Kramer)
Date: Fri, 16 Jun 2000 08:36:31 GMT
Subject: more on python and GUI speed
Message-ID: <jGl25.2323$3E6.42734@news1.alsv1.occa.home.com>

I know I am beating this subject to death but I moved over to python from
tcl to improve my programming skills (which are minimal), learn oop, and
still write gui programs with ease.

I have found the gui solutions to be a bit slow compared to tcl/tk.  Today I
found something interesting.. I put my computer under heavy load and ran a
simple tcl/tk app, python/tkinter app, and a wxPython app.. all were simple
drawing a line on a canvas type apps.

    The tkinter and wxPython apps were very choppy when the cpu was under
load from another process.. where the tcl/tk app saw almost no performance
hit at all.. it was still just as fluid as when my computer's cpu was idle.

Anyway.. I am leaning toward going back to tcl/tk and just learning [Incr
Tcl] to fill the oop gap.

-Daniel




From stephen at cerebralmaelstrom.com  Wed Jun 28 14:31:03 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 28 Jun 2000 18:31:03 GMT
Subject: list.pop([i]) and list.remove(x) w/ for loops
References: <8jdfb6$ugh$1@netnews.upenn.edu>
Message-ID: <Hvr65.97855$WS3.833210@typhoon.we.rr.com>

When you are iterating over a sequence using the for loop, you can't modify
the origional loop, otherwise you'll screw things up..

[a,b,c,d] -- when you start, you are at the first index.. which points to
'a'. When you remove that item, you don't actually increase your current
index..even though your list is now [b,c,d]... so 'for' will go to the next
index... which is 2.. which points to 'c', instead of 'b'

To get around this, iterate over a copy of thes equence.

for o in olist[:]:
    if(o.type == a):
        olist.remove(o)

The [:] slice makes a copy of the sequence.

--Stephen

David White <dwhite2 at blue.seas.upenn.edu> wrote in message
news:8jdfb6$ugh$1 at netnews.upenn.edu...
> I was getting some unexpected behavior when using list.remove(x),
wondering if
> someone can tell me what's going on.
>
> I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].  I wan't
to
> delete all the a objects out of the list while preserving the order of the
b
> objects.  My code looks like this:
>
> for o in olist:
>     if (o.type == a):
>        olist.remove(o)
>
> Now for some reason I get a left over a object that wasn't removed.
Anyone
> know why?  Using python 1.5.2.
>
> thanks,
> david




From barryhart at my-deja.com  Tue Jun 20 08:11:16 2000
From: barryhart at my-deja.com (barryhart at my-deja.com)
Date: Tue, 20 Jun 2000 12:11:16 GMT
Subject: Strange problem with PythonCOM
References: <8imq9s$i0u$1@nnrp1.deja.com>
Message-ID: <8inn0v$3o2$1@nnrp1.deja.com>

In article <8imq9s$i0u$1 at nnrp1.deja.com>,
  barryhart at my-deja.com wrote:
> I'm trying to use PythonCOM to call into TLBINF32.dll (the Microsoft
> type library info DLL), and I'm getting a strange error. My call to
> create the object - Dispatch("TLI.TLIApplication") - fails with "No
> such interface supported."

I found a partial answer to this myself - it appears that the
TLIApplication class does not support IDispatch, even though its main
interface derives from IDispatch! I determined this when I got the same
error creating the object from C++. This looks like an egregious error
to me!

There's still the question of how this works in VB. Maybe they're doing
something "questionable" under the hood that allows this to work.

I'm still interested in a Python-only solution to this. My workaround
has been to write a C++ COM component that creates the object for me
and does the work for me (although VB would've worked as well).

-----------------------------
Barry Hart


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


From stephen at cerebralmaelstrom.com  Sat Jun 24 04:27:37 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 24 Jun 2000 08:27:37 GMT
Subject: need insight...
References: <3954671F.51E53AE6@mindspring.com>
Message-ID: <Zh_45.85985$WS3.752637@typhoon.we.rr.com>

    The problem is that you really do not want to change the sequence you're
using in 'for' -- the documentation, if memory serves, shows an example of
precisely why. You'll want to make a copy of it to iterate through.. like;

for item in cwd[:]:
    ...
    cwd.remove(item)

Like that..see? The [:] slice copies the whole sequence.

--Stephen

David Broadwell <dbroadwell at mindspring.com> wrote in message
news:3954671F.51E53AE6 at mindspring.com...
> The code to follow, to what i know, is behaving erratically. it is
> erratic in that in the section after the print 'checking... statement
> the  'for item in directorylist:' seems to be skipping over entries in
> it's list for no apparent reason... (note: what is skipps changes if the
> list it is given chages, but from what i can tell, it always skips about
> half of the things on the list. i have snipped out as much of the code
> as i can and still leave it psudo functional..
>
> I am seeking help in some capacity, and if i have commited some grevious
> programming error, enough information to avoid it in the future..
>
> If it dosen't demonstate the same problem i will post the code entire
> (it is about 60 lines)
>
> thanks!
>
> # Code follows..
>
> import os
> import sys
>
> filetype = 'm3u'
> directorylist = []
>
> def doit():
>     cwd = []
>     print 'reading current directory'
>     cwd = os.listdir('.')
>     print 'building directorylist'
>     for item in cwd:
>         if (os.path.isdir(item)):
>            directorylist.append(item)
>            cwd.remove(item)
>     del cwd
>     print 'Checking for subdirectories to check.'
>     print directorylist
>     for item in directorylist:
>         print 'subdirectory ' + item + ' found, scanning.'
>         os.chdir(item)
>         subdir = []
>         subdir = os.listdir('.')
>         print subdir
>         for thing in subdir:
>             print thing,
>             if (thing[(len(thing)-3):] == filetype):
>                print 'Yup.'
>                # join item(path) to file
>                # add to filelist
>             else:
>                print 'Nope.'
>                subdir.remove(thing)
>         print subdir
>         del subdir
>         os.chdir('..')
>
> doit()
>
> --
>         ~We're all fallen angels who've forgotten how to fly~
>         Remember how: http://www.cyber-action.com/oed/dbroadwell.html
> --




From M.Strasser at myrealbox.com  Thu Jun  1 19:53:56 2000
From: M.Strasser at myrealbox.com (Michael Strasser)
Date: Thu, 01 Jun 2000 23:53:56 GMT
Subject: Installing numpy
References: <3910b8e1.0@news2.cluster1.telinco.net> <3911776F.25845625@oreilly.com>
Message-ID: <8h6t2h$lvj$1@nnrp1.deja.com>

In article <3911776F.25845625 at oreilly.com>,
  "Stephen R. Figgins" <fig at oreilly.com> wrote:
> I don't have a C compiler installed on my Windows system, so I haven't
> ever tried compiling the source code.
>
> Was there some reason you didn't want to use the precompiled version?

Yes. There is no zip file for NumPy 15.3, only a source tarball.

--
Michael Strasser
Brisbane, Australia


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


From think2 at pd.jaring.my  Thu Jun  8 13:52:52 2000
From: think2 at pd.jaring.my (Raymond Ng Tong Leng)
Date: Fri, 9 Jun 2000 01:52:52 +0800
Subject: Why can't I restore my redirected stdout?
Message-ID: <8hoo05$fc$1@news5.jaring.my>

I am trying to implement a verbose and quiet mode in my Python script. So I
wrote the following class:

class Quiet:
    def write(self, data):
        pass

Then I set sys.stdout to Quiet by:

>>> sys.stdout = Quiet()
>>> print 'hello'
>>>

'hello' doesn't come out which is what I want. But, when I tried to restore
sys.stdout's original settings:

>>> sys.stdout = sys.__stdout__
>>> print 'hello'
>>>

'hello' is not printed either! How do you restore sys.stdout to its original
settings? Also, I was wondering, are there other ways to suppress
sys.stdout? I'm just curious whether I could suppress sys.stdout without
redirecting to a class. It seems to me that I should be able to redirect
sys.stdout to a null device or something. Thanks!












From ullrich at math.okstate.edu  Sun Jun 18 11:44:13 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 18 Jun 2000 15:44:13 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <01bfd93c$66c03180$2ace8ad1@daves-dell>


William Dandreta <wjdandreta at worldnet.att.net> wrote in article
<sy535.11434$Xx5.530580 at bgtnsc06-news.ops.worldnet.att.net>...
> I recently read a message that suggested that using regular expressions
> might be faster than the find function.

	I would have guessed that if anything find would be faster, although
I certainly could be wrong. Did someone say that using a regular expression
would be _faster_, or did they actually say it was more _powerful_ or some 
such?

> I've heard the term regular
> expression used from time to time but I don't know what it means. Could
> someone give me a simple explanation?

	Regular expressions are a way of finding very general sorts of 
"patterns" in text - you could use a regular expression to search for
something like "any number of digits, followed by at least one letter,
only at the end of a line." See the docs on the re module for details.

	(There's at least one whole _book_ on regular expressions,
"Mastering Regular Expressions" or something.)

> Bill
> 
> 
> 


From maytagrepairman at yahoo.com  Wed Jun 21 18:06:46 2000
From: maytagrepairman at yahoo.com (Maytag Repairman)
Date: Wed, 21 Jun 2000 22:06:46 GMT
Subject: How to control Pmw.ScrolledText scrollbar position?
Message-ID: <8ire9f$rs4$1@nnrp1.deja.com>

I have this program that I've written which uses Pmw.ScrolledText.
There are buttons on my GUI which take data from entry fields and
insert, replace, and append data in a data structure.  The ScrolledText
is cleared and the updated data structure is displayed in a user-
friendly format.  Unfortunately, the scrollbar always ends up at the
top after this operation but the area of most interest to the user is
the point where the operation on the data structure took place.  I
can't figure out how to control the position of the scrollbar to move
it to the area of interest.

Yes, I do realize that I am doing this rather inefficiently by re-
creating the full text each time but it works well enough for now.

--
Rob Cupples


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


From cut_me_out at hotmail.com  Mon Jun 26 20:48:35 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Jun 2000 20:48:35 -0400
Subject: reload()'ing modules and how it affects references
References: <xjvn1k8grd3.fsf@gjallarhorn.ifi.uio.no>
Message-ID: <etdhfaghrto.fsf@w20-575-109.mit.edu>

Hi, Igor.

I posted something recently that does multiple reloads of all the
modules you specify in order to ensure that they are finally reloaded in
an order that respects their dependancies.  For each of the modules M,
it clears all the modules you've specified out of sys.modules, reloads
M, then looks at which of the other modules got reloaded to see what M
depends on.  After that, it does a crude topological sort on the
dependancy graph to decide what order they should be loaded in.  It
seemed to work on the cursory tests I made of it, but it would only help
for objects that actually get defined during one of the reloads.  For
instance, you would probably still have references to the old versions
of the classes in the __main__ namespace.

> X-news-hint: never quarrel with #:Erik

Erik Naggum?  That guy's a barrel of laughs. :)

Alex.


From cut_me_out at hotmail.com  Sat Jun 10 10:41:11 2000
From: cut_me_out at hotmail.com (Alex)
Date: 10 Jun 2000 10:41:11 -0400
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no> <m3hfb2vkdq.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <etdog59egzc.fsf@w20-575-109.mit.edu>

> A trivial example: wouldn't it be nice if when you hit C-a the point
> got moved to the start of the command line, rather than the leftmost
> column of the display?

You can already do this, using C-c C-a.  It doesn't work when you drop
in to pdb, though.

Noone-needs-to-know-elisp-because-all-the-trivial-stuff-has-been-
done-before-<wink>'ly yrs

P.S.  I have been thinking about getting emacs to interact more closely
with a python process, too.  There's tonnes of cool stuff you could do
if that were possible.

Alex.


From sdhyok at email.unc.edu  Wed Jun 14 17:48:45 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 17:48:45 -0400
Subject: How to prevent illegal definition of a variable in objects?
Message-ID: <8i8ujt$hop$1@news2.isis.unc.edu>

How can I prevent illegal definition of a variable to an object?
For instance,

class A:
    __init__(self):
            self.x = 1

    __setattr__(self, name, value):
            if name == "x":
                    self.x = value
            else:
                raise AttributeException

    __getattr__(self, name):
            if name == "x":
                return self.x
            else:
                raise AttributeException

>>> a = A()
>>> a.x = 10
>>> a.y = 20 # I want to raise an AttributeException.

Peter




From cut_me_out at hotmail.com  Tue Jun  6 22:03:52 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 22:03:52 -0400
Subject: Some questions about C extensions
Message-ID: <etdu2f69rhz.fsf@w20-575-109.mit.edu>

Hi.  I'm trying to write some C extensions, and have run across a couple
of mysterious things.  I was wondering whether anyone could explain
them:

-- Deallocation of memory for strings returned by python functions:
   Do I have to do it, or is it somehow taken care of?  If I have
   something like 

      if (!PyArg_ParseTuple (args, "s", &sequence)) {
           return NULL;}

   do I have to do a free (sequence); at some point?  I would have
   thought so, but it seems to be giving me a Seg fault.

-- Reference counts for elements of brand new tuples.  Why do I have to
   do a Py_INCREF on an object before I assign it to a new tuple with
   PyTuple_SetItem?  Since it seems as though you have to do a Py_INCREF
   every time anyway, is there a function like PyTuple_SetItem that will
   do it for you?  The comment in the C API manual about this is very
   cryptic, BTW.

Alex.


From thomas at xs4all.net  Mon Jun 19 15:18:12 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 19 Jun 2000 21:18:12 +0200
Subject: Why whitespace denotation of blocks doesn't work.
In-Reply-To: <394e5f98.844462828@nntp.interaccess.com>; from olczyk@interaccess.com on Mon, Jun 19, 2000 at 06:30:18PM +0000
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com> <39520785.690843718@nntp.interaccess.com> <3p0sksgnqhso5dsp1cmf5srffbeh42ov0g@4ax.com> <394e5f98.844462828@nntp.interaccess.com>
Message-ID: <20000619211812.C26436@xs4all.nl>

On Mon, Jun 19, 2000 at 06:30:18PM +0000, Thaddeus L. Olczyk wrote:

[Python's (metacomputational) speed]

> >The code can be typed in much faster; its SLOWER to type in the
> >braces.
> >
> Are you saying that you should choose a programming language based on
> how long it takes to type in things? If so then you should be using
> perl not python. The experts all agree that it is much faster to type
> in perl than python.

Not true. It may be possible to express certain ideas, operations,
algorithms in *less characters*, but that doesn't mean it's faster to type.
It may well be faster to type long words than to find the keys for $@{}[]->
;) And besides, like Eric Raymond said in his Linux Journal article (and
probably a lot of other times) the cool thing about Python is that you can
think it up almost as fast as you can type. (I agree, and I type pretty
fast, thanx to a good highschool teacher ;) This does not hold ture for
Perl (except perhaps for real hard-core hackers) where you constantly have
to think about how to write things.

> Me, I prefer a language which helps me write good code. Faster to type
> or slower to type doesn't make much difference to me. Of course if you
> want something that is faster to type, you can always go to perl.

I think the original author intended to say 'faster to write', instead of
'faster to type'. Writing implies typing (on computers, anyway, for those of
us who communicatie with the computer via a keyboard) but entails more than
just typing in the letters (or special characters ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From dsavitsk at e-coli.net  Mon Jun 19 13:28:58 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 19 Jun 2000 17:28:58 GMT
Subject: accessing access from Python on windows
References: <394E51D1.71382D39@nrc.ca>
Message-ID: <uLs35.8711$dF.408821@news1.rdc1.il.home.com>

i am trying to figure out why you would use odbc for access, unless you
planned to change your db and your os at some point.  DAO/Jet is certainly
easier, and ADO/OLEDB is more supported by M$ and will translate better to
other databases on win32.
from the one line of code it looks like you didn't use makepy yet. (Tools |
COM Makepy utility)
for ADO (ODBC/OLEDB) you will be better off using DSN-Less connection
strings as they are more portable, easier, more stable, and they run faster
(not a lot faster, but usually a little).

here's my $.02 on how easy dao is ...

import win32com.client
dbEngine = win32com.client.Dispatch(r"DAO.DBEngine.36")
db = dbEngine.OpenDatabase(DBName)
rs = db.OpenRecordset(TblName)

you can also access odbc using windll from www.nightmare.com

doug

"Marcin Kolbuszewski" <"Marcin.Kolbuszewski "@nrc.ca> wrote in message
news:394E51D1.71382D39 at nrc.ca...
> Hi there!
>
> I am running python 1.5.2 on Windows NT and am trying to
> access an Access database through ODBC. No lick though...:
>
> Traceback (innermost last):
>   File "C:\PROGRA~1\Python\Tools\idle\ScriptBinding.py", line 131, in
> run_module_event
>     execfile(filename, mod.__dict__)
>   File "D:\python\toc.py", line 3, in ?
>     import dbi, odbc
> ImportError: No module named dbi
>
> Where may I get the dbi and odbc modules?
>
> marcin Kolbuszewski




From thomas at xs4all.net  Wed Jun 21 02:26:22 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 21 Jun 2000 08:26:22 +0200
Subject: Python sockets/select
In-Reply-To: <8ipl0u$9l3$0@216.39.151.169>; from donn@oz.net on Wed, Jun 21, 2000 at 05:49:18AM +0000
References: <8ip6kt$1lc$1@bob.news.rcn.net> <8ipl0u$9l3$0@216.39.151.169>
Message-ID: <20000621082622.G26436@xs4all.nl>

On Wed, Jun 21, 2000 at 05:49:18AM +0000, Donn Cave wrote:
> Quoth "David Allen" <s2mdalle at titan.vcu.edu>:

> | When I select() sockets for reading, writing, and error conditions,
> | there are sockets (usually connected to clients who have already
> | quit or died unexpectedly) that don't show up in the error queue,
> | but continually flag select() as being readable.  When I read them,
> | of course, '' comes back.

> | For some reason, sometimes '' comes in for valid clients, so I 
> | can't just disconnect a client because I get that from them.  Is
> | there a way to detect which clients are alive and which aren't?
> | What causes a client to send '' when I didn't tell it to, according
> | to select()?  (The clients are also written in python)

> When you get 0 bytes back from a read, that's EOF (End Of File.)
> As you apparently surmised.  That is indeed a readable event, at
> least the first time.  I'd say behavior after the first 0 byte
> read is probably undefined, but anyway it's not going to do anyone
> any good.

Actually, there are some possibilities to get 0 bytes on a return without
hitting EOF: if you are using a wrapper object that possibly eats up some
bytes (for protocol negotiation, for instance.) Telnetlib is a good example
of this:

"""
It is possible to pass a Telnet object to select.select() in order to wait
until more data is available.  Note that in this case, read_eager() may
return '' even if there was data on the socket, because the protocol
negotiation may have eaten the data.  This is why EOFError is needed in some
cases to distinguish between "no data" and "connection closed" (since the
socket also appears ready for reading when it is closed).
"""

If you do something similar in your own sockets, this may well be the case.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From smigowsky at dspace.de  Tue Jun 20 09:16:42 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Tue, 20 Jun 2000 14:16:42 +0100
Subject: win32all-132 / Word Automation
Message-ID: <326536345498D311B3BC00105A39802A07473D@newsintern.dspace.de>

Hi,

probably there is a mix up between the generated wrappers.
In versions newer than 129, the parameter defines for calls
changed from tuple values to ints. 

You could try to delete the gen-py generated file and then do
the test again. The object model of word clearly states that
Variables.Item returns a dispatch pointer to a Variable object.

  Stefan 

-----Original Message-----
From: Mark Hammond [mailto:MarkH at ActiveState.com]
Sent: Tuesday, June 20, 2000 12:52 PM
To: python-list at python.org
Subject: Re: win32all-132 / Word Automation


"Nikolai Kirsebom" <nikolai.kirsebom at siemens.no> wrote in message
news:394f269c.580059221 at news.mch.sni.de...
> Mark,
> Thanks for your quick response.  It might be obvious ... but not
> really to me.

Hrm - not to me either!


> >>> print doc.Variables.Item(1)
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File
> "win32com\gen_py\00020905-0000-0000-C000-000000000046x1033x8x0.py",
> line 11402, in __str__
>     return str(apply( self.__call__, args))

Ahhh!  for some bizarre reason, it appears the print itself triggers this.
(Note that the exception is in the __str__ method, which is being used to
print it.

What does:
print repr( doc.Variables.Item(1) )

yield?

Mark.



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



From android at micron.net  Thu Jun 15 08:55:57 2000
From: android at micron.net (Peter Immarco)
Date: Thu, 15 Jun 2000 08:55:57 -0400
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com>
Message-ID: <3948D25D.7C83CD6@micron.net>

Courageous,

Tried Lisp for over 8 months, never did learn to like it.  I recognize
it's strong points but its just "not for me".  What's the LOC acronym
stand for?

Peter.

Courageous wrote:

> Re: python vs c++
>
> > However my C++ friends are badgering the crap out of me...
>
> Most of your C++ friends would be picking their tongues
> up off the floor if they saw the LOC of python programs
> in comparison to their C++ counterparts.
>
> Python may be too slow for some tasks, but in the area
> of programmer productivity, very little beats it, and
> arguably the environments that do (e.g., lisp) have too
> many other negatives associated with them.
>
> C/



From akuchlin at mems-exchange.org  Tue Jun  6 17:45:10 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 06 Jun 2000 17:45:10 -0400
Subject: Small bug in zlib module
References: <001601bfcfca$22907860$56a21c3f@versa> <393D29DE.FE761725@stud.ntnu.no>
Message-ID: <3dsnuqtrfd.fsf@amarok.cnri.reston.va.us>

Peter Schneider-Kamp <petersc at stud.ntnu.no> writes:
> Greg Chapman wrote:
> > I assume PyZlib_decompress should have a check for an empty input string and
> > (if found) should generate an appropriate exception.
> 
> On the current CVS version it immediately (as far as I can see without
> looping) returns with an error.

A puzzling question.  For decompression objects, feeding in an empty
string should certainly *not* produce an error, since it's perfectly
reasonable, if odd, to ask for more decompressed data without
providing more compressed material. It's not obvious whether this is
an error for the decompress function which takes a single string;
you're not providing any compressed data, so you get no decompressed
data.  But realistically, if you're passing "", it's probably a bug.

Similarly, if you parsed the Python source file "" into an AST, is it
an error or would you expect to get None or some similar vacuous tree?

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
If every man is supposed to think of sex once every nine minutes, what on
earth does he think of in the other eight?
  -- Peter Greenaway, introductory quotation from _8 1/2 Women_ (1999)




From jon_mclin at attglobal.net  Fri Jun  9 19:57:54 2000
From: jon_mclin at attglobal.net (Jon McLin)
Date: Fri, 09 Jun 2000 17:57:54 -0600
Subject: Programming Language Comparison
Message-ID: <39418482.B4E4A778@attglobal.net>

Apologies if this has been posted here before.

Found this very interesting study comparing (with a specific test case) many
characteristics of several languages, both system and scripting.
Includes C, C++, Java, Perl, Python, Tcl, and Rexx.  Python did well.

 http://wwwipd.ira.uka.de/%7Eprechelt/documents/jccpp_tr.pdf






From jwbnews at scandaroon.com  Fri Jun 16 16:27:51 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 16 Jun 2000 13:27:51 -0700
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com> <394A7763.DE2A771D@infocentr.ru>
Message-ID: <jwbnews-21EBF8.13275116062000@news.olympus.net>

In article <394A7763.DE2A771D at infocentr.ru>, Aleksey Kishkin 
<ax at infocentr.ru> wrote:

> Akira Kiyomiya wrote:
> > 
> > I wonder if Python runs faster than Perl.
> > 
> > Could someone give me some comparison and contrast with some languages 
> > such
> > as Perl, Tcl, JavaScript, etc?
> > 
> > I am just curious.
> > 
> > Thanks
> > 
> > Akria
> 
> I did. I found the programm (sieve) for java and rewrote this
> programm to all languages which I had at that moment.
> It worked 10 seconds and printed how much this task was solved.
> 
> (it means bigger amount is better)
> 
> TCL  - 13 times / 10 sec 
> perl - 64
> python - 36
> C without optimization - 6468
> C with optimization  - 19111
> Java without JIT (1.1.7) - 921
> Java with JIT (kaffe)  - 8230
> pascal (fpc)   - 4451
>  
> Well. that's all

Having encountered the infamous sieve years ago in an article in "Byte" 
which made Pascal on 6502 (1 mhz) look slower than Pascal on Z80 (2 
mhz), when in fact the reverse was true, I discount all such claims 
unless I see (and understand) the code.

Part of the problem in the Byte code was that runtime checking was on by 
default in the compiler on 6502, and off by default in the compiler for 
the Z80.  There were other issues as well.  (You also don't indicate 
what compilers are used for any of the above (unless fpc specifies a 
particular Pascal).

All that is separate from the fact that the sieve isn't real-world code 
(aside from classes and benchmarks, how often is it actually used)?  But 
if properly written, it is indicative.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From neilh at scintilla.org  Wed Jun 28 21:52:54 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 29 Jun 2000 01:52:54 GMT
Subject: simple question ... exec in W32?
References: <Nxu65.20928$Ip.511269@cac1.rdr.news.psi.ca>
Message-ID: <WZx65.981$Tb7.5974@news-server.bigpond.net.au>

> How to I start a W32 application, like Word etc, from a Python script? I'm
> actually trying to rewrite a backup script, which, additionally, does a
lot
> of non-backup processing overnight.

   There are several ways. I'd suggest ShellExecute from the Win32
extensions:

import win32api
win32api.ShellExecute(0,None,"winword.exe",None,"",1)

   Neil






From echuck at mindspring.com  Tue Jun  6 12:17:06 2000
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Tue, 06 Jun 2000 12:17:06 -0400
Subject: [ANNOUNCE] Webware for Python 0.3
Message-ID: <393D2402.31E06D1@mindspring.com>

Webware for Python is a suite of software components for developing web-based applications that are multi-threaded, object-oriented and cached. The suite uses well known design patterns such as servlets and server pages (as found in ASP, PHP and JSP).

Home page:      http://webware.sourceforge.net/
Download:       http://download.sourceforge.net/Webware/Webware-0.3.tar.gz
Release notes:  http://webware.sourceforge.net/Webware/Documentation/RelNotes-0.3.html

Some highlights:
  * There is an improved WebKit 0.3 with new features, refinements and bugfixes. 
  * There is a new PSP component (aka Python Server Pages) which is a WebKit plug-in that enables you to create server pages that freely mix HTML and Python in the same fashion as other popular server 
technologies such as ASP, PHP and JSP. 
  * The project is now fully hosted on SourceForge at http://webware.sourceforge.net including the CVS 
source code repository. You can anonymously checkout a snapshot of the repository if you want the latest, 
in-development version. 
  * Webware is now friendlier to Windows including both code and documentation. We now have developers using both Windows and UNIX/Linux. 


-Chuck


From thor at localhost.localdomain  Tue Jun 13 17:55:09 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 13 Jun 2000 21:55:09 GMT
Subject: Python text mode client of ICQ
Message-ID: <slrn8kdil7.8ug.thor@localhost.localdomain>

Python text mode client of ICQ

I'd like to know if somebody knows about such a beast. 
Yes, I don't want Licq, nor any graphical front end , because
I run an old version of Linux , and like simple the 
simplest thing running for these .

I guess it mustn't difficult to programm it. 

-- 
MGA


From bvdpoel at uniserve.com  Sat Jun 17 17:55:27 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 17 Jun 2000 14:55:27 -0700
Subject: tkinter: option_add() problem
Message-ID: <394BF3CF.450F776E@uniserve.com>

Using tkinter to modify the option database I CAN do things like:

	root.option_add( '*activeBackground', 'white')

However,

	root.option_add( '*Font', "family times -weight bold  -size 14" )

gives me the error message:

	TclError: expected integer but got "times"

the equivalent??? tcl:

	option add *Font {-family times -weight bold  -size 14}

works fine.

Help!


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



From dcalvelo at pharion.univ-lille2.fr  Thu Jun  8 11:23:35 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 8 Jun 2000 15:23:35 GMT
Subject: graph layout module?
References: <m3n1kwizcf.fsf@wilma.george.org>
Message-ID: <8hodpn$na2$2@netserv.univ-lille1.fr>

Harry George <hgg9140 at seanet.com> wrote:
: I have a shop scheduling task, thus roughly rectangular layouts
: with task bars and pred/succ arcs between tasks.  Any convenient tools
: for this?

If what you mean is *automatic* graph layout, there are free tools from AT&T, 
in a bundle called graphviz (ever heard of 'dot'?). They are available from 
http://www.research.att.com/sw/tools/graphviz/, under an Open Source licence.

: Probably need a simple gui for debugging, but final result is a 36
: inch printed plot.  I can imagine, e.g., generating XML for dia to
: visualize, then print to postscript.

The suite works by providing a graph description in a fairly easy-to-read
and easy-to-generate format (you define nodes and links mostly); there is
an interactive editor (it uses X & Athena widgets IIRC, so might not be useful
for NT) and commands for direct output to PS, imagemap, HPGL, PCL, MIF,...

: Needs to operate in both UNIX (AIX, SGI, Linux) and WinNT
: environments.

Source is available.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From thiele at muc.das-werk.de  Fri Jun 16 04:36:42 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Fri, 16 Jun 2000 10:36:42 +0200
Subject: Tkinter Widget command function call
Message-ID: <3949E71A.AF53A1CF@muc.das-werk.de>

Hallo!

I create some buttons dynamically in a for loop:

for i in range(0, n):
	MyButton = Button(frame, text="press me", command=self.CmButtonPressed) 
	MyButton.pack(side=TOP)	

And now I want to check witch button has called the CmButtonPressed -
function.

def CmButtonPressed(self, ???):
	print "Button ", buttonnumber, "pressed"

What is the best way to do this? Has anaboby an idea?

Thanks Thomas


From msoulier at nortelnetworks.com  Tue Jun 27 11:40:17 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 27 Jun 2000 15:40:17 GMT
Subject: compilation
References: <Pine.LNX.4.21.0006270751280.2223-100000@fep132.fep.ru>
Message-ID: <8jaht1$rui$5@bmerhc5e.ca.nortel.com>

In article <Pine.LNX.4.21.0006270751280.2223-100000 at fep132.fep.ru>, 
Oleg Broytmann wrote:

>#! /bin/sh
>
>if [ -z "$1" ]; then
>   echo "Usage: compyle file.py..."
>   exit 1
>fi
>
>TEMPLATE="from py_compile import compile; compile('"
>
>for file in $*; do
>   pgm=$TEMPLATE$file"')"
>   python -c   "$pgm" || exit 1
>   python -OOc "$pgm" || exit 1
>done
>
>   (I know I call python in the loop instead of looping inside python; I do
>this by purpose).

	Neat. So, once it's compiled into a pyc and pyo files, is there a way to
run those directly for better performance? I know that freeze links in the
interpreter. I guess I should use that if I want it. 

	Thanks,

	Mike

-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From dgobbi at minime.irus.rri.on.ca  Wed Jun 14 14:48:03 2000
From: dgobbi at minime.irus.rri.on.ca (David Gobbi)
Date: 14 Jun 2000 18:48:03 GMT
Subject: Python and Video for Windows?
References: <39476A9C.DFE84D0@stud.ntnu.no>
Message-ID: <8i8k13$6pi$1@panther.uwo.ca>

Peter Schneider-Kamp <petersc at stud.ntnu.no> wrote:
> I am currently working on a robot control project and I have
> to use a Video for Windows compatible video camera.

> Is there any possibility to get this working with Python?
> Or is there some C/C++/Java based solution which I could
> easily write an extension for?

> All I need is really the raw image data. No fancy control
> stuff or gamma correction.

> don't-wanna-do-it-in-C++---please-save-me-ly y'rs
> Peter

Swig is probably your best friend here.  I wrote a python interface
to the Matrox Imaging Library (i.e. for the Matrox Meteor frame grabber)
in a couple days that was essentially complete.  Of course, 
the Matrox Imaging Library has a coherent API, whereas VFW doesn't...

I can send you my SWIG .i files for the Matrox interface,
as well as some C++ code that uses VFW.  The C++ code is not
for a standalone app, unfortunately, it is a VTK class (and
I doubt that you want to build VTK + the VTK/Python interface
simply to grab an image into a Python string though that is
a possibility).

or-use-VFW-via-VTK-via-Java-via-jpython-ly y'rs
David

--
  David Gobbi, MSc                    dgobbi at irus.rri.on.ca
  Advanced Imaging Research Group
  Robarts Research Institute, University of Western Ontario


From igorr at ifi.uio.no  Tue Jun 27 06:08:41 2000
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 27 Jun 2000 12:08:41 +0200
Subject: reload()'ing modules and how it affects references -> thank you
References: <xjvn1k8grd3.fsf@gjallarhorn.ifi.uio.no> <slrn8lgn5p.ke4.thomas@centurion.xs4all.nl>
Message-ID: <xjvu2ef307q.fsf_-_@gladsheim.ifi.uio.no>

* Thomas Wouters

[snip]

Thomas, thank for the thorough explanation.
 


ivr
-- 
Simula-konsulent? Simsulent? "I eat Java for breakfast!".
                                           -- thorkild


From mmaul at nortelnetworks.com  Fri Jun  9 13:03:37 2000
From: mmaul at nortelnetworks.com (Michael Maul)
Date: Fri, 09 Jun 2000 13:03:37 -0400
Subject: Probably a dumb question
References: <394106F9.A58F87CE@micron.net> <8hr23k$bvk$1@nntp9.atl.mindspring.net>
Message-ID: <39412369.E499BC4B@nortelnetworks.com>

Well lets not forget jpython which compiles Python to JVM byte code
http://www.jpython.org


Aahz Maruch wrote:

> In article <394106F9.A58F87CE at micron.net>,
> Peter Immarco  <android at micron.net> wrote:
> >
> >Hi,  still waiting for my two books on Python from Amazon.  I know
> >Python is a full programming language so it can be used on web servers
> >for server side processing, but what about browser side processing?  Can
> >I also use it to create an 'applet' that runs in a compatible browser?
> >I'm guessing not likely because I don't think IE4+ or Netscape
> >Communicator support Python but I'm hoping I'm wrong.
>
> There's a rumor that a future version of Mozilla will have Python
> support (supposedly they're building in something similar to ASP with
> its switchable scripting engines).  Other than that, no.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "You could make Eskimos emigrate to the Sahara by vigorously arguing --
> at hundreds of screens' length -- for the wonder, beauty, and utility of
> snow."  -- PNH to rb in r.a.sf.f



From wjdandreta at worldnet.att.net  Wed Jun 21 20:35:06 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 22 Jun 2000 00:35:06 GMT
Subject: Determining Types in a String
References: <8inuov$8r2$1@nnrp1.deja.com>
Message-ID: <_ad45.14162$C44.818130@bgtnsc05-news.ops.worldnet.att.net>

Hi n,

Have you tried putting the types like 'string', 'int', etc,  as keys in a
dictionary something like this:

myTypes = {'int': doint, 'string': dostring, etc.}

fields = string.split(line)
try:
  for f in fields:
    myTypes[type(f)]
except:
  do something if type not listed


npk1977 at my-deja.com wrote in message <8inuov$8r2$1 at nnrp1.deja.com>...
>Hi all, I am trying to load a matrix of floats, ints and strings into
>an array.
>
>I have a method that works, but is slow, the psuedocode is:
>
>v = string.split(line)
>for items in v:
> Attempt to recast item as an int(), if that fails, catch the exception
>and try to recase as long(), keep doing this until I can go no further.
>
> 6 def Recast(v):
> 7         v = string.split(v)
> 8         for i in range(0,len(v)):
> 9                 try: v[i] = int(v[i])
>10                 except Exception:
>11                         try: v[i] = long(v[i])
>12                         except Exception:
>13                                 try: v[i] = float(v[i])
>14                                 except Exception:
>15                                         try: v[i] = str(v[i])
>16                                         except Exception:
>17 FAIL
>
>Thanks!
>n
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.




From eric at estinc.com  Wed Jun 14 13:42:05 2000
From: eric at estinc.com (Eric Lee Green)
Date: Wed, 14 Jun 2000 17:42:05 GMT
Subject: SV: Python Productivity over C++
References: <20000614085924.80042.qmail@hotmail.com> <20000614150508.A2932@fuf.sh.cvut.cz>
Message-ID: <3947C453.CF24E30F@estinc.com>

Michal Vitecek wrote:
>  this is highly debatable - properly written C/C++ source is pretty well
>  readable and self-documenting. similarly, badly written python source
>  makes it as unreadable as in other languages - there's not a point in
>  trying to prevent people from writing (un)readable source code.

It's harder to write unreadable Python source because you can't write code
like, hmm,

opendir(FOO,\"%s\");@files=readdir(FOO);foreach (@files) { if
(/^[^\\.].*\\.txt$/) { print \"%s/$_\\n\";};};closedir(FOO);

since Python enforces proper indentation and limits you to one statement to
line.

Yes, you can write unreadable Python code. But you have to work at it.

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From paul at prescod.net  Tue Jun  6 13:47:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 06 Jun 2000 12:47:52 -0500
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <393D3948.821E1DB6@prescod.net>

"Richard P. Muller" wrote:
> 
> ....
> And yet, I worry that using Python marginalizes one. Python is hard to
> beat for writing applications that run on a single computer. But
> that's yesterday's computing model.
> ....
> 
> I'm aware of CGI scripting with Python. It's great. But JavaScript
> pages load faster than CGI pages, and are easier to program, to
> boot. Plus, JavaScript can directly call Java Applet methods, enabling
> it to do functions that are simply impossible with CGI scripts.

Actually, Java applets were last year's programming model.  In fact,
Java applets are something of an endangered species.

Java is almost entirely used on the server side these days and Python is
just as good there as Java is. In fact, with Zope, Python is better.

Client-side Javascript is quite popular but in terms of volume of code
it is a drop in the bucket compared to what is going on on the server
side.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Alabama's constitution is 100 years old, 300 pages long and has more
than
600 amendments. Highlights include "Amendment 393: Amendment of
Amendment
No.  351", "Validation of Laws Regulating Court Costs in Randolph
County",
"Miscegenation laws", "Bingo Games in Russell County", "Suppression
of dueling".  - http://www.legislature.state.al.us/ALISHome.html



From mwh21 at cam.ac.uk  Fri Jun  9 14:09:02 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 09 Jun 2000 19:09:02 +0100
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net>
Message-ID: <m3em66vi9t.fsf@atrus.jesus.cam.ac.uk>

invalid.address at 127.0.0.1 (David) writes:

> And it's not like it adds new functionality nor improves code
> readibility.  It just saves the lazy programmer a whopping one or
> two keystrokes.

It can improve efficiency by reducing the number of temporaries
generated.  This matters if you're dealing with big matrices, for
example.

Cheers,
M.

-- 
  well, take it from an old hand: the only reason it would be easier
  to program in C is that you can't easily express complex  problems
  in C, so you don't.                   -- Erik Naggum, comp.lang.lisp


From mhagger at alum.mit.edu  Wed Jun  7 16:57:21 2000
From: mhagger at alum.mit.edu (Michael Haggerty)
Date: 07 Jun 2000 16:57:21 -0400
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net>
Message-ID: <xj0r9a9jjke.fsf@tempest.harvard.edu>

"Drew Fisher" <drew at level3.nospam.net> writes:
> Hello.  I am still learning Python, but I came across a need for
> generating dynamic graphs (network load average, uptime, boring
> stuff like this :).
> 
> Is the a module in Python that allows this?  If so, could someone
> point me to it?

Another possibility is my Gnuplot.py module.  It allows you to call
the gnuplot program from python in a fairly convenient way.  Gnuplot
can create output in a variety of formats.

    http://monsoon.harvard.edu/~mhagger/download/

Michael

-- 
Michael Haggerty
mhagger at blizzard.harvard.edu


From kpmurphy at my-deja.com  Thu Jun 29 15:15:04 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Thu, 29 Jun 2000 19:15:04 GMT
Subject: bbox?
Message-ID: <8jg776$ooa$1@nnrp1.deja.com>

does anyone know what bbox is?  I see it in the context of the
PhotoImage put(data, bbox).  Thanks,

Keith


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


From rmartin at objectmentor.com  Thu Jun 29 21:27:03 2000
From: rmartin at objectmentor.com (Robert C. Martin)
Date: Thu, 29 Jun 2000 20:27:03 -0500
Subject: OmPyUnit -- A python unit test framework.
Message-ID: <slnttnejjev80@news.supernews.com>

I have made our OmPyUnit unit test framework for python available in the
'downloads' section of http://www.objectmentor.com.  This framework is very
similar in appearance and operation to JUnit.  It uses a Tkinter based GUI
to track the status of the tests, keeps a scrolling list of double clickable
test failure line items, and reloads the program under test each time the
test button is pushed.  It is a considerable extension to the work of
Katharine Lindner.

--

Robert C. Martin    | "Uncle Bob"              | Training Courses:
Object Mentor Inc.  | rmartin at objectmentor.com | OOD, Patterns, C++, Java,
PO Box 85           | Tel: (800) 338-6716      | Extreme Programming.
Grayslake IL 60030  | Fax: (847) 548-6853      | http://www.objectmentor.com

"One of the great commandments of science is:
    'Mistrust arguments from authority.'" -- Carl Sagan




From delgado at olaf.de  Thu Jun  8 04:47:26 2000
From: delgado at olaf.de (Olaf Delgado)
Date: Thu, 8 Jun 2000 10:47:26 +0200
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
    <393EFF38.C023F10F@san.rr.com>
Message-ID: <uimnh8.9c.ln@olaf.de>

In article <393EFF38.C023F10F at san.rr.com>,
	Courageous <jkraska1 at san.rr.com> writes:
> 
>> 1. Everything is an object of a class that can be subclassed.
> 
> You want to subclass integers, why?

To change the default ordering, for example.

a-mathematician-will-find-a-good-reason-to-subclass-anything-ly y'rs
Olaf


From mjackson at wc.eso.mc.xerox.com  Tue Jun  6 21:47:06 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 7 Jun 2000 01:47:06 GMT
Subject: Nth digit of PI
References: <393C053A.58DD8220@home.com>
Message-ID: <8hk9iq$502$1@news.wrc.xerox.com>

Rajarshi Ray <rajarshi at home.com> writes:

> Besides the gee-whiz aspect of such algorithms, do they have any value?
> After all, what makes pi pi is ALL of its digits in a particular
> sequence.

Well, if you're captured by an alien who asserts that s/he has
calculated pi to a gazillion decimal places, and your survival depends
on determining whether the claim is true or not, you can check a few
digits at random and either be certain it's false or 99.99. . .% sure
it's true.  Never know when *that's* going to come in handy; in fact
IIRC this very plot was once used on McGyver.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	You can never solve all difficulties at once.
				- Paul Dirac




From ge at nowhere.none  Mon Jun 12 18:52:35 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 12 Jun 2000 22:52:35 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <hs815.297$i06.89565@ptah.visi.com> <20000612212452.P20700@xs4all.nl>
Message-ID: <TQd15.356$i06.100607@ptah.visi.com>

In article <20000612212452.P20700 at xs4all.nl>, Thomas Wouters wrote:

>> >Is it's base structure less related to any one specific operating
>> >system?
>
>> Perl isn't so much related to an OS, as it is to some other
>> utilities (awk,sed,grep) and languages (bourne shell).
>
>Yeah, right, and a chair isn't geared as much to a human as to a one-to-two
>meter tall, bi-legged creature with two arms, a flat back and bends in all
>the right places ;)
>
>I guess it depends on your definition of OS, and of UNIX ;)

Well, believe it or not, I actually did most of my initial
awk/sed/grep shellscript learning under VMS.  (For God's sake,
don't tell anybody...)

DEC used to have a product called DECShell than consisted of
the entire set of version 7 shell tools (including the Bourne
shell, sed, awk, lex, yacc, and a libc that did a fair job of
emulating the Unix system calls).  

Process creation under VMS involved a lot more overhead that
undre Unix, so shellscripts ran a bit slower than DCL command
scripts (DCL seemed to somehow run scripts without starting a
lot of new processes)

But still, I managed to write a set of shellscript utilitities
that cross-referenced and verified a whole bunch of things in a
1200 page SRS written in LaTeX.

-- 
Grant Edwards                   grante             Yow!  If I felt any more
                                  at               SOPHISTICATED I would DIE
                               visi.com            of EMBARRASSMENT!


From matt at virtualspectator.com  Mon Jun 12 06:46:37 2000
From: matt at virtualspectator.com (matt)
Date: Mon, 12 Jun 2000 22:46:37 +1200
Subject: lambda and if
In-Reply-To: <8i2c9l$b0r$1@news1.wire.net.au>
References: <8i2c9l$b0r$1@news1.wire.net.au>
Message-ID: <0006122253431T.01005@localhost.localdomain>

say you had the following

filenames = map(lambda x:x[2] + "_" + x[0] + "_" + x[1] + ".txt" ,items)

but really wanted
filenames = map(lambda x:x[2] + "_" + x[0] + "_" + (x[1] in ("big",) ? "big" :
"") + ".txt" ,items)

where (x[1] in ('big',) ? "big" : "") is a pseudo c++_pyhton expression that I
want to implement .....

i.e. the result is to append "big" to the filename being created if it is the
2nd element of item, otherwise don't append anything.

regards
Matt



From akuchlin at mems-exchange.org  Mon Jun 19 13:48:21 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 19 Jun 2000 12:48:21 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun 19)
Message-ID: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net>

    Guido's back from vacation, and buried alive in e-mail.  From
    python-dev: "It's rather late in the release cycle -- I'm trying
    to get alpha 3 released, and i'm still 1400 messages behind on
    my email."

    In a thread comparing the productivity of Python and C++, Martijn
    Faassen writes about Python OO style:
        http://x69.deja.com/getdoc.xp?AN=634383474

    In the same thread, David Bolen and Huaiyu Zhu write a script to
    verify that a class implements all the methods of a template class:
        http://x53.deja.com/getdoc.xp?AN=635109618

    Benyang Tang posts his collection of links related to Numeric
    Python:
        http://x62.deja.com/getdoc.xp?AN=635031042

    Sverre Stoltenberg announces JDBI-0.3, "a jpython module that can
    connect to any database that supports JDBC. I'm trying to make it
    DB-API spec v2.0 compatible."  
        http://www.terminator.net/jdbi/
 

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

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From pj at sgi.com  Fri Jun 30 16:47:27 2000
From: pj at sgi.com (Paul Jackson)
Date: 30 Jun 2000 20:47:27 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <m2lmznzk2t.fsf@mycroft.actrix.gen.nz> <slrn8lp673.u8u.thomas@centurion.xs4all.nl> <8jidt1$2k5s$1@noao.edu>
Message-ID: <8jj10v$drks$1@fido.engr.sgi.com>

From: rwolff at noao.edu (Richard Wolff)                                             |> "C was originally written for the PDP-11 under UNIX ... "
|>   The C Programming Language, by D. M. Ritchie, S. C. Johnson, M. E. Lesk,
|>   and B. W. Kernighan, Bell System Technical Journal, Vol 57, #6, 1978, Pg 1991

Further from that same article:

    The influence of BCPL on C proceeded indirectly through
    the language B, which was written by Ken Thompson in 1970
    for the first Unix system on the PDP-11.

See also the article:

    The Development of the C Language, by Dennis M. Ritchie
    http://cm.bell-labs.com/cm/cs/who/dmr/chist.html
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From geek+ at cmu.edu  Thu Jun  1 14:08:04 2000
From: geek+ at cmu.edu (Brian E Gallew)
Date: Thu, 1 Jun 2000 14:08:04 -0400
Subject: garbage collection and Tkinter
Message-ID: <200006011808.OAA26643@smtp3.andrew.cmu.edu>

I've got a Tkinter script that runs for a long time, creating and
destroying many windows (it's a zephyr client, if that means anything
to you).  Over time, this process grows to be excessivly large
(30-40MB before I get around to killing it).  I've satisfied myself
that it isn't a leak in my C module, as if I just do file logging it
doesn't seem to get so big.  

To finally get around to the point of all this rambling: is there a
relatively easy way to examine the garbage collector to see why it is
or is not collecting things?  I would not be at all surprised to find
circular loops in my Tk widgets, but I don't know where to find them...



-- 
=====================================================================
| JAVA must have been developed in the wilds of West Virginia.      |
| After all, why else would it support only single inheritance??    |
=====================================================================
| Finger geek at cmu.edu for my public key.                            |
=====================================================================
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 266 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000601/387bd3da/attachment.sig>

From erno at iki.fi  Mon Jun  5 22:23:40 2000
From: erno at iki.fi (Erno Kuusela)
Date: 06 Jun 2000 05:23:40 +0300
Subject: indirect references
References: <20000605160915.G11994@activestate.com> <XFMail.20000605161330.shaleh@valinux.com> <20000606012119.F469@xs4all.nl>
Message-ID: <ku66rnzgwj.fsf@lasipalatsi.fi>

>>>>> "Thomas" == Thomas Wouters <thomas at xs4all.net> writes:

    Thomas> On Mon, Jun 05, 2000 at 04:13:30PM -0700, Sean 'Shaleh'
    Thomas> Perry wrote:

    >> I knew that a string itself was immutable, I guess I was
    >> looking for a reference object of some kind (a la C pointers,
    >> C++/perl references).

    Thomas> Well, if you wanted a reference, you got it :)

    Thomas> foo = "string" bar = foo

if you had an immutable object in c or perl, and you wannted to share
a reference to it from 2 locations but change it from either one, you
would use a reference to a reference - but that doesn't exist in
python. so you have to use some kind of mutable container object.

  -- erno


From ralph.gauges at eml.villa-bosch.de  Tue Jun 20 05:54:31 2000
From: ralph.gauges at eml.villa-bosch.de (Ralph Gauges)
Date: Tue, 20 Jun 2000 11:54:31 +0200
Subject: jpython AND popen2 = error
Message-ID: <394F3F57.84B88C7F@eml.villa-bosch.de>

Hello,

I hope I am not in the wrong newgroup here, but I have a
question concerning pipes in jpython.
I just wrote a small class that lets me steer another
application through pipes. It works fine in cpython (1.5.2),
but when I try to run it in jpython (1.1) I get the error
message:

  File "/usr/lib/python1.5/popen2.py", line 78, in popen3
  File "/usr/lib/python1.5/popen2.py", line 17, in __init__
AttributeError: class 'org.python.modules.os' has no
attribute 'pipe'


Does anybody know, why jpython does not seem to have pipes?
Or am I doing something wrong here?

Thank you very much for the help

Ralph


From mhammond at skippinet.com.au  Fri Jun  9 10:36:07 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 09 Jun 2000 14:36:07 GMT
Subject: win32all-132 available
Message-ID: <rh705.459$cQ.7688@news-server.bigpond.net.au>

Ive just made win32all-132 available.  Everyone who thinks it isnt
terrible is welcome to upgrade :-)

This is the most recent build available for Python 1.5.2.

Major changes over 129 are largely stability related - this build has been
available to registered users for some time, and has proven to be very
stable.

There are no known serious bugs (although a few memory leaks have been
found since this release - get the CVS sources for the leak-free version
:-)

It can be found at
http://starship.python.net/crew/mhammond/win32/Downloads.html

Mark.




From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 14 10:06:08 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 14 Jun 2000 07:06:08 -0700
Subject: tkinter/windows/starting html
References: <39476C94.DA0AA2AD@stud.ntnu.no>
Message-ID: <028398e6.cd93152e@usw-ex0108-063.remarq.com>

Hi Peter,

I believe execpv replaces the current process, in this case the
Python interpreter.

I got your code working using os.spawnv but I had to use
c:\windows\command\start instead.

Richard

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From david_ullrich at my-deja.com  Tue Jun  6 10:39:40 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Tue, 06 Jun 2000 14:39:40 GMT
Subject: operator overloading
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com> <393C2386.307A160D@sightreader.com> <20000605235010.C469@xs4all.nl>
Message-ID: <8hj2f9$1rn$1@nnrp1.deja.com>

In article <20000605235010.C469 at xs4all.nl>,
  Thomas Wouters <thomas at xs4all.net> wrote:
> On Mon, Jun 05, 2000 at 03:02:47PM -0700, Ken Seehof wrote:
>
> > The following more closely conforms to these conventions:
>
> The return code of __cmp__ is hardly a convention! If you dont return
one of
> those values, in a consistent manner, you're going to confuse the
users of
> your class greatly. And possibly break things like list.sort()'s
default
> sort function.

     This reminds me of a question that I've never got around
to asking: Are there implicit requirements on __cmp__, like
the order it defines has to be transitive, __cmp(a,b)__ has
to be the opposite of __cmp(b,a)__, etc?

DU


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


From dieter at handshake.de  Mon Jun 19 13:15:10 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 19 Jun 2000 19:15:10 +0200
Subject: OT: for a software patent free Europe
Message-ID: <x766r5h9sx.fsf@lindm.dm>

Currently, software and human behaviour (business rules)
cannot be patented in Europe.

I think, this is good. It allows me to write software
(and give it away) without the need to spend some months
for patent investigations and money for legal advice.

Part of the European Commission now plans to introduce
patent practices known from the US in Europe. They include
patents for stand-alone software, algorithms and
human practices.

A group of European software developpers and companies
have joint forces to prevent this. We want to file
a petition to the European Parliament.

If you are interested and want to help, please visit:

	URL:http://petition.eurolinux.org/index.en.html


Dieter



From cpr at emsoftware.com  Tue Jun  6 20:55:32 2000
From: cpr at emsoftware.com (Chris Ryland)
Date: Wed, 07 Jun 2000 00:55:32 GMT
Subject: Python under VMware/NT?
References: <r3h%4.23$cT1.480@news>
Message-ID: <84h%4.24$cT1.639@news>

I should add: And, if so, what kind of performance differences did you note?

"Chris Ryland" <cpr at emsoftware.com> wrote in message
news:r3h%4.23$cT1.480 at news...
> Has anyone tried running Python in a virtual machine under VMware/NT,
> whether in Linux or NT/2K/9x?
>
> Odd question, I realize. ;-)
> --





From shapr at uab.edu  Thu Jun  1 20:22:03 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 02 Jun 2000 00:22:03 GMT
Subject: "non-essential difficulty" - syntax highlighting?
References: <D15A35708943D3119BA70060979C3C663828D5@PETRIFIED> <8ge7ka$ouo$1@slb6.atl.mindspring.net> <39351648.ED1F1288@uab.edu> <8h37ve$djt$1@slb7.atl.mindspring.net> <39352F52.B5A5B462@uab.edu> <8h4egl$pkc$4@newshost.accu.uu.nl> <393690B2.ED517A8C@uab.edu> <8h6pj4$ecq$3@newshost.accu.uu.nl>
Message-ID: <3937046D.73C41615@uab.edu>

Martijn Faassen wrote:

> > I was thinking of doing it after the Python Refactoring Browser is
> > finished, but if someone wants to join me in coding this up, I can
> > change priorities. I would guess it would take a month or two.
> 
> Any references to this browser? I did a few searches on the web but
> couldn't find anything, but interesting. :)

http://sourceforge.net/project/?group_id=4298
It's my fun project. I think Python needs a Refactoring Browser that
matches the original one written in Smalltalk. Feel free to join up :)

> Do you know about the compiler-SIG? One of the reasons it got started
> was exactly to create the infrastructure to create intelligent
> editors (such as IDLE); incremental parsers and all.

I've heard of it, but I haven't checked it out. I'll do so.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From tony at lsl.co.uk  Thu Jun 22 04:28:03 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Thu, 22 Jun 2000 09:28:03 +0100
Subject: boolean true and false values.
In-Reply-To: <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk>

Thinking about whether Python should have "true" booleans, rather than its
current behaviour, Albert Hofkamp wrote:
> I read a number of replies, but didn't read an answer that seems
> satisfactory. I ran in the same problem, and was also not able
> to find a elegant solution.
> Using 0 and 1 or `true` and None is not correct imho, there
> should be a native way to use true and false for indicating
> boolean values.

When I first started using Python, I also found this to be uncomfortable.
However, that was *my* problem, not Python's!

There seem to be two (equally valid, I hasten to add!) traditions of
handling booleans in programming languages. The (arguably) most obvious is
to have a discrete boolean type, and thus the logical operators return a
result of that type, and so on. Algol based languages tend to take this
approach, and I have always found it more natural (as a pedant).

The other tradition is that used by (for instance) the various Lisp
derivatives, which is to have "empty" quantities represent false, and
everything else represent true. This has various practical benefits, and I
hesitate to argue it is less "theoretically" good.

A language designer has to choose one or the other. Language users then get
to grumble if they don't like it. But both are eminently respectable, with a
long history.

C, of course, sort-of adopts the second mechanism, but I don't actually
believe that that is other than coincidentally related to Guido's decision
in Python - *I* think he knew what he was doing and made a decision based on
history to get the behaviour he wanted.

(This explanation does not, of course, work for my pet hate in Python, which
is that 0 at the start of a number means it is in octal (does *anyone* care
about octal anymore?) - but perhaps this is Guido's homage to the Amish idea
that any construction of mankind should contain a deliberate mistake, as
only god can make something perfect. It is, of course, a very *petty* pet
hate...)

[I guess my age is showing - I find it terminally (sorry - perhaps that
should be "terribly") amusing how many people think that C or C++ is an
ur-language on which all others must be based, because it is *so old* -
almost as funny as people who think Linux (or Windows) has been around a
long time...]

Tibs, in an awkward mood, feeling grumpy at all the people who *like* C
based languages - BCPL was so much more elegant...

--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Quilters: the one who dies with the most fabric wins
Programmers: the one who dies with the most programming languages wins?
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From chris at rpgarchive.com  Thu Jun 15 06:13:38 2000
From: chris at rpgarchive.com (chris)
Date: Thu, 15 Jun 2000 05:13:38 -0500
Subject: TCP packet size?
References: <CC59868A0A76D311B1F50090277B321C1F33BA@VRTMAIL>
Message-ID: <3948AC51.98E5EAEC@rpgarchive.com>

I'll try a more specific explanation.  My application is pretty much a
multi-user application similar to Net Meeting or a chat room.  One major
piece of data that the users are sharing is a xml document (in memory
not a file).  One application is the host who accepts TCP connections
from any number of clients.   If a user changes the xml document, those
changes are sent to the host and then resent to each client.  Again, all
connections are TCP.   A problem arises when I send a large message
(3000-4000 bytes). Despite the fact that I give a large buffer size to
socket.read() method, I often receive that message dissembled.  Also,
sometimes I'll find small messages combined in the same read() call.
The messages just aren't reassembled the way I thought TCP worked.  I'm
hoping to find a way to received one message at a time in its entirety,
or determine the start and end of my messages.    I'm not really using
xml as a network protocol, its just the data structure of the
application.  I'd be happy to show you my networking module if you like
(you probably just cringed :) )

One way I tried to solve my problem was to break my messages into
smaller (1000 byte) messages that included a header.  However, even
though I broke the message up, it was not received in the same format.
Some of the message were combined and broken at unpredictable positions.

FYI, this might be silly to add, but I'm not creating a new TCP
connection for every message. I'm using the same connection repeatedly.

Thanks for the help.  I'm kind of at my wits end.

Mike Fletcher wrote:

> Hmm, I'm very surprised you're needing to get down to this level when
> you're using something as "high level" as XML as your encoding.  XML
> is so inefficient as a network protocol that most people would just
> use standard TCP streams (i.e. open a stream socket and let TCP take
> care of the ordering/guaranteeing/decomposition/reassembly), instead
> of trying to send TCP packets as raw packets (heck of a lot of work
> for no real benefit I can see).  Or do you mean that you're doing a
> network sniff and are upset that the packets are getting broken up and
> reassembled?I do lots of work with UDP-based protocols where these
> issues come up, but since UDP is non-guaranteeing, you wind up
> explicitly dealing with the issues regardless (they're not easy,
> incidentally, lots of soft tradeoffs dependent on current network
> conditions, required QOS, priority among messages, etceteras).  I tend
> to favour breaking messages up more (I target 256 bytes, include 26
> byte UDP header), rather than less, as small UDP packets may get
> through a crowded network, while larger UDP messages will get dropped
> when load is high (or at least, that's what certain articles have me
> believe :) ).  To do it right, of course, you need a dynamically
> adapting system that increases message size when load is low (and does
> a thousand other optimisations, as well).If I've missed the point,
> bear with me and explain again.  Since I've not actually grokked what
> the "problem" being tackled is, it's quite likely I have missed the
> point one way or another.  Enjoy yourself,Mike
> -----Original Message-----
> From: chris [mailto:chris at rpgarchive.com]
> Sent: Thursday, June 15, 2000 2:53 AM
> To: Python List
> Subject: TCP packet size?
> Is there a way to ensure or predict packet size when using TCP?  I'm
> developing a Multi-user application that functions much like a
> muli-player game.  The applications are sharing xml data.  The xml
> data can get very big 3000-6000 bytes.  Unfortunately, when the size
> of the network message gets over about 2000, its to get broken up into
> smaller packets.  So, when I send a 3000 byte message it gets cut up
> into 2-3 packets.  Anther problem I get is when I send two small
> messages back to back, they are sometimes received together.  Is there
> something I'm missing here?
>
> I tried using headers and breaking up messages into smaller chunks,
> but the size of the receiving packets doesn't always match what I
> sent.  I'd appreciate any help or suggestions on how I can tackle this
> problem.  I'd be happy to send someone my network code.
>
> Thanks
>
> --
> chris davis
> chris at rpgarchive.com
>
> http://www.rpgarchive.com
> news and adventure database!
>
> http://www.openrpg.com
> open source role playing!
>
>

--
chris davis
chris at rpgarchive.com

http://www.rpgarchive.com
news and adventure database!

http://www.openrpg.com
open source role playing!


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/46ecaa46/attachment.html>

From godzilla at netmeg.net  Sat Jun 17 11:48:24 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sat, 17 Jun 2000 11:48:24 -0400
Subject: Problem displaying Numeric Array as image
Message-ID: <642A954DD517D411B20C00508BCF23B001296E89@mail.sauder.com>

Eric Frey said:
> I a trying to display a floating point Numeric (NumPy) array as a
> grayscale image via Tkinter.

this example code works for me. its an amalgam of stuff i have found
on the net... normalize takes a floating point image and scales it to
1 and returns byte values suitable for a greyscale image. with a
little cleaning up this should do what you want......

les schaffer

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

#!/usr/bin/python


from Tkinter import *
from ImageTk import PhotoImage
import Image

from Numeric import *
from RandomArray import random

def normalize(a):
    a = array(a, copy=1)
    a = (a - min(a.flat))/(max(a.flat) - min(a.flat)) 
    return ( a*254 ).astype('b')

def ArrayToImage(a, height=200, scale=1):
    a = normalize(a)
    i = Image.new("L", (a.shape[1], a.shape[0]), color=255)
    i.fromstring(a.tostring())
    if scale != 1:
	i = i.resize((i.size[0]*scale, i.size[1]*scale))
    return i

def ImageToArray(i):
    a = fromstring(i.tostring(), 'b')
    a.shape = i.im.size[1], i.im.size[0]
    return a


class MatrixDisplay:

    def __init__(self, master):

        frame = Frame(master)
        frame.pack()
        self.button = Button(frame, text="QUIT", fg="red", 
                             command=frame.quit)
        self.button.pack(side=TOP)
        self.test_B = Button(frame, text="Update", command=self.update)
        self.test_B.pack(side=TOP)
        self.img = Image.new("L",(100,100))
        self.photo = PhotoImage(image=self.img)
        self.canvas = Canvas(frame, width=100, height=100)
        self.canvas.create_image(0, 0, anchor=NW, image=self.photo)
        self.canvas.pack()

    def update(self):
        ra = fromfunction(self.func, (101, 101))
        #ra = random((101, 101))
        
        self.img = ArrayToImage(ra)
        #self.photo = PhotoImage(self.img)
        self.photo.paste(self.img)
        self.canvas.create_image(0, 0, anchor=NW, image=self.photo)
        self.canvas.update()

    def func(self, i,j):
        return cos( 4*2*3.1415926*i / 100. ) * cos( 4*2*3.1415926*(j / 100. ))
    
    

root = Tk()
app = MatrixDisplay(root)
root.mainloop()



From pete at visionart.com  Wed Jun  7 19:27:41 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 07 Jun 2000 16:27:41 -0700
Subject: Why not a Python Package Manager
References: <393f42e4.77383471@news.online.no> <393E8B7E.12751F79@visionart.com> <393ED13A.26AC0642@uab.edu>
Message-ID: <393EDA6D.3427CBE1@visionart.com>

> I agree. I agree with 'something like Debian' so much I think dpkg and
> friends is pretty much the way to go. Why not completely use the debian
> package manager?

don't leave us windows users out in the cold!
i'm not sure of the window'y'ness of the debian packages,
but perhaps dpkg can be applied to many platforms?


From dag at orion.no  Wed Jun 21 07:16:34 2000
From: dag at orion.no (Dag Sunde)
Date: Wed, 21 Jun 2000 11:16:34 GMT
Subject: "TypeError: read-only buffer, None" returned from file.write...
References: <MIS35.664$PQ4.13634@news1.online.no> <8ip2q3$2jrk$1@news.aros.net>
Message-ID: <mu145.13877$R7.2335108@juliett.dax.net>

That was absolutely correct!

Thank you...

Dag.

"Adam Ruth" <aruth at intercation.com> wrote in message
news:8ip2q3$2jrk$1 at news.aros.net...
> I get this all the time too, simple error.  Somewhere you assigned
sBuffer
> to None.  It happens to me a lot when I call functions that return
strings
> or return None when there is a problem.  You can confirm this by
writing:
>
> f.write(str(sBuffer))
>
> The file then should contain the word 'None'.
>
> --
> Adam Ruth
> InterCation, Inc.
> www.intercation.com
>
>
> "Dag Sunde" <dag at orion.no> wrote in message
> news:MIS35.664$PQ4.13634 at news1.online.no...
> > I have the following code-snippet that retuns a string:
> >
> >    result = m_XMLDoc.transformNode(xslDoc)
> >
> > ...and the string shows the expected result when printed,
> > but when I try to save the string to a file with the following
> > code:
> >
> >   f = open( outFile, 'w')
> >   f.write(sBuffer)
> >   f.close()
> >
> > I get this error:
> >
> > Traceback (innermost last):
> >   File "f:/apache/cgi-bin/getfinalconfirmation.py", line 142, in
> > saveConfirmedData
> >     f.write(sBuffer)
> > TypeError: read-only buffer, None
> >
> > Can anyone give me a hint?
> >
> > TIA...
> >
> > Dag Sunde
> >
> >
> >
>
>




From pehr at pehr.net  Tue Jun 27 15:50:19 2000
From: pehr at pehr.net (pehr anderson)
Date: Tue, 27 Jun 2000 19:50:19 GMT
Subject: A simple question
References: <8i5gg5$ip$1@nnrp1.deja.com>
Message-ID: <3959057A.944614CB@pehr.net>

"Michael A. Howard" wrote:
> 
> I am developing an application using Python/Tkinter that needs to look
> similar to a VB application (a language I am far more familiar with) as
> it will extend some functionality & we want the user interface to be as
> close as possible to the existing system.
> 
> What I need to do is have a tabbed switchboard type menu screen with
> buttons for each program that the user will interact with. This has
> already been developed and went far faster than doing the same thing
> with VB.
> From the switchboard the user will select the program they want to run
> and it should open a new window for data entry/edit.
> 
> None of the code examples I look at show me how to do this.
> 
> Am I missing something really simple or is this task beyond the
> capabilities of Tkinter or Python?
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


If you want to look like a VB app, check out wxWindows instead of
Tkinter.
I would prefer Tk for platform portability but wxWindows seems to be 
catching up fast.  WxWindows uses the native gui widgets underneath,
so on windows it looks like windows, and on gnome it looks like gnome.

I've only played around with wxWindows a little so far but I've been 
rather impressed with the state of development.  
Download it and check it out. I'd love to hear how it works for your
application.

On the third hand, if you really only want to use your app on windows, 
Buy a copy of the book "Python for Win32" which tells you the nuts and
bolts of building windows applications with python.
Not portable, but it may get you where you want to be the fastest.

	-pehr anderson


From ge at nowhere.none  Tue Jun 27 16:41:24 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 27 Jun 2000 20:41:24 GMT
Subject: more python evangelising...
References: <B57E791A.264E%eableson@mac.com>
Message-ID: <Uj865.1905$iN5.408850@ptah.visi.com>

In article <B57E791A.264E%eableson at mac.com>, Erik Ableson wrote:

>We are talking about MS - and they're more interested in making
>money in training packages, ms certification programs, etc.
>[...]
>As always these decisions are business decisions and in no way
>related to the relative technical merits of a product :)

Since MS is a publicly held company, the people who run it
have a fiduciary responsibility to the stockholders.  They are
obligated to make them as much money as they can (legally, that
is).  If they didn't, they could be held liable in court.

OTOH, if the charter of the company called for
socially/technically responsible behavior as a higher priority
than profits, then they could take lower profit, higher "social
good" directions.  But that's not what stockholders want.
Stockholders (at least in the US) only care about short-term
profits.  Your company's business is dropping flaming puppies
off the Empire State Building?  Fine -- as long as returns keep
going up...

If there was a large enough group interested in investing in
socially-conscious software companies perhaps things would be
different.  Perhaps.

-- 
Grant Edwards                   grante             Yow!  Two LITTLE black
                                  at               dots and one BIG black
                               visi.com            dot...nice 'n' FLUFFY!!


From smigowsky at dspace.de  Mon Jun 19 02:44:15 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Mon, 19 Jun 2000 07:44:15 +0100
Subject: Problem Installing/Using Windows Extensions
Message-ID: <326536345498D311B3BC00105A39802A074735@newsintern.dspace.de>

Hi,

the problem is not the dcom package but a DLL. Mircosoft removed an
interface
from the comcat.dll which the win32all package needs. 

Try to get hold of a comcat.dll version 4.71 and register it using
regsvr32.exe before
you install win23all. After the install you can register again the DLL which
is in your system32 
or system directory.  With this patch your packages will install. 

	Stefan

> -----Original Message-----
> From:	richard_chamberlain [SMTP:richard_chamberlain at ntlworld.com]
> Sent:	Saturday, June 17, 2000 10:59 PM
> To:	python-list at python.org
> Subject:	Re: Problem Installing/Using Windows Extensions
> 
> Hi David,
> 
> I had the same problem with a PC at work running W95, but not on any of
> the
> others. I finally narrowed it down to be the only one without Internet
> Explorer 5 installed. And from that I presume that DCOM needs to be
> installed.
> 
> Hope that helps,
> 
> Richard
> 
> 
> David Steffen <steffen at biomedcomp.com> wrote in message
> news:2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1 at lp.airnews.net...
> > Howdy:
> >
> >   I'm trying to use Pythonwin with Python 1.52 under Windows 95 and am
> > consistently running into a problem that I don't understand at all,
> > probably because I know relatively little about Windows.
> >
> >   When I selected "COM Makepy Utility" from the Tools menu of
> > Pythonwin, I got the following error message:
> >
> > """
> > >>> Failed to execute command:
> > from win32com.client import makepy;makepy.main()
> > Traceback (innermost last):
> >   File "C:\Program Files\Python\Pythonwin\pywin\framework\toolmenu.py"\
> > , line 103, in HandleToolCommand
> >     exec "%s\n" % pyCmd  File "<string>", line 1, in ?
> >   File "C:\Program Files\Python\win32com\client\makepy.py", line 46, \
> > in ?
> >     import genpy, string, sys, os, types, pythoncom
> >   File "C:\Program Files\Python\win32com\client\genpy.py", line 16, in ?
> >     import pythoncom
> >     Import Error: DLL load failed: A device attached to the system is \
> > not functioning.
> > """
> >
> >   I noted that what I had was version 1.25, and that version 1.32 was
> > up on Mark Hammond's site on starship python, so I installed the newer
> > version.  The installed completed, but with three error messages near
> > the end:
> >
> > """
> > Registering COM Server AXScript Engine
> >
> > Registration of the AXScript Engine COM server failed.
> >
> > Installation will continue, but this server will require manual
> > registration before it will function.
> >
> > exceptions.ImportError: DLL load failed: A device attached to the
> > system is not functioning.
> > """
> >
> > ...with the other two messages being similar ones about the Python
> > Interpreter COM server and the Dictionary COM server.
> >
> > I'd be delighted to RTFM, if I could find the relevant M (I've looked
> > in the included Python docs and on the python.org and starship python
> > websites).
> >
> > I'd be happy to register the servers manually, if that would solve the
> > problem and if I had a clue as to what than meant.
> >
> > If there is a device not functioning on my laptop, I'd love to know
> > that, but a disk scan indicates no problems and I haven't noticed any
> > other problems with my laptop.  I'm thinking I may be missing a DLL,
> > driver, or other software component; if anyone could suggest which one,
> > I'd be pleased as punch.
> >
> > Thanks in advance!
> >
> > -David Steffen-
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From kens at sightreader.com  Tue Jun 13 06:28:37 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 13 Jun 2000 03:28:37 -0700
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <394113F2.8D6D0833@lemburg.com> <8hr6ee$st8$1@sunnews.cern.ch>
Message-ID: <39460CD5.E7CAACF0@sightreader.com>



Wim Lavrijsen wrote:

> "M.-A. Lemburg" <mal at lemburg.com> writes:
>
> >reduces LOCs, scales well (from 5-100k LOCs
> >if needed -- which is roughly equivalent to 25-250k LOCs
> >in C++ from my own experience).
>
> This is interesting. Could other people confirm this? Ie. that the
> trade-off for Python is larger at small projects (factor ~5) than at
> medium sized projects (factor ~2.5)?
>
> And what is the cause of the difference?
>
> Best regards,
>       Wim Lavrijsen

I'd expect it to be the other way around.  The advantages of Python over
C++ for small projects (interactive interpreter, fast unit testing and
debugging, simpler syntax, no declarations) continue to be equally
relevant in large projects.  In addition, Python encourages component
oriented programming, which means it scales much better than C++.  In a
typical C++ project, it is difficult to make major changes, or even
small modifications, to the design and implementation of components
without having nasty chaos propagating throughout the rest of the
project.

In python, if you have ten different teams independently develop ten
modules after a one hour design meeting, you can figure they will
probably work together seamlessly.  Not likely in C++.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000613/2d089c4a/attachment.vcf>

From matt at virtualspectator.com  Wed Jun  7 01:47:46 2000
From: matt at virtualspectator.com (matt)
Date: Wed, 7 Jun 2000 17:47:46 +1200
Subject: win python and graphics
Message-ID: <0006071748250C.00959@localhost.localdomain>

Hi, I pretty much only use Linux, but have to do some things on windows.  In
particular I need to take screen shots of the desktop or particular windows.  I
would like do this using python rather than in borland C++.  Can anyone help out
here ?

regards
Matt




From gmcm at hypernet.com  Wed Jun 28 18:28:45 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 28 Jun 2000 22:28:45 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com>
Message-ID: <8F61B28C8gmcmhypernetcom@199.171.54.155>

Eric Lee Green wrote:

>Steve Juranich wrote:
>> I am *brand* new to python (as of Sunday), and I just got to the
>> section in the tutorial about classes.  I was wondering why there
>> really isn't such an idea as a "private" member of classes? 
>
>Mostly because Python doesn't "really" have classes. What Python
>"really" has are hash tables whose entries are objects (either methods
>or other data objects), and some syntactic sugar [...]

By this logic, all C++ has is pointers and some syntactic sugar.

>This is one reason why inheritance is so efficient in Python, BTW. All
>inheritence does is add into the hash table for the overall class object
>those entries that were in the base class object. Then when you create a
>new instance, you actually are just creating a copy of the overall class
>object, without having to refer back to the base object(s). 


Not positive I understand what you're saying, but reality appears to 
contradict it:

>>> class A:
...  arg = 1
...
>>> class B(A):
...  barg = 2
...
>>> class C(B):
...  pass
...
>>> c = C()
>>> c.arg
1
>>> c.barg
2
>>> c.__class__.__bases__[0].__dict__
{'barg': 2, '__doc__': None, '__module__': '__main__'}
>>> c.__class__.__bases__[0].__bases__[0].__dict__
{'arg': 1, '__module__': '__main__', '__doc__': None}
>>>

- Gordon


From alexandre.ferrieux at rd.francetelecom.fr  Tue Jun 27 09:10:06 2000
From: alexandre.ferrieux at rd.francetelecom.fr (Alexandre Ferrieux)
Date: Tue, 27 Jun 2000 15:10:06 +0200
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com>
Message-ID: <3958A7AE.7DAA@rd.francetelecom.fr>

Paul Duffin wrote:
> 
> [detailed explanation of stdio-buffering/ttys/pipes]
>
> you may want to look into Python versions
> of Expect. If they are not satisfactory then you could probably write
> a little Tcl/Expect wrapper around the application which forces the
> correct buffering and then control that using Python.

Not withstanding Tcl ;-) honesty commands to also mention the
existence of the 'pty' unix package, which is roughly "Expect minus
Tcl", and especially its 'nobuf' tool:

	nobuf cmd [args...]

starts cmd as a child, strongly held between ptys, so that the overall
command behaves properly (ie in unbuffered or line-buffered mode) betwen
pipes. If all you want is that, Expect is surely an overkill.

But ptys are no rocket science anyway. The nobuf above should be doable
in pure Python within 20 loc...

-Alex


From robin.escalation at ACM.org  Sun Jun 11 12:02:05 2000
From: robin.escalation at ACM.org (robin.escalation at ACM.org)
Date: Sun, 11 Jun 2000 12:02:05 -0400
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <0ud7ks4t2b1s2an6es8q42lc5bd2rp1oeu@4ax.com>

"Jeff" <no_spam at umich.edu> wrote:

>Pythonwin is terrible!

On the contrary I find it to be an essential part of development on
the Windows platform. Build 129 has an incredible folding editor,
multi-pane debugger, and configuration options. Mark Hammond has done
us an amazing service. Did you happen to notice that it was free?!

>If there are no answers to these questions, someone needs to work on
>pythonwin's "development" capabilities.

I believe the answer is that we all need to!

My abilities are not up to this particular task, but I will soon make
available some code I have been working on in another domain. Code
that would not yet be ready without PythonWin!

This is what open source is all about, no?

Yes, we all have improvements we'd like to see. Is any software ever
complete or perfect?

-----
robin                          robin at illusionsexeculink.com
media artist /                 remove illusions to reply
information architect          www.execulink.com/~robin/

The form of e-coli bacteria that causes geomagnetic storms on earth. 


From moshez at math.huji.ac.il  Sat Jun 10 13:09:30 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 20:09:30 +0300 (IDT)
Subject: Type problem on Windows
In-Reply-To: <eviltofu-B4ED2B.12472010062000@news.earthlink.net>
Message-ID: <Pine.GSO.4.10.10006102005580.6989-100000@sundial>

On Sat, 10 Jun 2000, Jerome Chan wrote:

> I am using Pythonwin with win32-129, MacPython 1.5.2c1, and Linux Python 
> 1.5.2.
> 
> While working on some socket stuff, I check the input parameters to some 
> function which expects a socket. The code I use is:
> 
> def sendData(self,data,s):
>   if type(s)!=socket.SocketType:
      ^^^^^^^^^

The answer is yet again "don't do that". Never ever check type sanity in
Python. If the object is good enough, your function will work with it. And
if it isn't, well, it will raise a TypeError or AttributeError itself.
Polymorphism rulez!

> On Unix and Mac, this works. On Win98, this bombs.
> 
> I get the following Exception Message
> 
> Windows98 gives me <type 'instance'>
> Unix and Mac gives me <type 'socket'>

Because Win98 apparently has an emulation layer so sockets can be used
portably. 

> Is there a workaround for this? And should this be reported as a bug?

Yes: just don't type-check
No: the socket module only promises to give you objects with the right
interface.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From rhicks at nospam.rma.edu  Thu Jun  8 08:55:19 2000
From: rhicks at nospam.rma.edu (Opus)
Date: Thu, 08 Jun 2000 12:55:19 GMT
Subject: GUI editor
References: <393f4568$0$25329@wodc7nh0.news.uu.net> <vskujs867ghjso3aqi2r38n5mei0bg4lhq@4ax.com> <m33dmowogo.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <XIM%4.2$RB1.1648@news>

If you want to leave Tkinter and go to wxPython...there is BOA CONSTRUCTOR.
It was pretty good a couple months ago but I know that right now they are
hard a work working the bugs out. I think a couple of Linux mags recently
gave short overviews of wxPython. When BOA matures and wxPython has docs...I
will be there.

Opus

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m33dmowogo.fsf at atrus.jesus.cam.ac.uk...
: Anders M Eriksson <anders.eriksson at morateknikutveckling.se> writes:
:
: > If you find any IDE/GUI editor that you like Please let me know! I
: > also searching...
:
: It's not quite what you're looking for (well, it's not an IDE), but
: you can get perfectly acceptable results very, very quickly (on Linux
: at least) using glade and gtk+.
:
: Last week, while avoiding revision, I wrote a panel applet that would
: display me the test scores in the England/Zimbabwe test at two minute
: intervals, added some options and a settings dialog in about an hour
: and 100 lines of code (plus a .glade file), having never done anything
: with the panel before (I've done a bit of gtk+ before, but not a lot).
:
: ObLinks:
:
:   http://galde.pn.org
:   http://www.gtk.org
:   http://www.daa.com.au/~james/pygtk/
:
: HTH,
: Michael
:
: --
:   This makes it possible to pass complex object hierarchies to a C
:   coder who thinks computer science has made no worthwhile
:   advancements since the invention of the pointer.
:                                        -- Gordon McMillan, 30 Jul 1998




From richard_chamberlain at my-deja.com  Mon Jun 26 10:25:49 2000
From: richard_chamberlain at my-deja.com (richard_chamberlain at my-deja.com)
Date: Mon, 26 Jun 2000 14:25:49 GMT
Subject: The automagic of Tkinter
References: <20000626135336.A8180@bork>
Message-ID: <8j7p5e$qia$1@nnrp2.deja.com>

Hi Egbert,

OK I'll let you get away with more than one
question per post, as long as they are not
difficult. ;-)

Your class Knop inherits from Button which in
turn inherits from Widget which inherits from
BaseWidget.

Button has the keyword argument master which
assigns None if no master is passed, which is
what you are doing. Widget passes this straight
to BaseWidget which does this:

if not master:
	if not _default_root:
		_default_root = Tk()
	master = _default_root

So Tkinter is kindly create a Tk() instance in
case you forgot, or were just being stubborn.

Here's mainloop:

def mainloop(n=0):
	_default_root.tk.mainloop(n)

so mainloop is gets called against _default_root
which just happens to be the Tk() instance that
Tkinter created for you.

When you create an a Tk() instance explicitly it
also calls it _default_root except this time you
get a nice convenient reference to it incase you
want to call methods against it.


Because most people are doing from Tkinter import
* you actually get the method mainloop() anyway.
So you needn't type root.mainloop() but just
mainloop() or call against any widget like you
did. It's probably not a good idea to do these
things though, unless no-one is ever going to
have to decipher your code.

What does knop mean anyway?

Richard












In article <20000626135336.A8180 at bork>,
  Egbert Bouwman <egbert at bork.demon.nl> wrote:
> Good afternoon,
> The following small script does what I intended
it to do:
> it produces a button, and that button
dissappears when i click on it.
>
>     from Tkinter import *
>     class Knop(Button):
>        def __init__(self,tekst="KNOPJE"):
>            Button.__init__
(self,text=tekst,command=self.quit)
>            self.pack
(side=TOP,fill=BOTH,expand=YES)
>     root=Tk()
>     k=Knop()
>     root.mainloop()
>
> However when I replace the last three lines
with only:
>     Knop().mainloop()
> then the script still works.
>
> This leaves me with a number of questions
(sorry, richard):
> - if 'root=Tk()' is dispensable, then what is
its builtin replacement
>   and what does that replacement do
> - up till now I saw a class object as a
template, that has to be
>   instantiated in order to be of any use.
However here it seems to
>   lead its own life.
> - probably 'mainloop()' plays a crucial role
here, but what are all
>   its functions ?  What is the difference
between
>   'root.mainloop()' and 'Knop().mainloop()'
> I promised Richard to pose only one question
per post, but these
> questions really are one big question about the
Tkinter mechanism.
> egbert
>
> --
> Egbert Bouwman - Keizersgracht 197 II - 1016
DS  Amsterdam - 020 6257991
>
==================================================
======================
>
>



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


From smnordby at yahoo.com  Thu Jun 29 18:03:27 2000
From: smnordby at yahoo.com (sn)
Date: Thu, 29 Jun 2000 15:03:27 -0700
Subject: Any other good tutorials?
References: <sl2i51h0e7f149@corp.supernews.com> <395B3DC7.D83FB4CE@gssec.bt.co.uk>
Message-ID: <395BC7AF.BCC2516@yahoo.com>

I started with Alan's tutorial, then GvR's, then the Learning Python book, so I
can recommend that path.  I just found How to Think like a Computer
Scientist-Python version:
http://yhslug.tux.org/obp/thinkCS/thinkCSpy/index.htm  It looks like a work in
progress.  I thumbed through Teach Yourself Python in 24 Hours at the bookstore
and wished it was out when I started (I may go back and buy it anyway).  When
Bruce Eckel's Thinking In Python (
http://www.mindview.net/Python/ThinkingInPython.html ) is done, it will
probably be a great resource too.

-Steve-

Alan Gauld wrote:

> SugarKing wrote:
> >      Are there any other good tutorials besides Guido van Rossum, I don't
> > think he's very explainatory. I like full explainations:)
>
> Mine is aimed at non programmers and tries to get them to the
> point where the official tut makes sense. It has a lot more
> explanation than the others mentioned so far.
>
> http://www.crosswinds.net/~agauld/
>
> Available as a zip from the site and prints out at
> around 90 pages.
>
> The book version will hopefully be out late this year...
>
> Alan g.
>
> --
> =================================================
> This post represents the views of the author
> and does not necessarily accurately represent
> the views of BT.



From 12klat at sightreader.com  Tue Jun  6 20:45:52 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 17:45:52 -0700
Subject: Why Do We Use This?
References: <8hjogr$d14@autodesk.autodesk.com>
Message-ID: <393D9B40.D086AF25@sightreader.com>

The "long" versions are not used in the way that you describe.  They are not
generally called directly, by name (e.g. x.__len__()), although they can be.

The missing concept here is called "operator overloading".  Operator
overloading gives you the ability to define the meaning of various
operations (such as length) by defining a function that will be called to
evaluate the operation.

Suppose you want to create a new class that behaves like a sequence.  For
example:

>>> x = MyClass()
>>> x[4] = 34.5
>>> n = len(x)

How would you go about writing MyClass?

The answer is that you define the "magic" functions that get called for
various sequence operations.

class MyClass:
    . . .
    def __len__(self):
        # . . . return the length
    def __setitem(self, i, v):
        # . . . set the value at position i

So __len__ gets called to evaluate len(x).

Operator overloading also is used for numeric classes, mappings (like
dictionaries), and also let you define the string representation and other
special effects.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

Akira Kiyomiya wrote:

> Okay, what is the difference between these?
>
> If these are same, whey do we bother using the "long" version such as
> __len__(a)??
>
> Sorry, the books I am reading do not explain very well about this issue.
>
> a = [1,2,3,4,5,6]
> len(a)                                # __len__(a)
> x = a[2]                            # __getitem__(a,2)
> a[1] = 7                            # __setitem__(a,1,7)
> del a[2]                            # __delitem__(a,2)
> x = a[1:5]                         #__getslice__(a,1,5)
> a[1:3] = [10,11,12]         # __setslice__(a,1,3,[10,11,12])
> del a[1:4]                        # __delslice__(a,1,4)
>
> Thanks
>
> Akira
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/f4bf4932/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/f4bf4932/attachment.vcf>

From M.Vitecek at sh.cvut.cz  Wed Jun 14 09:05:08 2000
From: M.Vitecek at sh.cvut.cz (Michal Vitecek)
Date: Wed, 14 Jun 2000 15:05:08 +0200
Subject: SV: Python Productivity over C++
In-Reply-To: <20000614085924.80042.qmail@hotmail.com>; from hungjunglu@hotmail.com on Wed, Jun 14, 2000 at 01:59:24AM -0700
References: <20000614085924.80042.qmail@hotmail.com>
Message-ID: <20000614150508.A2932@fuf.sh.cvut.cz>

Hung Jung Lu <hungjunglu at hotmail.com> wrote:
>>From: Michal Vitecek <M.Vitecek at sh.cvut.cz>
>> >    if ( a = getNumber() ) {...}
>>  i have no idea why you call this a stupid feature - IMO this all makes
>>  the program shorter and as such more readable. this is really something i
>>  hate about python, missing ++ and -- operators plus having to do the
>>  assignment stupidly before if comment - more & longer source.
>
>Hey, I can recommend you a great language for shorter and notations: Perl. 
>:) You've got all your lovely operators there. You are welcome to use it 
>instead of Python. :) Why use Python if you've got Perl? :)

 because perl is much less readable - we all know it.
 
>One thing we have learned from Python and Perl, it's that shorter notations 
>are not better. Proper longer notations (not those exaggerated notations 
>like in Cobol) are almost self documenting.

 this is highly debatable - properly written C/C++ source is pretty well
 readable and self-documenting. similarly, badly written python source
 makes it as unreadable as in other languages - there's not a point in
 trying to prevent people from writing (un)readable source code.

>Your experience with C++ certainly is highly limited. 
 
 :) one would never know what others are able to find out from a short
 email. i advice you not to be so certain about my C++ experience.

>Do you know that 
>Metrowerk compiler by default issues a warning when it encounters 
>assignments inside an "if" statement? Now, are you going to tell people in 
>Metrowerk that they are dumb, too? :)

 if(a = someFunction()) should generate a warning: 'warning: suggest
 parentheses around assignment used as truth value'

 if((a = someFunction())) shouldn't. 

 now - is it you who's the dumb here? :)

>> >C++ programmer get bitten all the time with the == && || operators, and
>> >refuse to acknowledge that there is a problem.
>>
>>    :) maybe C++ programmers like you?
>
>Wow wow, someone trying to lecture me about C++, you've got some guts. I 
>pushed out quite a few products distributed worldwide, managed teams of 
>programmers with several Ph.D. degree holders. Dear sir, what have you done 
>in your life?

 now your nose is so up the sky you should be happy it's not raining.
 i don't see why one should care to fight one another as to who has higher
 experience - it wouldn't change anything at all if your experience is
 better or worse than mine.

>>i personally never had any problem with ==, && and ||,
>
>You are a newbie, and it shows. :)

 again - give out some examples where you or your programmers had troubles
 with ==, && and ||. just smiling, considering me a newbie without knowing
 me and not showing any examples doesn't mark you high.

-- 
			fuf


------------------------------ na IRC -------------------------------------
 BillGates [bgates at www.microsoft.com] has joined #LINUX
 ...
 mode/#linux [+b BillGates!*@*] by DoDad
 BillGates was kicked off #linux by DoDad (banned: We see enough of Bill
          Gates already.)
 




From dbroadwell at mindspring.com  Wed Jun 28 22:51:54 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Wed, 28 Jun 2000 22:51:54 -0400
Subject: SystemError: bad argument to internal function
References: <8jcltq$1li$1@nnrp1.deja.com> <8jcma6$1v2$1@nnrp1.deja.com>
Message-ID: <395AB9CA.667A5BEA@mindspring.com>

> python.exe:
> sys.version: '1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)]'
> sys.platform: 'win32'
> os.name: 'nt'
That dosen't actually tell us platform.. my windows 98 says the 
same thing.

-- 
        ~We're all fallen angels who've forgotten how to fly~
        Remember how: http://www.cyber-action.com/oed/dbroadwell.html
--


From bwinton at tor.dhs.org  Fri Jun  2 19:11:59 2000
From: bwinton at tor.dhs.org (Blake Winton)
Date: Fri, 2 Jun 2000 19:11:59 -0400
Subject: Copyright changes with move from CNRI to BeOpen?
In-Reply-To: <000601bfcd16$7308e2c0$b42d153f@tim>
References: <000601bfcd16$7308e2c0$b42d153f@tim>
Message-ID: <200006022311.TAA16891@tor.dhs.org>

In comp.lang.python, you wrote:
>folderol

Do you read Miriam-Webster's Word Of The Day too, or is this just a
strange coincidence?

Later,
Blake.
-- 
6:43pm up 11 days, 9:53, 1 user, load average: 1.00, 1.00, 1.00



From bvdpoel at uniserve.com  Fri Jun 16 23:17:53 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Fri, 16 Jun 2000 20:17:53 -0700
Subject: Typing question
Message-ID: <394AEDE1.2E97621F@uniserve.com>

I'm a bit confused about how python decides on the type for a variable.
I was writing a simple program to parse a data file and print some
results. What I found was that if I did something like:

	a = 123
	print " %.2f " % a

All worked just fine.

However, when I read in a line of data from a file, and then converted
the data into a list using split() the types of the data (which are
numeric) stayed as 'string'. So, the resulting print() failed (actually
it was a bit more complicated since some arithmetic was involved as
well, but that obscures the point).

I see that what is happening: the data is read as a string, split()
converts a string into STRING list elements, and when I do something
like a=list[1] 'a' is just a pointer to the list element (I'm a bit
fuzzy on this last item).

So, I end up forcing the conversion using float() when assigning the
results of split() to the storage list. Is this the correct way to do
things in python?


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



From cjc26 at nospam.cornell.edu  Thu Jun 22 23:27:41 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 23 Jun 2000 03:27:41 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand> <8it320$g9$1@nnrp1.deja.com> <m3g0q5p7fp.fsf@gondolin.beleriand>
Message-ID: <slrn8l5meq.28cs.cjc26@localhost.mindriot.net>

* Steffen Ries <steffen.ries at sympatico.ca> menulis:
| 
| Unfortunately, this does not do what I need:
| >>> t='uid=joe, ou=orgunit, o=some\\, org'
| >>> pat.split(t)
| ['uid=joe', ' ou=orgunit', ' o=some\\', ' org']

You could use string.replace to change the \, sequence into a sequence
that won't appear in your input, then split on ,.  For example, if you
know that you'll never get any strings with "\@" in it, then you
could do:

t = 'uid=joe, ou=orgunit, o=some\\, org'
t1 = string.replace(t, r'\,', r'\@')
l = string.split(t1, ',')
l1 = map(lambda str: string.replace(str, r'\@', r'\,'), l)

(the last line goes through and changes "\@" back to "\,")


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From wjdandreta at worldnet.att.net  Tue Jun 20 18:36:36 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 20 Jun 2000 22:36:36 GMT
Subject: Dictionary vs if-elif?
Message-ID: <UlS35.15101$Xx5.781176@bgtnsc06-news.ops.worldnet.att.net>

Here is a code snippet
-------------------------------------------------------------------------
bogusCodes = { '1.':'', '31':'', '90':'', 'AZ':'', 'BH':'', 'C1':'',
               'CB':'', 'FR':'', 'HD':'', 'JF':'', 'KL':'', 'KU':'',
               'LU':'', 'ON':'', 'PF':'', 'Q3':'', 'UP':'', 'W4':'',
               'WO':''}

item  = data.readline()
while item:
  if (item[1:2] in digits) and (item[0:1] == 'D'):
    goodVend.write(item)
  elif vendors.has_key(item[0:2]):
    goodVend.write(item)
  elif bogusCodes.has_key(item[0:2]):
    pass
  else:
    missingVend.write('Vendor code %s is missing from file VendCode.txt\n' %
item[0:2])
  item  = data.readline()
---------------------------------------------------------------
originally I had it written like so:

if ...:
  pass
elif...:
  pass
etc.
-------------------------
another possibility is and if statement with a very complex condition:
if a or b or c ...:
  pass
else :
etc
---------------------
Is there a better way to do it?
Which of these three would be the preferred method?

Bill




From s2mdalle at titan.vcu.edu  Tue Jun 20 22:02:16 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Tue, 20 Jun 2000 21:02:16 -0500
Subject: Python segfault
Message-ID: <8ip7so$bv8$1@bob.news.rcn.net>

If there is a better place to forward this 'bug report' to, please
forward it if you know of it.  

First, here's a few details, and some gdb output, followed by
the code that caused this to hopefully shed some light:

[x at foobar /home/x/code/python/Grumble]$ rpm -q python
python-1.5.2-13
[x at foobar /home/x/code/python/Grumble]$ rpm -q tkinter
tkinter-1.5.2-13

[x at foobar /home/x/code/python/Grumble]$ python mumble.py
Segmentation fault (core dumped)

Here's some really ugly GDB output from the dump.....
[x at foobar /home/x/code/python/Grumble]$ ls -l core
-rw-------    1 x        x         7438336 Jun 20 20:48 core

[x at foobar /home/x/code/python/Grumble]$ gdb `which python` core
GNU gdb 4.18
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-redhat-linux"...
(no debugging symbols found)...
Core was generated by `python mumble.py'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /lib/libdl.so.2...done.
Reading symbols from /lib/libpthread.so.0...done.
Reading symbols from /lib/libm.so.6...done.
Reading symbols from /lib/libc.so.6...done.
Reading symbols from /lib/ld-linux.so.2...done.
Reading symbols from /usr/lib/python1.5/lib-dynload/selectmodule.so...done.
Reading symbols from /usr/lib/python1.5/lib-dynload/timemodule.so...done.
Reading symbols from /usr/lib/python1.5/lib-dynload/socketmodule.so...done.
Reading symbols from /usr/lib/python1.5/lib-dynload/stropmodule.so...done.
Reading symbols from /usr/lib/python1.5/lib-dynload/_tkinter.so...done.
Reading symbols from /usr/lib/libtix4.1.8.0.so...done.
Reading symbols from /usr/lib/libtk8.0.so...done.
Reading symbols from /usr/lib/libtcl8.0.so...done.
Reading symbols from /usr/X11R6/lib/libX11.so.6...done.
#0  0x400a95ae in chunk_alloc (ar_ptr=0xbfe0203c, nb=18086444) at malloc.c:2728
2728    malloc.c: No such file or directory.
(gdb) backtrace
#0  0x400a95ae in chunk_alloc (ar_ptr=0xbfe0203c, nb=18086444) at malloc.c:2728
#1  0x400a94be in malloc () at malloc.c:2181
#2  0x8072284 in PyString_FromString ()
#3  0x80593a8 in PyErr_SetString ()
#4  0x80516a6 in PyEval_EvalCode ()
#5  0x80541ab in PyEval_CallObjectWithKeywords ()
#6  0x8053dca in PyEval_CallObjectWithKeywords ()
#7  0x8065eab in PyInstance_New ()
#8  0x8053ed1 in PyEval_CallObjectWithKeywords ()
#9  0x8053dd8 in PyEval_CallObjectWithKeywords ()
#10 0x8059524 in PyErr_NormalizeException ()
#11 0x80595d3 in PyErr_NormalizeException ()
#12 0x80595d3 in PyErr_NormalizeException ()
#13 0x80595d3 in PyErr_NormalizeException ()
#14 0x80595d3 in PyErr_NormalizeException ()
#15 0x80595d3 in PyErr_NormalizeException ()
#16 0x80595d3 in PyErr_NormalizeException ()
#17 0x80595d3 in PyErr_NormalizeException ()
#18 0x80595d3 in PyErr_NormalizeException ()
#19 0x80595d3 in PyErr_NormalizeException ()
#20 0x80595d3 in PyErr_NormalizeException ()
#21 0x80595d3 in PyErr_NormalizeException ()
#22 0x80595d3 in PyErr_NormalizeException ()
#23 0x80595d3 in PyErr_NormalizeException ()
#24 0x80595d3 in PyErr_NormalizeException ()
#25 0x80595d3 in PyErr_NormalizeException ()
#26 0x80595d3 in PyErr_NormalizeException ()
#27 0x80595d3 in PyErr_NormalizeException ()
#28 0x80595d3 in PyErr_NormalizeException ()
#29 0x80595d3 in PyErr_NormalizeException ()
#30 0x80595d3 in PyErr_NormalizeException ()
#31 0x80595d3 in PyErr_NormalizeException ()
#32 0x80595d3 in PyErr_NormalizeException ()
#33 0x80595d3 in PyErr_NormalizeException ()
#34 0x80595d3 in PyErr_NormalizeException ()
#35 0x80595d3 in PyErr_NormalizeException ()
#36 0x80595d3 in PyErr_NormalizeException ()
#37 0x80595d3 in PyErr_NormalizeException ()
#38 0x80595d3 in PyErr_NormalizeException ()
#39 0x80595d3 in PyErr_NormalizeException ()
#40 0x80595d3 in PyErr_NormalizeException ()
#41 0x80595d3 in PyErr_NormalizeException ()
#42 0x80595d3 in PyErr_NormalizeException ()
#43 0x80595d3 in PyErr_NormalizeException ()
#44 0x80595d3 in PyErr_NormalizeException ()
#45 0x80595d3 in PyErr_NormalizeException ()
#46 0x80595d3 in PyErr_NormalizeException ()
#47 0x80595d3 in PyErr_NormalizeException ()
#48 0x80595d3 in PyErr_NormalizeException ()
#49 0x80595d3 in PyErr_NormalizeException ()
#50 0x80595d3 in PyErr_NormalizeException ()
#51 0x80595d3 in PyErr_NormalizeException ()
#52 0x80595d3 in PyErr_NormalizeException ()
#53 0x80595d3 in PyErr_NormalizeException ()
#54 0x80595d3 in PyErr_NormalizeException ()
#55 0x80595d3 in PyErr_NormalizeException ()
#56 0x80595d3 in PyErr_NormalizeException ()
#57 0x80595d3 in PyErr_NormalizeException ()
#58 0x80595d3 in PyErr_NormalizeException ()
#59 0x80595d3 in PyErr_NormalizeException ()
#60 0x80595d3 in PyErr_NormalizeException ()
#61 0x80595d3 in PyErr_NormalizeException ()
#62 0x80595d3 in PyErr_NormalizeException ()
#63 0x80595d3 in PyErr_NormalizeException ()
#64 0x80595d3 in PyErr_NormalizeException ()
#65 0x80595d3 in PyErr_NormalizeException ()
#66 0x80595d3 in PyErr_NormalizeException ()
#67 0x80595d3 in PyErr_NormalizeException ()
#68 0x80595d3 in PyErr_NormalizeException ()

(This just continues...I stopped scrolling at line #1053. 
They all say the same thing, i.e.
#1053 0x80595d3 in PyErr_NormalizeException ())

Now some code....a quick note on this, this uses Tkinter,
and all you need to do to replicate this, is go to the 
file menu, and choose the "Disconnect" option as the
very first thing you do, and it will segfault.  The disconnect
code is effectively trying to call a method "close()" on a 
object item that doesn't exist.  I think this is happening
at this line:

self.socket().close()

the socket() method fetches the socket item in the object.
At the point where this is happening, it doesn't exist.  But
python should spit out name error or no such attribute, not
die.  :)

Here's the code (there's more but to avoid burying you,
I'm cutting out the things that don't get called):

import select
import thread
import os
import time
import socket
import sys
import Pmw                      # Python Mega Widgets (for some GUI components)
import thread
import socket
import string
from types import *
from Tkinter import *
from GrumbleCommon import *
from GrumbleClient import *

class Mumble(GrumbleClient):
    def __init__(self):
        GrumbleClient.__init__(self)

        self.current_channel = None

        self.handlers = { "=cc"    : self.setCurrentChannel }

        self.window = Tk()
        self.frame = Frame(self.window)
        self.frame.pack(expand=1, fill='both')
        self.text = Pmw.ScrolledText(self.frame,
                                     hscrollmode = "dynamic",
                                     vscrollmode = "static")
       self.text.pack(side='top', fill='both', expand=1)
        self.bframe = Frame(self.frame)
        self.entry = Pmw.EntryField(self.bframe, command=self.insertData)
        self.OK = Button(self.bframe, text='Send', command=self.insertData)
        self.entry.pack(side='left', fill='both', expand=1)
        self.OK.pack(side='right')
        self.bframe.pack(side='bottom', fill='both', expand=1)

        self.make_menus()
        self.window.config(menu=self.menu)
    def make_menus(self):
        self.menu = Menu(self.frame)
        filemenu = Menu(self.menu)
        filemenu.add_command(label='Connect', command=self.new_connection)
        filemenu.add_command(label='Disconnect',
                             command=self.disconnect_wrapper)
        filemenu.add_command(label='Quit', command=self.quit)

        editmenu = Menu(self.menu)
        editmenu.add_command(label='Cut',   command=self.cut)
        editmenu.add_command(label='Copy',  command=self.copy)
        editmenu.add_command(label='Paste', command=self.paste)

        self.menu.add_cascade(label="File", menu=filemenu)
        self.menu.add_cascade(label="Edit", menu=editmenu)
    def copy(self):
        try:
            self.buffer = self.text.get(SEL_FIRST, SEL_LAST)
        except:
            pass
        return(None)

    def cut(self):
        try:
            self.buffer = self.text.get(SEL_FIRST, SEL_LAST)
            self.text.delete(SEL_FIRST, SEL_LAST)
        except:
            pass
        return(None)

    def insert(self, buffer, asis=None):
        if not buffer:
            return None
        # If asis is not true, then append a linebreak
        # if it is needed.  If asis is true, then leave it
        # 'as is'
        if not asis:
            try:
                if buffer[-1] != '\n' and buffer[-1] == '\r':
                    buffer = buffer + '\n'
            except:
                return None

        try:
            self.text.insert('end', buffer)
        except:
            print "Error inserting text into main widget!"
            pass
        return None

    def paste(self):
        if self.buffer != None and self.buffer != '':
            self.insert(self.buffer, 1)
            return(None)

    def quit(self):
        try:
            self.disconnect()
        except:
            pass
        self.window.destroy()
        sys.exit(0)
    def message(self, text):
        self.insert('\n****CLIENT:\n****' + text + '\n')
        return None

    def report_error(self, text):
        self.insert("\n****CLIENT ERROR:\n****" + text + '\n')
        return None
    def connect_wrapper(self, host=None, port=None, nick=None):
        self.message("Connecting to %s:%s..." % (host, port))
        self.connect(host, port, nick)

        thread.start_new_thread(self.reader, ())
        self.message('Connected to %s:%s' % (host, port))

    def insertData(self):
        text = self.entry.get()
        self.entry.setentry('')
        if text[0] == '=':
            retty = self.interpret(text)
            if retty:
                return None

        text = text + '\n'

        if self.current_channel:
            self.talk(self.current_channel, text)
        else:
            self.send(text)

x = Mumble()
mainloop() 

And here's the GrumbleClient.py file.  It segfaults when
self.disconnect() gets called, and that was inherited from
GrumbleClient:
import select
import thread
import os
import time
import socket
import sys
import string

from types import *

class GrumbleClient:
    GrumbleClientException = 'GrumbleClient'
    DEFAULT_BUFFER_LENGTH = 1024

    def __init__(self):
        self._socket    = None
        self._nick      = 'Guest'
        self._host      = 'localhost.localdomain'
        self._port      = 666
      self._connected = None
        return None

    def quit(self):
        return self.send_command('/quit')

    # Same as quit()
    def close(self):
        return self.quit()

    def part(self, channel):
        return self.send_command('/part %s' % channel)

    def change_nick(self, newnick):
        return self.send_command('/nick %s' % newnick)

    def topic(self, channel, newtopic=None):
        if newtopic:
            return self.send_command('/topic %s %s' % (channel, newtopic))
        else:
            return self.send_command('/topic %s' % channel)
        return None

    def join(self, channel):
        return self.send_command('/join %s' % channel)

    def kibitz(self, text):
        return self.send('kibitz!%s' % text)

    # Send a "talked" message (to a specific channel)
    def talk(self, channel, text):
        if (not channel) or not (type(channel) is StringType):
            err = 'Illegal channel argument to talk()'
            raise GrumbleClient.GrumbleClientException, err
        if (not text) or len(text) <= 0:
            return None
        return self.send('%s!%s\n' % (channel, text))

    # Read data off of the socket, up to len bytes
    def read(self, len=None):
        if len:
            return self.socket().recv(GrumbleClient.DEFAULT_BUFFER_LENGTH)
        else:
            return self.socket().recv(len)
        return None

    # Send data to the raw socket
    def send(self, data):
        try:
            self.socket().send(data)
        except socket.error, errstr:
            raise GrumbleClient.GrumbleClientException, errstr[1]
        return 1

    # SEGFAULT IS HAPPENING SOMEWHERE IN HERE
    def disconnect(self):
        self.quit()
        try:
            self.socket().close()
        except socket.error, errstr:
            e = 'Socket error: %s' % errstr[1]
            raise GrumbleClient.GrumbleClientException, e
        self.connected(None)
        return 1
    def socket(self, sock=None):
        if sock:
            self._socket = sock
        try:
            return self._socket
        except:
            raise GrumbleClient.GrumbleClientException, 'Socket undefined'

(lots of code clipped that isn't referenced by this bug)

I'm willing to send full source code to whoever wants/needs it.
I'm hoping that the code exerpts will be enough with the gdb
stuff to get at whatever is happening here.
While the coding isn't perfect, (this is a very early prototype,
and I'm a python beginner), I figure that the way it SHOULD be,
no matter HOW fucked up my programming is, segfaulting is
NEVER the correct response for the python interpreter.  From
the gdb output, it looks like the NormalizeException funciton
is calling itself until it croaks.
-- 
David Allen
http://opop.nols.com/



From petersc at stud.ntnu.no  Wed Jun  7 17:14:31 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 07 Jun 2000 23:14:31 +0200
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <393E5B21.B5F5B3C6@ecs.soton.ac.uk>
Message-ID: <393EBB37.E5FEEFE8@stud.ntnu.no>

Jacek Generowicz wrote:
> 
[lots of very nice translation summarized beautifully]

> That's about it . . . you haven't really missed much, except for the very first
> felicitation about which I still haven't made up my mind: did he mean to write
> what he wrote, or did the message get slightly mangled somewhere in the process .
> . . come on Peter, I think we should be told: was it your joke, or your Belgian
> friend's joke on you ?

You are allowed to make up your mind on your own. I'll
just supply you with some basic facts:

a) I did not make it up. My Belgian and two Dutch friends did.
b) I knew what it meant. I can understand that much.

assuming-that-it-is-the-Dutch-way-or-believing-Guido-to-have-
a-sense-of-humour-?-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From dgrisby at uk.research.att.com  Thu Jun  8 09:08:04 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 8 Jun 2000 13:08:04 GMT
Subject: Some questions about C extensions
References: <etdu2f69rhz.fsf@w20-575-109.mit.edu>
Message-ID: <8ho5rk$c71$1@pegasus.csx.cam.ac.uk>

In article <etdu2f69rhz.fsf at w20-575-109.mit.edu>,
 Alex  <cut_me_out at hotmail.com> wrote:

>-- Deallocation of memory for strings returned by python functions:
>   Do I have to do it, or is it somehow taken care of?  If I have
>   something like 
>
>      if (!PyArg_ParseTuple (args, "s", &sequence)) {
>           return NULL;}
>
>   do I have to do a free (sequence); at some point?  I would have
>   thought so, but it seems to be giving me a Seg fault.

No -- sequence points to the string which is inside the Python string
object.

>-- Reference counts for elements of brand new tuples.  Why do I have to
>   do a Py_INCREF on an object before I assign it to a new tuple with
>   PyTuple_SetItem?  Since it seems as though you have to do a Py_INCREF
>   every time anyway, is there a function like PyTuple_SetItem that will
>   do it for you?  The comment in the C API manual about this is very
>   cryptic, BTW.

You don't necessarily have to do Py_INCREF when you insert things in a
tuple. I often write code which does something like:

  PyObject *a, *b, *t;

  a = PyInt_FromLong(1234);
  b = PyString_FromString("Hello");
  t = PyTuple_New(2);
  PyTuple_SetItem(t, 0, a);
  PyTuple_SetItem(t, 1, b);

  /* ... do something with the tuple... */

  Py_DECREF(t);

That way, when the tuple is deleted, the int and string objects inside
it are also deleted.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From Michaell.Taylor at reis.com  Wed Jun 14 13:53:03 2000
From: Michaell.Taylor at reis.com (Michaell Taylor)
Date: Wed, 14 Jun 2000 13:53:03 -0400
Subject: stdout
Message-ID: <NDBBJMANPNIHDHDJJBHMKEGMCMAA.Michaell.Taylor@reis.com>


I am having a bit of trouble redirecting output from a file back to the
terminal.  I have:

import sys
f=open('c:\output.txt', 'w')
sys.stdout=f
blah..
blah..
blah..
sys.stdout='<stdout>'
print 'finished'


which does not work.  I've searched the docs but only see references for
directing to files (which is helpful) but none for redirecting to a screen.
The last two lines aren't doing the trick.  Any ideas?

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

Michaell Taylor
Senior Economist, Economics Research Group, Reis, New York City, USA





From cut_me_out at hotmail.com  Mon Jun  5 23:20:09 2000
From: cut_me_out at hotmail.com (Alex)
Date: 05 Jun 2000 23:20:09 -0400
Subject: SSL and POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com>
Message-ID: <etdog5fcx7a.fsf@w20-575-109.mit.edu>

What happens when you try to do a POST?

Alex.


From olivierS.dagenaisP at canadaA.comM  Fri Jun 30 08:45:28 2000
From: olivierS.dagenaisP at canadaA.comM (Olivier Dagenais)
Date: Fri, 30 Jun 2000 12:45:28 GMT
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr> <8jhuvn01vs0@news2.newsguy.com> <962367319.379353174@news.libertysurf.fr>
Message-ID: <ID075.11997$W35.349783@news20.bellglobal.com>

I usually don't use accessors to manipulate my member variables, unless my
accessors trigger some sort of "changed" flag, so that I know my object
needs to be updated or re-persisted.

--
----------------------------------------------------------------------
Olivier A. Dagenais - Carleton University - Computer Science III


"Jerome Quelin" <jerome.quelin at insalien.org> wrote in message
news:962367319.379353174 at news.libertysurf.fr...
> "Alex Martelli" <alex at magenta.com> wrote:
> >By calling the accessors, rather than accessing the data
> >directly, you ensure that if somebody overrides the
> >accessor functions in a derived class you'll be calling
> >the derived-class versions of those functions.  If that is
> >what you want (one direction along which you want to let
> >inheritors of your class extend or tweak behaviour), then
> >it's probably worth the overhead you mention.
> The class isn't to be derived any further. It's a class that I use in an
> application, and that may be imported in some other applications, but not
to be
> inherited.
>
> Then, is it better/cleaner to access private members with accessors or
not?
> Is it a matter of style? Or are there hidden caveheats using (or not
using)
> accessors? It _seems_ cleaner to use accessors, but man, it's quite
awkward.
>
> Jerome
> --
> jerome.quelin at insalien.org




From redaelli at inc.it  Fri Jun  9 17:56:18 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Fri, 09 Jun 2000 23:56:18 +0200
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl> <393D1155.D067A821@decrc.abb.de> <m33dmqygdj.fsf@atrus.jesus.cam.ac.uk> <slrn8jt5ck.779.jepler.lnk@potty.housenet> <8hocrk$na2$1@netserv.univ-lille1.fr>
Message-ID: <39416802.9820E032@inc.it>

It seems to be a locale related problem. I've been told that switching
locale stops the problem. But it's silly...
-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/


From R.Brodie at rl.ac.uk  Thu Jun 22 12:53:29 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Thu, 22 Jun 2000 17:53:29 +0100
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <03bf34ab.667a04d2@usw-ex0105-040.remarq.com>
Message-ID: <8itga8$ibk@newton.cc.rl.ac.uk>

"RickZ" <rzantowNOrzSPAM at usa.net.invalid> wrote in message
news:03bf34ab.667a04d2 at usw-ex0105-040.remarq.com...

> I'd like to get my employer to add Python to its "approved" list
> of development tools, too. But which Python? It's an extra
> handicap to say, "Well, the language now is changing constantly
> in ways that will potentially break what we develop now. And I
> am not at all sure what it will look like at this time next
> year.

I don't think you can consistantly say that about Python but not
Java. There was the bug AWT/Swing thing, and various other
deprecated features.

I think you are making too much of the incompatible change thing.
It's much easier to drop features where you have control of the
source: the fact that, with Python possible, compatibility problems
get discussed in public early on in the development cycle shouldn't
be seen as a problem.

You can always freeze the version you are using on the production
machines. Regression testing on N different JVMs isn't going to be
any easier.




From bill at libc.org  Wed Jun 28 19:34:23 2000
From: bill at libc.org (Bill Anderson)
Date: Wed, 28 Jun 2000 17:34:23 -0600
Subject: Which linux distros have Python by default?
References: <642A954DD517D411B20C00508BCF23B001296B43@mail.sauder.com>
Message-ID: <395A8B7F.6CAF87CC@libc.org>

Moshe Zadka wrote:
> 
> On Wed, 14 Jun 2000, Grant Edwards wrote:
> 
> > Nah, I wrote a TCL program once and swore never again.  Since
> > this is just a GUI front-end for some command line utilities,
> > the user who doesn't have Python/Tkinter can get along fine
> > without it.
> 
> Suggestion: write a Python module to wrap the command line utilities, and
> have two UIs available: a Python/Tkinter GUI and a cmd-based high-level
> command line utility.
> 
> Exercise for the advanced reader: write a Python/Curses, a Python/GNOME
> and a Python/KDE GUIs too <0.9 wink>


Extra credit: Write a How-To when done. :)


From nikolai.kirsebom at siemens.no  Wed Jun 21 06:59:37 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Wed, 21 Jun 2000 10:59:37 GMT
Subject: win32all-132 / Word Automation
References: <326536345498D311B3BC00105A39802A07473D@newsintern.dspace.de>
Message-ID: <39509f4c.72298439@news.mch.sni.de>

Is the change from tuple to int related to Python version 1.6 ?

I thought that versions 129 .. 131 of win32all were only for Python
1.6 - and that version 132 is only for Python version 1.5.2.

Deleting the generated (gen-py) files does not help.

Nikolai

On Tue, 20 Jun 2000 14:16:42 +0100, Stefan Migowsky
<smigowsky at dspace.de> wrote:

>Hi,
>
>probably there is a mix up between the generated wrappers.
>In versions newer than 129, the parameter defines for calls
>changed from tuple values to ints. 
>
>You could try to delete the gen-py generated file and then do
>the test again. The object model of word clearly states that
>Variables.Item returns a dispatch pointer to a Variable object.
>
>  Stefan 
>



From siva1311 at my-deja.com  Wed Jun 28 06:12:34 2000
From: siva1311 at my-deja.com (siva1311 at my-deja.com)
Date: Wed, 28 Jun 2000 10:12:34 GMT
Subject: multiple replacements
References: <8ja7ok$6be$1@nnrp1.deja.com>
Message-ID: <8jcj2b$vuc$1@nnrp1.deja.com>

Thank you all for the suggestions. It's nice to find people willing
to share their expertise.

-Brad


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


From skip at mojam.com  Fri Jun  9 16:47:33 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 9 Jun 2000 15:47:33 -0500 (CDT)
Subject: [Python-Dev] Re: Porting CPython
In-Reply-To: <Pine.GSO.4.10.10006092133190.28574-100000@sundial>
References: <8hrbph$a4n$1@nnrp1.deja.com>
	<Pine.GSO.4.10.10006092133190.28574-100000@sundial>
Message-ID: <14657.22501.30185.873713@beluga.mojam.com>

    >> - Excluding Modules
    >> 
    >> What is the build procedure for omitting certain built in C modules?
    >> (I"m talking about the stuff in Modules/)

    Moshe> Edit the "Setup.in" file. Instructions for how are inside the
    Moshe> file, but if you have any more specific questions, please ask
    Moshe> again.

Actually, I believe you should edit Setup.  The first time you build, Setup
will be generated from Setup.in, and then not modified later.  These days,
when turning on modules that are not on by default, I prefer to edit
Setup.local, even for modules in the core distribution.

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
On June 24th at 8AM, live your life for an hour as Ricky Byrdsong always lived
his - run/walk in the Ricky Byrdsong Memorial 5K:
    http://www.cararuns.org/raceinfo/otherevents.html
    http://www.chicagoaa.com/calendar/running.html#anchorjunerun




From wlav at hpatl26.cern.ch  Mon Jun 19 05:51:02 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 19 Jun 2000 09:51:02 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB333@ALTNET> <8hr8fj$fu$1@sunnews.cern.ch> <B9e05.1145$vp3.81025@news1.rdc2.tx.home.com> <8i7frt$o02$1@sunnews.cern.ch> <etS15.1090$OZ.300179@news1.rdc2.tx.home.com>
Message-ID: <8ikqe6$dun$1@sunnews.cern.ch>

"Rainer Deyke" <root at rainerdeyke.com> writes:
>Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote in message

>> Nonsense. You can specialize std::set with your own version of comparison
>> class instead of the standard less and you can make sure that your version
>> of it does not require said operators.

>True, but writing your own comparison class is at least as much work as
>implementing operator <.

Whether it is a lot of work depends on how many classes will use your new
version of less<>.

>Besides, IIRC you still need operator ==.

YDNRC. Perhaps you care to read clause 3 in paragraph 20.1 of the C++
standard to refresh your memory?

>> >Misleading because for many classes the compiler generated functions cause
>> >incorrect behavior.
>>
>> As I have explained in another post, 'many' is only true for buggy code
>> that is not exception safe.

>For some values of 'many'.  Smart pointers can help, but the smart pointer
>classes themselves almost always need explicit copy contructors and operator
>=,

Agreed. Then again, no-one should write his own smart pointer. (I admit,
I sinned, I wrote a smart pointer: it was copied, modified and proliferated
by several people. I learned, won't do it again.)

>and if you deal with a variety of objects and resources you'll need a
>variety of smart pointers.

Agreed.

>For example, COM objects would need their own smart pointer class. 

COM is a particular bad example as MS delivers several different smart
pointers for this technology and most textbooks on COM come with yet
another version.

Best regards,
        Wim Lavrijsen

(Who nows damn well that he just delivered another argument against being
able to write your own reference semantics.)


From contactXXX at redspice.com  Tue Jun  6 01:57:24 2000
From: contactXXX at redspice.com (contactXXX at redspice.com)
Date: Tue, 06 Jun 2000 05:57:24 GMT
Subject: Zope CGI
References: <3933c68a.4779639@192.168.0.1>
Message-ID: <393C9052.4CE904A4@redspice.com>

"Georg Gogo. BERNHARD" wrote:
> 
> Hello!
> Has anybody out there written a python program that can call a
> compiled CGI program? Is it possible to call a CGI program with
> Python/Zope in a way that its i/o gets redirected? Any solutions are
> welcome.
> pancakelizard at gmx.net

Do you want to set up a pipe and fork the cgi executable. you can dup
the stdin and stdout to the pipe descriptors. You can then
read the o/p from the cgi at the input end of your pipe from the python
program.


From david_ullrich at my-deja.com  Thu Jun 15 13:19:12 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 15 Jun 2000 17:19:12 GMT
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <20000614155102.E16232@freei.com>
Message-ID: <8ib35r$4ff$1@nnrp1.deja.com>

In article <20000614155102.E16232 at freei.com>,
  Sean Blakey <sblakey at freei.net> wrote:
> On Wed, Jun 14, 2000 at 05:48:45PM -0400, Daehyok Shin wrote:
> > How can I prevent illegal definition of a variable to an object?
> > For instance,
> >
> > class A:
> >     __init__(self):
> >             self.x = 1
> >
> >     __setattr__(self, name, value):
> >             if name == "x":
> >                     self.x = value
> >             else:
> >                 raise AttributeException
> >
> >     __getattr__(self, name):
> >             if name == "x":
> >                 return self.x
> >             else:
> >                 raise AttributeException
> >
> > >>> a = A()
> > >>> a.x = 10
> > >>> a.y = 20 # I want to raise an AttributeException.
> >
> > Peter
> >
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
>
> The code you posted seems to do what you want, provided you
> a)add the 'def' keyword to your function definitions
>     and
> b) define AttributeException (perhaps you meant AttributeError?)

   That's the way it seems all right, until you consider the
fact that the line "self.x = value" is going to call the
__setattr__ method... (I imagine this has been pointed out
already but I don't see it below - seems worth mentioning
explicitly for the benefit of anyone who doesn't see the
problem.)

> --
> Sean Blakey, sblakey at freei.com
> Software Developer, FreeInternet.com
> (253)796-6500x1025
> 	"Uncle Cosmo ... why do they call this a word processor?"
> 	"It's simple, Skyler ... you've seen what food processors do to
food,
> right?"
> 		-- MacNelley, "Shoe"
>
>


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


From wjdandreta at worldnet.att.net  Sun Jun 18 23:22:06 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Mon, 19 Jun 2000 03:22:06 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com>
Message-ID: <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net>

Hi Christian,

Christian Tismer wrote in message
<394D6579.8B2FF840 at appliedbiometrics.com>...

>Are you sure that you are talking about Python 1.2 ???
>Let me assume it is 1.5.2 - ok?


It is Python 1.2 because the later versions will not run on my machine.

>1) both joinfields and splitfields are deprecated. Please use
>   split and join instead.
>2) Is the split(join(...)) pattern necessary here? If the pattern
>   is a simple string, then you might be reasonably faster with
>   a single string.replace(str, oldstr, newstr) call.


I tried split and join first but the versions in strop have undocumented
parameters so I switched to splitfields, joinfields. Also, if you check the
source for string.py, split calls splitfields and join calls joinfields.

the replace function in string.py is simply

def replace (...

   """
   DOC String

  """

  return joinfields(splitfields(...)...)


>The time spent in comment lines is very small,and it is reasonably
>smaller if you use Stackless Python. Anyway, if you are concerned
>about multi-line comments, then try this out:
>Put a """ line in front of a longer couple of comments,
>and end it the same way.
>I should wonder if execution time would change reasonably.


The """ syntax would seem to be slower than putting # at the begining of
everyline because with # it only needs to test the first character of the
line, if it's # it can skip the rest of the line. But with """, it needs to
check every character after until it finds the matching """.


>If you move all of your comment text into the first line of your
>funcion, then it will become the DOC string, and it will cost
>you not a single cycle.

What we are talking about here is the DOC string in the Python source file
string.py. I will test it with your suggestions and see what happens.

Bill




































From thomas at cintra.no  Tue Jun 27 06:15:34 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 27 Jun 2000 10:15:34 GMT
Subject: Comments: full-text indexing of RDBMS
Message-ID: <395b7627.97196711@news.online.no>

Hi,

I`ve cooked up some plans for a simple, note! simple, full-text
index-engine and would like some comments. I got little or no
experience in the matter, some in database and a pretty happy
python-camper, so the things here might be dumb and ineffective, but
since it`s not outthere yet I have to do it myself.

I use PostgreSQL 7.0.2, Darcy`s database-module, latest version, all
under Linux.

I want to do simple searches like "Alien,Widescreen" and the result
should be all records were both "Alien" and "Widescreen" occurr.

This is how I thoguht it could be done :

I get a list of all the tables in a specified database from the
databasemodule, remove all database-system tables, goes thru each
table, select * from table. The result ends up in a dictionary-object.
I check all the values for plain text. Integers and numbers in general
are ignored. A function take the plain text and returns a list of
words found in that record. I put each word in a new table, gives it a
unique id. A record in PostgreSQL has a OID, an uniqe
object-identifier.  The OID is mapped against the ids of the words
that occurred in that record. 

When I look for a set of records I get an intersection of all the ids
where those words appears. I go thru the list of ids and fetches them
from the database.

I might have to keep track of which table the id appears in too. Don`t
know if PostgreSQL can fetch a record based on an oid alone or if the
python-module supports that function. 

The Python part should be pretty simple, but I`d like to hear how
others would do a thing like this too.

I can email any working code to interested parties.

Thanks.

Thomas


From jcm at bigskytel.com  Mon Jun 26 23:11:08 2000
From: jcm at bigskytel.com (David Porter)
Date: Mon, 26 Jun 2000 21:11:08 -0600
Subject: smtplib
In-Reply-To: <000801bfdfca$61227820$b6cb45c6@timaratz>; from timaratz@lisco.com on Mon, Jun 26, 2000 at 06:58:09PM -0500
References: <6mK55.1713$iN5.345307@ptah.visi.com> <000801bfdfca$61227820$b6cb45c6@timaratz>
Message-ID: <20000626211107.A18924@novara.avenue>

* Peter Timaratz <timaratz at lisco.com>:
> As a kindness to all the idiots like me perhaps this intro can be added to
> the smtplib module:
> 
> "This module implements SMTP and does not send e-mail. If you want to send
> e-mail with Python then learn SMTP. If you don't want to learn SMTP then you
> can use one of the many Perl modules available that don't require you to
> know SMTP."

Recommending a perl module would be strange. Then the 'idiots' would ask how
they import perl modules into python ;-) This would accomplish nothing. If
someone wants perl, I can hardly imagine them reading the source of the
python library. Not to mention that learning perl would require
substantially more effort than reading about the SMTP protocol. 



From wjdandreta at worldnet.att.net  Fri Jun  9 01:14:33 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 09 Jun 2000 05:14:33 GMT
Subject: Problem compiling source
Message-ID: <Z2%%4.1751$iy.137366@bgtnsc06-news.ops.worldnet.att.net>

This is the first time I have tried to post a message to a news group. I
hope it works.

I am trying to compile python 1.5.2 for DOS using Borland's C++ 4.5
compiler. I have run across the following problem:

off_t offset;

gives an error message
256: Undefined symbol 'off_t' in function file_seek

The problem seems to be that off_t is only defined if __FLAT__ is defined.

While __FLAT__ appears in many Borland C header files, I cannot find the one
where __FLAT__ is defined.

Can anyone tell me what __FLAT__ is for and which header it is defined in?

Bill Dandreta




From fiona at sitegnome.com  Mon Jun 26 22:17:08 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Tue, 27 Jun 2000 12:17:08 +1000
Subject: The automagic of Tkinter
References: <20000626135336.A8180@bork> <8j7p5e$qia$1@nnrp2.deja.com>
Message-ID: <39580EA4.6F818D5C@sitegnome.com>

> Your class Knop inherits from Button which in
> turn inherits from Widget which inherits from
> BaseWidget.

....

> What does knop mean anyway?

In dutch: 

knop == button

:-)


From kern at caltech.edu  Thu Jun 22 20:48:33 2000
From: kern at caltech.edu (Robert Kern)
Date: Thu, 22 Jun 2000 20:48:33 -0400
Subject: Countdown
References: <3952952D.88E3588D@asme.org>
Message-ID: <3952B3E1.DC36612B@caltech.edu>

Jeff Sandys wrote:
> 
> Python is really elegant when counting up:
> 
> for i in range(len(thing)):
>     print thing[i]
> 
> But when counting down looks ugly:
> 
> for j in range(len(thing)-1,-1,-1):
>     print thing[i]
> 
> Advice so far:
> Stop doing things backwards:
>     (customer request)
> Use reverse first:
>     (destructive, need to unreverse after)
> Live with it:
>     (bytecode doesn't care, but it is still ugly)
> Try range(len(thing)).reverse():
>     (doesn't work but):
>         rthing = range(len(thing))
>         rthing.reverse()
>         for k in rthing:
>             print thing[i]
>     (works, but also ugly)

So stick it in a function.

def revrange(*args, **kw):
    l = apply(range, args, kw)
    l.reverse()
    return l

--
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From frey at weasel.bme.unc.edu  Thu Jun  1 22:02:37 2000
From: frey at weasel.bme.unc.edu (Eric Frey)
Date: 2 Jun 2000 02:02:37 GMT
Subject: Compiling PIL under RH6.2
Message-ID: <8h74jt$d9r$1@news2.isis.unc.edu>

I have been trying to compile Python imaging library under RH6.2 (I was
unable to find a rpm). Unfortunately I am not at the system right now,
but when I do the make boot (using the appropriate makefile) I get
an error saying that it is unable to make /usr/lib/python/lib/config/Makefile
(or something like that). This sounds like a file which would be created
if I had compiled python myself from the sources (which I didn't do, I 
installed the RPMs). Is there anyone who has successfully compiled PIL
under RH6.2? Any hints as to what to do?

=======================================================================
Eric C. Frey                   
Assistant Professor 	         
Department of Biomedical Engineering and Department of Radiology
The University of North Carolina
Chapel Hill, NC


From jkraska1 at san.rr.com  Fri Jun  9 11:08:49 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 09 Jun 2000 15:08:49 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl>
Message-ID: <39410927.3B868E24@san.rr.com>

> > First off, any claim of an order of magnitude increase in productivity is
> > pure nonsense. See Brooks for an explanation or Boehms and McConnell for
> > hard data.
> 
> That's likely true. I do suspect there are more anecdotal examples of
> "I did the same project in Python and I finished in half the time!" compared
> to C++ than the other way around, but these aren't very scientific.

I've used C++ since 1985. Recent experiences with C++, Lisp, Java, and
Python all have me favoring Python for productivity. Is there an order
of magnitude increase in producitivity? Well base 2 maybe. :)

> >>2. How long did you use Python before you achieved increased
> >>productivity ?

About 2 weeks.
 
> > Up to now, I've been using C++ for scripting. I'm not proficient in Python
> > yet, so writing a Python script takes far longer than writing a C++ utility
> > and the Python code itself is larger than the equivalent C++ code.

It only takes about 2 weeks become proficient in Python.

> But then you must be very good at C++. :) (Hi Wim!) My experience was that I
> was faster than C++ after a couple of days already. Note that I haven't had
> any real experience with post-STL C++, which may or may not matter.

Stl? *S*tl? Could it be?!?!?! Satan!!! :^)

> It's perhaps just a difference in philosophies as well; C++ is a statically
> typed language and its proponents put a lot of emphasis on this kind of
> correctness. In Python you might come up with lots of correctness rules
> as well, but people don't, as the whole approach is different. People
> rarely care about things like public/private, even. This also means a lack
> of getters/setters in many cases; those will only be created if they
> are deemed necessary. This means encapsulation can be broken easily,
> but Python programmers don't care about this so much -- in practice often
> it is more of a problem to strictly preserve encapsulation (and type safety)
> than it is to fix any problems when and if they arise.

This flexibility of approach allows someone to concentrate more
on the problem domain and less on their language, IMO.


C/


From daniel at visionart.com  Fri Jun 16 04:09:39 2000
From: daniel at visionart.com (Daniel Kramer)
Date: Fri, 16 Jun 2000 08:09:39 GMT
Subject: tcl/tk vs. python/Tkinter - Example attached
References: <3946B170.17A2630@visionart.com> <8i8meh$cnc$1@msunews.cl.msu.edu>
Message-ID: <7hl25.2322$3E6.42905@news1.alsv1.occa.home.com>

> > Since I am new to this lang I was just wondering if there was a way to
> > get better python/TK performance... if anyone could try out the examples
> > and let me know.. for me the tcl/tk one feels much more responsive but I
> > don't have any hard numbers to prove that.
>
> > I have been reading that wxWindows is the much faster solution.. but I
> > am very familiar with TK and it's plenty fast for what I need.. but I
> > worry that the Tkinter ver won't be fast enough once I add alot more to
> > it.
>
>
> > -Daniel
>
>
>
>   Just adding my 2 cents here. I think I see  a very slight difference
>   on  my linux machine with all the latest versions of
>   interpreters. It's a celeron 366 with 128 Megs of RAM.
>   The python version seems just a tiny bit slower if I move the boxes
>   around very fast. And oh, the difference almost disappears for me on
>   commenting out the following line.
>
>      self.canvas.itemconfig(Node.currentNode[0] + "_base",fill = "red")
>
>      Could this change in color be causing the  perception of
>      jumpiness....?? The boxes always remain grey in the tcl version.
>
>
>       Chetan
>
Path: news!uunet!ffx.uu.net!newsfeed.easynews.net!easynews.net!newsfeed.germany.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!newsfeed.icl.net!newsfeed.direct.ca!newshub2.rdc1.sfba.home.com!news.home.com!news1.alsv1.occa.home.com.POSTED!not-for-mail
Xref: news comp.lang.python:99881


I think I have just confirmed to myself that it's quite a bit slower.. it's
at least more cpu dependent then tcl/tk.   I taxed my cpu very heavily with
another program (I started rendering an image w/a raytracer).. then I opened
both the tcl/tk and the python/tkinter versions of the canvas program.. the
tcl/tk showed minimal slowdown, it was still very fluid.. while the
python/tkinter one was very choppy in comparison..

It sounds like python is doing alot more work translating the info to tk.. I
wonder if this will improve in the future.. it's makes me hesitant to start
a project on python/tkinter.

-Daniel





From mnot at mnot.net  Tue Jun 27 14:40:17 2000
From: mnot at mnot.net (Mark Nottingham)
Date: Tue, 27 Jun 2000 11:40:17 -0700
Subject: ANN: cgi-buffer 0.1
In-Reply-To: <14680.56174.25987.149715@beluga.mojam.com>
References: <14680.56174.25987.149715@beluga.mojam.com>
Message-ID: <962131217.3958f51128dbc@mail.mnot.net>

Quoting Skip Montanaro <skip at mojam.com>:

> After taking a quick peek at the referenced web page for cgi-buffer, a
> couple things came to mind:
> 
>     1. Python 1.6 will support HTTP/1.1.  Don't know if this will affect
>        your library or not.

I've heard rumours of this, and have been meaning to check it out and give
feedback (part of my day job is HTTP compliance testing).

The question is, what do you mean by 'support HTTP/1.1'? That covers a lot of
ground. My library probably won't be affected, because it improves things for
1.0 clients as well, and the 1.1 mechanisms are probably not implemented. I'll
be writing up a paper explaining this When I Get Time (ha).


>     2. Python 1.6 will have a new module, atexit, that adds a little smidgen
>        of protocol on top of sys.exitfunc in the form of a register
>        function, thus allowin multiple modules to register exit functions
>        without stepping on each others' virtual toes.  If atexit gets loaded
>        and sees that sys.exitfunc is already defined (and not its own exit
>        func) it will simply add that function to the list of exit functions,
>        so your code should work unchanged *if* it gets there before atexit
>        does.

COOL! I was pondering on that just the other day.

Cheers,



From mal at lemburg.com  Fri Jun 23 11:29:11 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 23 Jun 2000 17:29:11 +0200
Subject: SAP R/3 and Python?
References: <C59F2D5004BDD311BEC500508B6FF2AD08E816@dewdfx16>
Message-ID: <39538247.28CFCB57@lemburg.com>

"Wegener, Joerg" wrote:
> 
> I think the best way to access R/3 is via a protocol called RFC (Remote
> Function Call) or via BAPIs (Business APIs). IIRC there is something called
> RFC-SDK which allows to write or generate code (I'm not sure here) which
> gives access to RFCs via DDE/OLE. Use this in combination with WIN32-Python.
> The downside is that you need much more knowledge about R/3 functions, but
> this is the best and most reliable way.
> 
> Hope this helps

Yes... at least it points into a direction which looks like a possible
solution.

Could it be that SAP (or a partner) already has such a Python
interface available ? I wouldn't want to reinvent the wheel
here and I assume this to be a highly non-trivial and time
consuming task.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/



From paul at fizzylab.com  Sat Jun 17 00:52:47 2000
From: paul at fizzylab.com (paul at fizzylab.com)
Date: Fri, 16 Jun 2000 21:52:47 -0700
Subject: broken examples in reference? Help??
References: <3949C65F.67A405F1@fizzylab.com> <u66rat92q.fsf@ctwd0143.fitlinxx.com>
Message-ID: <394B041F.28A50269@fizzylab.com>

you guessed it: i wasn't sufficiently caffienated and missed those parens.



David Bolen wrote:

> paul at fizzylab.com writes:
>
> > I am trying to work with python and it seems quite
> > straightforward most of the time, but I fear I may have a
> > broken binary or something else wrong.
>
> > I have typed verbatim an example from the python essential
> > reference and it doesn't seem to work: it just hangs. (...)
>
> Nope, it's just a typo - the line:
>
> > h.endheaders
>
> Should be a method call, so instead you should use:
>
>     h.endheaders()
>
> Written the way you have it, it's just going to represent the function
> object.  If you feed these statements in interactively, you'll see the
> output from the interpreter:
>
>     >>> h.endheaders
>     <method HTTP.endheaders of HTTP instance at 80eb90>
>
> so it's just printing the value of h.endheaders (the method reference)
> and not actually executing it, thus the headers are never terminated.
> Actually the statement doesn't really do anything, but it's the nature
> of the interpreter to print the value of any line that produces
> something other than None.
>
> My Python Essential Reference copy does have the () on this line, but
> I just bought it, so it may be a more recent printing than yours and
> perhaps it used to be a typo in the book.  (Or perhaps you just missed
> it when typing it in?  :-))
>
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/



From claird at starbase.neosoft.com  Wed Jun 21 15:44:58 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Jun 2000 14:44:58 -0500
Subject: snmp support in Python
References: <Pine.LNX.4.10.10006211416200.7273-100000@worf.mcnc.org>
Message-ID: <9B78E2D05837666B.7528A8300EF4F16B.0445A13F2004E0AD@lp.airnews.net>

In article <Pine.LNX.4.10.10006211416200.7273-100000 at worf.mcnc.org>,
oliver  <fwang2 at yahoo.com> wrote:
>Hi,
>
>Can someone update me the current status of SNMP support in Python? I
>remember a while ago, there were a few messages saying an development
>website setup on sourceforge, when I checked it, nothing was there yet.
>
>Are there any workable solution? how mature is it?
			.
			.
			.
Everything I know appears at 
<URL:http://starbase.neosoft.com/~claird/comp.lang.python/snmpy.html>.
I'll happily update that as I learn anything new.
-- 

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


From mturatti at yahoo.com  Thu Jun  1 04:42:56 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Thu, 01 Jun 2000 08:42:56 GMT
Subject: PIL problem: can't load PNG images
Message-ID: <8h57mg$jgi$1@nnrp1.deja.com>

Hi,
I just downloaded PIL 1.0 precompiled for Windows. When I try to load a
PNG file, I get this error:

>>> im = Image.open("Kvr-map-border.png")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "D:\Python\Lib\PIL\Image.py", line 894, in open
    raise IOError, "cannot identify image file"
IOError: cannot identify image file

But PNG file format is supported, isn't it? The zlib support lib is
installed and working.

Thanks,
Maurizio


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


From liang at cad.gatech.edu  Wed Jun  7 16:13:18 2000
From: liang at cad.gatech.edu (Shengquan Liang)
Date: 7 Jun 2000 20:13:18 GMT
Subject: SQL Variable Substitution Question
Message-ID: <8hmacu$ia9$1@news-int.gatech.edu>


'm trying to make a query using varible in a
Python program.

say i have yyyy = 1000
and i want to 

 select * from attt
 where bbbb = yyyy



in the documentation of DC Oracle, it says that
i shoul d use 'select * from attt'
		'where bbbb = :yyyy'

to make it work.

BUT, when i tried it this way, the 
iDC oracle reponded by:

  oci.err:(1008,'ORA-1008: not all variables bound\012')


what's the problem with that?



From cartho at mordor.ch  Wed Jun  7 18:42:11 2000
From: cartho at mordor.ch (Cyrille Artho)
Date: Thu, 08 Jun 2000 00:42:11 +0200
Subject: TypeError: unsubscriptable object
Message-ID: <393ECFC3.879BC357@mordor.ch>

Hi,
I am learning Python and liked it quite a lot so far, and I am now
implementing a small neural network simular in it. The graphical part in
Tkinter is now done, but I ran into the following error:
$ python nn.py
Traceback (innermost last):
  File "nn.py", line 81, in ?
    n.playRound()
  File "nn.py", line 73, in playRound
    self.W[i][j] = self.W[i][j]  + n * (
TypeError: unsubscriptable object

Line 74 continues the formula, which should be correct (at least
syntactically).

The initialization of the object where W is a member goes like this:

    def __init__(self):
        self.V = [0, 0, 0]
        self.W = [[0,0,0], [0,0,0], [0,0,0]]

self.W is defined (it works if I print its value just one line before
the error occurs), but I cannot change it. How should I change that
array? I tried it with simpler examples, where it worked (W was also a
member of an object, where I accessed it later - no problems). Why
doesn't it work now? I can send someone the entire script I have written
so far if it helps. Help would be greatly appreciated!
--
Regards,
Cyrille Artho - http://artho.com/ - Tel. +41 - [0]1 - 313 08 92
Better once than never, for never too late.
		-- Shakespeare, "The Taming of the Shrew"


From olczyk at interaccess.com  Sat Jun 17 19:04:06 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 17 Jun 2000 23:04:06 GMT
Subject: Python 3000
References: <394c2b15.634411296@nntp.interaccess.com> <20000617150333.T26436@xs4all.nl>
Message-ID: <395103d2.689896234@nntp.interaccess.com>

On Sat, 17 Jun 2000 15:03:33 +0200, Thomas Wouters <thomas at xs4all.net>
wrote:

>On Sat, Jun 17, 2000 at 07:49:49AM +0000, Thaddeus L. Olczyk wrote:
>> Is there a link where I can look up what's being considered for Python
>> 3000?
>
>Nope, because Py3K is just a name and an idea. A vision, so to speak, and a
>place to dump great ideas [for Python] that aren't possible without breaking
>old code. But, now that Guido & a couple of hardcore insides have moved/are
>moving to PythonLabs, we might just start seeing concrete ideas ;)

Yeah, but is there a place they kept a list of those ideas?


From tony at lsl.co.uk  Wed Jun 21 11:30:19 2000
From: tony at lsl.co.uk (Tony J Ibbs (Tibs))
Date: Wed, 21 Jun 2000 16:30:19 +0100
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <m3k8fk5stl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <000a01bfdb95$9bc210a0$f0c809c0@lslp7o.lsl.co.uk>

Someone (maybe Mikael Olofsson <mikael at isy.liu.se> - the attributions are a
bit tangled) wrote:
> Erm, isn't a pt as in 10 pt font the typographic unit 1/72 inch?

Well, that depends. Searching for "point size" in google led to one
explanation of the problem, at
http://www.vakcer.com/oberon/dtp/fonts/point.htm, which seems quite a good
description. I don't have my TeX book to hand (it's at home), but I'm fairly
sure it has another (similar) explanation therein.

(BTW, I 'love' (for some obscure sense of 'love') the typical confusion part
way down that page, where it says "As you can see, a point used in America
is smaller than that used in Europe", having already said that the points
used in UK and USA are the same...
...the page is presumably written by a USA citizen (either that or a
Euro-sceptic Brit who also confuses USA with America...))

Michael Hudson wrote:
> Well, sort of; the thing is the software doesn't know how big your
> monitor is physically, so can't work out what's it's own dpi is.
> You'd think a task as simple as "draw something this size" would be
> pretty simple after fifty years of computing, but you'd be wrong...
>
> (who's spent longer wrestling with this than he'd have liked to)

Ah, memories of *measuring* the size of early raster displays with a ruler
so we could set the pixel size correctly in our software (more years ago
than I want to remember, but the suppliers of the time thought 512 by 512
displays were good, whilst we (trying to display maps on the damn things)
knew they weren't...) (lest you think we were totally daft, they also didn't
actually SAY anywhere what their pixel size was, and we wanted the maps to
draw at something approaching the right scale!)

Tibs
--
Tony J Ibbs (Tibs)      http://www.tibsnjoan.demon.co.uk/
Give a pedant an inch and they'll take 25.4mm
(once they've established you're talking a post-1959 inch, of course)
My views! Mine! Mine! (Unless Laser-Scan ask nicely to borrow them.)




From petersc at stud.ntnu.no  Sat Jun  3 19:36:18 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 01:36:18 +0200
Subject: ftplib storbinary method interception
References: <39399FBA.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <39399672.B35B7193@stud.ntnu.no>

Bablos wrote:
> 
> Please don't tell me I'm going to have to rewrite chunks of ftplib.py
> - I'm well and truly a beginner with Python...

just define your own version of storbinary
in your module:

def mystorbinary(self, cmd, fp, blocksize):
  '''Store a file in binary mode and print stuff.'''
  self.voidcmd('TYPE I')
  conn = self.transfercmd(cmd)
  counter = 0
  while 1:
    print counter,"Bytes read yet."
    buf = fp.read(blocksize)
    if not buf: break
    conn.send(buf)
  conn.close()
  return self.voidresp()      

just-pass-your-FTP-object-as-the-first-parameter-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From wjdandreta at worldnet.att.net  Wed Jun 21 12:16:34 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 16:16:34 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <wk66r3cgek.fsf@turangalila.harmonixmusic.com>
Message-ID: <CT545.13337$C44.783317@bgtnsc05-news.ops.worldnet.att.net>

Hi Dan,

While it is true that I first learned C++ and more recently I have been
using Delphi, I am not concerned about how the computer lays out data in
memory but I am concerned about overhead.

I have already had one experience with a Python program taking 90 minutes to
process an 18 MB file. By making what I consider minor changes in the
program, it now takes 6 minutes.

I've come to the conclusion that Python is not an inherently slow language
but it is very easy to add overhead that makes Python programs run extremely
slowly. It is much more difficult to do that in the other languages that I
use. So for the present, I am being overly cautious about adding overhead.
As I become more familiar with Python, I'm sure that will dissippate.

Also, for the simple comma delimited ASCII files that I have been
manipulated, using 10 to 15 line Python scripts, creating objects seems like
over kill to me.

Bill

>I think it's exactly the opposite.  If you want to look things up by a
>name rather than a number, look them up by a name, not by a fake name
>that actually maps to a number.  Trying to shoehorn your 'Date, Foo,
>Bar' fields into an array is the contrivance.
>
>Perhaps you come from a background in languages like C?  Often it
>takes a while to get used to writing code the way you think, without
>worrying too much about overhead and exactly how the computer is
>laying out your data in memory.
>
>--
>                 Dan Schmidt | http://www.dfan.org
>Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html




From johann at physics.berkeley.edu  Mon Jun  5 15:49:06 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 05 Jun 2000 12:49:06 -0700
Subject: A Simple question about a slice of a string.
References: <8hgefr$sok$1@news2.acs.oakland.edu>
Message-ID: <mt8zwj7vt9.fsf@astron.berkeley.edu>

Jay R Wren writes:

> I'm attempting to accomplish grabbing the last 6 characters of sys.argv[0]
> so that I can compare it to another string.
[snip]
> myprogname = sys.argv[0][ len(sys.argv[0])-10:len(sys.argv[0])

> it works like this, but I think there is a better, more proper way of doing
> it.  can someone let me know.  Thanks.

myprogname = sys.argv[0][-10:]

is equivalent.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From gmcm at hypernet.com  Thu Jun 15 08:30:19 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jun 2000 12:30:19 GMT
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com> <01bfd2ed$42b67880$29cf8ad1@daves-dell> <39463BA7.43A3CD65@gssec.bt.co.uk> <39474D1D.F959969C@prescod.net>
Message-ID: <8F545A89Agmcmhypernetcom@199.171.54.154>

Paul Prescod wrote:

>Try double clicking a PDF document with a space in its path. Adobe
>Acrobat 4 will not find it. It's annoying as hell because Internet
>Explorer puts files from the Web in the "Temporary Internet Files"
>directory.

That's the installer's fault, not Acrobat 4's. Just (double)quote the %1 in 
the open action for the file type.

I get a big kick out of MS's habitual use of paths with embedded spaces. 
Sort of a barometer of hubris.

- Gordon




From just at letterror.com  Thu Jun 29 10:42:27 2000
From: just at letterror.com (Just van Rossum)
Date: Thu, 29 Jun 2000 15:42:27 +0100
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194> <ffcmlsc6p4lqohdd6tequdu918l0temns4@4ax.com>
Message-ID: <395B6027.B47094CA@letterror.com>

Toby Dickenson wrote:
> 
> gmcm at hypernet.com (Gordon McMillan) wrote:
> 
> >I think the "relative import" feature should be shot in the head (2 posts
> >today), but unfortunately most packages rely heavily on it.
> 
> A relative import is just as useful as a relative directory: it allows
> intra-module imports, without each module needing to know where it is
> packaged.
> 
> The real problem is that relative imports and absolute imports are
> spelled the same way.

That's an interesting way of looking at it.

<silly-proposal>
If relative imports as we know them are shot in the head, an
alternative spelling for relative imports could be invented.
Eg. with a leading period:

import .some.submodule
</silly-proposal>

Just


From petersc at stud.ntnu.no  Wed Jun 14 18:05:03 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Thu, 15 Jun 2000 00:05:03 +0200
Subject: boolean true and false values.
References: <3947f922.425016156@nntp.interaccess.com>
Message-ID: <3948018F.B82F7F7F@stud.ntnu.no>

"Thaddeus L. Olczyk" wrote:
> 
> at the moment I am using
> false=(1==0)
> true=(1==1)
> but would like something more elegant, or at least less sloppy.

well, of course:
false = 0
true = 1

(or
 false = None
 true = [[]])

Not that the first one is equivalent to your definition.
>>> 1==0
0
>>> 1==1
1

what-d-ya-mean-by-sloppy-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From aahz at netcom.com  Fri Jun  9 11:22:28 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jun 2000 15:22:28 GMT
Subject: Probably a dumb question
References: <394106F9.A58F87CE@micron.net>
Message-ID: <8hr23k$bvk$1@nntp9.atl.mindspring.net>

In article <394106F9.A58F87CE at micron.net>,
Peter Immarco  <android at micron.net> wrote:
>
>Hi,  still waiting for my two books on Python from Amazon.  I know
>Python is a full programming language so it can be used on web servers
>for server side processing, but what about browser side processing?  Can
>I also use it to create an 'applet' that runs in a compatible browser?
>I'm guessing not likely because I don't think IE4+ or Netscape
>Communicator support Python but I'm hoping I'm wrong.

There's a rumor that a future version of Mozilla will have Python
support (supposedly they're building in something similar to ASP with
its switchable scripting engines).  Other than that, no.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From smst at bigfoot.com  Thu Jun  8 05:30:16 2000
From: smst at bigfoot.com (Steve Tregidgo)
Date: Thu, 08 Jun 2000 10:30:16 +0100
Subject: TypeError: unsubscriptable object
References: <393ECFC3.879BC357@mordor.ch> <8hmp6i$ets$6@newshost.accu.uu.nl> <393F488B.37F13682@mordor.ch>
Message-ID: <393F67A8.1791A54D@bigfoot.com>

Cyrille,

Cyrille Artho wrote:
> 
[code snipped]

# Problem line:
>         P = 1 / (1 + math.exp(-sum))

# Exception raised by attempts to acces P[i]
>                 self.W[i][j] = self.W[i][j]  + n * (
>                     r(S[i] - P[i]) - alpha * (1 - r)
>                     * (S[i] - (1 - P[i]))) * I[j]

There's your problem -- on line 43 you assign a number to
P, which you later try to access as if it were a sequence
object.

Python reports the error (on physical line 74) as being on
line 73, which is how it sees the whole statement.

HTH,
Steve
-- 
Steve Tregidgo
Software Developer
http://www.businesscollaborator.com



From fiona at sitegnome.com  Sat Jun 10 04:54:16 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 10 Jun 2000 08:54:16 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 10th, 2000
Message-ID: <20000610085416.10017.qmail@synop.com>

Hi Guys,

Only a couple of entries into http://python.faqts.com tonight.

F.


## New Entries #################################################


-------------------------------------------------------------
Can I transform pyc files to py files?
http://www.faqts.com/knowledge-base/view.phtml/aid/3639
-------------------------------------------------------------
Fiona Czuczman
Michael Hudson

There's decompyle at 

http://www.csr.uvic.ca/~aycock/python/content.html#de-about


-------------------------------------------------------------
If given a slice s[a:b] can I assume index "b" means upto but not including the index "b"?
Why is s[a:a] always '' (empty)?
http://www.faqts.com/knowledge-base/view.phtml/aid/3640
-------------------------------------------------------------
Fiona Czuczman
David Bolen,Sean Blakey

Precisely - that's required or else the invariant wouldn't be true. It's 
also generally a very useful condition since it fits the base 0 
addressing nicely, and helps simplify various computations of pieces of 
strings and lists that otherwise would be adjusting computed offsets by 
1 a lot.

For one example, if a slice was inclusive then s[0:n] (or s[:n]) where
n was len(s) would have an upper bound one too large.

> I guess another way to ask my question is:
> Why is s[a:a] always '' (empty) ?

Because that particular slice doesn't cover a full element.  It's sort 
of like a piece of the string just before the 'a'th element.  For 
mutable objects (like lists) you can actually assign to a slice like 
this to insert elements into the list at the point just before element
'a'.

Probably the easiest way to make sense of slice semantics is to think of
the slice indexes as counting the borders between values.  For example,
s[0:] is the slice from before the first character on. s[2:2] is the
"space" number two, which is between the elements of index 1 and two.

You have probably seen magic acts where a person is locked into a
segmented box.  The magician then places blades between the various
segments.

Think of a python sequence as being like this box.  Slice indexes do not
count the sub-boxes, but the spaces between the boxes where blades can
be inserted.  You can take your slice from above the head to just above
the ankles (body[0:-1]), or from the neck to below the feet (body[1:]),
or from the neck to the ankles (body[1:-1]).  If you insert two blades
at the neck (body[1:1]), you end up with nothing between them.

Slice indexes and direct indexes are different.  Perhaps it would help
to visualize your string like this:

0   1   2   3   4   5   6
| a | b | c | d | e | f |
  0   1   2   3   4   5

With this in mind, let's try to explain your examples:
s[2:2] == ''
The slice from separator 2 to separator two is a zero-length string.
This is the slice starting at the separator between 'b' and 'c', and
ending at the same place.

s[:2] + s[2:] == s      (a very nice feature of python, BTW)
Everything up to separator 2, plus everything after separator two.

In summary, and to actually get around to your question, s[:b] means up
to separator b, which is just before index b.  Similarly, s[b:] means
everything after separator b, which is just before index b.


## Edited Entries ##############################################


-------------------------------------------------------------
I am looking for a treeview widget.  Is there one available?
http://www.faqts.com/knowledge-base/view.phtml/aid/3595
-------------------------------------------------------------
Fiona Czuczman, Richard Chamberlain
Richard Chamberlain

Idle uses a tree widget in 0.5 that you could look at.

You could also look at:

http://www.magicnet.net/~gcash/tree.html
or
http://www2.zyvex.com/OpenChem/index.htm


-------------------------------------------------------------
How can I trap a keyboard interrupt using Python?
Ctrl-Break does not raise KeyboardInterrupt, how can it be caught?
http://www.faqts.com/knowledge-base/view.phtml/aid/1424
-------------------------------------------------------------
Nathan Wallace, James Spears
Oleg Orlov

import sys

def Worker(argv):
  while 1:
    pass

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






From shapr at uab.edu  Thu Jun 15 20:42:41 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 16 Jun 2000 00:42:41 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com> <8i6u39$re3$1@nnrp2.deja.com> <8i8le9$b6u$1@nnrp1.deja.com> <8i95iq$o3a$1@nnrp1.deja.com> <8ibju8$i13$1@nnrp1.deja.com>
Message-ID: <39497854.A2CDF245@uab.edu>

millja at my-deja.com wrote:
To: python-list at python.org
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta3
Precedence: bulk
List-Id: General discussion list for the Python programming language <python-list.python.org>

> 
> Here is an example of what I mean.
> 
> <SCRIPT RUNAT = SERVER LANGUAGE=Python>
> def Test4(number):
>         return number + 3, number + 4
> </SCRIPT>
> Our numbers of the day are
> <%
> =Test4(10)
> %>.
> 
> I get a Type Mismatch, An unhandled data type was encountered error.
Path: news!uunet!ffx.uu.net!newsfeed.easynews.net!easynews.net!newsfeed.germany.net!news.vas-net.net!newsfeed.icl.net!colt.net!newspeer.highwayone.net!newsfeed1.telenordia.se!algonet!newsfeed1.funet.fi!newsfeeds.funet.fi!nntp.inet.fi!central.inet.fi!inet.fi!read2.inet.fi.POSTED!not-for-mail
Xref: news comp.lang.python:99850

I can only offer guesses, I've never done ASP w/PythonScript
Have you tried turning your number into a string before you return it?

<SCRIPT RUNAT = SERVER LANGUAGE=Python>
def Test4(number):
        return str(number + 3, number + 4)
</SCRIPT>

Do you know exactly where the error is happening? Is it only in the
returned value?

Try some experimental incremental scripts like this:

<SCRIPT RUNAT = SERVER LANGUAGE=Python>
def Test5(number):
        return (1,'I can return numbers<BR>')

</SCRIPT>

<SCRIPT RUNAT = SERVER LANGUAGE=Python>
def Test6():
        return 'I can return strings.<BR>'
</SCRIPT>

<SCRIPT RUNAT = SERVER LANGUAGE=Python>
def Test7(handyString):
        return 'I can pass and return %s.<BR>' % handyString
</SCRIPT>

<%
=Test7('strings')
%>

I'm sure you can come up with more things to test... When you test each
thing your code does separately, you can usually find the problem.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
 squirrels furnished by Paul E Black, the world is safe


From anders.eriksson at morateknikutveckling.se  Thu Jun  8 08:37:50 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 08 Jun 2000 14:37:50 +0200
Subject: GUI editor
References: <393f4568$0$25329@wodc7nh0.news.uu.net> <vskujs867ghjso3aqi2r38n5mei0bg4lhq@4ax.com> <393F8D02.9A87A6EF@uab.edu>
Message-ID: <9k4vjs0klckrvba86usus8ckpgmt4b6ca5@4ax.com>

On Thu, 08 Jun 2000 12:10:30 GMT, Shae Erisson <shapr at uab.edu> wrote:

>Anders M Eriksson wrote:
>
>> 
>> There are alot of 'pre-alpha' versions of GUI's and editors but non
>> that works.... (please correct me on this!)
>
>I'm confused. Are you talking about an editor for building GUIs, or are
>you talking about IDLE/PythonWin/etc.?
>

In my own very unclear way I was talking about building GUI's. Eighter
using a GUI editor or an IDE the has the GUI editor build in.


>> If you find any IDE/GUI editor that you like Please let me know! I
>> also searching...
>
>If you just want a nice way to edit Python source, check out IDLE. If
>you look in your python install directory, it should be under
>Tools/idle/

Yes I know. I'm using IDLE some of the time, then I switch to
PythonWin, and then I use UltraEdit. None of these products are any
good at GUI's though. IDLE just quits when I end my Tkinter
application, PythonWin hangs when I run Tkinter within it, etc ...

Creating GUI's by hand instead of by eye isn't my favourite occupation

// Anders



From eric at estinc.com  Wed Jun 14 13:48:50 2000
From: eric at estinc.com (Eric Lee Green)
Date: Wed, 14 Jun 2000 17:48:50 GMT
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com> <8i86sf$e7c$0@216.39.141.36> <1VO15.607$i06.193222@ptah.visi.com>
Message-ID: <3947C5E8.4CE1032E@estinc.com>

Grant Edwards wrote:
> I'm going to distribute a utility written in Python/Tkinter as
> a configuration/administration utility that goes with a Linux
> device driver.  I was just wondering how many customers are
> likely to have Python/Tkinter available.  I doubt that anybody
> is going to install Python just to run this program.

Red Hat and Caldera certainly will, since their system administration tools
are written in Python. 

But I would strongly suggest using TCL/Tk instead if it is a simple little
configuration program, because I don't know of any Linux distribution that
does not come with TCL/Tk -- amongst other things, 'make xconfig' needs TCL/Tk
in order to work. TCL is brain-dead, but for simple stuff it's usable. 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From daniellundqvist at my-deja.com  Wed Jun 28 04:18:51 2000
From: daniellundqvist at my-deja.com (daniellundqvist at my-deja.com)
Date: Wed, 28 Jun 2000 08:18:51 GMT
Subject: Extending python
Message-ID: <8jccdb$qoh$1@nnrp2.deja.com>

Hi,

Is there a way to create python classes in C and make them behave like
ordinary classes,can be inherited and so on?

Kind regards,
Daniel Lundqvist


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


From bvdpoel at uniserve.com  Tue Jun 20 18:45:50 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Tue, 20 Jun 2000 15:45:50 -0700
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net> <j8y35.1071$ul.67488@news1.rdc2.tx.home.com> <394EB843.EEB3B68F@uniserve.com> <m3g0q856w6.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <394FF41E.E4FE1BCD@uniserve.com>

Michael Hudson wrote:
> 
> Bob van der Poel <bvdpoel at uniserve.com> writes:
> 
> > Yes, but I'm using __name to keep the variable private/static to a
> > particular instance of the Goption().
> 
> This is a bit of a wild guess but are you suffering from
> instance/class confusion?
> 
> Cheers,
> M.

More than likely.... Perhaps I should have made my example and
reason more clear.... I have a simple (tkinter) program which does some
datafile searches using either grep or agrep. I thought it'd be nice to
be able to change the options fo grep/agrep so I wrote a little function
to do this. Problem with this is that I ended up with 2 nearly identical
functions (one for grep, another for agrep); and I found that I needed
some global variables to store the current and default settings.

so, being a newbie to the wonderful world of classes, objects and other
things I don't understand I figured that I'd do this as a class. To
create the class I do:
        grep_opts = Goption('-ih', "Grep")

which creates a class instance. I can (hopefully) now do the following:

        grep_opts.get() to get the current setting
        grep_opts.change() to call the gui to change the current
setting.

So, here's the simple thingie I came up with. Note, that this uses the
dialog class from tkinter to create the dialog window (and this is where
my problems come in.

class Goption:
        __searchName = ""
        __optDefault = ""
        __optSetting = ""


        def __init__(self, title, setting ):
                self.__searchName=title
                self.__optDefault=setting
                self.__optSetting=setting

        def get(self):
                return self.__optSetting

        def change(self):
                class MyDialog(tkSimpleDialog.Dialog):

                        def body(self, master):
                                Label(master, text="%s Options:" %
Goption._Goption__searchName).grid(row=0)
                                self.e1 = Entry(master)
                                self.e1.insert(0,
Goption._Goption__optSetting)
                                self.e2 = Button(master, text='Default',
                                        command=self.default)
                                self.e1.grid(row=0, column=1)
                                self.e2.grid(row=0, column=2)
                                return self.e1 # initial focus

                        def default(self):
                                self.e1.delete(0,END)
                                self.e1.insert(0,
Goption._Goption__optDefault)

                        def apply(self):
                                self.result = self.e1.get()

                a = MyDialog(root, title="FindSong %s Options" %
self.__searchName)
                if a.result:
                        self.__optSetting = a.result


The Goption._Goption_??? calls are from another members suggestions. But
no joy...Inserting some print lines shows that
Goption._Goption_optSetting are
undefinded. So:

Am I on the right track...or is my concept of how to do this just plain
dumb?

Am I missing something really obvious on how to get the values I need?


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



From mrfx at etang.com  Wed Jun  7 07:52:55 2000
From: mrfx at etang.com (mrfx)
Date: Wed, 7 Jun 2000 19:52:55 +0800
Subject: is there any gui builder like glade for wxwindows
Message-ID: <8hmn32$kk0$1@news.cz.js.cn>

Glade is a good gui builder for GTK,
is there any gui builder like glade for wxwindows?




From igorr at ifi.uio.no  Mon Jun 19 11:11:36 2000
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 19 Jun 2000 17:11:36 +0200
Subject: split() with a separator set
Message-ID: <xjvr99tyac7.fsf@gjallarhorn.ifi.uio.no>

Hi,

I've just started learning Python, so, please bear with me. 

string.split() function takes (amongst other things) a separator
argument, and the given string is split according to this separator.
But what if I have several one-character separators? Obviously:

# input looks like this: 'foo:bar|baz/zot'
# I want [ 'foo', 'bar', 'baz', 'zot' ]
string.split( src, ":|/" )

doesn't work.

I've written my own function to solve the problem:


import string

def splitsepset( src, sepset = string.whitespace ):
    reslist = []
    frompt = topt = 0
    inword = 0
    i = 0

    while i < len( src ):
        if src[i] in sepset:
            if inword:
                reslist.append( src[ frompt:topt ] )
                inword = 0
            frompt = topt = i+1
        else :
            inword = 1
            topt = topt + 1
        
        i = i + 1
    # end while

    return reslist
# end mysplit


but it seems just too awkward to use that.

Is there a better (that is first cleaner, then faster) way to split a
string into fields according to a one-character separator _set_ (for
instance, as string.split( s ) does, except that I want to specify
something else than string.whitespace)? And as a follow-up -- how do I
split a string according to a set of _multicharacter_ separators?


Thanks in advance,



ivr, willing to read the proper section of the FAQ.
-- 
Simula-konsulent? Simsulent? "I eat Java for breakfast!".
                                           -- thorkild


From bob at slim.instantappointment.com  Sat Jun 17 13:06:39 2000
From: bob at slim.instantappointment.com (Bob Erb)
Date: 17 Jun 2000 17:06:39 GMT
Subject: Help: 'AttributeError: wm_title' Error starting IDLE
Message-ID: <slrn8knc4g.3u3.bob@slim.instantappointment.com>

I can't run IDLE. It bombs with 'AttributeError: wm_title'.
Complete traceback is below.

- Linux 2.2.10
- Python 1.5.2
- IDLE 0.5

Any ideas? All suggestions appreciated. Thanks.

Traceback (innermost last):
  File "/usr/lib/python1.5/site-packages/idle/idle.py", line 3, in ?
    PyShell.main()
  File "/usr/lib/python1.5/site-packages/idle/PyShell.py", line 720, in main
    shell = PyShell(flist)
  File "/usr/lib/python1.5/site-packages/idle/PyShell.py", line 288, \
in __init__
    OutputWindow.__init__(self, flist, None, None)
  File "/usr/lib/python1.5/site-packages/idle/OutputWindow.py", line 15, \
in __init__
    apply(EditorWindow.__init__, (self,) + args)
  File "/usr/lib/python1.5/site-packages/idle/EditorWindow.py", line 169, \
in __init__
    self.saved_change_hook()
  File "/usr/lib/python1.5/site-packages/idle/EditorWindow.py", line 431, \
in saved_change_hook
    self.top.wm_title(title)
AttributeError: wm_title



From bvdpoel at uniserve.com  Sat Jun 17 14:34:53 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 17 Jun 2000 11:34:53 -0700
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com> <slrn8kn881.sdl.nobody@isis.visi.com> <394BA666.E5F3B19B@uniserve.com> <15O25.1379$i06.308670@ptah.visi.com>
Message-ID: <394BC4CD.6432CBBF@uniserve.com>

Grant Edwards wrote:
> 
> On Sat, 17 Jun 2000 09:25:10 -0700, Bob van der Poel <bvdpoel at uniserve.com> wrote:
> 
> >> Yes, I know.  The call to BitmapImage() works fine, the error occurs when I
> >> try to use the bitmap in a button like:
> >>
> >>    Button(root, bitmap='queen')
> >>
> >> This is where the error occurs -- it doesn't matter if I use the name I
> >> specified in the call to BitmapImage or let Tk generate one and retrieve it
> >> with str().  What I can't figure out is how to use bobpic as the bitmap for
> >> a Button.
> >>
> >
> >More from my snippet:
> >
> >       l = Label(w, image=bobpic, borderwidth=4, relief='sunken')
> 
> Yes, using it as the widget's image works, but I'm trying to
> use it as the bitmap for the widget.

Oh, ummm did you try:

	 Button(root, bitmap=queen)

without the quotes around the name? Remember, 'queen' is a string, not
the name of the variable where the data is stored. Also, don't let this
bite you: if this is a function then the variable has to declared global
otherwise when the function exits the variable (and the display)
disappear.

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


From jtschirr at engineering.uiowa.edu  Thu Jun 15 22:50:44 2000
From: jtschirr at engineering.uiowa.edu (Juerg Tschirren)
Date: Fri, 16 Jun 2000 02:50:44 GMT
Subject: identifying operating system
Message-ID: <39499924.3BC84686@engineering.uiowa.edu>

Is there a way how I can identify the operating system my Python script
is running on (e.g., Linux, Windows, Mac, ...)?

Juerg


From neilh at scintilla.org  Mon Jun 12 02:23:30 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 12 Jun 2000 06:23:30 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <Cl%05.2665$cQ.22917@news-server.bigpond.net.au>

> In many Perl vs. "other scripting language" discussions, I have often
> seen Perlers state that those who dislike Perl's syntax and structure
> are often non-Unix types, and thus are not used to common Unix
> concepts. And, since Perl was born as a Unix admin tool, us "Windows
> types" will likely be lost.

   I think that was part of my problem with Perl. While I can use Unix, its
at a low level of competence - I count myself lucky to survive a session
with vi without pulling the power cord to stop the torture. Coming from an
RSX/11M+, Mac, OS/2, Windows background I found the syntax choices in Perl,
which probably appear obvious to native Unixers, bizarre and difficult to
reason with. Python does not try to use much 'clever syntax' at all making
it much easier for me.

   This leads to not liking the 'augmented assignment' addition. The only
examples of augmented assignment that are used often enough to have their
own special syntax are addition and subtraction. Even for these, I'd prefer
to see syntax that looks like function calls similar to Modula 2.

> So my question is this: Is Python a more "Windows-friendly" language?
> Is it's base structure less related to any one specific operating
> system? Is learning Python an exercise in learning the language, and
> not a Unix tutorial? Can I be successful even if I do not use Unix ;)

   Its more OS-neutral than Windows friendly and there are plenty of us
Windows and Macintosh users here to prove it.

   Neil





From ajwilson at ezy.net.au  Fri Jun  2 21:45:12 2000
From: ajwilson at ezy.net.au (A J Wilson)
Date: Sat, 3 Jun 2000 11:45:12 +1000
Subject: Combining two graphic images
Message-ID: <959996628.860207@centauri.ezy.net.au>

Hi

I would like to combine two BMP images so that one overlays the other.  Does
anyone know how this could be done using Python.

Thanks
Austin




From millja at wwc.edu  Mon Jun 26 12:41:19 2000
From: millja at wwc.edu (millja at wwc.edu)
Date: Mon, 26 Jun 2000 16:41:19 GMT
Subject: Python COM does not seem to like me.
References: <8j0ddv$fqb$1@nnrp1.deja.com> <39541965.45DCE978@san.rr.com>
Message-ID: <8j813b$if7$1@nnrp1.deja.com>

Thanks!  I still cant get the import issue taken care of though.  I
have changed the PYTHONPATH key in the registry (WinNT server 4.0) to
include the dir in which the modual is contained in
c:\InetPub\BRC\case13.py.  Then I include it using from case13 import
*.  Is there something Im doing incorrectly?  Any ideas?  Thanks again.
jason



In article <39541965.45DCE978 at san.rr.com>,
  Courageous <jkraska1 at san.rr.com> wrote:
>
> > First off, is there something I dont know about the import
statement?
>
> The python module must either be in your PATH or your PYTHONPATH
> for python to be able to import the module, IIRC.
>
> > TypeError: cant multiply sequence with non-int
> > The statment is:
> > CM1 = roll[1] * roll[2]
> > I am pretty much positive roll[1] and roll[2] are ints.
>
> You'd know for sure if on the line before you wrote:
>
> print `roll[1]` + " " + `roll[2]`
>
> I suspect you're going to be surprised by the results.
>
> C/
>


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


From dcalvelo at pharion.univ-lille2.fr  Thu Jun  8 11:07:32 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 8 Jun 2000 15:07:32 GMT
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl> <393D1155.D067A821@decrc.abb.de> <m33dmqygdj.fsf@atrus.jesus.cam.ac.uk> <slrn8jt5ck.779.jepler.lnk@potty.housenet>
Message-ID: <8hocrk$na2$1@netserv.univ-lille1.fr>

jepler epler <jepler.lnk at lnk.ispi.net> wrote:

: I'm guessing that 'import gnome.ui' sets the locale, including the
: numeric format:
: 	$ env LC_NUMERIC=en_US python -c 'import gnome.ui; print float("1.1")'
: 	1.1
: 	$ env LC_NUMERIC=de_DE python -c 'import gnome.ui; print float("1.1")'
: 	Traceback (innermost last):
: 	  File "<string>", line 1, in ?
: 	Value Error: invalid literal for float(): 1.1

Same here. Debian woody Linux. gnome-libs-1.2.1, python-gtk 0.6.6 and python-gnome debian package 1.0.53.

: Jeff

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From sbender at harmony-ds.com  Tue Jun  6 12:40:08 2000
From: sbender at harmony-ds.com (Scott Bender)
Date: Tue, 6 Jun 2000 12:40:08 -0400
Subject: Capture stderr/stdout from python embedded python?
References: <c_Q_4.1728$JC.3933905@news.abs.net>
Message-ID: <rP9%4.1785$JC.4049116@news.abs.net>


OK, I'm a loser. If I would have read the archives first, I would have seen
that I can reassign sys.stdout and sys.stderr.

thanks anyway,
- Scott

"Scott Bender" <sbender at harmony-ds.com> wrote in message
news:c_Q_4.1728$JC.3933905 at news.abs.net...
>
> Hi,
>
> I need to be able to capture/intercept any output by python in a program
> that embeds python. There is no stdout/stderr visiable to the user, so I
> need to present any messages in a window. Any easy way to do this? I have
no
> problem with modifying the python source. Can I just modify mywrite() in
> sysmodule.c? Or are there many other places I would need to do this?
>
> It seems to me that this would be a common problem for embeded python apps
> that run in a GUI environment. There should be a standard way to hook
this??
>
> thanks,
> - Scott
>
>
>




From just at letterror.com  Tue Jun 20 03:22:22 2000
From: just at letterror.com (Just van Rossum)
Date: Tue, 20 Jun 2000 08:22:22 +0100
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <NGz35.3149$qz4.24314@news-server.bigpond.net.au>
References: <l03102800b573e58c0e58@[193.78.237.139]>
Message-ID: <l03102800b574c9a15f7a@[193.78.237.171]>

At 1:21 AM +0000 20-06-2000, Neil Hodgson wrote:
>> ... (In my MacPython IDE the user can choose
>> between the spaces and pixels as a visual size unit when setting the
>> indentation level...)
>
>   What has been the feedback on defining space in pixel terms?

Erm, I can't remember any...

>   The thing that stopped me doing it for Scintilla was that it would not
>work very well with the text magnification command. Maybe it should be
>defined in terms of x-width which doesn't change as much as the width of a
>space which varies a lot between fonts.

You mean the width of the character 'x'? Eek! (Besides, I don't see how the
x-width changes any less that the space-width across fonts...)

Why not use points, as in pointsize? A 10 pt font will look like 20 pt when
viewed at 200% magnification, so a 10 pt indent will look like 20 pt at
200%. Seems pretty straightforward to me.

(Due to historic narrowmindedness, a point is the same as a pixel under
MacOS, no matter what physical resolution the display has; I believe
Windows has (had?) the same problem, except that a pixel is 72/96th
points...)

Just





From jae at ilk.de  Fri Jun  2 21:00:29 2000
From: jae at ilk.de (Juergen A. Erhard)
Date: Sat, 03 Jun 2000 03:00:29 +0200
Subject: String.join revisited (URGENT for 1.6)
In-Reply-To: <8h320n$6fo$1@newshost.accu.uu.nl> (m.faassen@vet.uu.nl)
References: <28052000.5@sanctum.jae.ddns.org> <8h15qr$rje$1@news.udel.edu> <mtzop76b2z.fsf@astron.berkeley.edu> <39346FAC.F2797EA5@cosc.canterbury.ac.nz> <8h320n$6fo$1@newshost.accu.uu.nl>
Message-ID: <03062000.1@sanctum.jae.ddns.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>>>>> "Martijn" == Martijn Faassen <m.faassen at vet.uu.nl> writes:

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

    >> Indeed - there seem to be just as many subjective arguments
    >> in favour of any of the alternatives, so the safest thing
    >> to do is leave the darn thing alone!

    >> I know we're promised that string.join will still be there,
    >> but if ''.join exists, people will start using it, and it
    >> won't be long before it's unremovable. Too Many Ways To Do It...

    Martijn> Agreed. So keeping string.join() would seem to be the
    Martijn> best alternative, definitely (or possibly making this a
    Martijn> builtin, so you don't have to import to use it, making it
    Martijn> import-equivalent with the new .split()).

    Martijn> a = "foo bar baz"  
    Martijn> b = a.split(" ")
    Martijn> assert a == sjoin(b, " ")

    Martijn> versus:

    Martijn> a = "foo bar baz"
    Martijn> b = a.split(" ")
    Martijn> assert a == " ".join(b)

    Martijn> Perhaps we should've left 'split' alone too. :)

I'm in favor of giving the mythical Sequence type (which I *hope* will
be created when the types/classes split will be healed (yes, healed))
a `join' method.

It would seem (at least to me) to be the nicest (and cleanest) design
to be able to do:

   "some string".split().join()

Bye, J

PS: I don't like `sjoin', by the way... looks much too ugly for such a
beautiful language as Python.

- -- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
     MARS: http://members.tripod.com/Juergen_Erhard/mars_index.html
   GNU Privacy Guard (http://www.d.shuttle.de/isil/crypt/gnupg.html)
             pros do it for money -- amateurs out of love.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: Use Mailcrypt and GnuPG <http://www.gnupg.org/>

iEYEARECAAYFAjk4WKsACgkQN0B+CS56qs2xCwCggqYM/LxYramyyOX9YLDoAWVZ
seAAoIXl6eIVdxM8/+h5YTS9c0ug3Yo8
=QcMC
-----END PGP SIGNATURE-----



From pecora at anvil.nrl.navy.mil  Sun Jun 11 09:26:54 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Sun, 11 Jun 2000 13:26:54 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <m33dml6p34.fsf@brant.geog.ubc.ca>
Message-ID: <110620000926542025%pecora@anvil.nrl.navy.mil>

In article <m33dml6p34.fsf at brant.geog.ubc.ca>, Phil Austin
<phil at geog.ubc.ca> wrote:

> Peter Schneider-Kamp <petersc at stud.ntnu.no> writes:
> 
> > I have to create a tuple (of two integer values) in a
> > C extension. At the moment I use the following (rather
> > ugly in my eyes) construction:
> > 
> 
> [snip]
> 
> > Is there "A Better Way(TM)"?
> > 
> 
> If you're able to use the CXX extension, the sequence is:
> 
> Tuple t(2);
> t[0]=Int(a);
> t[1]=Int(b);
> 
> If the conversion fails, Int will throw an exception that
> can be caught on the python side.

Phil,

What about reference counting for these CXX examples?


From tismer at tismer.com  Sun Jun  4 17:21:04 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 04 Jun 2000 23:21:04 +0200
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com>
Message-ID: <393AC840.9D50C196@tismer.com>


Courageous wrote:
> 
> 1. For slice operations, ilow/ihigh indices which undeflow
> or overflow the list in question are reset to the edges
> of the list. Why was it that PyExc_IndexError wasn't
> thrown instead?

Slice operations are defined to never fail.

> 2. In slice expressions of the form s[i:j], what was the
> design decision that lead to s[i:i] always returning an
> empty list and s[i:i+1] selecting one item. At first
> glance, it would appear that s[i:i] is the obvious and
> intuitive correct expression to return just one item on
> a slice, where s[i:i+1] ought to select two items and so
> forth. Was there some particular situation which arose
> which required this slightly counterintuitive idiom?

All Python indexing is like [a, b), so you have a right-open
interval. It may look a little counter-intuitive at first
glance, but finally it is not at all.
The number of elements covered in a slice operation is
always the difference between the bounds for normal cases.
Normal cases means positive indexes which are inside the
list's size, of course.
Becoming used to it takes a bit of re-thinking, but finally
I believe it greatly reduces the number of off-by-one errors.

advocating-ly - y'rs - chris

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



From ame at swipnet.se  Thu Jun 22 17:24:44 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Thu, 22 Jun 2000 23:24:44 +0200
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net> <3gh3lsgg9dtkfgrsq1ml2i9jcfkknrb6ei@4ax.com> <Qfq45.15292$C44.879449@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <tm05lskrbk1if0h39m3o7fee7j64ushhkg@4ax.com>

On Thu, 22 Jun 2000 15:27:44 GMT, "William Dandreta"
<wjdandreta at worldnet.att.net> wrote:

>Right now we are using a Dbase file with proprietary indexes. The .lib that
>we use for access is written in assembly language so it is restricted to
>DOS.
>

shelve will not do.

>Eventually we plan to migrate to Linux. So I need something platform
>independent.
>

Why don't you set up a Linux server and a TCP/IP net with DOS clients.
Then you could have the database on the server and you would have much
more to choose from!

>I checked the website for MetaKit. There are no DOS binaries and I searched
>the site for DOS but nothing useful showed up so I lost interest.
>

No I also checked. There is no DOS version, there is the source code
so You could maybe create one...

>What's wrong with xDB?
>
It's probably nothing wrong with xDB, but I think it's rather a big
task to port it to DOS and create a Python Wrapper. So maybe there are
other solutions that are better?

As I said earlier maybe a Linux server and DOS client is the way to
go?

// Anders



From nascheme at enme.ucalgary.ca  Thu Jun  8 13:44:38 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Thu, 08 Jun 2000 17:44:38 GMT
Subject: wishlist Python3K: subclass ints
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <393EFF38.C023F10F@san.rr.com> <393FD0C7.D70479A2@visionart.com>
Message-ID: <slrn8jvmu6.1an.nascheme@cranky.arctrix.com>

Pete Shinners <pete at visionart.com> wrote:
[subclassing types]
>i think this is really important also. i'd like to see all
>python objects given the same 'value'
>
>a small sampling of reasons to make all types (yes, even ints)
>real class types that can be subclassed
>
> - "Debug Int" class. want to make an int that prints to
>   a log every time it gets assigned?

Your mental model of Python is flawed I believe.  It is
impossible to "assign" to an int.  Variables are bindings not
storage locations.  ints are immutable.

> - Just the fact that i could subclass any "C" object would
>   make it worth the while.

This should be fixed when the type/class distinction is fixed.

    Neil

-- 
"If you're a great programmer, you make all the routines depend on each
other, so little mistakes can really hurt you." -- Bill Gates, ca. 1985.


From moshez at math.huji.ac.il  Sat Jun 10 05:32:19 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 12:32:19 +0300 (IDT)
Subject: Damnation! (was: Re: Python3k extended grammar)
In-Reply-To: <87n1kuowrf.fsf@globalreach.net>
Message-ID: <Pine.GSO.4.10.10006101231580.3669-100000@sundial>

On 10 Jun 2000, Phil Fraering wrote:

> Guido van Rossum <guido at python.org> writes:
> 
> > Python is not a democracy.  Voting doesn't help.  Crying may... :-)
> 
> What about large numbers of small unmarked bills with serial
> numbers not in sequence?

Sure, it's called the Python Consortium <wink>
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From paul at prescod.net  Thu Jun 22 14:58:52 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 22 Jun 2000 13:58:52 -0500
Subject: XML-code examples
References: <394f7bc2.365514882@news.online.no>
Message-ID: <395261EC.4976AED1@prescod.net>

Thomas Weholt wrote:
> 
> Hi,
> 
> Is there more simple XML code out there? I mean the HowTos took about
> ten minutes, and it didn`t cover much.
> 
> If somebody has some tips or code on how to traverse a DOM-tree,
> processing nodes based on tagname, and especially, based on relations
> with other nodes ( i.e. who`s your daddy, here`s your sister etc ),
> that would be great.

Well, at its simplest, you can do something like this:

for node in root.getElementsByTagName( "*" ):
    if node.tagName=="...": doSomething()
    elif node.tagName=="...": doSomethingElse()
    elif node.tagName=="...": doSomethingElseAgain()

I am trying to get people to test out and "vote" for a new library that
I want to put into Python 1.6. It is half-way between SAX and DOM. It is
efficient like SAX because you only ever work with a "peep hole" of your
document. It is convenient like DOM because you have access to your
parent nodes (but not your sibling nodes). If you want access to your
child nodes, you can ask it, but you are responsible for the performance
implications for building a tree (e.g. you don't want to do it on the
root of a 1GB document).

If you have a little time to experiment with something that may be part
of Python soon (or in the distant future...depending) then I would
appreciate you trying it out and seeing if it solves your problem as
conveniently and efficiently as I hope it will.

http://www.prescod.net/python/pulldom.html
-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"So there you have it folks, the Millenium: warring and whoring but
never boring." - Dennis Miller



From mfkilgore at graycouncil.com  Sun Jun  4 16:26:03 2000
From: mfkilgore at graycouncil.com (Mike Kilgore)
Date: Sun, 04 Jun 2000 20:26:03 GMT
Subject: PyXML-0.5.4 and Site-Packages
References: <392fa459.764858@news.c2i.net>
Message-ID: <vXy_4.7360$NT4.240673@newsread1.prod.itd.earthlink.net>

I am having the same problems on Windows... and must be missing something
very basic.  The setup.py install option does not appear to "install" to my
python site directory instead placing everything in build?  Any help would
be appreciated

Thanks


<thomas at cintra.no> wrote in message news:392fa459.764858 at news.c2i.net...
> Hi,
>
> I`ve tried to install PyXML-0.5.4 ( actually PyXML-0.5.3 and
> python-xml-0.5.1.rpm too with same result ) on my Mandrake 7.0
> Linux-box without luck.
>
> No erros during python setup.py build or setup.py install. The
> xml-folder has a __init__.py file with the __all__ : .... -line.
>
> When I try to use anything but the xml-folder, like
> "from xml.sax import *"
> I get a
> "no module named sax"-error.
>
> Then I tried to add a xml.pth in the site-packages-folder, looking
> like :
>
> xml
> xml/parsers
> xml/sax
> xml/unicode
> ...
>
> No change.
>
> I`ve tried to install this package before without luck. I just gave up
> in the end. But now it`s a vital part of a project, not some private
> messing around, so I need it to work.
>
> Is there a way around the site-package method of doing it?
>
> Any tips, hints?
>
> Thanks.
>
> Thomas
>




From davec999 at my-deja.com  Mon Jun 26 09:27:40 2000
From: davec999 at my-deja.com (David Currie)
Date: Mon, 26 Jun 2000 13:27:40 GMT
Subject: Encryption modules
Message-ID: <8j7lnq$9jq$1@nnrp1.deja.com>

Hi.  Does anyone know where to get the cryptographic modules written by
Andrew Kuchling that are referenced in the Python library
documentation?  The link given for them is on Starship and seems to be
down.

I also need a module that does md4 encryption, but that is not
mentioned in the description for this package.  Does this also come
with md4 support, or do I need another module (if so which one, link,
etc.. please).

For the record, I need these modules because I'm trying to implement a
module that does NTLM "authentication" (the M$ IIS<->IE proprietary
authentication) so that I can talk to an IIS-based HTTP proxy in a LAN
from Python.  If there is a module that already does this, does anyone
know a link to save me the trouble??

Thanks

Dave


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


From aa8vb at yahoo.com  Tue Jun  6 06:35:46 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 06 Jun 2000 06:35:46 -0400
Subject: Newbie question.
In-Reply-To: <393CE029.27B87C91@dulcesol.es>; from pablo.prieto@dulcesol.es on
 Tue, Jun 06, 2000 at 12:27:37PM +0100
References: <393CE029.27B87C91@dulcesol.es>
Message-ID: <20000606063546.A111435@vislab.epa.gov>

Pablo Prieto:
 |Hi all of you there!

Hi!

Try this:

 |How can I make the variable AYS global to setEnviroment, so I'd be able
 |to return "NUMBER 9" in getEnviroment?. (Don't say return "NUMBER9".
 |I'll get annoyed :))
...
 |AYS = ""
 |
 |def setEnviroment():
    global AYS              # <-----------------------------
 |  AYS = "NUMBER 9"
 |
 |def getEnviroment():
 |  return AYS


Now, I'm not saying that's the best way to provide this interface.
But it fixes your example.  Another idea:

   class Environment:

     def Set( val ):
       self.ays = val

     def Get():
       return self.ays

Now:

   env = Environment()
   env.Set( "NUMBER 9" )
   print env.Get()

-- 
Randall Hopper
aa8vb at yahoo.com



From bkc at Murkworks.com  Mon Jun 19 11:23:12 2000
From: bkc at Murkworks.com (Brad Clements)
Date: Mon, 19 Jun 2000 11:23:12 -0400
Subject: Python Programmer wanted. Sydney Australia
References: <8ik45k$rd0$1@nnrp1.deja.com>
Message-ID: <dQq35.29$pa6.2496@news.clarkson.edu>

com
<stalacticus at my-deja.com> wrote in message
news:8ik45k$rd0$1 at nnrp1.deja.com...

> considered. Come to the city of which Lonely Planet says ".you would
> have to die and go to heaven before you see a more spectacular setting
> for a city." Please send r?sum?s to jobs at memetrics.com.


That sounds like something Ian Wright would say.






From johnm at magnet.com  Tue Jun  6 13:16:23 2000
From: johnm at magnet.com (John Mitchell)
Date: Tue, 6 Jun 2000 13:16:23 -0400
Subject: PyOpinion: Does Python Programming Marginalize You?
In-Reply-To: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <Pine.SGI.4.05.10006061254220.22654-100000@lemur.magnet.com>

Alright, I'll bite.  Background: I'm a the senior software engineer at a
100-person web company, and I've been doing web for ~4 years, python for
almost as long.

On Tue, 6 Jun 2000, Richard P. Muller wrote:

> Does Programming in Python Marginalize You?
> 
> And yet, I worry that using Python marginalizes one. Python is hard to
> beat for writing applications that run on a single computer. But
> that's yesterday's computing model. The future of computing, heck, the
> present of computing, is in writing programs that run through a web
> browser. And the fact of the matter is that 99% of web browsers
> surfing the internet run Java and JavaScript, which gives these
> languages a huge advantage over Python, an advantage that Python's
> superior design may not be able to overcome.

Yesterday's computing model?

In our business, *no one* cares about client-side Java -- too much
start-up time, and it looks like heck. You didnt mention ShockWave and/or
Flash, which are client-side programmable.  Using JavaScript is more or
less a hack -- you cant guarantee it's enabled, so any checking or
computation must be re-done on the server side.  There are *hundreds* of
nasty bugs and incompatabilities with the different browser's JavaScript
implementations, and if you get a client-side problem they wont show up on
your logs...


> I'm aware of CGI scripting with Python. It's great. But JavaScript
> pages load faster than CGI pages, and are easier to program, to
> boot. Plus, JavaScript can directly call Java Applet methods, enabling
> it to do functions that are simply impossible with CGI scripts.

No one uses CGI anymore, FastCGI et al are faster, although CGI can be
quite quick.  One site we run that gets ~1M hits/day has a one-line Perl
script on the home page!

I'd disagree with the "faster" bit, also the "easier" bit.  Having
client-side JavaScript call client-side Java is cool, no doubt, but rather
difficult in a Internet situation where you cant tell people which browser
& Java flavor to use, and cant detect client-side errors!
(Intranet: maybe.)

> I'm also aware of the JPython module, which promises to enable
> us to write applets in Python instead of Java. But who among us has
> actually *used* JPython? Or, for that matter, who has come across a
> random web page on the net that uses JPython applets? 

JPython is stunning.  It works, and works well.  It's already found a few
niches, one of which being testing Java apps.  Java's lack of
interactivity is a bummer (although: check out BeanShell:
http://www.beanshell.org/ )

> Do I recommend that students today learn Python? Certainly

Check out Guido's "Computer Programming for Everybody", which discusses
programming as a tool for teaching thinking, in addition to the normal
uses: http://www.python.org/cp4e/


> If we don't want to see Python wind up like Ada, what should we do?
> I've been teaching a 6-lecture short-course on Python here at Caltech,
> in an effort to introduce more people to the language. I'm thinking
> about learning enough JPython to give an additional lecture on that
> language. Hopefully, if enough people use JPython for writing Applets
> we can help establish it as a reasonable alternative to Java.


Go learn JPython, although that wont take long.

There are other client-side language alternatives: ShockWave/Flash are
excellent.  I'll *never* write another applet if I can help it.  
(Although you can compile a shockwave movie to Java if you want to...)


I make my money writing web applications, and I disagree with your
client-side emphasis.  Server-side mostly we use Java, but that's because
of buzzword-compliance.  It's a tremendous pain in the ass.

For heavy string-formatting programs with loose specifications, Python
will beat the tar out of Java every time.  Matter of fact, I cant really
think of a niche for Java.  Hmm.  Maybe I should beat management with the
cluestick again.


- j





From eviltofu at rocketmail.com  Mon Jun 26 07:32:26 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Mon, 26 Jun 2000 11:32:26 GMT
Subject: Differentiating between WinNT and Win32
Message-ID: <eviltofu-F9E289.07322926062000@news.earthlink.net>

Is there a way to do this in Python? I went DejaDiving and found 
something called platform.py but the link to it on starship is dead. 
Anyone has any clues?


From 12klat at sightreader.com  Wed Jun  7 01:00:11 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 22:00:11 -0700
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <m3wvk2wy1u.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <393DD6DA.92794707@sightreader.com>



Michael Hudson wrote:

>        <snip>

> Wait?  I think Python is getting there ... it's far more "in the
> mainstream" now than it was a year ago, and I think the trend is
> continuing.  Use Python, be successful with Python - these are the
> best adverts surely?  I don't think there's a real need for a
> Java-style hype monster.
>
> optimistically-ly y'rs
> Michael

Why not use Python, be successful with Python, and have a Java-style hype
monster too?

We need a Hype Monster!  Otherwise how can we make everyone admit that
Python is by far the best language in every conceivable category and no
other languages have any reason to exist except maybe C but only for writing
extensions and maybe English too for mailing lists and stuff and that way
Microsoft which will die because of Open Source and Netscape will have to
get rid of Java and use Python instead and everyone will be very happy.

I think Python has a good chance at World Domination.  True, it is competing
with firmly entrenched standards, but standards change, and it's foolish to
underestimate the roll of education.  If students use Python in class, they
will continue to use it in the real world.  It is becoming easier every day
to compete with the giants, both in languages and operating systems (look
out, Microsoft) because people have finally learned how to write portable
open architecture software.  Eventually the best stuff will float to the
top, though this was not always the case.

Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/a7f9ea6d/attachment.vcf>

From akira.kiyomiya at autodesk.com  Tue Jun  6 12:47:54 2000
From: akira.kiyomiya at autodesk.com (Akira Kiyomiya)
Date: Tue, 6 Jun 2000 09:47:54 -0700
Subject: Space Sensitive?
Message-ID: <8hj9vv$1kd@autodesk.autodesk.com>

I hear Python is space sensitive, however, "Learning Python" from O'Reilly
and "Python Essential Reference" from New Riders do not cover this topic.

Could someone explain this one with a nice simple example why Python is
space sensitive?

Thanks     Akria




From fiona at sitegnome.com  Tue Jun 20 23:11:07 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 21 Jun 2000 03:11:07 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 20th, 2000
Message-ID: <20000621031107.6024.qmail@synop.com>

Hello Python Folk!

Below are the entries I've entered into http://python.faqts.com today.

cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
What module would I use to make serial I/O?
http://www.faqts.com/knowledge-base/view.phtml/aid/3888
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

Have a look at ftp.python.org/pub/python/contrib/sio-151.zip


-------------------------------------------------------------
Is there a Python module that has functions in it to manipulate Dbase tables and indexes?
http://www.faqts.com/knowledge-base/view.phtml/aid/3890
-------------------------------------------------------------
Fiona Czuczman
Anders M Eriksson

You don't say which version of dBASE and not on which platform you
are, so let's assume that you are using Windows 9x/NT/2000.

The only real alternative today is to use ODBC!

The win32-all comes with an ODBC modules that works OK, if you want to
spice it up a bit get mxODBC (seach on Vaults of Parnassus)


-------------------------------------------------------------
How can I convert a hex number to an integer?
http://www.faqts.com/knowledge-base/view.phtml/aid/3886
-------------------------------------------------------------
Fiona Czuczman
Jarkko Torppa, Thomas Wouters

>>> eval('0x32')
50
>>> import string
# Forces base to 16(hex)
>>> string.atoi('32',16)
50
# Guesses base from leading characters
>>> string.atoi('0x32',0)

Note that in Python 1.6, you can do the above two things like this:

>>> int('0x32',16)
50
>>> int('0x32',0)
50


-------------------------------------------------------------
How do I get a file list via an ftp connection?
http://www.faqts.com/knowledge-base/view.phtml/aid/3887
-------------------------------------------------------------
Fiona Czuczman
Thomas Weholt

I think you could do either a ls or dir, like so :

ftp = ftplib.... # open connection
file_list = ftp.dir()
or file_list = ftp.ls()

Just off the top of my head. Check the source of ftllib, it`s almost
self-explanatory.


-------------------------------------------------------------
Is anyone aware of a Computer Algebra System (CAS) i.e a mathematical software that is able to derivate, integrate, simplify, find limits, analytically - in python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3885
-------------------------------------------------------------
Fiona Czuczman
Konrad Hinsen, Jonathan Epstein, Matt Feinstein

Pythonica isn't there yet, but it's an interesting start. More at:

       http://www.strout.net/python/pythonica.html

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

Matlab has an interface to Maple (their "Symbolic Toolbox", I think).

It might be useful to study their API, and emulate it in a reasonable
(and hopefully legal) fashion in Python.

It's pretty simple, at least from the user's point of view. You
declare your symbolic variables a la

sym x y z;

and then just go ahead and use the symbolic variables in the
interpreter in algebraic or arithmetic expressions the same way you
would use numerical expressions-- all the arithmetic and algebraic
operators used by the interpreter are appropriately overloaded. There
is an accompanying suite of functions and state variables that control
how the symbolic stuff is done. There is also an 'Extended' symbolic
toolbox that gives the user direct access to the Maple kernel.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I get the current date and time?
How do I print out the current time?
http://www.faqts.com/knowledge-base/view.phtml/aid/3600
-------------------------------------------------------------
Mike Hostetler, Fiona Czuczman
http://www.python.org/doc/current/lib/module-time.html

I didn't think that the official documentation was very clear on this 
(but it is there), so I'm adding it here.

If you do

>> import time
>> print time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(time.time()))

You'll see something like:

2000-06-08 15:05:09

To dissect this:

time.time() gets the time in seconds since the epoch

time.localtime(secs) creates a tuple

time.strftime(string,tuple) creates a string of the time, using the 
formatting in "string" (that is what the "%"'s are doing there).
To see the full table of formatting options, see 
http://www.python.org/doc/current/lib/module-time.html






From rupole at compaq.net  Tue Jun 13 22:21:32 2000
From: rupole at compaq.net (Roger Upole)
Date: Wed, 14 Jun 2000 02:21:32 GMT
Subject: Problems with COM
References: <Wey15.3428$UM2.51434@weber.videotron.net>
Message-ID: <M_B15.88$18.48718@pouncer.easynews.com>

What does the SearchDocuments function look like ?
Couldn't you just create the object in that function, and
then return object.count to the calling function ?

BTW, resultSpace=win32com.client.Dispatch actually
assigns the Dispatch function to resultspace, so you are
passing a function rather than a COM object.

                Roger Upole

"Jean-Louis Couturier" <jean-louis.replacebylastname at ixiasoft.com> wrote in
message news:Wey15.3428$UM2.51434 at weber.videotron.net...
> Problem : I need to access an object created by an OUT variable in a
> function call to a COM object.
> Code example :
>
> import win32com.client
> import pythoncom
>
> _testQueryString="""<?xml version="1.0" encoding="UTF-8"?><query
> VERSION="1.0" RESULTSPACE="R1"><andkey><key
>
NAME="Date"><interval><start><date><year>1996</year><month>January</month><d
>
ay>4</day></date></start><end><date><year>1997</year><month>December</month>
> <day>30</day></date></end></interval></key><key NAME="Full
> Text"><elem>commission</elem></key></andkey></query>"""
>
> def debug(msg):
>     print "[DEBUG]",msg
>
> def get_docbase_service(servername="jeanlouix", docbasename="TEXTML
Demo"):
>     debug("Connecting to COM server")
>     clientService=win32com.client.Dispatch("TextmlServer.ClientServices")
>     debug("Connecting to server "+servername)
>     serverService=clientService.ConnectServer(servername)
>     debug("Connecting to DocBase "+docbasename)
>     docBaseService=serverService.ConnectDocBase(docbasename)
>     return docBaseService
>
> def query_server(docBaseService, queryString):
>     searchService=docBaseService.SearchServices
>     resultSpace=win32com.client.Dispatch
>     debug("Querying... ")
>     searchService.SearchDocuments(queryString,resultSpace)
>     return resultSpace.Count
>
> def main():
>     docBaseService=get_docbase_service()
>     print query_server(docBaseService,_testQueryString)
>
> if __name__=="__main__":
>     print "Testing pyTEXTMLServer"
>     main()
>
> My example above gives the following error on the line "return
> resultSpace.Count":
> AttributeError: Count
> The resultSpace object is a COM object that cannot be created beforehand.
> It has to be placed in a VARIANT that is fed to the function
> SearchDocuments.  How do I do this?  I tried creating a Dispatch object
> without specifying a type but it doesn't seem to work.
>
> Thank you in advance for any solution or direction to relevant
> documentation.
> I will be happy to provide more details if I have not been precise enough.
>
> Jean-Louis Couturier
> Developer
> Ixiasoft
>
>
>
>




From pablo.prieto at dulcesol.es  Tue Jun 27 09:18:24 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Tue, 27 Jun 2000 14:18:24 +0100
Subject: Unregistering COM Classes
References: <iUL45.23838$ds.694847@newsread2.prod.itd.earthlink.net> <8F5CEEDFAgmcmhypernetcom@199.171.54.195> <9b855.28036$ds.808114@newsread2.prod.itd.earthlink.net> <8F5DEAB46gmcmhypernetcom@199.171.54.155>
Message-ID: <3958A9A0.78FECF28@dulcesol.es>

Hi evrybody there!

I've made a class for registering as a COM Server.
One of the functions is this:

	def getVersion(self):
		return "VERSION 2"


I've put this two sentences in a class 

if __name__=='__main__':
	import win32com.server.register
	win32com.server.register.UnregisterClasses(testAuto)
	win32com.server.register.UseCommandLine(testAuto)
	
So, I hoped that all the modifications to the class testAuto would be
updated.
But when I call from VB:
	Set o = CreateObject("Python.TestAuto")
	MsgBox o.getVersion

I get in the screen "VERSION 1"

That was the message in getVersion the first time I built the class.
So,
How can I tell python to rebuild with the new code without having to
define a new class?.

Thanks in advance.

Pablo.



From donn at oz.net  Wed Jun 28 01:25:48 2000
From: donn at oz.net (Donn Cave)
Date: 28 Jun 2000 05:25:48 GMT
Subject: Controlling an external program with Python
References: <3958C648.1D76@rd.francetelecom.fr> <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com>
Message-ID: <8jc28s$8eb$0@216.39.151.169>

Quoth Alexandre Ferrieux <alexandre.ferrieux at rd.francetelecom.fr>:
| Paul Duffin wrote:
|>> existence of the 'pty' unix package, which is roughly "Expect minus
|>> Tcl", and especially its 'nobuf' tool:
|>>
|>>         nobuf cmd [args...]
|>>
|>> starts cmd as a child, strongly held between ptys, so that the overall
|>> command behaves properly (ie in unbuffered or line-buffered mode) betwen
|>> pipes. If all you want is that, Expect is surely an overkill.
|>>
|> 
|> Where can I get the 'pty' package from ?
|
| Last time I checked it was a pty-4.0.tar.gz .
| I find it currently with FTPSearch at
| ftp://ftp.sunsite.org.uk/Mirrors/wuarchive.wustl.edu/languages/c/unix-c/kernel/pty4.tar.gz
| .
|
| Surely more recent/portable variants must exist...

Not of Dan Bernstein's pty, but the MIT Kerberos 5 implementation
of all things has a portable pty library.  One unfortunate effect
of its association with Kerberos is some really silly export problems.
They appear to have thought about breaking it out, and may already
distribute it in other places, I don't know.

|>> But ptys are no rocket science anyway. The nobuf above should be doable
|>> in pure Python within 20 loc...
|
|> While ptys are not rocket science they are also not particularly portable,
|> Expect handles all of this for you.
|
| In unix, AFAIK the only nonportable part is the *naming* of the pty
| devices. That kind of decision is fairly trivial to make in
| sh/awk/perl/python/tcl...

The Berkeley pty, with the kind of naming scheme you allude to, is
already supported in the "pty" module.  It's supposedly tested on
Linux, and I'd expect it could work on FreeBSD/NetBSD/etc., and
probably lots of others like Digital UNIX.

But it's far from universal, and some of the alternatives do use
special functions that would need to be supported in C.

On the bright side, I am optimistic that the horrors of past
pseudotty implementations are fading into oblivion and the
survivors are fewer and cleaner.  So if you have looked at
Expect's pseudotty support and run away screaming, I think
there is no reason a module written today would have to be
that messy.

|> Windows is also a real problem as it
|> does not have ptys and needs to be approached in a completely 
|> different way.
|
| Yes but Expect for Windows is a totally different distrib too :-(
|
| Hence the suggestion to consider writing a specific Python module (if
| there isn't already - I've just arrived in town): for unix it just opens
| the proper devices and substitutes the popen2() pipes with the masters;
| for windows more trickery is needed (namely, behaving as a debugger to
| trap IO calls!), but should be doable in win32all, shouldn't it ?

Good idea, especially if the author isn't going to run away after a
couple of months and leave it to gather mold.

	Donn Cave, donn at oz.net


From JamesL at Lugoj.Com  Fri Jun  9 23:07:55 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Fri, 09 Jun 2000 20:07:55 -0700
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <3941B10B.8E6B2B41@Lugoj.Com>

Steve Mullarkey wrote:
> 1. What productivity increase do you achieve ?

I find it easier and quicker to lay out high level design and in
experimenting with finding the best class/data schema. And I can then run
it. The syntactical constructs of C++ tend to distract from the design
thought processes. All in my humble opinion of course (and I think that is
what you were asking for).

I wouldn't say I am many times faster at this with Python than C++, just
that the Python syntax seems to be more pleasurable to work with and that I
*feel* I make good progress with it. It is subjective I'm afraid.

P.S. Used C for about 20 years, C++ for about 4.


From mwh21 at cam.ac.uk  Sun Jun  4 16:01:46 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 04 Jun 2000 21:01:46 +0100
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk> <393A3A9A.FB9F2E37@stud.ntnu.no> <393A4C60.MD-1.4.4.angband@blueyonder.co.uk> <m3d7lxxngf.fsf@atrus.jesus.cam.ac.uk> <033501bfce78$ba0d62a0$6401a8c0@home>
Message-ID: <m38zwlxljp.fsf@atrus.jesus.cam.ac.uk>

Nooo, the point I was trying to make is that a cross platform library
hs to be able to carry out all its actions on every platform it is
meant to run on (if they only work on certain platforms you haven't
really got very far!).

Cheers,
M.

"Darrell Gallion" <darrell at dorb.com> writes:

> Check out os.py in the standard libs.
> It configures for each OS.
> >>> os.name
> 'nt'
> 
> --Darrell Gallion
> 
> From: "Michael Hudson" <mwh21 at cam.ac.uk>
> > 
> > Because I suspect the behaviour common to all the platforms that
> > Python runs on is so restricted as to be useless.
> > 
> 
> 
> 

-- 
  Indeed, when I design my killer language, the identifiers "foo" and
  "bar" will be reserved words, never used, and not even mentioned in
  the reference manual. Any program using one will simply dump core
  without comment. Multitudes will rejoice. -- Tim Peters, 29 Apr 1998


From joseph at src.no  Thu Jun  1 13:20:31 2000
From: joseph at src.no (Joseph Santaniello)
Date: Thu, 1 Jun 2000 19:20:31 +0200
Subject: Currency format for floats?
Message-ID: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no>

Hi,

I have an SQL query which I can format to give me a nice value like
23,342.40 which Python squashes into 23342.4 when I stick it into a
list element. Is there a nice way to format floats so they look like
(US) currency?

Thanks,

Joe



From jcw at equi4.com  Fri Jun  9 07:56:49 2000
From: jcw at equi4.com (Jean-Claude Wippler)
Date: Fri, 09 Jun 2000 13:56:49 +0200
Subject: Why is Python so slow?
References: <Pine.GSO.4.10.10006091112080.21387-100000@sundial>
Message-ID: <jcw-A1148D.13564909062000@news.nl.net>

On Fri, 9 Jun 2000, William Dandreta wrote:
> Why does it take Python 50 times longer than C?

Because Python is a high-level interpreted language.  As others have 
pointed out, the 50x figure you state is pessimistic.

Absolute speed is a confusing metric - fast enough is what matters.

Like every scripting language, Python lets you code at more than one 
level of abstraction:  take the slow part, recode in C/C++, and Python 
will always be fast enough.  Given that 80% of the time is spent in 20% 
of the code, just figure out what the 20% are, then rethink/recode.

Now compare the total time to complete the job:  all in C/C++ vs. all in 
Python and then (maybe) recode some of it.  The second wins hands down.  
Always - because there is (much) less code to write to do the job.

So one answer to your question is: who cares?

-- 
-jcw


From thomas at xs4all.net  Thu Jun  8 16:00:32 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 8 Jun 2000 22:00:32 +0200
Subject: My wishlist for Python3K
In-Reply-To: <8hogq2$cce$1@slb6.atl.mindspring.net>; from dalke@acm.org on Thu, Jun 08, 2000 at 10:19:18AM -0600
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <20000608122619.I469@xs4all.nl> <8hogq2$cce$1@slb6.atl.mindspring.net>
Message-ID: <20000608220032.B20700@xs4all.nl>

On Thu, Jun 08, 2000 at 10:19:18AM -0600, Andrew Dalke wrote:

> Thomas Wouters :
> >On Thu, Jun 08, 2000 at 01:47:14AM +0000, Huaiyu Zhu wrote:
> >> My wishlist for Python3K (This is not to start a flame war):

> >> 2. Augmentation operators: +=, *=, etc.  Dot operators .*, ./, etc (for
> >>    numerical computation to distinguish matrix and elementwise
> >>    operations).

> >This needn't wait for Py3K, and in fact I'm writing code for this right
> >now,

> Would there be a problem with the ambiguity of "3./2" being parsed
> as either
>   '3.' '/' '2' == 1.5
>   '3' './' '2' == whatever ./ does to scalar integers, probably
>                   make it be regular division == 1
>   ?

I'm afraid I was a bit careless in the quoting. I am not implementing the
'dot operators' for numerical computation, I'm only implementing augmented
assignment (+=, >>= and friends.) But for what it's worth, I think the 'dot
operators' as using a dot are out -- there is no way the parser can
distingish between float-creation, attribute-lookup and matrix-operations,
and still remain as obvious as it is now. At least, I dont think so.  Using
another character would be possible, of course ;)


Hacking-ly yr's,

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From dana at oz.net  Mon Jun 12 02:11:44 2000
From: dana at oz.net (dana_booth)
Date: 12 Jun 2000 06:11:44 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <8i1uv0$hki$0@216.39.141.21>

mkx at excite.com wrote:

mec: In many Perl vs. "other scripting language" discussions, I have often
mec: seen Perlers state that those who dislike Perl's syntax and structure
mec: are often non-Unix types, and thus are not used to common Unix
mec: concepts. And, since Perl was born as a Unix admin tool, us "Windows
mec: types" will likely be lost.

mec: So, I come now to Python. After reading much about it's ease of use
mec: and readability advantages, I see it may be a better alternative. In
mec: most cases, I can actually get a concept of what the code is doing,
mec: without picking up a reference book - after only a week of
mec: experimentation.

mec: So my question is this: Is Python a more "Windows-friendly" language?

Okay, I couldn't leave this alone. :)

I think "Windows-friendly" is relative. I know Windows only programmers who
can out-Perl just about anyone. (FWIW, I'm in the UNIX crowd. :)

Anyway, the reason to choose to learn Python is not because you use Windows,
but because it's a great language for either platform. Learning the concept
of objects is a snap, coding with Python. The code is so clean and
structured... How many times have you written something in Perl, just to
have some Perl hacker go out of his way to show you up by doing the same
thing in less code? :) Doesn't happen with Python, there's a right way, and
there's the way that don't work. Fortunately, the right way is really easy.
Use one of the graphical modules to construct graphical applications...

Yesterday, in this NG, I asked about a Python interface for MySQL. I had a
reply later in the day, grabbed the module, and before I went to bed, was
communicating with my company's database with a graphical client made with
Python. The same code worked from my kids Windows computer, after I
installed the MySQL module there, too.

Can't say enough about the language... Way too easy to learn. In just a few
hours of farting around with a reference book on your lap, you can be making
useful stuff. Try "Python - Essential Reference" by David M. Beazley. Great
book.

-- 

-----
Dana Booth <dana at oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From andres at corrada.com  Thu Jun  8 15:03:14 2000
From: andres at corrada.com (andres at corrada.com)
Date: Thu, 8 Jun 2000 15:03:14 -0400
Subject: Why can't I restore my redirected stdout?
In-Reply-To: <8hoo05$fc$1@news5.jaring.my>; from Raymond Ng Tong Leng on Fri, Jun 09, 2000 at 01:52:52AM +0800
References: <8hoo05$fc$1@news5.jaring.my>
Message-ID: <20000608150314.A26601@corrada.com>

On Fri, Jun 09, 2000 at 01:52:52AM +0800, Raymond Ng Tong Leng wrote:
> I am trying to implement a verbose and quiet mode in my Python script. So I
> wrote the following class:
> 
> class Quiet:
>     def write(self, data):
>         pass
> 
> Then I set sys.stdout to Quiet by:
> 
> >>> sys.stdout = Quiet()
> >>> print 'hello'
> >>>
> 
> 'hello' doesn't come out which is what I want. But, when I tried to restore
> sys.stdout's original settings:
> 
> >>> sys.stdout = sys.__stdout__
> >>> print 'hello'
> >>>
> 
> 'hello' is not printed either! How do you restore sys.stdout to its original
> settings? Also, I was wondering, are there other ways to suppress
> sys.stdout? I'm just curious whether I could suppress sys.stdout without
> redirecting to a class. It seems to me that I should be able to redirect
> sys.stdout to a null device or something. Thanks!
> 

sys.stdout and sys.__stdout__ point to the same object. Save the default
sdtout before you redefine it:

>>> import sys
>>> sys.stdout
<open file '<stdout>', mode 'w' at 80ce940>
>>> sys.__stdout__
<open file '<stdout>', mode 'w' at 80ce940>
>>> default = sys.stdout
>>> class Quiet:
...     def write(self, data):
...             pass
...
>>> sys.stdout = Quiet()
>>> sys.stdout
>>> print 'hello'
>>> sys.stdout = default
>>> sys.stdout
<open file '<stdout>', mode 'w' at 80ce940> 

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From chibaA at TinterlogD.Tcom  Thu Jun  1 22:28:05 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Fri, 02 Jun 2000 02:28:05 GMT
Subject: scripts not completing...
Message-ID: <39371b5e.376157555@news1.on.sympatico.ca>

Other than having a runaway script... is there any other routine
reason why a script would not terminate properly?  (I'm running python
scripts as CGI through apache).  I noticed that script processing gets
really slow after a while, and noticed that there were still processes
still running with that script.  

Is there anyway to make the system automatically kill those processes
(apache, python scripts) which have been idle for a long time (and
probably hanging, or running away)?

TIA

kc


From thomas at xs4all.net  Sat Jun  3 17:47:11 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 3 Jun 2000 23:47:11 +0200
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
In-Reply-To: <slrn8jj4ot.om.thor@localhost.localdomain>; from thor@localhost.localdomain on Sat, Jun 03, 2000 at 09:18:09PM +0000
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <3937D345.CC0935D5@uab.edu> <39385DC2.BC0B4853@stud.ntnu.no> <slrn8jj4ot.om.thor@localhost.localdomain>
Message-ID: <20000603234710.Q469@xs4all.nl>

On Sat, Jun 03, 2000 at 09:18:09PM +0000, Manuel Gutierrez Algaba wrote:

> Seguro que si, quien sabe, a lo mejor mi perro sabe el holandes. :P

Eso es divertido, porque mi pescado preferido habla espa?ol :)

> Por cierto, vaya ca?a que va a ser el tener a Guido trabajando
> al 100 % para python. Espero que no se les vaya la mano a?adiendo
> nuevas caracteristicas a python y lo estropeen.

Dado el ese Guido y la otra gente que trabaja en Python (deja para darles un
cierto cr?dito tambi?n! :-) han evitado de estropear Python hasta ahora,
deja el corte ellos una cierta holgura y espera y ve cuidadosamente. Dudo
que la arruinar?n eso f?cilmente.

Hey, Martijn, misschien dat er ergens een aangepaste babelfish draait die
*wel* nederlands<->engels kan :)

Multilingual-ly yr's,
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From cfelling at iae.nl  Fri Jun  9 21:07:52 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 10 Jun 2000 03:07:52 +0200
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl>
Message-ID: <8hs4d8$33b$1@animus.fel.iae.nl>

Thomas Wouters <thomas at xs4all.net> wrote:

> As promised, here is my patch to add augmented assignment (+=, -= and
> family) to Python. It's still Work In Progress, and probably requires a
> rewrite of some magnitude before Guido will accept it.

Great! Though I'll probably only use it once it's part of the main distro:(

> The 'ad' in __add_ad__ is what Guido proposed, and I have to say I dont
> particularly like it. Making it '__iadd__' or perhaps '__inplace_add__'
> seems more obvious, and more in keeping with '__radd__'. But perhaps Guido
> meant something entirely different from this patch.

When Guido wrote this I had the urge to reply, but as he stated he wouldn't
read any follow-ups I let it pass. But now it's here again, this ugly use
of abbreviations, so here I go...

What about __add_is__? This sounds like addish, a kind of add; it resembles
its use, plus followed by is; would prevent confusion of where to place the
r like in __r_inplace_add__ versus __inplace_radd__; and still is concise:)

-- 
groetjes, carel


From thomasm at legend.co.uk  Tue Jun 20 04:39:35 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Tue, 20 Jun 2000 09:39:35 +0100
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk> <8imc1f$181$1@slb6.atl.mindspring.net>
Message-ID: <394f2d55.0@katana.legend.co.uk>

> >I notice that if a directory is root owned and chmoded 700 (let say
/test)
> >and if it contain a directory called test2
> >os.path.isdir("/test/test2") return 0 if called by a "normal user"
> >I was expecting an exception.
>
> Why were you expecting an exception?

I created few directories for my application as "myself" but I am running
the script as a cgi.

drwx------ thomas thomas  [somwehere in the world]/test
drwx------ thomas thomas  [somwehere in the world]/test/test2

my web server and I are different users,  as "nobody" have no read/go
through access on the parent (test) directory I am checking, it can not even
try to perform any action on the child one (test2).

Nothing with the current return code indicate that the test failed.
I was just guessing : return 0 means no directory present, not enable to
perform the requested action ..

Thomas







From siddi at penguinpowered.com  Fri Jun 16 07:48:38 2000
From: siddi at penguinpowered.com (Juerg Schulthess)
Date: Fri, 16 Jun 2000 13:48:38 +0200
Subject: freeze, standalone?
Message-ID: <394A1416.ABBE428C@penguinpowered.com>

I'm trying to use the 'freeze' and 'standalone' utilty to build
standalone executables under Win32. The executables produced with
'standalone' work fine. Now I'd like to burn the executable on CD. This
adds a 'readonly' attribute to the executable and this seems to change
the behaviour of the exe completely - it exits immediately and does
nothing and exits without any output. no errors, nothing! This efect can
be easily simulated by adding the 'readonly' attribute in the file
properties. Has anyone succeeded in writing standalone executables
produced from Python code to CDs that still work from the CD???
Any help is appreciated!

-- Juerg Schulthess



From thomas at xs4all.nl  Fri Jun 30 09:08:55 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 30 Jun 2000 13:08:55 GMT
Subject: Is this a leaker?
References: <GH075.17918$NP5.455672@newsread2.prod.itd.earthlink.net> <0S075.17928$NP5.457944@newsread2.prod.itd.earthlink.net>
Message-ID: <slrn8lp6ul.u8u.thomas@centurion.xs4all.nl>

On Fri, 30 Jun 2000 13:00:44 GMT, Arint? <shouldbe at message.com> wrote:
>Let me expand this a little more
> PyObject* deqme(Py_BuildValue("i",0));
> PyObject* comm=PyObject_GetItem(args, deqme);
> Py_XDECREF(deqme);

>comm isn't decref until later should I wait to decref deqme after I do
>comm??

No. 'deqme' is simply used to find 'comm', but isn't necessary to keep
'comm' valid. And if it was, 'comm' would INCREF deqme itself. (Or, in the
extreme case, it would be documented as 'Stealing' the reference to 'deqme')

>Also, here is a painful question Who is responsible for
>return Py_BuildValue("i",0);
>I skimmed thru python's source and it seems they Incref's on return values
>so now it seems it is leaked.

This depends on how you document your function :) If you document it as
returning a borrowed reference, the caller should not DECREF the returned
value. If you document it as returning a new reference, it should take care
of the DECREF itself. In this case, it doesn't make sense to return a
borrowed reference: If you do a DECREF on the fresly-created
Py_BuildValue()-object, it'll end up destroyed !



From cut_me_out at hotmail.com  Wed Jun  7 10:05:18 2000
From: cut_me_out at hotmail.com (Alex)
Date: 07 Jun 2000 10:05:18 -0400
Subject: Calling external programs: ioredirection
References: <393e4e6f.72239515@nntp.interaccess.com>
Message-ID: <etdg0qp37tt.fsf@w20-575-109.mit.edu>

Check out os.popen.  It will let you do what you are asking with grep.
As for strings of pipes, I have never had much luck getting two-way
communication between python and a process going, but you might want to
check out the popen2 module.

Alex.


From chibaA at TinterlogD.Tcom  Sun Jun 25 17:02:00 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Sun, 25 Jun 2000 21:02:00 GMT
Subject: http redirect line in Python dynamic html script.?
Message-ID: <395671e6.929362672@news1.on.sympatico.ca>

Sorry if this isn't the right group...  but it's python related.

Anyone know the line to redirect to another script when creating
python dynamic web-pages?  This is the scenario:

A script (let's say "script.py") runs, and needs to call another
python script which will end up displaying results to http.  I could
call it using:
	print "Location: next_script.py?arguments\n\n"
But this presents two problems.  One, it's terribly slow (because it
has to display the (invisible) URL, and restart the python
interpreter; and second, I can only call the next_script.py using the
GET method (not POST) - so the arguments are visible on the URL line
of the browser.

So... is there a way to call the next script, using a line other than
"Location: next_script.py?arguments\n\n"??  If not, then is there a
way to invoke "next_script.py" from script.py AND pass those arguments
to it?  (I can't seem to find my way around the Python Docs...).

Thanks!




From contactXXX at redspice.com  Tue Jun  6 01:32:22 2000
From: contactXXX at redspice.com (contactXXX at redspice.com)
Date: Tue, 06 Jun 2000 05:32:22 GMT
Subject: How do I create extensions to Python in C?
References: <BlAW4.3473$DQ3.245148@news1.atlantic.net> <392BE0ED.12E4DA0E@cfdrc.com>
Message-ID: <393C8A74.F2FAAF76@redspice.com>

 like lyle pointed out swig is a very good tool. This is an easier route
than learning the whole shooting match which is very process oriented
and can be automated which is what swig does


From porter at et.byu.edu  Wed Jun 21 18:10:50 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Wed, 21 Jun 2000 16:10:50 -0600
Subject: curses input with threads
In-Reply-To: <394AF6C8.FFCBF182@houston.rr.com>
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <394AD545.AC791A2B@san.rr.com> <394AF6C8.FFCBF182@houston.rr.com>
Message-ID: <Pine.GHP.4.21.0006211541260.6090-100000@don.et.byu.edu>

I have a multi-threaded program whose interface is done with curses.  It
pretty much just runs on its own, and displays its progress on the
screen.  I want to add a pause function.  I tried creating a separate
thread that runs stdscr.getch() to catch the keyboard entries.  But when I
do that, the entire program seems to pause to wait on that
getch() function.  

Is there any way to get input without freezing the other threads?

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------




From elengvenis at uswest.net  Tue Jun  6 19:36:17 2000
From: elengvenis at uswest.net (Eric Lengvenis)
Date: Tue, 6 Jun 2000 18:36:17 -0500
Subject: Perl and NT/2000 resources please
Message-ID: <zXf%4.522$t56.6648@news.uswest.net>

I currently use Perl to administer NT. I mostly do simple things like change
settings on every user on the network in one fell swoop. Or to allow call
center to unlock accounts without explicitly giving the Help Desk Tech the
rights to do this. Or to explicitly check for and stop processes before a
scheduled reboot of the NT servers (And you all know they need regular
rebooting to stay healthy).

I would like to learn Python, but the resources I see don't cover this type
of thing. I am no programming guru, which is why Perl's large support
community has been such a boon. Could someone point me in the right
direction? I am dying to give Python a real world try.

Also, my limited Perl ability has garnered some attention in my department
and they want me to write the connectivity between our intranet and the SQL
server our HR dept. uses. I told them yes, but I have never written such a
thing before. Any advice on where an all NT guy like me could pick this up
for Python?

Thanks for any advice.

Eric Lengvenis

PS, I am not adverse to *NIX, but my department doesn't pay me to work on
it.




From xklbr at hotmail.com  Tue Jun 13 21:03:49 2000
From: xklbr at hotmail.com (Dirk Xklbr)
Date: Tue, 13 Jun 2000 20:03:49 -0500
Subject: Web Session
Message-ID: <8i6lg3$htd$1@news.mty.itesm.mx>

I need to make a cgi program that makes a session in web (like a email web
session), where do I start from?, do I need cookies, how does it work?
Thanks





From thomas at cintra.no  Tue Jun 13 04:35:56 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 13 Jun 2000 08:35:56 GMT
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no> <3945e37b$0$489$7f31c96c@news01.syd.optusnet.com.au>
Message-ID: <3947efe3.198635993@news.online.no>

On Tue, 13 Jun 2000 17:44:52 +1000, "Jason" <it at onestopent.com.au>
wrote:

>hi thomas
>thanks for your help again...
>here's the error message i get when i run your script in IDLE
>the listing is the first in the txt file
>can you help pls....
>does this script make the output file a tab file..i can't see any tab in the
>script at all...can you let me know pls...
>
>thanks
>jason
>
>FORM: ['Popular Recording']
>CMTYPESWITCH: ['Popular']
>RELOCORIGINAL: ['8F2C014FEFE7A61DCA2568D600267EBD']
>CMCOUNTRY_ORIGIN: ['']
>CMGENRE_CODE: ['']
>CPTITLE: ['LIQUID']
>CPARTIST: ['RECOIL']
>CPOARTISTS: ['']
>CMCATNUMBER: ['CDSTUMM173']
>CMAPN: ['5016025611737']
>CMARIAPRICECODE: ['C1145']
>CMARIAMEDIACODE: ['21']
>CMARIADISTRIBUTORCODE: ['MDS']
>CMARIAPACKAGECODE: ['']
>DMRELEASE: ['19/06/2000']
>cMDateFormat: ['19/06/2000 12']
>DMDELETE: ['']
>cMDeleteStatus: ['No']
>CMLISTPTRACKS: ['']
>CMLISTPARTISTS: ['']
>RELOCMEDIA: ['0FE68A46C3E72687CA2568D6002C9FAD']
>CURMPRICE: ['21.71']
>CMTYPE: ['Compact Disc']
>CMRECORDCOMPANY: ['MUSHROOM DISTRIBUTION SERVICES']
>CMPACKAGE: ['']
>CMARIADISTRIBUTORHOUSECODE: ['BFM']
>cDistributorHouse: ['BFM']
>CMLOCALE: ['Y']
>$UpdatedBy: ['CN=PPT/OU=AEROS/O=JUKEBOX']
>$Revisions: ['05/05/2000 06']
>Traceback (innermost last):
>  File "C:\PROGRA~1\PYTHON\TOOLS\IDLE\ScriptBinding.py", line 131, in
>run_module_event
>    execfile(filename, mod.__dict__)
>  File "C:\WINDOWS\Desktop\aeros1.py", line 6, in ?
>    print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
>IndexError: list index out of range
>
>

Look at the error-message. IndeError: list index out of range. It`s
referring to the line above, 
print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`

The variable pair have to parts, accessible by using an index, like
pair[0] would get you the first item in pair, pair[1] the second etc.
In the line that gives you the error you try to access and print an
item that doesn`t exists. That`s why it says "IndexError". Before
trying to access the pair variable you should check to see that it
actually has some contents, like so:

if len(pair) == 2:
	print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`

Then it will only try to access and print if the pair-variable has to
items in it.

Hope that helps.

Thomas



From jkraska1 at san.rr.com  Sat Jun  3 15:23:39 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 03 Jun 2000 19:23:39 GMT
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com>
Message-ID: <39395BE1.B52756D7@san.rr.com>

> I'm against a linked list deque implementation for Python.
> But I'm plus 1 on doing it as an extension of lists.

For what it's worth, I'm implementing a dynarray which
leaves buffer room on both the head and tail as we speak.
Note that this slows down random access ever-so-slightly,
as you have to have a "virtual index" to account for the
actual and current positions of the "beginning" of the list.
However, the cost is very small: an extra add on every
access.

My current algorithm is as follows:

1. Store a "virtual index" into the list, allowing buffer
   room at the head.

2. On an insert at the head, simply decrement the virtual
   index and insert.


3. On an insert operation at the head, if the virtual index
   would go negative on the insert, double the allocation
   space of the array and place the actual used space of
   the array 1/3 of the way into the array.

4. On removal at the head, simply increment the virtual
   index.

5. If the virtual index becomes larger than 2/3 of the
   allocated space, move the actual used space of the
   array 1/3 of the way into the array (this last is to
   avoid an array which would grow infinitely on append
   to tail and removal from head).

6. The growth rate of the array will always be 100%
   increase (as compared to python's 50/500 heuristic,
   IIRC).

7. You'll be able to specify an initial size for the list.

This technique should have the same complexity as python
lists with the additional benefit of O(1) removal/insert
at the head without sacrificing O(1) random access. I'll
do a complete performance test as before, of course.

As per usual, the source will be entered into the public
domain.

C/


From jeffrey at digicool.com  Tue Jun 27 11:44:43 2000
From: jeffrey at digicool.com (Jeffrey P Shell)
Date: Tue, 27 Jun 2000 11:44:43 -0400
Subject: Darwin (Not Python, yet...)
References: <Pine.A32.3.90.1000624142526.18059A-100000@elvis.med.Virginia.EDU> <8j7skg$1f$1@bob.news.rcn.net> <Pine.A32.3.90.1000626195403.20434A-100000@elvis.med.Virginia.EDU>
Message-ID: <8jahjq$fd8$1@bob.news.rcn.net>

>  A binary package for Mac OS X Server from Scott Anguish.
>  It doesn't seem to be a Debian package, which is what Darwin wants.
>  I would guess it used whatever packaging NeXT used.
>  Unpacking in manually and trying to run python I get:

Yes, it's the MacOS X Server package format, which is the old installer for
OpenStep/NeXTStep/Rhapsody.  MacOS X Server is running a much older kernel
(Mach 2.5, older BSD stuffs), so NO MacOS X Server binaries even run on
MacOS X.  A lot has changed between MacOS X Server and the current Darwin
based builds of MacOS X.

This binary installation was just plain vanilla python --  no threads, no
shared library support.  Hence, it can't be used for Zope.

>  The other version is an OS X Source distribution by Serge Beauchamp
>  which says that it's the "Same as the standard unix distribution except
>  that I changed the Make files so that it can successfully compile
>  and install on a HFS+ volume."

I got this unpacked on an HFS+ volume, and there was a python and a
libpython1.5.a in there prebuilt (again, not the shared library support i
was going for).  I tried to reconfigure it in a way that Zope seems to
prefer, and it didn't work.  (that's when I remembered I had Python sources
on my UFS partition, so I just went over there and rebuilt and got it mostly
working).

>      BTW:  <http://www.mit.edu/people/wsanchez/papers/USENIX_2000/>
>       is a paper on "The Challenges of Integrating the Unix and Mmac OS
>       Environments"  which has some interesting info about some of the
>       differences in HFS+/UFS filesystem semantics and MacOS/Unix
>       file semantics.

I've glanced at this.  Again, I'm hoping someone smarter with makefiles and
configure scripts and other joys of unix can read documents like this and
get some changes into the Python core set.  I would love to see a
threaded/shared lib ready Python in the Darwin base.  (btw:  JPython works
under MacOS X.  I get a lot of leak errors though, i think it's due to the
JVM still being in development).

>   This version actually has the binaries included and they do run
>   under Darwin! Problem is, it is distributed as a .zip file, which
>   (I think) unpacked under MacOS9 (/w Stuffit Expander) with the
>   wrong line endings, so that some of the imports don't seem to
>   work properly. ( I think this is what's wrong -- I've seem similar
>   symptoms on Mac when unix .py files get transfered as binaries.)

Find the .zip file (if Stuffit didn't destroy it) and use the 'unzip'
command in Darwin to unpack it.

>   This zip file doesn't seem to be in gunzip format so I couldn't
>   unzip it from Darwin to get the right line endings.

Hmm.  unzip worked in DP4.

>   I'll try to find another way to unpack the .zip file and try again.
>   Still ironing out a few kinks, but it's nice to know that the
>    binaries are compatible.

It's nice to see interest in Darwin.  I'm disappointed in the basicness of
the binaries though, but i am glad people are taking time to make them.

Does Darwin follow the MacOS X directory structure convention?  (ie,
/Local/Users/Bob/
/System/Library/Frameworks/...
)?





From jajs at NOSPAMretemail.es  Mon Jun 12 20:52:07 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Tue, 13 Jun 2000 00:52:07 GMT
Subject: "import foo" and "from foo import bar"
Message-ID: <slrn8kaa4u.1r0.jajs@Pascual.es>

I have a question regarding the difference between those two, apart from 
having to type the module name before the symbol in the
first case.

If I am only using one or two simbols of a module (example: string): 

It's better to write the "from string import join" form or it's exactly
the same (in memory use and performance) that "import string"?


From jin.choung at gte.net  Thu Jun 15 01:29:34 2000
From: jin.choung at gte.net (jin choung)
Date: Thu, 15 Jun 2000 05:29:34 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net> <uitvbluqx.fsf@ctwd0143.fitlinxx.com>
Message-ID: <2RZ15.3850$Gp6.258234@dfiatx1-snr1.gtei.net>

oh absolutely!

and your advice is immediately acknowledged and heeded.  i'm not even very
certain that i even know what OO really is!  so yeah, much learning must be
accomplished before i can speak of any of this in any kind of credible
manner!

yes, it was my assumption that C++ was the OO god but it does not surprise
me at all that i am way off on it :)  but for the c++ book at least, it does
not assume any kind of knowledge of any abstract concepts and it starts from
the foundation of absolute ignorance.

i'd have it no other way!

jin




From gmcm at hypernet.com  Tue Jun 20 09:43:20 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 20 Jun 2000 13:43:20 GMT
Subject: ReRe: Kjbuckets-Module, Python 1.6 and NT
References: <394E4B02.5F4A0AB4@szs.ira.uka.de> <8inn9m$3q5$1@nnrp1.deja.com>
Message-ID: <8F596620Agmcmhypernetcom@199.171.54.155>

spex66 at my-deja.com wrote:

[snip]

>compiled the code without an error or a warning, into kjbuckets.dll and
>dropped it in the Python/DLLs directory
>
>then I get:
>>>> from kjbuckets import *
>Traceback (innermost last):
>  File "<pyshell#1>", line 1, in ?
>    from kjbuckets import *
>ImportError: dynamic module does not define init function
>(initkjbuckets)
>
>any ideas??? 

Did you compile in debug? Don't, unless you have python_d.exe and 
python15_d.dll.

- Gordon


From root at rainerdeyke.com  Sat Jun 10 21:45:36 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 11 Jun 2000 01:45:36 GMT
Subject: namespace question
References: <eviltofu-8E2F45.12073510062000@news.earthlink.net>
Message-ID: <4bC05.3319$vp3.359825@news1.rdc2.tx.home.com>

Jerome Chan <eviltofu at rocketmail.com> wrote in message
news:eviltofu-8E2F45.12073510062000 at news.earthlink.net...
> If I do the following:
>
> import socket
>
> socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
>
> how do I call a function from the socket module later on?

Technically correct answer (assuming that you don't use the name _socket):

_socket = socket
import socket
# call socket function here
socket = _socket
del _socket


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




From gmcm at hypernet.com  Tue Jun 13 16:05:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 13 Jun 2000 20:05:32 GMT
Subject: How do you FTP ASCII if ftplib strips CR?
References: <8i5jhd+h2lu@eGroups.com>
Message-ID: <8F52AEEE6gmcmhypernetcom@199.171.54.155>

[posted and mailed]

al at servana.com wrote:

>If you look at the FTPLIB docs, it states that RETRLINES will strip 
>CR when passing the text to the RETRLINES callback.
>
>So, how do you put the CR back in if you want them?

[snip]

>ourFTP.retrlines('RETR ' + filename, open(os.path.join(localdir, 
>filename), 'w').write)

How about:

lines = []
ourFTP.retrlines('RETR ' + filename, lines.append)
open(os.path.join(localdir, filename), 
    'w').write(string.join(lines,'\n'))

or:

f = open(os.path.join(localdir, filename),'w')
wrtr = lambda ln, f=f: f.write(ln+'\n')
ourFTP.retrlines('RETR ' + filename, wrtr)

- Gordon


From matt at virtualspectator.com  Mon Jun 12 06:57:57 2000
From: matt at virtualspectator.com (matt)
Date: Mon, 12 Jun 2000 22:57:57 +1200
Subject: Python Productivity over C++
In-Reply-To: <394484C0.A5E412D9@san.rr.com>
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <85snuj346q.fsf@one.net.au> <394484C0.A5E412D9@san.rr.com>
Message-ID: <0006122258481V.01005@localhost.localdomain>

hmmm, yes a python -> STL  without memory leaks .... that would be nice.



On Mon, 12 Jun 2000, Courageous wrote:
> > My first impression is that its dynamic typing, uncluttered syntax and
> > ample built-in types encourages more "natural" designs. It makes
> > coding easier and more fun (which is important to me). Even as a
> > newcomer, I feel that I'm using the language with more pleasure and
> > confidence rather than carefully paying attention to all its
> > idiosyncracies (C++) or forcing my design to fit the paradigm
> > ("classical OO" / Eiffel). I find that Python translates thought into
> > code without too much fuss, and virtually no linguistic cruft. But is
> > there more to it than that?
> > 
> > What is it about Python that encourages the "component" style you
> > described?
> 
> For one, treating each file/module as a distinct object is
> componentizing in itself, but this is neither here nor there.
> I find that you observations are closer to the most important
> truth than not: python is syntactically lightweight, and yet
> has rich expressive power.
> 
> Python is a simply wonderful prototyping environment, IMO.
> Now only if it had an optimizing compiler...
> 
> C/
> -- 
> http://www.python.org/mailman/listinfo/python-list



From teyc at bigfoot.com  Wed Jun  7 18:42:13 2000
From: teyc at bigfoot.com (Chui Tey)
Date: Thu, 8 Jun 2000 08:42:13 +1000
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net>
Message-ID: <8hmii1$ob7$1@bunyip.cc.uq.edu.au>

There is also a gdchart module for python as well as gdchart extension for
Zope.
http://www.fred.net/brv/chart/


Drew Fisher <drew at level3.nospam.net> wrote in message
news:393d0c07$0$17334 at news.denver1.Level3.net...
> Hello.  I am still learning Python, but I came across a need for
generating
> dynamic graphs (network load average, uptime, boring stuff like this :).
>
> Is the a module in Python that allows this?  If so, could someone point me
> to it?
>
> Thanks in advance,
> Andrew Fisher
>
>




From richard_chamberlain at ntlworld.com  Mon Jun 19 02:07:12 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 19 Jun 2000 07:07:12 +0100
Subject: tkinter: option_add() problem
References: <394BF3CF.450F776E@uniserve.com>
Message-ID: <cDi35.2869$_55.58179@news2-win.server.ntlworld.com>

Hi Bob,

try

root.option_add('*Font','Times 14 bold')

Richard

Bob van der Poel <bvdpoel at uniserve.com> wrote in message
news:394BF3CF.450F776E at uniserve.com...
>
> Using tkinter to modify the option database I CAN do things like:
>
> root.option_add( '*activeBackground', 'white')
>
> However,
>
> root.option_add( '*Font', "family times -weight bold  -size 14" )
>
> gives me the error message:
>
> TclError: expected integer but got "times"
>
> the equivalent??? tcl:
>
> option add *Font {-family times -weight bold  -size 14}
>
> works fine.
>
> Help!
>
>
> --
>    __
>   /  )      /         Bob van der Poel
>  /--<  ____/__        bvdpoel at uniserve.com
> /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
>




From dana at oz.net  Sat Jun 10 20:18:44 2000
From: dana at oz.net (dana_booth)
Date: 11 Jun 2000 00:18:44 GMT
Subject: MySQL module
Message-ID: <8hult4$t1a$0@216.39.141.21>

I've asked this before, but never did get a good response, so I hope that
one of you reading this might be able to help...

The MySQLmodule1.4 does not compile properly on any platform. I've tried
removing the '&'s on the three lines which puke during compile. On
UNIX-like's Python won't make install when having done that, on Windows,
it'll compile, but Python chokes when trying to import.

I've even tried using mSQL and PostgreSQL, on which the modules compile
fine, but I need to use MySQL, the only backend whos Python module seems to
be flawed in some way. :)

Anyway do any of you know of, or have made yourself, a module for MySQL?

-- 

-----
Dana Booth <dana at oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From pete at visionart.com  Tue Jun 20 18:28:41 2000
From: pete at visionart.com (Pete Shinners)
Date: Tue, 20 Jun 2000 15:28:41 -0700
Subject: remote execution
Message-ID: <394FF019.CC0A13BE@visionart.com>

i pair of related questions here.

i'm interested in finding a way to run remote commands
on different IRIX/Linux machines on our LAN. Usually
i will be running commands that run between 5 minutes
to several hours. (ie, not quick little routines)

of importance to me are the ability to monitor the
remote stderr/stdout as well as the ability to terminate
the remote process. oh, and getting the return exit code
will be needed.

there are two methods i can think of. the first one is
simply piping an "rsh" command reading the output. i
should be able to get the output and exit status from
the pipe, but i'm curious if i can send the terminate
process easily this way.
the other method would be 'cool!' but i'm not sure how
possible it is. it would simply be to run python code
on the remote machine.

now i need to be able to do this between a lot of
different machines, so something with minimal/no
installation of every machine is somewhat important.

it seems like i could make do using the "RSH" and other
unix style tools, but i wanted to make sure i wasn't
missing something great before i get into this.

actually the ability to run remote python code would
not be that beneficial, since 99.9% of the time it would
simply be something like os.system(command)

thanks!


From nickb at earth.ox.ac.uk  Thu Jun  1 09:50:48 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Thu, 01 Jun 2000 14:50:48 +0100
Subject: Interprocess comunication
References: <8F456BAC0lminutieidit@192.6.1.10> <3934f634.704981920@news.u-psud.fr> <8F45B7D7Clminutieidit@192.6.1.10>
Message-ID: <39366A38.262E0E07@earth.ox.ac.uk>

Luca wrote:

>
> >On Wed, 31 May 2000 10:04:21 +0200, lminuti at eid.it (Luca) wrote:
> >
> >This is generally an operating system dependent problem. Except if you
> >goes to distributed objects protocols like CORBA.
>
> I work with linux 2.2.x the portability of the code is not very important
> for me. I search an easy way to send a message from a program to another
> (both coded by me).
>

You might want to try the Medusa stuff  - asyncore/asynchat.py - in the
standard library.

These modules allow fully asynchronous comms between independant progs on
the same or different machines via  pc-name:port-number channels. - pretty
powerful stuff.

cheers
Nick/Oxford





From tim.lavoie at mts.net  Mon Jun 19 15:08:03 2000
From: tim.lavoie at mts.net (Tim Lavoie)
Date: Mon, 19 Jun 2000 19:08:03 GMT
Subject: SOCK_RAW with python socket library?
Message-ID: <slrn8ksreq.i2.tim.lavoie@goibniu.beyondtv.net>

Hello,

I've been tinkering with some socket programming lately, basically trying
out ideas I get from browsing W. Richard Stevens' "UNIX Network
Programming". Anyway, I'd like to play in Python too, so I'm using the
socket library functions. When I try to use SOCK_RAW though, I get:

   socket.error: (93, 'Protocol not supported')

I understand the "operation not permitted" if I try it as a non-root user,
but why this? Is SOCK_RAW not supported yet? The docs mention that it isn't
generally useful, but it is quite specifically useful for say, ping.


  Thanks,
    Tim Lavoie

-- 
One doesn't have a sense of humor.  It has you.
                -- Larry Gelbart


From jkraska1 at san.rr.com  Fri Jun 23 02:34:21 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 23 Jun 2000 06:34:21 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <03bf34ab.667a04d2@usw-ex0105-040.remarq.com> <8itgs2$4f0$1@slb6.atl.mindspring.net>
Message-ID: <39530583.FBD1A556@san.rr.com>

> I don't think this is a big issue, in the end.

It's especially not, because you always have your older interpreter,
and so therfor select your own migration time. 



C/


From mhammond at skippinet.com.au  Fri Jun 16 23:03:39 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 17 Jun 2000 03:03:39 GMT
Subject: win32gui question - is my window the foreground window?
References: <39437A27.1020403@bigpond.com>
Message-ID: <fUB25.699$qz4.6155@news-server.bigpond.net.au>

"Linus Chang" <Linus_Chang1 at bigpond.com> wrote in message
news:39437A27.1020403 at bigpond.com...

> However, win32gui only gives me SetForegroundWindow, not Get...

The next win32gui has all these.  In fact, it appears from my CVS logs
that these are in win32ll-132.

Mark.





From jkraska1 at san.rr.com  Sun Jun  4 16:31:47 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 20:31:47 GMT
Subject: Um? *boggle* Silly question about listobject.c
References: <000e01bfcddf$d2e09d40$1d2d153f@tim> <tolot.09oq@utolot.toppoint.de>
Message-ID: <393ABD58.EA62E75D@san.rr.com>

> int main(void)
> {
>   printf("%d %d %d\n",sizeof(short),sizeof(int),INT_MAX);
>   return 0;
> }
> -- EOF --
> 
> (0;0)13.SRam:> spam
> 2 2 32767
> 
> (Ok, I admit using a switch to get short==int ;-)

:)

I'm so used to using long when I mean 32 bits, I'd actually
forgotten that int was 32 bits most of the time. Forgetting
simple things like this can make you feel realy dumb. :)-


C/


From thollowe at opentext.com  Wed Jun 14 15:34:26 2000
From: thollowe at opentext.com (Skip Hollowell)
Date: Wed, 14 Jun 2000 15:34:26 -0400
Subject: Piping output to a web browser...
Message-ID: <8i8moh0q8l@enews3.newsguy.com>

A while back I saw a thread about keeping a web coonection alive (too bad I
saw it 4 hours AFTER I came up with the same general plan).  But I am still
struggling with how I can pipe the output to the screen on a streaming
basis.  The code I have seems to cache it all up and pushes it to the screen
all at once once I return out of the function.  Any ideas?

    print "Content-type: text/html"
    print
    sys.stdout.flush()
    print "<TITLE> Hello, World!</TITLE>"
    sys.stdout.flush()
    print "<PRE>Hello, World!"
    sys.stdout.flush()
    os.system ('date > /tmp/z.z')
    # This following routine takes a while to run but will delete the
/tmp/z.z file when it is done.
    os.system ('PATH=' + mypath + ';'  + support.ottestindexsitefile +
fFlags + ' >/tmp/x.x 2>&1 & ')
    sys.stdout.flush()
    while 1:
        try:
            #s = os.stat ('/tmp/z.z')
            f = open ('/tmp/z.z')
        except IOError:
            print "Process Done."
            break
        print ('Processing...')
        f.close()
        os.system ('sleep 2;date')
        sys.stdout.flush()
    os.system ('rm /tmp/z.z')

    return 'ZZZ'





From trentm at activestate.com  Tue Jun 13 00:33:56 2000
From: trentm at activestate.com (Trent Mick)
Date: Mon, 12 Jun 2000 21:33:56 -0700
Subject: Newbie questions
In-Reply-To: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
Message-ID: <20000612213356.B12900@activestate.com>

On Mon, Jun 12, 2000 at 09:10:35PM -0700, jim roush wrote:
> I'm just learning Python.  The following code does not work.  It's
> supposed to calculate the circumference after the user enters the
> radius.  The program displays the prompt, but when I enter the radius
> and the press enter, nothing happens.
> 
> FYI, I'm using Python 1.5.2 on WinNT.
> 
> pi = 3.141592654
> radius = input("Enter is the radius: ")
> circ = 2 * pi * radius
> print circ
> 
> 

It works for me. Sorry I am no help here.


BTW,
>>> import math
>>> math.pi
3.1415926535897931


Trent

-- 
Trent Mick
trentm at activestate.com



From samwun at yahoo.com  Sat Jun 10 22:02:22 2000
From: samwun at yahoo.com (Sam Wun)
Date: Sun, 11 Jun 2000 12:02:22 +1000
Subject: python DBI?
Message-ID: <3942F32E.B4751F35@yahoo.com>

Where can I download python DBI for MYSQL?

Thanks
Sam.




From stephen at cerebralmaelstrom.com  Sat Jun 17 11:51:55 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 17 Jun 2000 15:51:55 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <v8N25.63502$WS3.627109@typhoon.we.rr.com>

    I discovered Python a bit over a year ago, and have been using it
relatively constantly in idle projects. Some are simple, some are not.
Currently i'm working on one that's not.

    Initially, the indentation thing struck me as weird. My main complaint
about it is not that it's difficult to maintain proper indentation, but that
if you're scrolled far down in nested 'if', for instance, it's sometimes
hard to see which indentation-level is for which 'if'... but then i'll just
split-my-screen and scroll up.

    I've never encountered any of your problems. Everything that you've
described, though, i've done with C++'s braces.. forgot one here, added one
here.. etc...etc... The difference is that in Python, it's actually EASY to
just scan your code and see something indented improperly.

    As for your first point.. isn't it just as easy to scan for '__debug__'?
Heck, make your editor hilight it specially. Then just do your:

if __debug__:
    print Whatever

--S

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
news:394ac7ea.609024015 at nntp.interaccess.com...
> After using Python for a while I've come to the conclusion that
> using whitespaces to determine the begining and ending of blocks
> is bad for three reasons.
>
> Before I list those three reasons let me make two points. People say
> that the indentation requirements don't matter because you generally
> indent the same way. That may be true, but do you think it is a good
> idea to make conventions requirements ( remember that conventions
> start flame wars )? The second is that modern editors with good
> indentation rules ( like emac's python mode ) make indentation based
> coding easy. The thing is that by doing this you are abrogating
> control flow to the editor. When it was the editor indenting code
> which was well defined then there was no problem. By having
> the code depend on indentation, and the editor do the indentation
> for you, you give the editor responsiblity in how the code will flow.
>
> Now on to my points.
> 1) The first is rather stylistic and may not appeal to most, but I
> have often found it useful. When debugging code I often put in
> code to help me figure out whats happening. I also avoid indenting
> such code so that it easy to remember where it is later when I want to
> take it out. An example ( assuming I use begin and end to denote
> blocks ):
>
> def Average():
> begin
>       sum=0
> print "number of numbers=",len(numbers) #illegal
> i=0
>       for n in  numbers:
>       begin
> print "n=", n #illegal
>             sum=sum+n
> print "sum=",sum #illegal
>  i=i+1                     #illegal
> if sum>MaxNumber*I:#illegal
>


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


>
> You can immediately see code that you used for debugging and remove
> it. However right now writing code that way makes it illegal.
>
> 2) You can easily insert a tab introducing a bug.  Consider the
> following code ( similar to a bug I recently fixed ):
>
> def FindInvoice(idnum):
>       for invoice in invoices:
>             if invoice.id=idnum:
>       return invoice
>             return None # should be indented one less tab.
>
> ....
>
> for lines in file.readline():
>      idnum=lines[:6]
>      invoice=FindInvoice(idnum)
>      if not invoice:
>               invoice=InvoiceClass(idnum)
> invoices.append(invoice)
> ....
> Initially the line in error was indented correctly, but someplace
> along the line the extra tab was introduced. A lot of effort then had
> to go in to finding this bug. With begin...end or braces this is less
> likely to happen.
>
> 3) Finally. Python is a language where it is almost easy to write
> extensible programs. To illistrate what I mean consider this example
> of a python program I recently wrote:
> There is a C++ project which involves many project files and workspace
> files ( the project files are roughly equivalent to makefiles, the
> workspace files primarily are used to couple projects and define
> dependencies ). Version 1 simply read all the project files and
> generated a call to compile each one with all different configuartions
> (debug , release, alpha, beta etc..) . Since this built everything
> even if it was already built, and since this had to be run several
> times ( sometimes compiles would fail because a project that the
> present project was dependent on wasn't compiled ), I made version
> two. Version 2 read the workspace file and organised the projects
> into workspaces. Version 3 passed the string to a second script so
> that the second script could figure out if this was something that
> should be built ( for example whether a standard Win32 build or a
> WinCE build should be done ). Version 4 tracked dependencies more
> carefully ( it did a topological sort and orgainsied the build
> according to it ). Since the core stuff should be built first, I had
> to find out a way to figure out which was core. I used a definition of
> the core stuff is the stuff with the longest chain of dependencies.
> That was Version 5.
>
> As you can see I started with a simple program and made it more
> complex as I went on.  That's what I mean by extendible. Python is
> almost good at it. Almost. But that damned code indentation dependence
> backs its ugly head. In the process of extending code, I have to move
> blocks of code around. Sometimes put it in a loop, sometimes add an if
> clause to the top. Sometimes extract code into a subroutine. That
> means that I have to reindent the block. So as I reindent I have to
> try to remember how the indenting went. Something I stink at.
> Emacs also doesn't reindent correctly. So I have to go through the
> code, and try to figure out how it works. This distracts from my
> rewrite ( this also applies to refactorings ). I would rather think
> about how new code is supposed to work rather than how old code
> is supposed to work.
>
> Anyway I hope this causes TPTB to think about how blocks should be
> handled in Pyhton 3000.
>
>
>
>
>




From jkraska1 at san.rr.com  Mon Jun 12 02:33:05 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 12 Jun 2000 06:33:05 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au>
Message-ID: <394484C0.A5E412D9@san.rr.com>

> My first impression is that its dynamic typing, uncluttered syntax and
> ample built-in types encourages more "natural" designs. It makes
> coding easier and more fun (which is important to me). Even as a
> newcomer, I feel that I'm using the language with more pleasure and
> confidence rather than carefully paying attention to all its
> idiosyncracies (C++) or forcing my design to fit the paradigm
> ("classical OO" / Eiffel). I find that Python translates thought into
> code without too much fuss, and virtually no linguistic cruft. But is
> there more to it than that?
> 
> What is it about Python that encourages the "component" style you
> described?

For one, treating each file/module as a distinct object is
componentizing in itself, but this is neither here nor there.
I find that you observations are closer to the most important
truth than not: python is syntactically lightweight, and yet
has rich expressive power.

Python is a simply wonderful prototyping environment, IMO.
Now only if it had an optimizing compiler...

C/


From torppa at polykoira.megabaud.fi  Tue Jun 20 06:35:12 2000
From: torppa at polykoira.megabaud.fi (Jarkko Torppa)
Date: 20 Jun 2000 10:35:12 GMT
Subject: converting hex number contained in a string to an integer
References: <8ing7d$o6r$1@nnrp2.deja.com>
Message-ID: <8inhd0$7s$1@news.kolumbus.fi>

In article <8ing7d$o6r$1 at nnrp2.deja.com>, ejfc at my-deja.com wrote:
>Hello all,
>
>I have a string containing a number. This is an
>hex number. How can I convert this to an integer?
...
>How can I do this? I am sure it is really easy,
>but I am just beginnig with
>Python.

>>> eval('0x32')
50
>>> import string
# Forces base to 16(hex)
>>> string.atoi('32',16)
50
# Guesses base from leading characters
>>> string.atoi('0x32',0)
50

-- 
 Jarkko Torppa                torppa at staff.megabaud.fi
  Megabaud Internet-palvelut


From wtanksle at dolphin.openprojects.net  Mon Jun  5 23:43:22 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 06 Jun 2000 03:43:22 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl>
Message-ID: <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net>

On 5 Jun 2000 12:01:04 GMT, Martijn Faassen wrote:
>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>> On 30 May 2000 10:59:24 GMT, Martijn Faassen wrote:
>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>[snip]
>>>> Your rhetorical questions answer themselves.  Why should a programmer
>>>> enforce these things?  Because it's stupid to NOT enforce them.  So
>>>> enforce them already!

>>>That's nice to say, but you don't agree with that yourself if you think
>>>Python's use of indentation is a good thing. It can help a lot if the
>>>language makes it impossible to be inconsistent. Unless you think the
>>>_only_ advantage of Python's indentation strategy is the lesser line 
>>>count.

>> This is a classical example of false dichotomy ;-).

>It's not.

It actually is.  You're claiming that I hold one of two opinions; I hold
neither one.

>> I liked Python's
>> indentation not because it enforced consistency, but rather because it
>> makes code look good.  In fact, it could be argued that Python does _not_
>> enforce consistency; the programmer is free to choose any indentation for
>> each block, so long as each block can be parsed (which does require
>> internal consistency).

>"""
>I liked Python's
>case sensitivity not because it enforced consistency, but rather because it
>makes code look good. In fact, it could be argued that Python does _not_
>enforce consistency; the programmer is free to choose any case-spelling for
>each variable name, as long as each variable name can be parsed.
>"""

But now you've converted my paragraph from a true statement to a false
one.  The programmer is not in fact able to choose any case-spelling for
each variable name; they must use the same one which was originally used,
unless they wish to create or access another variable.

>Consistency in style happens to one of the most important factors that
>makes code look good (on the syntactic level). Your indentation looks
>better in Python because Python complains loudly if you don't indent properly.
>That's not the most important thing though; you can indent properly in
>any language. It's just that everybody indents properly in Python, and
>more or less in the same way.

Again, Python code looks good not because consistent indentation is
enforced, but rather because indentation alone is used to draw a tree
structure.

Consider the following example of inconsistent indentation:

if something:
       do_it()
else:
 dont_do_it()

Python will happily accept this.  It'll even happily accept this for long
blocks, here it can be inconvenient.  This makes the code a bit harder to
read, but it'll still not bad -- because Python expresses a 2-dimensional
structure (the syntax tree) in a 2-dimensional form.

Most ugly code I've seen isn't ugly because it has poor indentation
(although I have seen some); most of it's ugly because of horrible naming,
or bad structuring, or ...

>> More recently I've come to realise that the reason Python indentation
>> looks good is that Python, like most modern languages, is a
>> tree-structured language, and so is best represented in two dimensions.
>> Most modern languages, however, attempt to represent themselves in one
>> dimension: an ASCII stream which ignores lines and whitespace.

>Well, why shouldn't a programmer enforce such indentation in any
>language?

Of course they should.  Lisp programmers get really mad if people post
non-standard formatted code on c.l.lisp.  EMACS enforces it when writing,
and angry programmers enforce it when reading ;-).

>Indentation looks good in any language, not just Python, so
>that's not really an argument for Python's indentation style.

Well, there you go -- my point exactly.  I was saying the same for case
(which Python enforces much more consistently than it enforces
indentation).

In fact, the nice thing isn't that Python enforces indentation consistency
(because it doesn't really, any more than the minimum): the nice thing is
that Python doesn't throw anything more in your way than you really need
to parse the tree.

>Besides, if you ignore case you ignore that words in Python are spelled
>consistently in case, which is *also* a property of lots of modern languages.
>Sometimes not enforced, but still a desired property of the programmer.

But in case-sensitive languages, words can be spelled *differently* in
case: NO is not the same as No or no.  If the language were
case-insensitive this would be impossible.

In fact, it's conceivable to have a case-insensitive language which prints
an error whenever an alternate case spelling is used.  I think that would
be silly, but it would seem to fit what you want FAR better than a
case-sensitive language ever could.

So if you wrote:

  no = 1
  No = 2 # error!
  print NO # error!

I don't like that -- but it proves that case insensitivity can be more
powerful in terms of enforcing style.  It also seems to indicate to me at
least that I don't want a language enforcing THAT much style, any more
than I'd want Python to require exactly one tab or two spaces or any other
fixed amount for indentation.  Thank you, I'll decide that on my own.

>> So I don't really care about languages enforcing consistency; if I need
>> it, I'll enforce it myself.

>A computer language would be worth less if it brought no consistency at
>all.

Granted.  But this doesn't seem to justify your slippery-slope treatment --
we don't need to enforce your amount of consistency just because no
consistency is bad.  We already enforce _some_.

>I want my language to enforce the amount of parameters I pass to
>a function, unless I overrule it myself. I want my language to consistently
>enforce namespaces. I want my language to language to be consistent in
>the way I can call functions (foo(1, 2) and not also foo 1, 2). I want
>my language to be consistent in the way I spell keywords (I don't
>want to see code with IF and ELIF all over the place while other modules
>use 'if' and 'elif'). I want the language to be consistent about rejecting
>things that make no direct sense ( 1 + "foo"..see also the discussion on
>1/2).

And I like polymorphism (which implies typechecking).  And a lot of other
things, all of which I hope to see.  I'm curious to see whether other
people like them, and I'll defend them -- but unlike most of the people
arguing about case sensitivity, an ENORMOUSLY minor issue, I won't
consider them prerequisites to using the new Python -- I'll use the new
Python if it feels like an improvement.

>According to your argument, you will just enforce this consistency
>yourself. The problem is that, *even* if everybody enforces self
>consistency, people will enforce *different* consistency and their
>coding style will vary over time. From a maintenance and communication
>point of view, this is bad. 

And for case sensitivity it _just doesn't matter_.  The only people who
are strongly helped or hurt by case sensitivity appear to be newbies, and
they're being hurt.

>> Take a lesson from the people complaining
>> about Python's indentation: one of the leading false accusations is that
>> it's inflexible, that it won't allow people to choose thier own
>> indentation.

>I don't see how this maps to the case sensitivity argument at all. 
>Python's indentation *is* less flexible in what code-layouts are
>allowed, and it *is* a silly accusation.

It may or may not be silly -- but it's _definitely_ false.  Python is
_not_ inflexible; you can lay your code out in any way you want, so long
as it's in the same tree shape as its semantic structure implies.

I don't think it's silly, either; I've complained enough about languages
which don't do this right, such as Make.

There are two obviously terrible errors: inconsistency and inflexibility.
There's no reason to assume that avoiding one will force everyone into the
other, and I'm not making that mistake about you.

>So I don't see why I should take a lesson from this complaint or those
>people. Besides, people who make this silly accusastion generally use
>languages which *are* case sensitive. So I don't see how this relates to
>anything.

Are you trying to miss the point?  I don't think it's working, because you
previously argued on the point.  I'll confess that you're trying hard,
though -- three consecutive sentances which miss the point in three new
ways.

>Martijn

-- 
-William "Billy" Tanksley


From aiglauer at iname.com  Thu Jun  8 15:26:57 2000
From: aiglauer at iname.com (Alexis Iglauer)
Date: Thu, 8 Jun 2000 21:26:57 +0200
Subject: How to prevent a web client from timing out
References: <8hgfn8$3uq$1@nnrp1.deja.com> <393BCA8B.FC062262@rubic.com> <8hh066$vor$10$1@news.t-online.com> <393C1660.6A22F08D@rubic.com>
Message-ID: <8horqc$ld5$16$1@news.t-online.com>

It takes a non-technical girlfriend to solve this :)  I simply write
something to the webpage every 60 secs or so (either &nbsp; or number of
iterations or whatever).  Works like a charm, too.

I might still create a calculation server though, although that will have to
wait until I have more time on my hands again.

Thanks for all the help
aei

"Jeff Bauer" <jbauer at rubic.com> wrote in message
news:393C1660.6A22F08D at rubic.com...
> Alexis Iglauer wrote:
> > The job is different every time, so a cron solution
> > wouldn't work.  The problem is that when the client
> > times out the server kills the python process doing
> > the number crunching
>
> I didn't suggest cron.  (Someone else may have.)
>
> You can basically have a separate process already
> running, waiting for a number-crunching request.
> It is independent of the client, so if the client
> times out, nothing happens to the long-running
> process.
>
> Assume the long-running process (LRP) is threaded,
> and it can report back the status of the operation
> when your client requests it.
>
> Scenario:
>
> 1. Client (cgi script) submits a number-crunching
> request to LRP.
>
> 2. LRP responds back to client, "I'm working
> on it.  Press HERE to check on status."
>
> 3. User presses HERE button and another cgi script
> calls the LRP, which replies, "Gimme a minute,
> will ya?  Jeez."
>
> 4. User presses HERE button again.  Cgi script
> calls the LRP which has now finished the calculation.
> It responds, "The answer is 42."
>
> > So how do I stop apache from killing the process?
>
> Because Apache didn't launch the process, Apache
> won't kill the process.  You do the LRP independent
> of Apache.
>
> > How well does xmlrpc work in the real world?
>
> I don't know about the "real" world, but I've
> created an artificial world where it works pretty
> well.
>
> But for the technique described above, you don't
> *have* to use xmlrpc. Any similar technique (e.g.
> DOPY, Fnorb, ILU) should also prove satisfactory.
>
> Jeff Bauer
> Rubicon Research
>




From clive at pisquaredoversix.force9.co.uk  Sun Jun 11 07:10:54 2000
From: clive at pisquaredoversix.force9.co.uk (Clive Tooth)
Date: Sun, 11 Jun 2000 12:10:54 +0100
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <8hqhhi$396$1@mail.pl.unisys.com> <8huc4t$gu7$1@slb2.atl.mindspring.net>
Message-ID: <394373BE.6E3C4ABB@pisquaredoversix.force9.co.uk>

Kent wrote:

> Clive Tooth <clive at pisquaredoversix.force9.co.uk> wrote in message
> news:8hqhhi$396$1 at mail.pl.unisys.com...
> > Robert Israel wrote in message <8houlu$j9s$1 at nntp.itservices.ubc.ca>...
> >
> > >[...]
> > >There are two different algorithms here, for two different problems.
> > >
> > >The BBP algorithm can find a hexadecimal (or more generally, base
> > >2^k) digit of pi, faster than any known algorithm that would find
> > >all the digits up to this one.  This is the exciting one.
> >
> > Although BBP do mention that "This algorithm is, by a factor of
> > log(log(log(n))), asymptotically slower than the fastest known algorithms
> > for generating the n-th digit by generating all of the first n digits of
> > log(2) or pi."
> 
> I'm not getting this. It's "faster than any known algorithm that would find
> all the digits up to this one", but it's "asymptotically slower than the
> fastest known algorithms
>  for generating the n-th digit by generating all of the first n digits of
>  log(2) or pi". How can it be both faster and asymptotically slower? Does
> this mean "faster for small enough values of n, but slower for large enough
> values"? If so, about how big are "small enough" and "large enough"?

David Bailey, Peter Borwein and Simon Plouffe talk about this half way
down page 8 of "On the Rapid Computation of Various Polylogarithmic
Constants" (or half way down page 670 of "Pi: A Source Book").

Their algorithm requires that arithmetic (in particular, multiplication)
be done with a precision of O(log(n)) digits. Calculating all the digits
of pi requires doing arithmetic with a precision of O(n) digits.
Depending on the hardware capabilities of the computer, and the value of
n, it is quite possible that multiplication with a precision of
O(log(n)) digits is just the "ordinary" hardware multiplication. In this
case the BBP algorithm will be faster than than any known algorithm that
would find all the digits up to n.

For huge values of n, where we might be using Strassen-Sch?nhage
multiplication, even to attain a precision of O(log(n)) digits, then
computing _all_ the digits would be (just) faster.

As to what is "small enough" and what is "large enough" this would
depend on the capabilities of the hardware and the details of the
programming of the algorithms. I would guess that n being "a few
million" would be small enough for any likely hardware and likely
programming.

Note that the _memory_ requirements for the two methods are vastly
different. About O(n) bits for all n digits, and about O(log(n)) bits
for just the nth digit.

-- 
 
Clive Tooth
http://www.pisquaredoversix.force9.co.uk/
End of document


From mhammond at skippinet.com.au  Fri Jun  9 03:10:43 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 09 Jun 2000 07:10:43 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <TL005.123$cQ.2301@news-server.bigpond.net.au>

"Jeff" <no_spam at umich.edu> wrote in message
news:8hp2ku$a25 at news.or.intel.com...

> If there are no answers to these questions, someone needs to work on
> pythonwin's "development" capabilities.

You can take your demands and stick them where the sun dont shine,
sunshine!





From robin at jessikat.co.uk  Mon Jun 26 04:26:28 2000
From: robin at jessikat.co.uk (Robin Becker)
Date: Mon, 26 Jun 2000 09:26:28 +0100
Subject: Python class tree viewer (with wxPython)
References: <3951AA42.13155B89@coolest.com>
 <yBGXTvAmEcU5Ew62@jessikat.demon.co.uk> <3956D541.4471726E@coolest.com>
Message-ID: <9CDkBCA0OxV5Ewje@jessikat.demon.co.uk>

In article <3956D541.4471726E at coolest.com>, Alexander Schonfeld
<alex at coolest.com> writes
>Which algorithm is that? :)

There are many; I remember one from Dr Dobbs last year, August I think.
They used a variant of the spring model. Try searching graph+drawing :)

>
>I wish I had payed more attention in alg analysis now.
>
>Alex.
>
>Robin Becker wrote:
>> 
>> In article <3951AA42.13155B89 at coolest.com>, Alexander Schonfeld
>> <alex at coolest.com> writes
>> >Hi,
>> >
>> >I wrote (I mean hacked a demo...for wxOGL) to show class trees of bunches of
>> >python files.  You can then move them around and take screen shots and bask 
>in
>> >the glory that is your python classes.
>> >
>> >To use:
>> >       python pyinherit.py d:\cvs\Zope2\lib\python\Products\LoginManager
>> >       python pyinherit.py [specific .py files and dirs... whatever you want]
>> >Dirs are recursively traversed with 'walk'... looking for py files.
>> >etc...
>> >
>> >It requires wxPython (a gui lib) from:
>> >       http://wxpython.org/
>> >
>> >Please expand on this if it looks interesting, and email me a copy.
>> >
>> >Thanks,
>> >
>> >Alex.
>> >
>> ...
>> Wow! Now all we need is to apply that algorithm which tries to draw the
>> diagram decently in planar form.
>> --
>> Robin Becker
>
>--
>--
>--> Software ---> http://Coolest.com
>-----> Art -----> http://Sk8r.net
>-------------------------------------

-- 
Robin Becker


From moshez at math.huji.ac.il  Sat Jun  3 01:46:19 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 3 Jun 2000 08:46:19 +0300 (IDT)
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
In-Reply-To: <8h6pa1$ecq$2@newshost.accu.uu.nl>
Message-ID: <Pine.GSO.4.10.10006030844560.14749-100000@sundial>

On 1 Jun 2000, Martijn Faassen wrote:

> This again proves my Python-Dutch connection conspiracy theory: all
> Python programmers really also speak Dutch.
> 
> First an individual with a german-sounding name posting from Norway
> posts Dutch, and then someone with an eastern european sounding name
> posting from the UK understands it.

Martijn, do you happen to know how much time do I have to learn dutch
before....ummmm....I can sleep more easily?

or-is-haskell-enough-ly y'rs, Z.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From fdrake at beopen.com  Mon Jun  5 11:39:26 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Mon, 5 Jun 2000 11:39:26 -0400 (EDT)
Subject: Bad Math Module Notation re:  atan2
In-Reply-To: <pcovgzstc9b.fsf@math.ntnu.no>
References: <010620001459577031%pecora@anvil.nrl.navy.mil>
	<pcovgzstc9b.fsf@math.ntnu.no>
Message-ID: <14651.51630.701544.672038@cj42289-a.reston1.va.home.com>

Harald Hanche-Olsen writes:
 > Yes.  The documentation (meaning the library reference) is wrong, but
 > the implementation is right:

  Ok, I've taken a look at the current version of the documentation:

	http://www.python.org/doc/current/lib/module-math.html

This appears to be correct judging from the content of this
discussion.  Perhaps you're using an older version of the
documentation?  The current documentation was released 22 March 2000.
You can read or download the current version at:

	http://www.python.org/doc/current/

The HTML version of the package can be used to update documentation
that shipped with the Windows installer.


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From al_amerin at yahoo.com  Tue Jun 13 23:53:58 2000
From: al_amerin at yahoo.com (Al-Amerrho H. Amerin)
Date: Tue, 13 Jun 2000 22:53:58 -0500
Subject: stringobject question
Message-ID: <394701D6.37E6F560@yahoo.com>

Is this a bug or a feature?:

Python 1.6a2 (#4, Jun  9 2000, 02:22:41)  [GCC 2.95.2 19991024
(release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)

>>> ''.isdigit()
1
>>> ''.isspace()
1


Shouldn't both string methods return 0 when the string is empty?



From un4e at rzstud1.rz.uni-karlsruhe.de  Wed Jun 21 04:59:16 2000
From: un4e at rzstud1.rz.uni-karlsruhe.de (Bjoern Giesler)
Date: 21 Jun 2000 08:59:16 GMT
Subject: Strange Tkinter trouble
References: <8inqbe$rgu$1@news.rz.uni-karlsruhe.de> <8iokrd$r7d$1@nnrp1.deja.com>
Message-ID: <8iq054$gdr$1@news.rz.uni-karlsruhe.de>

Hi,
gregholmes at my-deja.com wrote:
: I'm not sure if this is the problem,

No. :-(

: from Tkinter import *
: root=Tk()

This is the line that doesn't work. It breaks in the call to Tk().

Regards,
		--Bj?rn


From JoeSmith at bogusaddress.com  Fri Jun  9 18:01:17 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Fri, 09 Jun 2000 22:01:17 GMT
Subject: ActiveX scripting - does it work?
Message-ID: <NOd05.12902$e5.188416@news1.frmt1.sfba.home.com>

I have the following file hello.wsf:
<package>
<job id="testp">
<?job debug="true"?>
<script language="python">
import sys
print 'hello world'
print 'line 2'
</script>
</job>
</package>

It prints out " hello world" (notice beginning space).  It does not print
out "line 2".
Then cscript crashes and brings up the "Application Error" dialog box
because it memory at 0x00000024 was referenced.
Then I see a VC++ run time error dialog box "abnormal program termination".
Then I see "Fatal Python error: PyEval_ReleaseThread: wrong thread state" in
the console window.

I have no problem getting python to execute in ASP pages.  Is it possible to
use python in one of these wsf files?  If so, are there any examples?





From aahz at netcom.com  Fri Jun  2 19:27:35 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jun 2000 23:27:35 GMT
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de> <3938090C.9FB5D057@stud.ntnu.no> <8h9682$v4h$1@nntp9.atl.mindspring.net> <39385BAC.771293BE@stud.ntnu.no>
Message-ID: <8h9ft7$6is$1@slb6.atl.mindspring.net>

In article <39385BAC.771293BE at stud.ntnu.no>,
Peter Schneider-Kamp  <peter at schneider-kamp.de> wrote:
>Aahz Maruch wrote:
>> 
>> Let's be technically precise here: an exception is raised and caught by
>> the "for" handler; the for loop is then broken (precisely as if the
>> break statement was executed).  Like any other caught exception, the
>> exception is thrown away (not "error is cleared").
>
>Okay, then we will talk source. From dist/src/Python/ceval.c of the
>current CVS tree (haven't checked, but should be similar with 1.5.2):
>
>static PyObject *
>loop_subscript(v, w)
>	PyObject *v, *w;
>{
>	PySequenceMethods *sq = v->ob_type->tp_as_sequence;
>	int i;
>	if (sq == NULL || sq->sq_item == NULL) {
>		PyErr_SetString(PyExc_TypeError, "loop over non-sequence");
>		return NULL;
>	}
>	i = PyInt_AsLong(w);
>	v = (*sq->sq_item)(v, i);
>	if (v)
>            return v;
>	if (PyErr_ExceptionMatches(PyExc_IndexError))
>		PyErr_Clear();
>	return NULL;
>}
>
>Note the line PyErr_Clear() which is a call into
>dist/src/Python/errors.c
>where it is defined as PyErr_Restore(NULL,NULL,NULL) which just puts
>NULL
>values into the type, value and traceback fields. In my opinion this
>(besides the name) qualifies as "error is cleared".
>
>am-I-missing-something-obvious-ly y'rs

Yes and no.  For reference, here's some code I posted earlier that is
precisely equivalent to a for loop:

i = 0
while 1:
  try:
    object.__getitem__(i)
  except IndexError:
    break
  i = i + 1

Now, in Python terms (not C), what would you say happens to the
exception at the break?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From just at letterror.com  Mon Jun 19 11:22:00 2000
From: just at letterror.com (Just van Rossum)
Date: Mon, 19 Jun 2000 16:22:00 +0100
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <wvn35.2720$qz4.21769@news-server.bigpond.net.au>
References:  <FVN15.3523$227.83704@nnrp1.uunet.ca><8i8bi2$1s8$1@news.wrc.xerox.com><slr
 n8kg937.328.hzhu@rocket.knowledgetrack.com><ud7ljln92.fsf@ctwd0143.fitlinx
 x.com><slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com><l03102800b570d8042c72
 @[193.78.237.143]> <l03102800b57372d82b38@[193.78.237.140]>
Message-ID: <l03102800b573e58c0e58@[193.78.237.139]>

I wrote:
[something about spaces for indentation with proportional fonts.]

Mark Hammond wrote:
>I agree 100% with your sentiments!
>
>But Im not sure I understand the above.

Oops, I snipped it...

>If you indent with leading spaces, then presumably all leading spaces will
>be rendered in the same font.  So even for proportional fonts, the code
>should align correctly with spaces or tabs.

Yeah, it'll be correct, but my point is that the amount of (visual)
indentation then depends on the font instead of a user preference. To me
this makes the mere idea of having to choose how many spaces to use for
indentation complete nonsense. (In my MacPython IDE the user can choose
between the spaces and pixels as a visual size unit when setting the
indentation level...)

>For alignment of anything other than the first non-space character (eg,
>attempting to align a table, or a parameter list that spans
>multiple-lines, for example), then if you use tabs, you are back in the
>same position that started this thread - you are making some assumption
>about the location of the tab-stops, or an assumption about the width of
>the previous characters.

Yep.

>So I simply can not see a solution for ensuring _all_ code is visually
>indented correctly when using purely tabs with a proportional font.

True.

>Let me know if I am wrong - I use a proportional font myself, and added a
>quick toggle-between-fixed-and proportional command to Pythonwin purely to
>get around this problem - Im forced to switch to fixed-point and using
>spaces for non-leading indents...

That's often what I do, too... However, I never align multi-line arg lists
the way pymode (and I think IDLE/PythonWin) does it; I always indent
continued arg lines with two extra tabs. Not only because of the problem
you describe but also because I like it better that way. Most of the time
when the arguments don't fit comfortable on a single line, I don't split
them up by using a single line per argument, but rather have a bunch of
args per line. This split is rather arbitrary and to me aligning the Nth
argument with the first doesn't make a whole lot of sense to begin with.

I also try to avoid aligning comments like this:

	a = 12              # a comment
	yabbadabbadoo = 13  # another comment

by doing

	# a comment
	a = 12
	# another comment
	yabbadabbadoo = 13

Unless you comment every single line this is quite doable (and likable!).

Tables indeed remain a problem with proportional fonts...

Just





From jared at tgflinux.com  Tue Jun 13 18:34:17 2000
From: jared at tgflinux.com (Jared Lee Peterson)
Date: Tue, 13 Jun 2000 15:34:17 -0700
Subject: python with Postgresql using mxODBC
Message-ID: <3946B6E9.7519DC50@tgflinux.com>

I was wondering if anyone out there could point me to some info of how
to set up python to use postgresql through mxODBC.  Or if there is
another odbc module that works that would be great as well.  Thanks for
any help that anyone can offer

Jared



From neilh at scintilla.org  Thu Jun 29 20:34:25 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 30 Jun 2000 00:34:25 GMT
Subject: Bug in PythonWin??
References: <8jfsv4$g53$1@nnrp1.deja.com>
Message-ID: <lWR65.1686$Tb7.11974@news-server.bigpond.net.au>

> Is this a bug or a misconfiguration (in PythonWin or NT itself)?  Does
> anyone else out there manage to run this on a similar configuration
> with no problems?

   This is a bug that was tracked down after the release of 132. Mark knows
what caused this so there should be a fix in the next release.

the-beer-holder-on-the-keyboard-leaves-no-room-for-accents-ly y'rs

   Neil






From jkraska1 at san.rr.com  Mon Jun 12 23:34:29 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 13 Jun 2000 03:34:29 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu> <8i3kbi$3a0$1@slb1.atl.mindspring.net>
Message-ID: <3945AC6C.36BD396A@san.rr.com>

> >Then I move to C.  The process took about twice as long as the C++
> >attempt; not counting the time it took for me to write miscellaneous
> >libraries (stack, memory allocation), I noticed that I was spending far
> >more time hunting pointer bugs and syntax errors than writing new
> >code. oThis was the longest port; it took me about four times longer than
> >it did in Python.

Although, to be fair, any seasoned C programmer has his or her
own preferred set of utility functions for doing hashtables,
stacks, lists, and so forth. I have a set of macro-expansion
tools for that that are far more efficient than the standard
template library could ever pray to be (written by a nut who
would meticulously process to assembly and analyze the number
and nature of assembly instructions issued for the equivalent
C... he was really freaky :).

Fair point with the pointer bugs, though.

That's not something that can even happen in Python, really. :)-

C/


From gmcm at hypernet.com  Wed Jun 14 08:44:54 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Jun 2000 12:44:54 GMT
Subject: TCP packet size?
References: <CC59868A0A76D311B1F50090277B321C1F33BA@VRTMAIL> <3948AC51.98E5EAEC@rpgarchive.com> <8i7snf$r2m@newton.cc.rl.ac.uk>
Message-ID: <8F53571B2gmcmhypernetcom@199.171.54.194>

Richard Brodie wrote: 

>
>"chris" <chris at rpgarchive.com> wrote in message
>news:3948AC51.98E5EAEC at rpgarchive.com... 
>
>>The messages just aren't reassembled the way I thought TCP worked.
>>I'm hoping to find a way to received one message at a time in its
>>entirety, or determine the start and end of my messages
 
>Everyone programming to the TCP/sockets API gets burnt sooner or later
>by this. Curse three times, dance round the room widdershins, then put
>a data length field in your message header. Check Stevens or any other
>decent introduction to socket prograaming for details.

And use a buffer to reassemble the data into discrete messages, which seems 
to be the thing you want to avoid. Here's one pattern you can use (you need 
something similar on the write side, too):

def getmsg(self):
  if self.buff is None:
    raise IOError, "read past end of file"
  while 1:
    msglen = int(self.buff[:5])
    if len(self.buff) > msglen:
      msg = self.buff[:msglen]
      self.buff = self.buff[msglen:]
      return msg
    input = self.sock.recv(8192)
    if not input:
      # socket closed
      msg = self.buff
      self.buff = None
      return msg
    self.buff = self.buff + input

- Gordon


From jvickroy at sec.noaa.gov  Tue Jun 13 10:09:53 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Tue, 13 Jun 2000 08:09:53 -0600
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu>
Message-ID: <394640B1.A972B37C@sec.noaa.gov>

Would the Event class in the threading module be of use to you?

Michael Vanier wrote:

> Python gurus,
>
> Apologies if this has come up before...
>
> I wrote a little python script using the thread module the other day.  I
> wanted to be able to set a flag in the parent thread that would be read in
> the child thread.  So I set a flag variable (an integer in this case) in the
> parent, created the child thread, and then checked the value of the flag
> every time through a loop in the child thread to see if it had changed.  It
> turns out that the child thread apparently made a copy of the flag, because
> when it changed in the parent it didn't change in the child.
>
> To fix this, I changed the flag to be a list with one integer element.  Now,
> when I change the list element in the parent, the child sees it too.  I
> gather that what this means is that there is a fundamental distinction
> between atomic types like integers and reference types like lists or
> dictionaries.  However, since python treats everything as a reference this
> seems inconsistent to me.  So what I want to know is: what is the cause of
> this behavior, and what is the justification for it?  Or could it be a bug?
>
> FWIW this is on a RH Linux 6.1 system running python 1.6a2 with pthreads (I
> assume).
>
> TIA,
>
> Mike
>
> --------------------------------------------------------------
> Mike Vanier     mvanier at bbb.caltech.edu
> Department of Computation and Neural Systems, Caltech 216-76
> GNU/Linux: We can't lose; we're on a mission from God.



From cut_me_out at hotmail.com  Tue Jun  6 15:51:58 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 15:51:58 -0400
Subject: PyRun_SimpleString error handling
References: <393D062D.8413AC36@ncsa.uiuc.edu> <etdem6a4v8k.fsf@m56-129-16.mit.edu> <393D47C9.763628AF@ncsa.uiuc.edu>
Message-ID: <etdd7lu4mg1.fsf@m56-129-16.mit.edu>

Thanks.

Alex.


From moshez at math.huji.ac.il  Tue Jun  6 13:35:47 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 6 Jun 2000 20:35:47 +0300 (IDT)
Subject: PyOpinion: Does Python Programming Marginalize You?
In-Reply-To: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <Pine.GSO.4.10.10006062029570.1650-100000@sundial>

On Tue, 6 Jun 2000, Richard P. Muller wrote:

> And yet, I worry that using Python marginalizes one. Python is hard to
> beat for writing applications that run on a single computer. But
> that's yesterday's computing model. The future of computing, heck, the
> present of computing, is in writing programs that run through a web
> browser.

Ummmm....it seems the more high-end web applications are more server-based
then client-based. And in that area Python's got one *huge* killer app --
Zope. There is only one (big) problem with Zope -- it's woefully
undocumented. One thing that would boost Python acceptance almost
overnight would be better documentation for Zope. I hear Zope developers
are working on it, but I don't know the current status.

JPython is yet another killer application, at least in principle, for some
of the server-side scripting. The big problem here seems to be maturity --
and that can only improve with age.

In short, my view of the future is not as bleak as yours -- I see
companies that crop up and stand by Python, I see Python mentioned more
and more in developer-oriented web-sites, etc. Just last week I gave two
Python lectures here in Israel.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From egbert at bork.demon.nl  Tue Jun 27 04:51:48 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Tue, 27 Jun 2000 10:51:48 +0200
Subject: The automagic of Tkinter
In-Reply-To: <39580EA4.6F818D5C@sitegnome.com>; from Fiona Czuczman on Tue, Jun 27, 2000 at 12:17:08PM +1000
References: <20000626135336.A8180@bork> <8j7p5e$qia$1@nnrp2.deja.com> <39580EA4.6F818D5C@sitegnome.com>
Message-ID: <20000627105147.A2664@bork>

Hi Richard,
Thanks for your illuminating replies. Actually I missed the first one,
so Fiona could beat me in translating Knop.

This the advantage of speaking a minority language:
if Guido decides to make Python case insensitive, 
and button=Button() doesn't work anymore, 
you can use Knop=Button()

The wonderful (I mean very user friendly) pipermail archive 
recovered your mail.
egbert
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From fellowsd at cs.man.ac.uk  Thu Jun 22 11:35:56 2000
From: fellowsd at cs.man.ac.uk (Donal K. Fellows)
Date: 22 Jun 2000 15:35:56 GMT
Subject: Regex engines
References: <394c2b15.634411296@nntp.interaccess.com> <8igpab$7tf$1@slb2.atl.mindspring.net> <3950B2BE.AF756B9F@hursley.ibm.com> <3dd7laes4x.fsf_-_@mems-exchange.org>
Message-ID: <8itbos$plp$1@m1.cs.man.ac.uk>

In article <3dd7laes4x.fsf_-_ at mems-exchange.org>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
> * Tcl uses a Unicode-aware engine written by Henry Spencer, but it's
> not available as a standalone library.  At the Ottawa Linux conference
> last July I asked about releasing it, but have seen no sign of it
> almost a year later.

There shouldn't be a problem with releasing it by just ripping it
straight out of the Tcl sources, as it is not tightly bound to the
rest of Tcl.  Indeed, apart from a bit of code relating to dynamic
string management and some code relating to unicode-aware character
management, Henry's RE code is completely separate from Tcl (well, it
is in Tcl8.3.0 which is the last version I've got a handy source tree
to grep through for at the moment... :^)

Donal.
-- 
Donal K. Fellows    http://www.cs.man.ac.uk/~fellowsd/    fellowsd at cs.man.ac.uk
-- I may seem more arrogant, but I think that's just because you didn't
   realize how arrogant I was before.  :^)
                                -- Jeffrey Hobbs <jeffrey.hobbs at scriptics.com>


From aahz at netcom.com  Wed Jun 14 11:34:29 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 15:34:29 GMT
Subject: tuple question
References: <3947255E.7CF77923@softhome.net>
Message-ID: <8i88m5$si3$1@slb7.atl.mindspring.net>

In article <3947255E.7CF77923 at softhome.net>,
etoffi  <etoffi at bigfoot.com> wrote:
>
>admittedly this is a very [insert descriptive word here] question
>to ask, but what is a tuple and why is it used (as opposed to an
>array, etc). 

>From my POV, the primary purpose of tuples is for dictionaries: they
allow you to do things like

foo = {}
foo[(1,2)] = 3

This essentially permits the use of multi-column primary keys (to use
database terminology).  In addition, as Stephen Hansen pointed out,
tuples are a fair bit more efficient than lists.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From rcc at nospamthanks_jennic.com  Fri Jun  9 07:18:51 2000
From: rcc at nospamthanks_jennic.com (Robert Cragie)
Date: Fri, 09 Jun 2000 11:18:51 GMT
Subject: Definite Newbie Question....
References: <q_z%4.828$223.8755@news1.tor.primus.ca>
Message-ID: <newscache$4wwvvf$znn$1@jenpc07.jennic.com>

If you want to run it in the IDLE IDE, type the following at the interpreter
prompt:

import <saved file name>

where <saved file name> is, er, the name of the file you saved, *without*
the .py extension. Note that if you edit the file and save it again, to run
it again from the interpreter prompt, you have to reload it:

reload(<saved file name>)

Or as Shae Erisson has already suggested (which I personally find easier):

python <saved file name>.py

at a command prompt.

Robin Porter <mogul at primus.ca> wrote in message
news:q_z%4.828$223.8755 at news1.tor.primus.ca...
|     How is it that you run code once you enter it?  I am using Python
1.5.2
| in Win 9x.  After inputing the code in the Python GUI, and saving it, I
have
| been trying to run it by pressing F5 or pressing edit (top menu) and then
| selecting RUN at the bottom.  Please tell me what I am doing wrong?
|     Thank you so much for your help.
|
| Robin
|
| Here is the code:
|
| TaxRate  = 0.23
| MinPay   = 300.0
| RegHours = 40.0
|
| TaxPayable = 0
| Overtime = 0
| OvertimeHours = 0.0
|
| HoursWorked = float(raw_input("Please enter the hours worked per week? "))
| PayRate = float(raw_input("Please enter the rate of pay? "))
|
| if HoursWorked > RegHours:
|     OvertimeHours = HoursWorked - RegHours
|     Overtime = (OvertimeHours) * (PayRate*1.5)
|
| GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime
|
| if GrossPay > MinPay:
|   TaxPayable = (GrossPay-MinPay)*TaxRate
|
| NetPay = (GrossPay-TaxPayable);
|
| print "The gross pay is %f" % GrossPay
| print "The net pay is %f" % NetPay
|
|
|
|




From hat at se-46.wpa.wtb.tue.nl  Thu Jun  8 07:06:31 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 8 Jun 2000 11:06:31 GMT
Subject: WxPython & MOTIF (Not GTK)
References: <8hlur3$8fg$1@nnrp1.deja.com> <k3y%4.1$RB1.1185@news>
Message-ID: <slrn8jv331.9ef.hat@se-46.wpa.wtb.tue.nl>

On Wed, 07 Jun 2000 20:15:12 GMT, Opus <rhicks at nospam.rma.edu> wrote:
>You would have to go to wxWindows.com or some such to see if there is a
>wrapper...

Actually, it is www.wxwindows.org

wxPython is also dioscussed there (although not in combination with
Motif afaik.) If you really want this, and it does not exist, then you
have to do it yourself, and create Python bindings for wxwindows with
Motif (and create wxMotifPython). If you are in luck, this may be a
largely automagic process by using SWIG (I have no personal experience
with SWIG though).


-- 
Albert
---
Look ma, windows without Windows !!


From hughett at mercur.uphs.upenn.edu  Thu Jun  1 16:48:36 2000
From: hughett at mercur.uphs.upenn.edu (Paul Hughett)
Date: 1 Jun 2000 20:48:36 GMT
Subject: Styling Temporary Code
References: <3936A9BB.FBEF9201@visionart.com>
Message-ID: <8h6i74$pu2$1@netnews.upenn.edu>

Pete Shinners <pete at visionart.com> wrote:
: I've been developing in C for several years and have
: really been enjoying python recently.

: one thing i do frequenctly in C is to place temporary
: "debug/doublecheck/sanity" type code while i'm developing
: it. i always put this code in with no indenting, so it's
: easy to spot, work with, and cleanup. it looks something
: like this...


: int myfunc(int nullargs)
: {
:     int newval = do_fancy_stuff(nullargs);
: cout << "newval = " << newval << endl;
:     return more_fancy(newval);
: }


In C I habitually add a comment /* DEBUG */ to such code; then I
can use the search command in the editor to find all instances quickly.
A possible translation into Python would be


: int myfunc(int nullargs)
: {
:     int newval = do_fancy_stuff(nullargs);
:     cout << "newval = " << newval << endl;  # DEBUG
:     return more_fancy(newval);
: }

for a single line, or something like

: int myfunc(int nullargs)
: {
:     int newval = do_fancy_stuff(nullargs);
      # DEBUG
:     cout << "newval = " << newval << endl;
      # /DEBUG
:     return more_fancy(newval);
: }

if you want to more explicitly mark the exact beginning and
end of the test code.

Paul Hughett


From thiele at muc.das-werk.de  Wed Jun  7 12:34:19 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 07 Jun 2000 18:34:19 +0200
Subject: color of listbox entries (Tkinter)
Message-ID: <393E798B.A5DF784E@muc.das-werk.de>

Hallo!

I want to show entries in a listbox with different (foreground)-color.
For example, the first entry is red, the second green, the third blue
and so on.


entrylist = ["one", "two", "three", "four"]
color = 0
for entry in entrylist:	
	if color==0: 
		self.Listbox["foreground"] = "#f00"	#red
	else: 
		self.Listbox["foreground"] = "#0f0"	#green
	self.Listbox.insert("end", entry)
	if color == 0: color = 1
	else: color = 0

In this example all odd numbers shall be red and all even numbers green.
But this doesn't work. (all entries are green)

Has anyboby an idea how this can be realized?
I want to have "clickable"-Widgets. -> Labels cannot be used.

Thanks Thomas


From robin at jessikat.demon.co.uk  Sun Jun 11 18:16:04 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 11 Jun 2000 23:16:04 +0100
Subject: Will logging system be useful for python programmers ?
References: <39437BAA.2DC4CE90@mail.ru> <39438DDA.4DD32FCF@stroeder.com>
Message-ID: <B9zbMKAk+AR5EwWi@jessikat.demon.co.uk>

In article <39438DDA.4DD32FCF at stroeder.com>, Michael Str?der
<michael at stroeder.com> writes
>"Igor E. Poteryaev" wrote:
>> 
>> Will logging system be useful for python programmers ?
>
>Yes.
>
>Best would be if it's added to the standard Python distribution in
>the future.
>
>Ciao, Michael.
medusa contains a log implementation. Zope expands on it as well.
-- 
Robin Becker


From pinard at iro.umontreal.ca  Wed Jun 14 08:05:56 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 14 Jun 2000 08:05:56 -0400
Subject: Which linux distros have Python by default?
In-Reply-To: Michael =?iso-8859-1?q?Str=F6der's?= message of "Wed, 14 Jun 2000 10:36:49 +0200"
References: <uKs15.437$i06.133733@ptah.visi.com>
	<39474421.E63F098B@stroeder.com>
Message-ID: <oqhfawxyaj.fsf@titan.progiciels-bpi.ca>

[Michael Str?der]

> S.u.S.E. comes with packages for Python, Tkinter and other handy
> Python modules.  Well, the person who installs it has to choose it in
> the install application called YaST (series d).

Yes, but you choose Python explicitly only for simpler profiles.
Python gets installed automatically in more sophisticated profiles.
If you tune a simple profile yourself (which is what I usually do), you
merely confirm once the installation of Python, when a dependency of any
package you add to the profile.  This is more and more likely, as Python
gets more often needed in packages.  In my last SuSE installation, a few
days ago, I did not have to explictly select neither Python nor `pygtk'.

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





From djmoore at uh.edu  Wed Jun 28 18:58:00 2000
From: djmoore at uh.edu (David Moore)
Date: Wed, 28 Jun 2000 17:58:00 -0500
Subject: win32ras.pyd import fails
Message-ID: <kn0lls06khej25cp2gbm8bv7p0812ao6pp@4ax.com>

I figured this one out, but I thought I'd save the
next [purblind fool] newbie some distress:

With Python 1.5.2 and PythonWin build 132 installed,
attempting to import win32ras fails with:
 "ImportError: DLL load failed: The specified
  module could not be found."

Well, doggon, python/win32 is in PythonPath, and
the win32ras.pyd is /right there/, I can see it,
and I can import win32process and win32security,
and it worked at home, so what's the dang deal here 
at the office?

AFter thrashing about for, well, way too long,
I thought to try the same thing under python.exe.

Ah. "The dynamic link library RASAPI32.dll could not
be found in the specified path."

Of course. I don't have RAS installed here at the 
office.  Fortunately, I just wanted it for study,
so I guess I can wait till I get home.

Onward through the fog....

-- 
Dave Moore == DJMoore at UH.EDu == I Speak For Me.
"...what we have here is not merely a one-time
lapse but rather a chronic silliness...."
                            -- Edward R. Tufte


From thomas at xs4all.net  Fri Jun  9 10:39:38 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 9 Jun 2000 16:39:38 +0200
Subject: PATCH: Augmented assignment
Message-ID: <20000609163937.D20700@xs4all.nl>

As promised, here is my patch to add augmented assignment (+=, -= and
family) to Python. It's still Work In Progress, and probably requires a
rewrite of some magnitude before Guido will accept it.

The patch can be found here:
http://www.xs4all.nl/~thomas/augmented_assign.diff

What it does:

It adds support for 'in-place' binary operands, using the same syntax as C,
Perl and many more languages. In short, the new syntax

x += 1

is the same as 

x = x + 1

for immutable types. For mutable types, it's more like

x.add(1)

if the mutable type defines it like that. For classes, it tries to call
'instance.__add_ad__', which should return the resulting object -- which
could be self, or a new object, to simulate mutablility and immutability,
respectively. If the class does not define __add_ad_, regular __add__ is
attempted, and if that fails, __radd__ on the right hand side of the
expression.

The patch implements these operands (and their overloaders):

 +=	(__add_ad__)
 -= 	(__sub_ad__)
 *= 	(__mul_ad__)
 /= 	(__div_ad__)
 %= 	(__mod_ad__)
 **= 	(__pow_ad__)
 |= 	(__or_ad__)
 &= 	(__and_ad__)
 ^= 	(__xor_ad__)
 >>= 	(__rshift_ad__)
 <<= 	(__lshift_ad__)

This patch is based on Michael Hudsons' patch that added augmented
assignment only to builtin-types (and only the syntax, not the in-place
modification), and comments from Michael, Guido and Tim. It's quite an
improvement over Michaels' patch in that it adds overloading, and
inplace-operator-bytecodes, so that the new code is actually 'thread safe'.
I'd like to think that this patch is also easier to understand, because
there is no need to manipulate the stack in strange ways to get the right
operands in the right spot for the next operation ;)

The 'ad' in __add_ad__ is what Guido proposed, and I have to say I dont
particularly like it. Making it '__iadd__' or perhaps '__inplace_add__'
seems more obvious, and more in keeping with '__radd__'. But perhaps Guido
meant something entirely different from this patch.

BUGS:
I have tested the patch, by running all the tests in the test suite (which
includes an aug_assignment test) and by running PySol with the new python,
and it hasn't found any killing bugs (at least, not in this patch ;)

However, I'm fairly certain I screwed up the INCREF/DECREF count in a couple
of places, so it's quite probable that this patch makes Python the Exxon
Valdez of bytecode interpreters.

Also, the builtin mutable types do not (and can not) actually do in-place
modification, currently; Python falls back to the normal 'add' operations,
so 

x = [1,2,3]
x += [1,2,3]

and 

x = [1,2,3]
x *= 2

both work exactly as expected, but they create new objects, instead of
modifying x. See the TODO list.


TODO:
- add in-place variants of binaryfunctions to the PyNumberMethods struct, so
  builtin and extention types can provide them

- supply in-place binary functions for the builtin types

- Add proper prototypes to the proper places to avoid warnings

- Add support for the new opcodes to Lib/dis.py

- Add support for the new Grammar to parsermodule.so

- eradicate bugs

- Make Grammar more readable by splitting 'exprlist' and ajust compile.c
  accordingly. (Wanted to do that right away, but couldn't be bothered with
  figuring out how to modify compile.c to understand the extra branch, at
  that moment ;)

- write JPython version (I haven't looked at this yet, but I doubt I know
  enough Java. Volunteers greatly appreciated ! :-)

- write documentation (See JPython ;)

- Add support for in-place operators to the std. library, where it makes
  sense.

PROGRAMMER NOTES:

To implement this, I've had to add support for 2-argument bytecodes. (It was
either that, or add 11 times as many opcodes (a total of 99, that is), or
create a PyObject only intended for passing an argument to a bytecode -- not
very nice. I didn't want to risk pushing an int onto the
object-pointer-stack by pretending it is a pointer.) 2-argument bytecodes
may come in handy in other situations, I do think, but I'm not saying this
is the Best Solution.

I've also added 9 new opcodes:

/* Opcodes for augmented assignment (+= and family) */
/* Slice/subscr opcodes take the specific operation as argument */
#define GETSET_SLICE    150     /* Augmented operation to perform on stack */
/* Also uses 151-153 */
#define GETSET_SUBSCR   155     /* "" */

#define HAVE_2ARGUMENT  156

/* vrbl-assignment take two arguments: operation to perform, and index into
   name list */
#define GETSET_NAME     156     /* arg1: operation, arg2: index in namelist */
#define GETSET_ATTR     157     /* "" */
#define GETSET_GLOBAL   158     /* "" */
#define GETSET_FAST     159     /* arg1: operation, arg2: local vrbl number */

The GETSET_ name is only the best I could come up with while writing it,
perhaps something else would be more obvious :P

The patch also adds these functions to the 'number protocol' API:

PyObject * PyNumber_InPlaceAdd ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceSubtract ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceMultiply ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceDivide ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceRemainder ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlacePower ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceLshift ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceRshift ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceAnd ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceXor ( PyObject *o1, PyObject *o2 );
PyObject * PyNumber_InPlaceOr ( PyObject *o1,  PyObject *o2 );

Which are InPlace versions of the normal functions.

QUESTIONS

Having written it this far, I'm getting a bit stuck. Perhaps a more
experienced python-patcher can give me a hint on these things:

- Is it safe to add new members to the PyNumberMethods struct ? How is
binary compatibility achieved ?

- Should the patch for 2-argument opcodes be sent in seperately ? Or should
I forget about 2-argument opcodes and use the object stack instead ?

- Is it likely this patch gets accepted as it is (but finished), or is there
something grossly wrong with it ? The purpose of this exercize, extreme fun
though it is, really is to get the patch accepted, so if I'm on the wrong
track, please let me know ;-P

The last two issues are not exactly urgent, as I wont submit the patch 'for
real' until I've added the first four entries in the TODO list at least
(unless someone influential suggests I do it anyway), but if I have to
change the entire approach, the sooner the better ;)

Unforuntately, while I was writing this mail, my Palm Vx *finally* arrived,
so I'm not very likely to spend the weekend on this patch ;-) But, for the
adventurous, the curious and the innovative, here is the URL to the patch,
v0.3. Thanx to Michael Hudson for providing the hints on how to proceed, and
Tim for explaining what Guido *really* meant ;-)

http://www.xs4all.nl/~thomas/augmented_assign.diff

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From eddthompson at ssi.parlorcity.com  Wed Jun  7 22:08:41 2000
From: eddthompson at ssi.parlorcity.com (Blu Dragoon)
Date: Thu, 8 Jun 2000 02:08:41 -0000
Subject: function and library listing
Message-ID: <393f4787$0$25327@wodc7nh0.news.uu.net>

Is there a site that list what functions and or libraries  are availible.
Not nesseccarily all of them just most of them.





From olczyk at interaccess.com  Wed Jun  7 17:14:50 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 07 Jun 2000 21:14:50 GMT
Subject: CPAN for python?
Message-ID: <393fbb15.100053125@nntp.interaccess.com>

Is there anything like CPAN for python?


From neilh at scintilla.org  Wed Jun 14 09:18:01 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 14 Jun 2000 13:18:01 GMT
Subject: Can't execute an external program !
References: <3947831B.73DEB063@informactis.com>
Message-ID: <dCL15.4673$cQ.38267@news-server.bigpond.net.au>

> I've tried this :
>
>    > f = open('./var/fiche-concessionnaire.rtf', 'w')
>   >  f.write(tpl.data)
>   >  f.close()
>    > commands.getoutput('c:\\windows\\write.exe
> ./var/fiche-concessionnaire.rtf')
>
> It always reply "Command Unknown" even when I try to use the DIR
> command.

# Module 'commands'
#
# Various tools for executing commands and looking at their output and
status.
#
# NB This only works (and is only relevant) for UNIX.

   It'd be nice if commands was fixed to work on Windows.

   Neil





From alex at magenta.com  Mon Jun 12 06:25:52 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 12 Jun 2000 12:25:52 +0200
Subject: Probably a dumb question
References: <394106F9.A58F87CE@micron.net> <8hr23k$bvk$1@nntp9.atl.mindspring.net>
Message-ID: <8i2e0e01rrn@news2.newsguy.com>

Aahz Maruch <aahz at netcom.com> wrote in message
news:8hr23k$bvk$1 at nntp9.atl.mindspring.net...
> In article <394106F9.A58F87CE at micron.net>,
> Peter Immarco  <android at micron.net> wrote:
> >
> >Hi,  still waiting for my two books on Python from Amazon.  I know
> >Python is a full programming language so it can be used on web servers
> >for server side processing, but what about browser side processing?  Can
> >I also use it to create an 'applet' that runs in a compatible browser?
> >I'm guessing not likely because I don't think IE4+ or Netscape
> >Communicator support Python but I'm hoping I'm wrong.
>
> There's a rumor that a future version of Mozilla will have Python
> support (supposedly they're building in something similar to ASP with
> its switchable scripting engines).  Other than that, no.

If Python is installed (as an Active Scripting compliant language) on
the client machine running the browser, it will be usable as the
scripting language (just like any other Active Scripting compliant
language would be) from Internet Explorer (4 and later, at least).

See Hammond and Robinson's "Python Programming on Win32", O'Reilly.

This is not very useful for the general case (since you don't know
whether the client machine does have Python installed, etc, etc),
but it can come in handy on an intranet (where you do have some
degree of control on what's installed client-side).


Alex





From smnordby at yahoo.com  Tue Jun 27 17:15:25 2000
From: smnordby at yahoo.com (sn)
Date: Tue, 27 Jun 2000 14:15:25 -0700
Subject: multiple replacements
References: <8ja7ok$6be$1@nnrp1.deja.com>
Message-ID: <3959196C.2FA738F5@yahoo.com>

How about using a dictionary and a loop:

replaceme = {'foo1': 'bar1', 'foo2': 'bar2', 'foo3': 'bar3'}
for key in replaceme.keys():
    srch = key
    rplc = replaceme[key]
    contents = string.replace(contents, srch, rplc)

-Steve-


siva1311 at my-deja.com wrote:

> Hi,
> I am a newbie Python programmer and was wondering the best way to make
> multiple replacements of text strings in a file. I have 10-15 strings to
> replace and right now I am doing this like:
>
>   source = open(source_file,'r')
>   contents = source.read()
>   source.close()
>
>   contents1 = string.replace(contents, 'foo1', 'bar1')
>   contents2 = string.replace(contents1, 'foo2', 'bar2')
>   ...
>   contents10 = string.replace(contents9, 'foo10', 'bar10')
>
>   dest = open(dest_file, 'w')
>   dest.write(contents10)
>   dest.close()
>
> I know this must be a terrible kludge.
>
> Using sed I can use a single statement to carry out the replacements:
>     sed "s/foo1/bar1/g
>          s/foo2/bar2/g
>          ...
>          s/foo10/bar10/g" source_file > dest_file
>
> so I assume there must be a much better method than I am currently using
> in my Python version.
>
> Any suggestions for improvement to the Python code are appreciated.
>
> -Brad
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From matthias.huening at univie.ac.at  Mon Jun 19 10:03:44 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 19 Jun 2000 16:03:44 +0200
Subject: standalone programs
References: <8iknh9$2a3t$1@walton.maths.tcd.ie>
Message-ID: <8il8vp$2t96$1@www.univie.ac.at>

You should try Gordon McMillan's "Installer". It's at:
http://starship.python.net/crew/gmcm/distribute.html

Matthias

--
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/dt/





From shapr at uab.edu  Thu Jun  8 08:06:07 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 08 Jun 2000 12:06:07 GMT
Subject: function and library listing
References: <393f4787$0$25327@wodc7nh0.news.uu.net> <slrn8jumb2.qlt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <393F8BFF.5C66480B@uab.edu>

Remco Gerlich wrote:
> 
> Blu Dragoon wrote in comp.lang.python:
> > Is there a site that list what functions and or libraries  are availible.
> > Not nesseccarily all of them just most of them.

If you want a listing of freely distributed functions and libraries that
don't come with the base Python install, check out parnassus:

http://www.vex.net/parnassus/
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From thomas at xs4all.net  Mon Jun 12 21:19:58 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 03:19:58 +0200
Subject: "import foo" and "from foo import bar"
In-Reply-To: <slrn8kaa4u.1r0.jajs@Pascual.es>; from jajs@NOSPAMretemail.es on Tue, Jun 13, 2000 at 12:52:07AM +0000
References: <slrn8kaa4u.1r0.jajs@Pascual.es>
Message-ID: <20000613031958.R20700@xs4all.nl>

On Tue, Jun 13, 2000 at 12:52:07AM +0000, Juanjo ?lvarez wrote:

> I have a question regarding the difference between those two, apart from 
> having to type the module name before the symbol in the
> first case.

> If I am only using one or two simbols of a module (example: string): 

> It's better to write the "from string import join" form or it's exactly
> the same (in memory use and performance) that "import string"?

There certainly is a difference, but it isn't very large. Both statements
load the 'string' module into memory, both exactly in the same way. After
they load it in, the module shows up in sys.modules. The rest depends on
the exact statement.

'import string' does something like this:

string = sys.modules['string']

Whereas 'from string import join' does something like this:

join = sys.modules['string'].join

If there is any performance difference, it's going to be in the use of
'join'. If you use 'string.join' everywhere, Python will have to first look
for 'string', which will be in the global namespace, and then search that
for 'join'. If you use 'join' everywhere (either by using 'from string
import join' or by doing an explicit 'join = string.join') Python only has
to search the global namespace.

Of course, if you're that keen on performance, you probably want to pass
'join' into the functions that use it most, so that it gets turned into a
fast local variable index instead of a slow global namespace lookup ;)

Bottom line: In performance it has no real impact. Use whichever you deem
most readable: having to append a function or variable with the module name
can really clutter your code. On the other hand, by loading functions and
variables into your own namespace, you might shadow builtin functions or
variables (or worse, your own !) and cause great confusion when trying to
find out where a function came from.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From neilh at scintilla.org  Thu Jun  8 20:05:47 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 09 Jun 2000 00:05:47 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <vxW%4.1319$iJ6.12966@news-server.bigpond.net.au>

> 1. How do you kill an infinite loop program without killing the pythonwin
> environment?

   You can try right-clicking on the little green python down in the lower
right hand side of the screen and then choosing "Break into running code".
This normally works, but sometimes it doesn't.

   Neil





From aahz at netcom.com  Fri Jun 16 16:02:53 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jun 2000 20:02:53 GMT
Subject: random # errors?
References: <8h5m0n$k1o$1@news2atm.raleigh.ibm.com> <8h6dor$10r$1@slb3.atl.mindspring.net> <8h8h2g$nia$1@nnrp1.deja.com>
Message-ID: <8ie15d$mis$1@nntp9.atl.mindspring.net>

In article <8h8h2g$nia$1 at nnrp1.deja.com>, Arint?  <jamarijr at my-deja.com> wrote:
>
>Here is the calling code.
>
>def biko(w,h):
>        len = w * h
>        bytes = len/8
>        cvt=0
>        ret = ""
>        gen = whrandom.whrandom()
>        for idx in range(bytes):
>                cvt = gen.randint(0,255)
>                ret = ret +chr(cvt)
>        return ret

First of all, if len>8000 at all frequently, the "ret=ret+chr()" call is
going to get somewhat costly, because you copy/create a new string each
time.

Secondly, I still can't figure out what the problem is, partly because I
don't know whether it's dying on the call to whrandom.whrandom() or the
call to gen.randint().

Side note: you should not use whrandom, you should use random.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From liam at gpu.srv.ualberta.ca  Sat Jun 10 00:43:27 2000
From: liam at gpu.srv.ualberta.ca (L Stewart)
Date: 10 Jun 2000 04:43:27 GMT
Subject: readlines() question
References: <Pine.GSO.4.10.10006092103340.28247-100000@sundial>
Message-ID: <8hsh1f$h0c$1@pulp.srv.ualberta.ca>

Thank you all for your answers! Now I have another one :)

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> [Aahz] 
> > Yes, it's closed.  The anonymous object created by open() gets
> > deallocated after the assignment; part of the cleanup is to silently
> > call close().

> While what Aahz said can't be said to be untrue, it's hard to call it
> true, either. That is what happens in the *current* *CPython*
> implementation. In no way does the language require it, and specifically,
> for JPython, it is not true.

Again just out of curiosity, what does JPython do? (CPython is the
standard python distribution or another one?)

But-that-was-two-questions-ly yours,

Liam

-- 
Liam Stewart
E-Mail: liam at ualberta.ca

"It is not enough to be busy, so are the ants. The question is: what are
we busy about?"
-Henry David Thoreau


From thomas at xs4all.nl  Tue Jun 27 03:50:31 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 27 Jun 2000 07:50:31 GMT
Subject: reload()'ing modules and how it affects references
References: <xjvn1k8grd3.fsf@gjallarhorn.ifi.uio.no>
Message-ID: <slrn8lgn5p.ke4.thomas@centurion.xs4all.nl>

On 26 Jun 2000 21:43:52 +0200, Igor V. Rafienko <igorr at ifi.uio.no> wrote:

>[the article is a little bit long -- sorry]

>I'm trying to understand how reload() works. "Learning Python" and
>Python's documentation warn against the dangers of combining 
>from ... import ... and reload(), so that is fine, I use 
>"import modulename".

The problem with 'from module import something' and 'reload()' is not the
magic word 'from': It's the fact that 'from module import something' is a
combination of 'import module' and 'something = module.something'.

The best way to look at this is probably by thinking of the reload()ed
module as a *different* module, which just happens to have the same name.

Lets see:

>import sys
>import fooclass
>
>ml = [ fooclass.Foo, fooclass.Foo() ]
>while 1:
>    print ml[0].Y, ";", ml[1].Y
>    next = sys.stdin.readline()
>    if not next: break
>    else:
>        exec next
>    print ml[0].Y, ";", ml[1].Y

># fooclass.Foo.Y is printed, runtime1.py import's fooclass
>$ ./runtime1.py 
>5 ; 5

You start runtime.py, which loads fooclass, and stores a reference to
fooclass.Foo *and* an instance of fooclass.Foo in a list. The instance of
fooclass.Foo keeps its own reference to fooclass.Foo, by the way, in the
.__class__ attribute.

>			# here, I've changed fooclass.Foo.Y to 6 in emacs
>reload( fooclass )
>5 ; 5
>5 ; 5

Right. You reload the module, but the list containing the class and the
instance point to the *old* value. You don't point to a name, in Python, you
point to an object, and the object hasn't changed -- the old name simply
points to a new object.

>			# now re-assign the element explicitly
>ml[0] = fooclass.Foo
>6 ; 5
>6 ; 5
>			# and the instance
>ml[1] = fooclass.Foo()
>6 ; 6
>6 ; 6

These work, of course, because you point the variables to the new objects by
hand.

>(Apparently it does not matter whether the class reference/class
>instance reference is stored in a variable or a list)

Nope. Nor does it matter wether it's done in a tuple, a dictionary, a
builtin type that holds references, an extention type someone made that
holds references, or a user-defined class (which can hold references as
well.)

A variable is simply a container for a single reference. 'something =
fooclass.something' simple takes the reference that is stored in
'fooclass.something', and assigns it to 'something'. It doesn't look at the
value, it doesn't keep tabs on 'fooclass.something' to see if it's changed.
A list is very similar, except that it can be a container for an arbitrary
number of references, and is itself stored in a variable.

>Now, for the questions:

>1) What is the most elegant way of assigning all references to a
>certain class to a new definition of this class after a reload? (I can
>keep a dictionary/list of all the references, and assign them
>explicitly right after the reload() call, but it just looks too ugly).

There isn't really one. What helps for functions is if you do the 'import' +
assignment inside functions, so that they always have the newest version.
This doesn't help for classes defined in fooclass, however. You can change
an instance's __class__ attribute to point to a new parent, but you'll have to
keep a list of the instances first ;-P

reload() isn't really intended to change the behaviour of whole programs in
mid-run. You can do it, but only if you consider all information in the
relevant module as 'volatile' -- you have to do a full lookup each time. And
that includes not using classes defined in that module.

>I suppose that it is impossible to make an instance of a class
>suddenly "become" an instance of the new definition of the same class,
>so that is fine.

Nothing is impossible in Python ! :-)

>>> class A:
...     pass
>>> class B:
...     pass
>>> a = A()
>>> a
<__main__.A instance at 80c9790>
>>> a.__class__
<class __main__.A at 80c97f0>
>>> a.__class__ = B
>>> a
<__main__.B instance at 80c9790>

(You knew you could do your little while 1: readline program above in the
interpreter instead, right ? Without having to code a small interpreter
using exec, and with the ability to peek at more things ? :)

The problem with above code is that it skips B's __init__ method, so it
might not be properly initialized.

>2) When I call a method in class Foo that looks something like this:

>def __repr__( self ):
>    print <In Foo: Foo.Y ==", Foo.Y, "; self.Y == ", self.Y, ">"

>after a reload the changes to Foo.Y affect Foo.Y but _not_ self.Y.
>That is, the output from the test similar to the one above looks like
>this (the function I call is __repr__):

>$ ./runtime1.py 
>6 ; 6 ; <Foo instance: self.x == 3, Foo.Y == 6, self.Y == 6>
>reload( fooclass )    
>	# now Foo.Y == 5 in the module file
>6 ; 6 ; <Foo instance: self.x == 3, Foo.Y == 5, self.Y == 6>
>6 ; 6 ; <Foo instance: self.x == 3, Foo.Y == 5, self.Y == 6>

>Why does it happen -- I thought that Foo.Y and self.Y were supposed to
>refer to the same object from within a class method? Am I missing
>something essential?

Yes :) Think of it in references. 'self.Y' searches itself for
a Y attribute, and if it can't find it, searches its own __class__
attribute, which points to a class object, for the attribute. But as we've
seen above, self.__class__ points to the *old* Foo.

Foo.Y, however, first searches for Foo. It isn't defined in the __repr__
function, so it's searched for in the global namespace, the fooclass module,
where Foo is defined. But this is the *new* Foo class, so it contains the
new value.

Think of reload creating a new module with stupidly the same name (silly
reload function, why does it do that ? Confusion galore !) and all will be
well ;-)

reload()'s-a-bitch-and-then-you-marry-one-ly y'rs,
	Thomas.


From samwun at yahoo.com  Sat Jun 10 23:40:30 2000
From: samwun at yahoo.com (Sam Wun)
Date: Sun, 11 Jun 2000 13:40:30 +1000
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21> <3942DC44.4A85EEA7@rubic.com>
Message-ID: <39430A2E.99CE3FC@yahoo.com>

I've tried to use command python build.py to build the module, but it
complains the _mysqlmodule.so is not found...

Do you know how to fix this problem?

Thnaks
Sam
Jeff Bauer wrote:

> dana_booth wrote:
> > Anyway do any of you know of, or have made yourself,
> > a module for MySQL?
>
> Use Andy Dustman's module:
>
>   http://dustman.net/andy/python/MySQLdb/
>
> Jeff Bauer
> Rubicon Research



From stevee at recombinant.demon.co.uk  Tue Jun 27 13:58:03 2000
From: stevee at recombinant.demon.co.uk (Stephen D Evans)
Date: Tue, 27 Jun 2000 18:58:03 +0100
Subject: float literal too large
References: <ulsntz2kra.fsf@runner.princeton.idaccr.org>
Message-ID: <3958EB2B.E42B7EF7@recombinant.demon.co.uk>

I found this one some time ago and reported it. Guido van Rossum and  Tim Peters
followed up on this one. It is still open as of 27 June 2000.

http://www.python.org/python-bugs/open?id=245;

print eval("float(1.0e-309)") # Ok == 0
print float("1.0e-309") # ValueError: float() literal too large: 1.0e-309

Apparently this is due to a difference between the C implementations of strtod()
and atof() on some platforms.

The quick fix I used was:

        # s is string
        try:
            value = float(s)
        except ValueError:
            import sys
            if sys.platform == 'win32' and sys.version[:5] == "1.5.2":
                # floating point values smaller than DBL_MIN
                # cause an exception
                # approx 2.2e-308 and smaller

                # according to Guido van Rossum this occurs on Solaris too.
                # the method that causes the exception used strtod()
                # whereas the following uses atof()
                value = eval("float(" + s + ")")
            else:
                # behaviour may be different on other platforms/versions
                raise

I did this so that I do not forget to change it when the Python version changes.


Stephen D Evans
stevee at recombinant.demon.co.uk


"Victor S. Miller" wrote:

> I just wrote a python program to read output produced by a Fortran
> program that somebody had written.  Much to my surprise I got the
> exception:
>
> float() literal too large.
>
> The literal in question was '0.23489798479873E-2325' (this was
> generated on a Cray computer).  I couldn't find anything in the
> documentation about this restriction.  Even more puzzling was the fact
> that eval() worked just fine (it gave the answer 0.0 which is
> perfectly acceptable to me, given that the platform on which I was
> running had less precision than the platform that generated it).  Now,
> on general grounds I would rather convert this using float, because I
> had a statement like
>
> a[i,j] = float(s)
>
> where a is a numerical python array.  So, is there some way around
> this?  How do I find out what the restrictions on literals are?
>
> I'm running:
>
> Python 1.5.2 (#4, Nov 23 1999, 14:57:03)  [GCC 2.7.2.2] on sunos5
> --
> Victor S. Miller     | " ... Meanwhile, those of us who can compute can hardly
> victor at idaccr.org    | be expected to keep writing papers saying 'I can do the
> CCR, Princeton, NJ   | following useless calculation in 2 seconds', and indeed
>     08540 USA        | what editor would publish them?"  -- Oliver Atkin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000627/bb552712/attachment.html>

From neilh at scintilla.org  Tue Jun 13 03:23:19 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 13 Jun 2000 07:23:19 GMT
Subject: Testing for a file in python?
References: <Pine.GSO.4.10.10006130850160.20303-100000@sundial>
Message-ID: <Hjl15.3536$cQ.29422@news-server.bigpond.net.au>

> So Perl is the only language that has it. I don't think awk has it,
> and I'm certain neither guile nor Tcl have it. You have a really strange
> definition of "most".
> ...
> BTW:
> You didn't mention Ruby (which I think does not have it), Rebol (ditto),
> and others.

   Or DCL, Rexx, AppleScript, or VBScript. None of these use -f.

   Not all scripting languages originated on *x.

   Neil





From kens at sightreader.com  Tue Jun 20 17:47:55 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Jun 2000 14:47:55 -0700
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <394F0C4D.3911FF90@commsecure.com.au>
Message-ID: <394FE68A.D3AB1FF8@sightreader.com>

If anyone has questions about the quality and quantity of technical support for
Open Source software, drag them by the ear to this newsgroup and show them some
examples of multiple high quality responses in under an hour, 24 hours a day, 365
days a year!

In many closed-source companies, technical support is a major expense, and that
expense must be kept under control.  As a result, tech support reps are often
underpaid trainees.  One of the major benefits of going Open Source is that you
get lots of high quality support engineers working for you for free!

Ray Loyzaga wrote:

> Thomas Weholt wrote:
> >
> > Hi,
> >
> > I want to convince alot of my co-workers that OpenSource software,
> > especially Python, is ready for big business. ( is it ?! )
> >
> > How do I sell an OpenSource solution like Python to hardheaded
> > closedsource-fanatics? I know they care about money, so that would be
> > my best angle. If anybody has any good examples of companies using
> > Python, or similar technology, with profit, that would be great.
>
> Having been involved in several Python based Online share trading systems for
> banks I have had quite a deal of experience in these sorts of "debates".
> The interesting thing that emerged under any serious questioning is
> how appalling their experiences have been with commercially bought
> and supported software. They all look towards the OpenSource software
> with great disdain, and argue that support is non-existent, however
> when you ask about software they have paid for, they frequently
> tell of documented "features" that they expected to use during the
> project that never worked. And this is over many years, they then
> address this by buying another package which purports to solve
> these issues. At this point when you mention that the great support
> that they should expect in commercial software doesn't seem to amount
> to much, they start making excuses for the manufacturer ... really
> weird!
>
> By way of example, we managed to use Python/Linux/MySQL with a very
> conservative international bank and after a year of 7x growth, no
> problems have emerged with the solution. The bank initially developed
> with MS SQLserver, but it never worked so we swapped it out to a MySQL
> system in the last few weeks of the project (since we were using Python,
> we had enough time up our sleeves to help their side of the
> project too!)
> >
> > To be more exact :
> >
> > My mission is to swap expensive, oversized databases solutions like MS
> > SQL 7.0 with PostgreSQL 7.0, crappy inconsistent script-languages like
> > VBScript with Python and MS SiteServer / IIS with Zope, Linux with
> > Windows  2000/ NT 4.0. I argue that the load on the system will never
> > be to great for PostgreSQL to handle as supposed to MS SQL-server,
> > VBScript, it sucks and they agree to that, Linux instead of Windows
> > will save lots of cash, so will the "reduced" cost of PostgreSQL
> > instead of MS SQL 7.0.
> >
> > The scenario above is normal in my experience, and I`d appreciate any
> > feedback from people with similar ideas. Related Case-studies would
> > also help alot.
> >
> > The thing is that these people don`t thrust OpenSource. They laugh at
> > script-languages in general, and the idea of sharing code is even more
> > stupid.
>
> An alternative is to get your Visa in order and move to a new employer
> that knows a good thing ... we are always looking ...
>
> They can laugh at Python all they like, I prefer it that way, but in
> a competitive environment we have been able to produce highly maintainable
> mission critical system in an order of magnitude less time with
> fewer errors in a very compressed development timeframe. I consider
> it to be a very serious competitive advantage, so caught in two minds
> as to whether educating companies such as these as to the merits of
> Python and OpenSource tools.
>
> >
> > The growth of OpenSource, Python, Linux etc. depends on how well it
> > can be a suitable substitution for ClosedSource-alternatives. A more
> > agressive form of propaganda, showing off successful solutions, WITH
> > benchmarks, results of stress-tests etc. is vital. Right now there are
> > alot of talk about how great Python are, but few actual test of how
> > well Python performes under pressure, and not just Python alone, but
> > in a more complex environment, like in Zope or using Python as
> > interface to a RDBMS.
> >
> > I need selling points for OpenSource, the idea and philosophy,
> > examples of successful technology and eh .. moral support. I feel like
> > I`m in the lions den right now. ( " Think of a happy place. Calm blue
> > ocean, calm blue ocean ..." )
> >
> > I truly love Python and would like to see it get the credit it
> > deserves.
> >
> > Best regards,
> > Thomas Weholt
>
> See www.hsbcinvestdirect.com.au and www.quicktrade.com.au for examples
> of full Python share trading systems, and a few more to follow soon ...
> Total lines of code including charting and full failover redundant servers
> (done at the application level) 3200, simple enough for even a bank to
> maintain.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000620/47c44d82/attachment.vcf>

From wware at world.std.com  Mon Jun 26 10:44:32 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 26 Jun 2000 14:44:32 GMT
Subject: Final garbage collection?
References: <962028703.1248416950@news.libertysurf.fr>
Message-ID: <FwrnM8.MMA@world.std.com>

Jerome Quelin (jerome.quelin at insalien.org) wrote:
> Is
> it that Python does not have a _final_ garbage-collection that can
> handle cycles and break them?

That's correct. Garbage collection in Python is accomplished by
reference counting. When a gets a pointer to b, b's reference count
is incremented, and vice versa. If you read the Extending and
Embedding Tutorial you can find out more about this. It's a simple
system but it means that circular references like these do not get
collected. For the time being, it's good Python programming practice
to avoid circular references in your code.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From jamarijr at my-deja.com  Fri Jun  2 10:41:23 2000
From: jamarijr at my-deja.com (Arinté)
Date: Fri, 02 Jun 2000 14:41:23 GMT
Subject: random # errors?
References: <8h5m0n$k1o$1@news2atm.raleigh.ibm.com> <8h6dor$10r$1@slb3.atl.mindspring.net>
Message-ID: <8h8h2g$nia$1@nnrp1.deja.com>


> What platform is this on?  What are lines 30-40 in Beer?
NT 4.0.

Here is the calling code.

def biko(w,h):
        len = w * h
        bytes = len/8
        cvt=0
        ret = ""
        gen = whrandom.whrandom()
        for idx in range(bytes):
                cvt = gen.randint(0,255)
                ret = ret +chr(cvt)
        return ret


Thanks

> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>


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


From dan at netrics.com  Mon Jun 19 15:02:43 2000
From: dan at netrics.com (Daniel Gindikin)
Date: Mon, 19 Jun 2000 15:02:43 -0400
Subject: command-line parser
Message-ID: <20000619150243.A13173@netrics.com>

I once wrote a command line parser, its at

http://members.home.com/gindikin/dev/python/cmdline

Check it out, it may give you some ideas.
-- 
Dan



From thiele at muc.das-werk.de  Tue Jun 13 11:01:45 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 13 Jun 2000 17:01:45 +0200
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net>
Message-ID: <39464CD8.38FB435B@muc.das-werk.de>

>
> Yup.  And that's precisely why Python looks better and better as your
> programs get larger.  Python allows you to concentrate on your problem
> rather than the language.  Python literally allows you to develop the
> code for your algorithm in realtime, rather than designing your
> algorithm and *then* writing the code to match your algorithm.
>

The only difference I see between developing algorythms in C++ and Python
is that python needs no time for compiling. But the advantage of C++ is that I
get
the most error messages during compilation (-> earlier!)

And I think the developing overhead (algorithms, class design) is larger than
translation into a special language.





From aahz at netcom.com  Thu Jun 29 11:01:42 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jun 2000 15:01:42 GMT
Subject: traceback string
References: <8jfnn5$1fpo$1@news2atm.raleigh.ibm.com>
Message-ID: <8jfocm$24n$1@nntp9.atl.mindspring.net>

In article <8jfnn5$1fpo$1 at news2atm.raleigh.ibm.com>,
Arint? <shouldbe at message.com> wrote:
>
>How do you print the traceback string from an exception
>try:
>   whatever()
>except:
>   print sys.exc_traceback
>
>gives me <traceback object at 351900>

Use the traceback module.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From skip at mojam.com  Thu Jun 29 10:54:00 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 29 Jun 2000 09:54:00 -0500 (CDT)
Subject: list.pop([i]) and list.remove(x) w/ for loops
In-Reply-To: <Pine.GSO.4.10.10006290852230.27161-100000@sundial>
References: <8jdfb6$ugh$1@netnews.upenn.edu>
	<Pine.GSO.4.10.10006290852230.27161-100000@sundial>
Message-ID: <14683.25352.255209.986017@beluga.mojam.com>

    >> I was getting some unexpected behavior when using list.remove(x),
    >> wondering if someone can tell me what's going on.

    Moshe> Short answer: never ever modify a list you're iterating on. Make
    Moshe> a copy first.

In situations where copying the list is impractical, you can iterate over
the list in reverse order:

    for i in range(len(olist)-1,-1,-1):
	if o.type == a:
	    del olist[i]

-- 
Skip Montanaro, skip at mojam.com, http://www.mojam.com/, http://www.musi-cal.com/
On Phil Jackson's ability to manage multiple Lakers superstars, Shaquille
O'Neal said: "He's got the rings.  I listen to the man with the rings."



From bvdpoel at uniserve.com  Fri Jun 30 16:50:55 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Fri, 30 Jun 2000 13:50:55 -0700
Subject: [Tkinter] Entry widget and textvariable option?
References: <962394390.813366342@news.libertysurf.fr>
Message-ID: <395D082F.97FC7CD7@uniserve.com>

Jerome Quelin wrote:
> 
> Hi all,
> 
> I can't manage to use the textvariable option in an Entry widget. And
> I can't find docs on this... And since we can't take references to
> variables in Python, I don't see how to use this feature?
> 
> When running the following code:
> 
> >>> from Tkinter import *
> >>> Entry(width=10, textvariable=spam).pack()
> 
> spam is to be defined, since Python interpolates it in the previous
> statement. And that is definitely _not_ what I want Python to do! I
> want the content of the Entry to go into the variable spam... In perl,
> I would use '\$spam' to take a reference to the variable, and in Tcl
> I'd use 'spam' to indicate a variable name, without interpolating
> it. But what should I do in Python? 'spam' interpolates the content of
> the spam variable, and we can't take variable references in Python...
> Is it that we can't use the textvariable option in Python/Tkinter?
> 
> Any help welcome.
> 
> Thanks,
> Jerome
> --
> jerome.quelin at insalien.org


This should be put into a FAQ...

The variable has to be a Tk type. So, do something like:

	spam=StringVar()

and update with

	spam.set("Monty")

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


From jrwren at ice.acs.oakland.edu  Fri Jun  2 17:12:29 2000
From: jrwren at ice.acs.oakland.edu (Jay R. Wren)
Date: Fri, 2 Jun 2000 17:12:29 -0400
Subject: A Simple question about a slice of a string.
Message-ID: <8hgefr$sok$1@news2.acs.oakland.edu>

I'm attempting to accomplish grabbing the last 6 characters of sys.argv[0]
so that I can compare it to another string.
If 'myprog' is called as /usr/local/bin/myprog.  It could be called from any
path, as long as the 'myprog' portion is there, certain things will
execute.(this is used as a symlink)

myprogname = sys.argv[0][ len(sys.argv[0])-10:len(sys.argv[0])

it works like this, but I think there is a better, more proper way of doing
it.  can someone let me know.  Thanks.

-Jay




From jkraska1 at san.rr.com  Wed Jun 21 01:08:43 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 21 Jun 2000 05:08:43 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com>
Message-ID: <39504E63.691C86FB@san.rr.com>

>     Initially, the indentation thing struck me as weird. My main complaint
> about it is not that it's difficult to maintain proper indentation, but that
> if you're scrolled far down in nested 'if', for instance, it's sometimes
> hard to see which indentation-level is for which 'if'... but then i'll just
> split-my-screen and scroll up.

It would be nice if an editor had a vertical guide, much as some
of the drawing programs do. You can click and drag the guide around,
leaving a vertical line drawn across your editor, allowing you to
move up and down in the file to see what matches that indentation
level.

C/


From debl.nospammm at world.std.com  Wed Jun  7 11:08:26 2000
From: debl.nospammm at world.std.com (David Lees)
Date: Wed, 07 Jun 2000 15:08:26 GMT
Subject: ANNOUNCEMENT: The Python Imaging Library, version 1.1 (june 7, 2000)
References: <014a01bfd089$d32b4af0$0500a8c0@secret.pythonware.com>
Message-ID: <393E6560.FB3F2B55@world.std.com>

Is there a precompiled version for Win95/98 users?

David Lees


PythonWare wrote:
> 
> yet another release from the labs:
> 
>     The Python Imaging Library (PIL) adds image processing capabilities
>     to your Python interpreter. This library supports many file formats,
>     and provides powerful image processing and graphics capabilities.
> 
>     Version 1.1 moves the Tkinter support to a separate binary module,
>     and fixes a couple of problems caused by incompatible changes in
>     Python 1.6.
> 
>     Get your copy of the source kit here:
> 
>         http://www.pythonware.com/downloads.htm#pil11
> 
>     For more information, including support options, see the product
>     site:
> 
>         http://www.pythonware.com/products/pil/
> 
> enjoy,
> the pil team
> 
>     "Secret Labs -- makers of fine pythonware since 1997."


From maxm at normik.dk  Fri Jun  9 03:24:59 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Fri, 9 Jun 2000 09:24:59 +0200 
Subject: SV: getting a URL
Message-ID: <7BD10B680501D411B9DF009027E06F32011ABE@EXCHANGE>

Fra: johnvert at my-deja.com [mailto:johnvert at my-deja.com]

Here's a small snippet i sometimes use.

it's not to safe though as there's no exception handling

########################################################

def HTTPGet(URL, page):
    h = httplib.HTTP(URL)
    h.putrequest('GET', page)
    h.putheader('Accept', 'text/html')
    h.putheader('Accept', 'text/plain')
    h.endheaders()
    errcode, errmsg, headers = h.getreply()
    #print errcode # Should be 200
    f = h.getfile()
    data = f.read() # Get the raw HTML
    f.close()
    return data

htmlPage = HTTPGet('www.python.org', '/doc/current/lib/module-urllib.html')

#####################################################

urllib is probably even simpler.



From yawyi at gmx.de  Sun Jun  4 08:49:31 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Sun, 4 Jun 2000 14:49:31 +0200
Subject: auto-debugging features?
References: <4.3.1.2.20000603191943.01621258@phear.dementian.com>
Message-ID: <393a5078$0$1608@personalnews.de.uu.net>

Steven D. Arnold <stevena at permanent.cc> wrote:
>I'd like a generic system-wide method of logging the activities of my 
>programs for debugging purposes.

[snip]

>Something that would be cool in such a debugging info utility:
>
>	- Notification on entry to a function.
>	- On entry to a function, list the parameters and values for the params.
>	- When data is changed, note the before and after values.
>	- When an if statement is executed, list the expressions that were 
>compared and the values of each.

Once apon a time I struggled with debugging and white-box
testing in an embedded interpreter. After a while of experimenting,
I came up with a little hack based on pythons sys.setprofile hook.
Once engaged, the module prints all function calls (with
arguments), returns and exceptions, plus optional timestamps.

While it was useful then, I admit I haven't used it myself ever
since. If you're still interested, you can find it in the bowels
of the Python FTP site:

ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/System/wbtrace.README
ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/System/wbtrace.py

Johannes



From m.faassen at vet.uu.nl  Wed Jun 14 07:12:27 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 14 Jun 2000 11:12:27 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <8i7par$l92$1@newshost.accu.uu.nl>

jin choung <jin.choung at gte.net> wrote:
> howdy all,

> i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
> tried to pick up python before moving on to c++.  however, i was trying to
> work my way through the book LEARNING PYTHON and while it says that there is
> no need to have prior experience in languages like c or c++, i really found
> that that was not true.

> anyone else feel this way?

I think that this would be more a problem of the book (Learning Python)
than a problem with the language. "Learning Python", while it's fine for
people who have significant experience with another language, really should've
gotten a different title, in my opinion. It's *not* good for people with
little or no experience with programming. I've heard both "The Quick
Python Book" and "Python in 24 hours" are much better in that respect.

"Learning Python" assumes too much that a newbie programmer really doesn't
know. I've tried to teach Python to someone with that book, but the book
was just too much for her. Too many concepts assumed that she didn't
know about. That said, it has worked for others.

> i'm currently working my way through a 24 hour c++ book and i think this is
> going to work for me.

Oh, you are so wrong. :) I mean, you may pick up some simple C++, but C++
is definitely a way more complicated language than Python. You just picked
up a book that works better for you than the Python book, but the language
is far far more complicated. It takes many people a year to get proficient
in C++; even experienced programmers. It rarely takes an experienced
programmer longer than a month to get proficient with Python, especially if
that programmer has experience with object oriented concepts.

> oh, well, will be back to contribute here in a couple
> of months... in the meantime, i'll just leech valuable nuggets of wisdom
> from you guys.

I do suggest you take a look at some other Python books first, and then
take a look at some other C++ books. C++ is really not an easy language
to pick up. And it's really not necessary to learn C or C++ before you
can learn Python -- the other way around is indeed an easier path to take.

Regards,

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


From richard_chamberlain at ntlworld.com  Fri Jun 23 02:22:55 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 23 Jun 2000 07:22:55 +0100
Subject: How to convert a string to hex?
References: <3952EC0D.D5CF9066@esec.com.au>
Message-ID: <wdD45.7813$_55.169378@news2-win.server.ntlworld.com>

Hi Sam,

import string
b = "6b584"
i=string.atoi(b,16)
print hex(i)

0x6b584

Richard


Sam Wun <swun at esec.com.au> wrote in message
news:3952EC0D.D5CF9066 at esec.com.au...
> Suppose here is:
>
> b = "6b584"
>
> I would like to b to hex. Does anyone know how to do that in python?
>
>
> Thanks
> Sam.
>
>




From trentm at activestate.com  Sun Jun 18 18:11:25 2000
From: trentm at activestate.com (Trent Mick)
Date: Sun, 18 Jun 2000 15:11:25 -0700
Subject: command-line parser
In-Reply-To: <kdgqkscjmqa1aqs78l6grsmkonfmjcpsb6@4ax.com>
References: <kdgqkscjmqa1aqs78l6grsmkonfmjcpsb6@4ax.com>
Message-ID: <20000618151125.A22384@activestate.com>

Cool. You know of the getopt module right? You could either just use *it* or
use it to simplify your implementation here. Some comments below.


On Sun, Jun 18, 2000 at 09:36:29PM +0000, Ben Wolfson wrote:
> """
> arglist.py
> 
> A utility for retrieving arguments passed via the command line.
> """
> 
> class Error(Exception):
>     pass
> 
>     
> class Arguments:
>     """
> Instantiate with tuples of acceptable options.  Tuples should have the form
> 
>     (short_form, long_form, accepts_value, default_value)
> 
> Either short_form or long_form can be None or a false value, if there is no
> short
> or long equivalent of the option, but not both (obviously).  If
> accepts_value is true
> and default_value is None, the instance will assume that a value is
> necessary, and
> raise Error if the option but no value is present.  Short_form and
> long_form should
> _not_ be passed with leading dashes.
> 
> __getattr__ is used to find out if an option is passed.  If the instance is
> instantiated as
> 
> >>> arg = arglist.Arguments(
>         ('h', 'help', None, None),
>         ('p', 'port', 1, 5555),
>         ('l', 'lalala', 1, None),
>         ('a', 'anotheropt', 1, 21)
>         )

I have often thought that getopt is nice but it would be nice to some kind of
OO approach where each "option" is like an object in that you can run its
"handle" method that will the processing/setup required for that options.

This could fit into your module by changing the above tuples into contructors
of an "Option" base class, or something like that

class OptionSpec:
	def __init__(short_form, long_form=None, accepts_value=0,\
		default_value=None, description=None):
		#...
	def handle():
		#...
	def help():
		"""print out standard formatted help for this option

		This could be helpful for creating automatic usage command line
		output."""
		#...

then 


>>> arg = arglist.Arguments(
        OptionSpec('h', 'help'),
        OptionSpec('p', 'port', 1, 5555),
        OptionSpec('l', 'lalala', 1),
        OptionSpec('a', 'anotheropt', 1, 21)
        )

A user could then also derived from the Option base class to have it do
special things.



Could


> 
> and the command line options are --port -l --anotheropt=haha
> 
> >>> arg.process(['--port', '-l:2', '--anotheropt=haha'])
> 
> then the following should result:
> 
> >>> arg.h
> >>> print arg.h
> None
> >>> arg.help
> >>> print arg.help
> None
> >>> arg.port
> 5555
> >>> arg.p
> 5555
> >>> arg.l
> '2'

Could maybe add type convertion, which would require type specification in
the Option contructor, so that the user could require that an integer or a
floating-poitn value be passed in.

<snip>



Cheers,
Trent

-- 
Trent Mick
trentm at activestate.com



From colinmeeks at home.com  Tue Jun 27 23:08:47 2000
From: colinmeeks at home.com (ColBoy)
Date: Wed, 28 Jun 2000 03:08:47 GMT
Subject: Using zipfile.py
Message-ID: <3%d65.680$ZI2.26352@news1.rdc1.on.wave.home.com>

Could anybody give me a dummies guide to using the zipfile.py module.  Just
a couple of examples of viewing a zips contents and extracting all of a zips
contents.  I know this is probably dead simple, but I'm still working my way
around Pythons syntax for stuff.  Many thanks in advance.

Regards

Colin




From Sara.Fowler at usa.alcatel.com  Thu Jun  8 09:23:49 2000
From: Sara.Fowler at usa.alcatel.com (Sara Fowler)
Date: Thu, 08 Jun 2000 08:23:49 -0500
Subject: Successful Solaris 2.6 Compile???
Message-ID: <393F9E65.80E9C383@usa.alcatel.com>

Has anyone successfully compiled the Python Interpreter on a Solaris
2.6?  If so, do you know where DIR is declared?

Thanks!

Sara
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Sara.Fowler.vcf
Type: text/x-vcard
Size: 330 bytes
Desc: Card for Sara Fowler
URL: <http://mail.python.org/pipermail/python-list/attachments/20000608/68fb5b12/attachment.vcf>

From claird at starbase.neosoft.com  Thu Jun 29 11:10:04 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 29 Jun 2000 10:10:04 -0500
Subject: Moving functionality between languages (was: Controlling an external program with Python)
References: <nz855.1583$iN5.265679@ptah.visi.com> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com>
Message-ID: <1542456BE0832BEE.8A959B400FAE5129.26A3D7DF9AA9B305@lp.airnews.net>

In article <395B32D1.B926FDB3 at hursley.ibm.com>,
Paul Duffin  <pduffin at hursley.ibm.com> wrote:
			.
			.
			.
>I agree which is why I suggested that rather than rewrite Expect in Python
>which is no more *orthogonal* than Expect in Tcl why not do as I suggested
>and work with Don Libes on seperating the pseudo tty control stuff into
>a seperate library which both Tcl and Python can use. Everyone would
>gain and noone would lose, the same could be done for Tk.
>
>      +-----------+    +----------+
>      | TclExpect |    | PyExpect |
>      +-----------+    +----------+
>
>       +------------------------+
>       |  Pseudo TTY management |
>       +------------------------+
			.
			.
			.
>This sort of arrangement is only possible when people who are interested in
>Expect functionality can transcend any language *issues*, by accepting that 
>the language is basically irrelevant apart from the effect in may have on
>the required functionality and work together.

Of course.  And I like to think of myself as a leader
in proclaiming this sort of language irrelevance.

I believe I've heard Don express surprise at Expect's
career.  He thought people would steal the ideas for
other (languages') libraries.  What he least expected
(sic), if I understand it correctly, was that the whole
world would be content to invoke Expect as the standalone
executable which we know utterly dominates its use.

How can we write new scenes for this play?  Alex consi-
stently looks for canonical decompositions, Paul says,
the code is already written and just needs a bit of
fitting, Jean-Claude Wippler scorns language issues as
purely epiphenomenal, I plead with the camps to play
nicer, ...  Is there one definite programming task we
can tackle to demonstrate what we're talking about?
MetaKit and Monoid are coming along at their own pace.
How much of a difference to Python will TkGS make?
Would Pythoneers like help plagiarizing Henry Spencer's
Unicode-aware RE engine?  Do we want good control of COM
through the network from Linux?  Should we get together
on something seriously trivial, like serial port manage-
ment or free-disk-space utilities or such?

If I were more current on biology, I'd probably draw
some conclusion from what works in transgenic
engineering.
-- 

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


From tismer at tismer.com  Sat Jun  3 12:27:16 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sat, 03 Jun 2000 18:27:16 +0200
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use>
Message-ID: <393931E4.4283BD7F@tismer.com>


Russell Turpin wrote:
> 
> Christian Tismer wrote:
> > Please note that there is an alternative to using doubly linked
> > lists ..
> 
> My miscommunication. When I said deque, I didn't mean linked
> lists, but just a vector that is allocated with spare space
> at both ends, at least whenever elements are added to or
> removed from the start. There's really no reason to copy or
> reallocate a vector, every time the first element is
> deleted or inserted.  I think vector is the clear choice
> over linked list as a default implementation; but I still
> want to be able to add and remove from both ends efficiently.

The first message in this thread starts with
"""
A C-native bidirectionally linked list (deque) was
tested versus the performance of native python lists
as a LIFO. Items were appended to the end of each
data structure, but always removed from the beginning.
"""

and it was clearly meant this way along all of the discussion.
I'm not riding dead horses when everybody but you got a
different idea of what we're talking about.
I was looking for a message where I could hook my
point to, and you'r was the one with the "deque wish" :-)

So, repeating my self:
I'm against a linked list deque implementation for Python.
But I'm plus 1 on doing it as an extension of lists.

ciao - chris

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



From kbelli at coe.neu.edu  Thu Jun  1 15:13:14 2000
From: kbelli at coe.neu.edu (Kimberly Belli)
Date: Thu, 1 Jun 2000 15:13:14 -0400
Subject: Styling Temporary Code
In-Reply-To: <3936A9BB.FBEF9201@visionart.com>
References: <3936A9BB.FBEF9201@visionart.com>
Message-ID: <Pine.SV4.4.10.10006011507030.2989-100000@Energy.coe.neu.edu>


On Thu, 1 Jun 2000, Pete Shinners wrote:

> I've been developing in C for several years and have
> really been enjoying python recently.
> 
> one thing i do frequenctly in C is to place temporary
> "debug/doublecheck/sanity" type code while i'm developing
> it. i always put this code in with no indenting, so it's
> easy to spot, work with, and cleanup. it looks something
> like this...

Why not use the method the librarians used to make us do so we knew where
to put the books back on the shelves?

Yeah, it might not be as pretty, but it's a suggestion ;)

 int myfunc(int nullargs)
 {
     int newval = do_fancy_stuff(nullargs);
#  
     cout << "newval = " << newval << endl;
# 
     return more_fancy(newval);
 }

...or with comments....

 int myfunc(int nullargs)
 {
     int newval = do_fancy_stuff(nullargs);
# I think maybe I have a clue
     cout << "newval = " << newval << endl;
# phew! back to stuff I know.
     return more_fancy(newval);
 }

Good luck finding something that works!

Kim.



From dbroadwell at mindspring.com  Sat Jun  3 03:22:25 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sat, 03 Jun 2000 03:22:25 -0400
Subject: file.seek misbehaving?
References: <203da3d8.c492c8fd@usw-ex0103-019.remarq.com>
Message-ID: <3938B231.4FA3148F@mindspring.com>

Run a virus scan?

Note:files on unix and NT are stored in differnt native formats.

What is the differential i their seeks?
ther is a 2 character difference in the line lengths of the sustems as
on unix, a new line is a CR/lf (carriage return/line feed) and on nt and
windows platforms is just a CR.

cheesecake wrote:
> 
> I've run into a problem that is perhaps not a Python-specific
> bug. I was processing some log files using Python, when I
> noticed some mysterious bugs in certain files. I tracked down
> the problem to the seek method misbehaving.
> 
> Some of the files I worked with were files created in NT or
> saved in NT using a text editor. The problem files were copied
> over ftp from a unix machine to the NT machine on which I work.
> None of the files that were saved in NT (ie, NT native files)
> had any problems.
> 
> It seems that seek doesn't work correctly because it's not clear
> how large the file really is. I copied all the contents of one
> bad file into a new file, and the new file was over 5KB bigger
> according to NT. The new file works just fine with the seek
> operations. But when I read() both files into memory, they're
> identical strings and have the content-length of the original
> file.
> 
> Could the problem be that the differences between the NT file
> system and the unix file system of the other computer are not
> properly handled because I'm running on NT, so the Python
> interpreter figures it should assume everything is NT?
> 
> Let me summarize the problem with some code:
> fd1 = open("somefile.txt", "r")  # somefile.txt is a unix file
> fd2 = open("someother.txt", "r") # someother.txt is NT
> lines1 = []
> lines2 = []
> lines1[0] = fd1.readline()
> lines2[0] = fd2.readline() # lines1[0] == lines2[0]
> pos1 = fd1.tell()
> pos2 = fd2.tell() # at this point pos1 == pos2, like it should be
> lines1[1] = fd1.readline()
> lines2[1] = fd2.readline() # lines1[1] == lines2[1]
> fd1.seek(pos1)
> fd2.seek(pos2) # pos1 == pos2 (still)
> lines1[2] = fd1.readline()
> lines2[2] = fd2.readline() # lines1[2] != lines2[2] !!!!?
> 
> What gives here? That's like having file.seek(file.tell())
> moving you somewhere else.
> 
> Anyway, I'd like to find out if anyone else has had such
> problems or knows what the solution is. I don't read this ng, so
> please send replies to my email address at townsend at beehive.de.
> 
> TD
> 
> * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
> The fastest and easiest way to search and participate in Usenet - Free!

-- 
 --
	Got CODE?
 --


From hzhu at knowledgetrack.com  Fri Jun 23 18:30:26 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 23 Jun 2000 22:30:26 GMT
Subject: Examples wanted by newbie
References: <39543021.52CECB68@wanadoo.nl>
Message-ID: <slrn8l7pdo.uk.hzhu@rocket.knowledgetrack.com>

On Fri, 23 Jun 2000 23:50:57 -0400, Max Heijndijk <cchq at wanadoo.nl> wrote:
>Hello, does anyone know where to find *simple* python example programs ?
>Tutorials never work for me without examples.
>I want to convert my large shell scripts to python.
>

If you want a quick introduction to features of python I always recommend
http://www.idi.ntnu.no/~mlh/python/instant.html which you can follow by copy
and paste. 

For shell scripts it really depends on what they do.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From joseph at src.no  Fri Jun  2 12:38:17 2000
From: joseph at src.no (Joseph Santaniello)
Date: Fri, 2 Jun 2000 18:38:17 +0200
Subject: Help with SSL in 1.6
Message-ID: <Pine.LNX.4.21.0006021832560.13804-100000@lade.src.no>

Hi,

I've just built OpenSSL and Python 1.6a for the express purpose of being
able to use urlopen or similar to grab some ssl web-pages. But I can't get
it to work. I get name errors when I try to use HTTPS or anything it seems
because the "if hasattr(socket, "ssl"):" condition never seems to be
true. And urlopen says https is an unknown protocol. What have I missed in
setting this up so that I can grab ssl pages?

Thanks,

Joe



From jmg at ecs.soton.ac.uk  Wed Jun  7 10:24:33 2000
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: Wed, 07 Jun 2000 15:24:33 +0100
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net>
Message-ID: <393E5B21.B5F5B3C6@ecs.soton.ac.uk>

Aahz Maruch wrote:

> Okay, okay, a joke's a joke, but if people are actually going to write
> full paragraphs in Dutch, it'd be nice if someone wrote translations for
> the small minority of us who don't know Dutch.

Brief summary thus far:

It all started with Peter Schneider-Kamp congratulating Guido on `taking the
first step towards divorce'.  This led Martijn to mention his conspiracy theory,
and, later, to wonder (in Dutch) why so many people on this NG appear to speak
the language.

Shae suggested that not only people who speak Dutch are interested in Python,
which engendered the reply that all Dutch speakers should be interested in
Python.

Konrad Hinsen then emphasized that we only *appear* to speak Dutch, and
hypothesized that we are all bots, written by a single Dutch programmer; or maybe
that there is only one person posting to this NG, under hundreds of
noms-de-plume, in order to create the impression that Python is popular . . . the
thrust of his argument was that there are numerous rational explanations, which
do not require the invocation of cosmic forces. Martijn, earlier, expressed his
chagrin (in English) at this type of explanation.

Konrad's misspelling of the Dutch word `meerdere' (numerous) making it sound much
more like a French expletive led to to an observation about said
slip-of-the-tongue.

That's about it . . . you haven't really missed much, except for the very first
felicitation about which I still haven't made up my mind: did he mean to write
what he wrote, or did the message get slightly mangled somewhere in the process .
. . come on Peter, I think we should be told: was it your joke, or your Belgian
friend's joke on you ?

Jacek



From Ed_Taekema at continuus.com  Thu Jun 22 23:27:52 2000
From: Ed_Taekema at continuus.com (Ed Taekema)
Date: Fri, 23 Jun 2000 03:27:52 GMT
Subject: Binary Informix Module for *nix???
Message-ID: <20000623.3275200@localhost.localdomain>

I am looking for a way to get a prebuilt python module that will
connect to a run time informix server (Informix Dynamic Server 2000
Version 9.20.UC1 and Informix 7 I believe) and need this for various
flavours of Unix & NT (solaris, linux, win32 etc).  All I've seen are
various packages which seem to require a full installation of informix
to build (ie esql etc) which I don't have.  There also appears to be a
prebuilt binary package for NT but no Unix.  Is anyone aware of any
solutions short of spending the $$ on a full informix licence?

Thanks for your help,

Ed Taekema
Taekema at sympatico.ca


From brlspam at sperience.com  Fri Jun  2 08:37:39 2000
From: brlspam at sperience.com (brlspam at sperience.com)
Date: 02 Jun 2000 08:37:39 -0400
Subject: Using python (or Tcl) on the web
References: <MPG.1365adcfa8470eef9896aa@news> <BC4B61746187C623.3FA0E7D58DF4EA7E.ECC1D97F577DEDE5@lp.airnews.net> <slrn8g28ah.cc.amphgobb@eccie.blackflag.net> <L3iZ4.154057$681.2782248@news-east.usenetserver.com> <8h4eeu$bpg$1@saltmine.radix.net>
Message-ID: <nm9k8g8l0mk.fsf@department-of-alchemy.mit.edu>

cobrien at Radix.Net (Cary O'Brien) writes:

> In article <L3iZ4.154057$681.2782248 at news-east.usenetserver.com>,
> J.M. Ivler <ivler at basecamp1.netquest.net> wrote:
> >
> >And what about TCL users?
> >
> 
> Well, they've got mod_dtcl and aolserver.
> 
> Python has got mod_python and pyapache.
> 
> Look, this is mature technology. Can't everyone just get along?
> 
> (Plus you forgot embedded perl - embperl).

Java users have Servlets/JSP.  Scheme users have BRL (implemented with
servlets).  We could probably keep going.

-- 
Bruce R. Lewis				http://brl.sourceforge.net/


From nadavh at envision.co.il  Wed Jun 28 11:16:35 2000
From: nadavh at envision.co.il (Nadav Horesh)
Date: Wed, 28 Jun 2000 17:16:35 +0200
Subject: Python vs. Matlab
Message-ID: <395A16D3.A7F97C9B@envision.co.il>

Matlab is very handy because its impressive arsenal of plotting and
visualization tools. Some time ago I ported a wave propagation
simulation program from Matlab to Python+Numpy and observed the
following:

  1. Numpy was twice as fast (mostly 2D FFT)
  2. The Python memory requirement were about half (Matlab half speed
     observed when there was no memory deficiency, when running full
     scale simulation, the excess of paging made Matlab 5--10 times
     slower on a 256  MB machine).
  3. Python code was much more extensible, not sup rising considerint
     Matlab archaic Fortran-like syntax.

But for quick and dirty calculations with plotting, Matlab is still
better, since all its tools are accessible without any extra
programming.

    Nadav.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nadavh.vcf
Type: text/x-vcard
Size: 360 bytes
Desc: Card for Nadav Horesh
URL: <http://mail.python.org/pipermail/python-list/attachments/20000628/2e303300/attachment.vcf>

From alan.gauld at gssec.bt.co.uk  Tue Jun  6 12:16:25 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Tue, 06 Jun 2000 17:16:25 +0100
Subject: Spanish tutor available
Message-ID: <393D23D9.2E7D65B0@gssec.bt.co.uk>

My online Programming/Python tutor now has 
a Spanish version in production. The first 
pages are there and more will follow.

Credits to Martin Pozzi for the translation.

It's at:

http://www.crosswinds.net/~agauld/spanish/index.htm

Hope it helps someone, thanks again to Martin.

Alan G.

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


From thomas at xs4all.nl  Fri Jun 30 21:06:50 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 1 Jul 2000 01:06:50 GMT
Subject: UserDict and UserList issue (bug?_
References: <395D333D.9240593C@email.sps.mot.com>
Message-ID: <slrn8lqh0n.gqv.thomas@centurion.xs4all.nl>

On Fri, 30 Jun 2000 16:54:37 -0700, Norman Shelley
<Norman_Shelley-RRDN60 at email.sps.mot.com> wrote:
>in UserDict and UserList there is one or more places where a call to
>self's __class__() method is made, e.g.
>
>def __getslice__(self, i, j):
>  i = max(i, 0); j = max(j, 0)
>  # This class and its subclasses need more arguments then 0
>  userlist = self.__class__()
>  ....

>This blows causes major problems when either of these two classes are
>subclassed and the class which subclasses them requires creation
>arguments.  I think maybe a copy of self instead of a __class__() may be
>a more robust solution.

And how would you get a copy of self without creating a new instance of
self.__class__ ? You usually create a copy of a list by doing:

    mycopy = mylist[:]

Which calls __getslice__, which is exactly what you are complaining about
:-)

If you want to subclass UserList, either make sure you provide defaults for
all arguments to the initializer, or make sure you override all places where
self.__class__() may be called without your required arguments. #1 is
probably the best (and most portable) solution, and not just because of the
above issue: providing defaults for all initializer arguments allows your
object to be pickled, for instance ;)

But I dont know why you need those arguments, so YMMV.

Regards,
	Thomas.


From eddthompson at ssi.parlorcity.com  Wed Jun  7 22:35:31 2000
From: eddthompson at ssi.parlorcity.com (Blu Dragoon)
Date: Thu, 8 Jun 2000 02:35:31 -0000
Subject: physical hardware accesss
Message-ID: <393f4dd2$0$2877@wodc7nh6.news.uu.net>

If I wanted to would be able to probe a serial, parallel, or USB port
with python.
Open(/dev/cua1)
Write(/dev/cua1)

something to that effect?




From stephen at cerebralmaelstrom.com  Thu Jun  1 22:30:31 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Fri, 02 Jun 2000 02:30:31 GMT
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com>
Message-ID: <b%EZ4.31465$WS3.367841@typhoon.we.rr.com>

    That license isn't viral (like GPL :)). You have to include it in your
documentation, explaining thus that you're using Python with legal
permission..and that you can do anything you want with it :)

--S

<t_e_sanders at my-deja.com> wrote in message
news:8h6rbt$ksv$1 at nnrp1.deja.com...
> Howdy. I'm interested in using Python within a commercial software
> package. Python's license seems to be really good about this, but I
> have a question about the copyright notice:
>
> > Permission to use, copy, modify, and distribute this software and its
> > documentation for any purpose and without fee is hereby granted,
> > provided that the above copyright notice appear in all copies and
> > that both that copyright notice and this permission notice appear in
> > supporting documentation...
>
> Do the terms of Python's distribution license require me to allow other
> users to copy and distribute my software? I would guess not, but
> inclusion of this notice in my package explicitly states so. There's
> obviously something I'm missing, (one of the downsides of being a
> developer, rather than a lawyer? :) Can anyone shed some light on this?
>
> Thanks much,
> Thomas Sanders
> t.e.sanders at usa.net
>
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From pfo at informactis.com  Wed Jun 14 09:05:31 2000
From: pfo at informactis.com (patrick fox)
Date: Wed, 14 Jun 2000 15:05:31 +0200
Subject: Can't execute an external program !
Message-ID: <3947831B.73DEB063@informactis.com>

I am working with RTF files and I'd like my WORDPAD to appear once the
work is done.

I've tried this :

   > f = open('./var/fiche-concessionnaire.rtf', 'w')
  >  f.write(tpl.data)
  >  f.close()
   > commands.getoutput('c:\\windows\\write.exe
./var/fiche-concessionnaire.rtf')

It always reply "Command Unknown" even when I try to use the DIR
command.

Q: How can I get this working ?

Thanks.



From petersc at stud.ntnu.no  Sat Jun  3 03:38:19 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 03 Jun 2000 09:38:19 +0200
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de> <3938090C.9FB5D057@stud.ntnu.no> <8h9682$v4h$1@nntp9.atl.mindspring.net> <39385BAC.771293BE@stud.ntnu.no> <8h9ft7$6is$1@slb6.atl.mindspring.net>
Message-ID: <3938B5EB.2A0BAFAF@stud.ntnu.no>

Aahz Maruch wrote:
> 
> i = 0
> while 1:
>   try:
>     object.__getitem__(i)
>   except IndexError:
>     break
>   i = i + 1
> 
> Now, in Python terms (not C), what would you say happens to the
> exception at the break?

The exception gets caught and ignored, so I would rather say
it is "thrown away" than that the "error is cleared".

The point is that the standard for-loop actually used by
the original poster is implemented in C (not in Python). So
to understand why this implementation of the for-loop does
what it does I have to look at the C level. Of course giving
an explanation as an equivalent Python code snippet (like the
one above) is a nice way to put it. But I still don't see
where you object to the term "error is cleared" when that is
what actually happens in the Python interpreter.

maybe-my-problem-is-that-I-use-the-source-as-my-documentation-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From cbbrowne at news.hex.net  Thu Jun  8 22:13:17 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Fri, 09 Jun 2000 02:13:17 GMT
Subject: Using python with Corba
References: <eieJ4.8408$H7.704991@brie.direct.ca> <871z4awiz0.fsf@parus.parus.no> <39410efe.18494828@nntp.interaccess.com>
Message-ID: <1pY%4.164477$MB.3073078@news6.giganews.com>

Centuries ago, Nostradamus foresaw a time when Thaddeus L. Olczyk
would say: 
>On 13 Apr 2000 09:36:19 +0200, Jon K Hellan <hellan at acm.org> wrote:
>>"Fanny Yeung" <fannyy at idmail.com> writes:
>>> Hi everybody,
>>> 
>>> I am new python and I am particular interest in use python with corba. After
>>> some research, I find out omniOrb provide interface 'omniOrbpy'. Any one
>>> know where I can find it ?
>>
>>http://www.uk.research.att.com/omniORB/omniORB.html
>>
>>> Also, besides omniOrb2, is there any other Corba that Python can use ?
>>
>>Fnorb is almost pure Python.
>>
>>ILU does a lot of languages.
>>
>>ORBit is extremely fast.
>>
>>> Anyone know TAO can be used in Python ?
>>
>>Don't think so. But a Python ORB can interoperate with TAO.
>>
>>Regards
>>
>>Jon Kare
>
>They are supposed to but (sigh ) most orbs have many parts of the
>standard unimplemented yet. Do you know if Fnorv,ILU, OmniORb
>or ORBit (didn't know ORbit supported python) can interoperate with
>TAO?

I can't speak for TAO and ORBit, but there are not one, but _TWO_
implementations of "ORBit support for Python."

<http://theopenlab.uml.edu/pyorbit/>
<http://projects.sault.org/orbit-python/>

I prefer the latter, but have gotten both to function...
-- 
aa454 at freenet.carleton.ca - <http://www.hex.net/~cbbrowne/corba.html>
Warning: Dates in calendar are closer than they appear. 


From ge at nowhere.none  Fri Jun 30 11:08:42 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 30 Jun 2000 15:08:42 GMT
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <yYM65.2213$iN5.491623@ptah.visi.com> <395C73E9.839ABEA6@hursley.ibm.com>
Message-ID: <_J275.2322$iN5.531627@ptah.visi.com>

In article <395C73E9.839ABEA6 at hursley.ibm.com>, Paul Duffin wrote:
>Grant Edwards wrote:
>> 
>> In article <39588741.FB7EB5FE at hursley.ibm.com>, Paul Duffin wrote:
>> >Donn Cave wrote:
>> >>
>> >> Quoth ge at nowhere.none (Grant Edwards):
>> >> | Or you can roll your own using the popen2 module.
>> >
>> >I think that you will find that Expect offers a lot more than a simple
>> >pipe to the executable.
>> 
>> Yes, I know. That's where the "roll-your-own" bit comes in. If
>> all you _want_ is a simple pipe to the executable, then expect
>> isn't really the right choice.
>
>If all you _want_ is a simple pipe then you may be disappointed if it turns
>out that what you _need_ is proper tty support. You are correct that a 
>simple pipe is often enough, I was trying to explain that you cannot
>roll your own version of Expect using a simple pipe.

Yea, the whole stdout buffering thing can be a headache.

-- 
Grant Edwards                   grante             Yow!  .. If I cover this
                                  at               entire WALL with MAZOLA,
                               visi.com            wdo I have to give my AGENT
                                                   ten per cent??


From chafik at nevrax.com  Tue Jun  6 05:18:40 2000
From: chafik at nevrax.com (Sameh chafik pro)
Date: Tue, 6 Jun 2000 11:18:40 +0200
Subject: Need help on Python/C++
References: <8gltp7$f7f$1@gw.localdomain> <p6qitvx4aes.fsf@informatik.hu-berlin.de>
Message-ID: <8hifbl$k55$1@gw.localdomain>

That  is okay, but why i do to translate a class write in C++. The
problem that the standard way is to use the PyTypeObject and define the
constructor as a function, but python don't recognize the translation as
class but as type object.morality if i do the translation of a class "X" it
is'nt posible in python to use it as a derivate class:
class A(X):
    def doSomething(self):
    .....
    ....

is not possible. Because X is a function.That seem insignificant but i want
to do an efficient translation between C++ code and Python.



"Martin von Loewis" <loewis at informatik.hu-berlin.de> a ?crit dans le message
news: p6qitvx4aes.fsf at informatik.hu-berlin.de...
> "Sameh chafik pro" <chafik at nevrax.com> writes:
>
> > are there someone how know all aboute the data member of the structur
> > PyTypeObject?
>
>
> See Include/object.h:PyTypeObject.
>
> Regards,
> Martin
>




From rdejuana at my-deja.com  Thu Jun 22 18:16:11 2000
From: rdejuana at my-deja.com (rdejuana at my-deja.com)
Date: Thu, 22 Jun 2000 22:16:11 GMT
Subject: extending python problem
Message-ID: <8iu37c$jsa$1@nnrp2.deja.com>

I'm trying to extend python with c.  I trying to
get the progam at the tutorial to work. but every
time i try to import it it tells me it cant find
the init function.  This is the error:

>>>import libPYSTAF
Traceback (innermost last):
  File "./libPYSTAF.py", line 3, in ?
    import libPYSTAF.so
ImportError: dynamic module does not define init
function (initlibPYSTAF)

This is my code:

#include <Python.h>

void initlibPYSTAF();

static PyObject* libPYSTAF_system(PyObject
*self,PyObject * args)
{
    char *command;
    int sts;

    if (!PyArg_ParseTuple(args, "s", &command))
        return NULL;
    sts = system(command);
    return Py_BuildValue("i", sts);
}

static PyMethodDef libPYSTAFMethods[] =
{
    {"system",  libPYSTAF_system, METH_VARARGS},
    {NULL,      NULL}        /* Sentinel */
};

void
initlibPYSTAF()
{
    Py_InitModule("libPYSTAF", libPYSTAFMethods);
}

I'm new to coding, so i probably forgot something
stupid.  If you stop it please tell what it is.

thanks

-rod

rdejuan at us.ibm.com



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


From jkraska1 at san.rr.com  Mon Jun 12 23:34:29 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 12 Jun 2000 23:34:29 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com> <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu> <8i3kbi$3a0$1@slb1.atl.mindspring.net>
Message-ID: <642A954DD517D411B20C00508BCF23B00129688E@mail.sauder.com>

> >Then I move to C.  The process took about twice as long as the C++
> >attempt; not counting the time it took for me to write miscellaneous
> >libraries (stack, memory allocation), I noticed that I was spending far
> >more time hunting pointer bugs and syntax errors than writing new
> >code. oThis was the longest port; it took me about four times longer than
> >it did in Python.

Although, to be fair, any seasoned C programmer has his or her
own preferred set of utility functions for doing hashtables,
stacks, lists, and so forth. I have a set of macro-expansion
tools for that that are far more efficient than the standard
template library could ever pray to be (written by a nut who
would meticulously process to assembly and analyze the number
and nature of assembly instructions issued for the equivalent
C... he was really freaky :).

Fair point with the pointer bugs, though.

That's not something that can even happen in Python, really. :)-

C/



From jcashman55 at my-deja.com  Thu Jun  8 12:12:59 2000
From: jcashman55 at my-deja.com (jcashman55 at my-deja.com)
Date: Thu, 08 Jun 2000 16:12:59 GMT
Subject: Win32net.NetServerEnum Unicode problem
References: <8hlnd2$lrm$1@nnrp2.deja.com> <erz%4.2026$wp.166067@pouncer.easynews.com>
Message-ID: <8hoglp$54l$1@nnrp1.deja.com>

I have Hammond's Python on Win32 book, and it doesn't list another
parameter relating to domain name.  Plus it doesn't seem to work.
Complains that the system call level is not correct, which is usually
when I don't have the right number of params or type of params.

jim

In article <erz%4.2026$wp.166067 at pouncer.easynews.com>,
  "Roger Upole" <rupole at compaq.net> wrote:
> It looks like you are missing the domain name, which should
> be right after SV_TYPE_ALL.
>         Roger Upole
>
> <jcashman55 at my-deja.com> wrote in message
> news:8hlnd2$lrm$1 at nnrp2.deja.com...
> > Having trouble with the following line:
> >
> >         entries, total, resume= \
> > win32net.NetServerEnum('Dopey', \
> > 102,win32netcon.SV_TYPE_ALL, 0, 4096)
> >
> > When I run code that contains this line, I get:
> >
> > TypeError: The object can not be converted to a
> > Unicode object
> >
> > I have run some of the the other win32net calls
> > without a problem.  Anyone have an idea?
> >
> > Thanks.
> >
> > Jim
> >
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


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


From israel at math.ubc.ca  Thu Jun  8 16:11:42 2000
From: israel at math.ubc.ca (Robert Israel)
Date: 8 Jun 2000 20:11:42 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com>
Message-ID: <8houlu$j9s$1@nntp.itservices.ubc.ca>

In article <8hogt2$57t$1 at nnrp1.deja.com>,  <david_ullrich at my-deja.com> wrote:

>    ??? In one of the references cited the authors say

>"The computation can
>be achieved without having to compute the preceeding digits. We claim
>that the algorithm has a more theoretical rather
>than practical interest, we have not found a faster algorithm, nor have
>we proven that one does not exist."

>    Regarding the "have not found a faster algorithm": Are
>you certain they're saying that none of this is any faster
>than finding the first n digits, or do they mean that they
>have not found a faster "digit-extraction" algorithm than
>the original Bailey-Borwein-Plouffe algorithm?

>    The other reference begins

>"David Bailey, Peter Borwein and Simon Plouffe have recently computed
>the ten billionth digit in the hexadecimal
> expansion of pi. They utilized an astonishing formula:

>[astonishing formula goes here]

> which enables one to calculate the dth digit of pi without being forced
>to calculate all the preceding d-1 digits."

>    If these algorithms _really_ take as much time as finding
>all the digits, as you seem to be saying, it's hard to see why
>they'd have everybody so excited (or rather why they _did_ have
>everybody so excited back when this was news). It's also hard to
>see why people would suggest that one point to the digit-extraction
>algorithms would be to be able to check the accuracy of calculations
>of _all_ the digits.

There are two different algorithms here, for two different problems.

The BBP algorithm can find a hexadecimal (or more generally, base
2^k) digit of pi, faster than any known algorithm that would find
all the digits up to this one.  This is the exciting one.

Plouffe has another algorithm that can find a digit in an arbitrary
base.  This takes C*n^3*log(n)^3 time for the n'th digit, so it is not
faster than other methods that would compute the first n digits, but it
does take very little memory.  This one has "a more theoretical rather
than practical interest".

Whether there is an algorithm for the n'th decimal digit that performs
as well as BBP is still an open question.

Robert Israel                                israel at math.ubc.ca
Department of Mathematics        http://www.math.ubc.ca/~israel 
University of British Columbia            
Vancouver, BC, Canada V6T 1Z2






































From mhammond at skippinet.com.au  Fri Jun  9 20:34:46 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 00:34:46 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no> <8hrhqa$afm$1@news1.xs4all.nl>
Message-ID: <G2g05.688$cQ.9254@news-server.bigpond.net.au>

"Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
news:8hrhqa$afm$1 at news1.xs4all.nl...

Thanks for your nice comments about Pythonwin!

[BTW - Im not at all fazed by the start of this thread - I know lots of
people enjoy Pythonwin!  But Ive been waiting for an opportunity to use
that "where the sun dont shine, sunshine" line for a while now :-]

> The lookups aren't that great - they only find what you've already used
> in that file. I'd like loopup to look in every file that's included for
> completion. But I'm not going to complain - it's good for my memory not
> to have it ;-).

This is amazingly hard to do correctly - and the hard bits are not related
to Pythonwin!

The problem can be expressed without any regard for a GUI.  If it can be
solved generically, Pythonwin, IDLE and any other IDEs could use it to.
Indeed, I did the original IDLE CallTips for Pythonwin, then ported it to
IDLE.  If I could get Pythonwin's AutoComplete to give good results most
of the time,  would happily port that to IDLE too (incentive for
non-Windows users to think about this:).

The problem is simply this:  Given an arbitary location in an arbitary
source file, give me a representation of the Python object at that
location.  This representation must have enough info to deduce the
signature (if a callable object) and all attributes suitable for an
auto-complete type list.  The object itself would be perfect, but this is
not practical.

Further, this must be done without any Python introspection capabilities.
Introspection could be used only for already imported code.  Executing
something like "CallTips" or "AutoComplete" should never execute users
code, as the side effects can't be predicted.  This restriction generally
means you can't rely on introspection at all.  Further, introspection can
only solve the easy 20% of the problem :-)  And to further complicate
matters, thus must work on syntactically incorrect code (the syntax is
_never_ correct when autocomplete or calltips are actually displayed!)

So if anyone can come up with generic code that can do this, all current
and future IDEs for Python could benefit.

Mark.




From tanzer at swing.co.at  Mon Jun  5 02:06:25 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 05 Jun 2000 08:06:25 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!) 
In-Reply-To: Your message of "02 Jun 2000 21:05:15 +0200."
             <8h90hb$5e1@news.Informatik.Uni-Oldenburg.DE> 
Message-ID: <m12yq1p-000wcEC@swing.co.at>

Johannes Nix <Johannes.Nix at mail.uni-oldenburg.de> wrote:

> Russell Wallace <rwallace at esatclear.ie> writes:
> 
> > It seems to me that the important argument in this context is "case
> > insensitivity will break millions of lines of existing code, and perhaps
> > split the user base if a significant number of existing users refuse to
> 
> It is a trivial task to convert case-sensitive identifiers in unique
> case-insensitive identifiers in case of a name clash.

Unfortunately, it isn't.

One of Python's big strengths are features like `eval' and `exec'.
It's impossible to auto-convert the case of identifiers built
by a program dynamically. 

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From digitig at cix.co.uk  Fri Jun 16 19:34:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 17 Jun 2000 00:34 +0100 (BST)
Subject: Where are the math functions?
References: <20000616125043.A19134@better.net>
Message-ID: <memo.20000617003443.42607B@digitig.compulink.co.uk>

In article <20000616125043.A19134 at better.net>, parkw at better.net (William 
Park) wrote:

> What do you mean "harder to cite"?  

It's easier -- perhaps I should say more credible -- to reach over to my 
bookcase and type in "Press, W. H., S. A. Teukolsky, W. T. Vetterling, and 
B. P. Flannery (1992). Numerical Recipes in C, Cambridge University Press, 
Cambridge", than to put "A library I have on my disk and I expect is on 
the web somewhere". Shipping product I don't care, but academic 
dissertations are not the real world -- or at least, are a /different/ 
real world.


From hniksic at iskon.hr  Sat Jun 10 06:00:59 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 10 Jun 2000 12:00:59 +0200
Subject: Python for Windows -- what to install?
References: <000001bfd253$ce6f9180$5b2d153f@tim>
Message-ID: <dppuppx3c4.fsf@mraz.iskon.hr>

Thanks for the detailed reply, Tim!

"Tim Peters" <tim_one at email.msn.com> writes:

> > I'd like him to be able to use the stuff that is normally distributed
> > with Python 1.5.2, plust IDLE under Tkinter.
> >
> > I assume I'll need `py152.exe' for the basic functionality.
> 
> That contains all of the above -- jusr run it.

Excellent!

> I strongly recommend one deviation from the defaults, though: when
> it asks for the Python installation directory, force it to
> "c:\python15" (where "c:\" is the root of his boot drive, almost
> certainly exactly "c:\").  The default is some monster with embedded
> spaces in its name, and that just makes life harder.

OK.

> > Do the above files include documentation, or do I have to download
> > it separately?
> 
> It's all in py152.exe.

What is the format of the documentation?

> A probable problem and a suggestion: depending on which version of
> Windows he's running and the phase of the moon, the Scriptics Tcl
> installer (which comes with and is invoked by py152.exe -- one big
> reason that's bigger than the Unix Python distributions) may not set
> up Tcl correctly.  The cure for that is to get rid of pre-existing
> tcl*.dll and tk*.dll files, and/or to set the PATH in his
> autoexec.bat to include Tcl's bin directory.  After Tcl is working,
> IDLE will work too, and at that point you should visit python.org
> and grab a more-recent version of IDLE (IDLE has improved a lot
> since the 1.5.2 release).

Will do.  That's why I'm asking all this in advance -- once I burn the
CD and ship it via mail, there will be no more chance to visit
www.python.org for additional stuff.


From breiter at usf.Uni-Osnabrueck.DE  Mon Jun 26 07:28:36 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 26 Jun 2000 11:28:36 GMT
Subject: Boa stalled? (GUI editor for wxPython)
References: <8iqbv4$4p$1@newsserver.rrzn.uni-hannover.de>
    <3952A499.1339701@e.co.za>
    <8ivup5$dsq$1@newsserver.rrzn.uni-hannover.de>
    <3953935C.826AF3FB@e.co.za>
Message-ID: <8j7ep4$8km$2@newsserver.rrzn.uni-hannover.de>

In article <3953935C.826AF3FB at e.co.za>,
	Riaan Booysen <riaan at e.co.za> writes:
> 
> 
> Bernhard Reiter wrote:
>> I did not find any progress in the CVS.
>> Is there a reason for not leeting people see the development drive?
>> 
>> So can I download a CVS version with all the improvements?
> 
> Well for awhile I had a few problems and I wasn't sure that the
> changes were 'improvements'.

Well, that is what CVS is for anyway. :)

> I'm hopefully updating the CVS this weekend, you're not the only
> one asking ;)

Please do so. (Untill now I do not see changes in the CVS toplevel
bos directory.)

	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From ilnodo.it at ilnodo.it  Sat Jun 10 13:09:20 2000
From: ilnodo.it at ilnodo.it (ilnodo.it)
Date: Sat, 10 Jun 2000 19:09:20 +0200
Subject: novity
Message-ID: <8htsld$cde$12@fe1.cs.interbusiness.it>

                      www.ilnodo.it





From jmassung at magpiesystems.com  Wed Jun 21 11:19:17 2000
From: jmassung at magpiesystems.com (Jeff Massung)
Date: Wed, 21 Jun 2000 09:19:17 -0600
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie>
Message-ID: <t7545.239$ZR1.2944@news.uswest.net>

Please, this interests me -- I've had differention equations, but that's as
far as I go. In theory, pi (and e) are infinite, but has that been proven
that there is not end to their decimals?

Jeff




From ldw at us.ibm.com  Fri Jun 23 10:24:26 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Fri, 23 Jun 2000 09:24:26 -0500
Subject: Dictionary class in C++
Message-ID: <8ivrup$nqe$1@news.rchland.ibm.com>

I am writing a python program to process some traces in preparation for
running the traces through a model written in C++.  Python's dictionary
class seems to give me what I'm looking for on the python side of the
equation and I am trying to translate the dictionary to C++ constructs.

Looking through the IBM Open Class library (I haven't upgraded my compiler
to one that can do STL yet) I find a number of "collection classes".  The
first level of organization is into Flat Collection Classes or Trees.  Of
the two Flat seems to be right.  The next level of organization is a table
too complicated to reproduce here.  The net is that it looks to me like the
Dictionary is: Unordered and Keyed.   Given that the corresponding
collection class is one of four posibilities depending on the element and
key characteritics:

Element equality and element uniqueness = Map
Element equality and no element unizueness = Relationship.
No Element equality and element uniqueness = Key Set.
No Element equality and no element uniqueness = Key Bag.

Can anyone help me sort the above out - which is the python dictionary?

Larry




From donn at u.washington.edu  Mon Jun 26 12:26:51 2000
From: donn at u.washington.edu (Donn Cave)
Date: 26 Jun 2000 16:26:51 GMT
Subject: determining an assigned IP address
References: <39576449.CFC3E4EC@nih.gov> <8j6lnu$jhd$1@nnrp1.deja.com>
Message-ID: <8j808b$7f1a$1@nntp6.u.washington.edu>

Quoth Jonathan Epstein <Jonathan_Epstein at nih.gov>:
| The following Perl code works:
|
|   use Sys::Hostname;
|   use Socket;
|
|   $ip=gethostbyname(hostname());
|   $theip=inet_ntoa($ip);
|
| It should be relatively straightforward to convert this to Python.

Indeed, just omit half of it!  The inet_ntoa() in particular is
already done by socket.gethostbyname().  hostname() is
socket.gethostname().  Usually this will even return the address
you want.  When it doesn't, see my previous follow-up.

	Donn Cave, donn at u.washington.edu


From wware at world.std.com  Wed Jun 21 18:05:56 2000
From: wware at world.std.com (Will Ware)
Date: Wed, 21 Jun 2000 22:05:56 GMT
Subject: Dumb overloading question
Message-ID: <FwIypw.KG@world.std.com>

I'm feeling a little idiotic and hope somebody can point out the dumb
thing I'm doing here. I'm trying to write two SGML parsers that translate
my source document to an HTML page, and a set of slides in Postscript.
I wrote the HTML generator first, so I have:

---- begin s2html.py ----
import sys, sgmllib
class HTMLParser(sgmllib.SGMLParser):
    def handle_data(self, data):
        self.mydata = self.mydata + data
    # then define many functions that make calls to
    # self.handle_data, for example...
    def start_source(self, attrs):
        self.handle_data('<html>')
    def end_source(self):
        self.handle_data('</body></html>')
# other extraneous stuff elided for brevity
---- end s2html.py ----

Then I decided that the Postscript generator should start off using
most of the same behavior.

---- begin s2ps.py ----
import sys, s2html
class SlideParser(s2html.HTMLParser):
    def __init__(self):
        s2html.HTMLParser.__init__(self)
        self.inSlide = 0
    def handle_data(self, data):
        if self.inSlide:
            self.mydata = self.mydata + data
    def start_slide(self, attrs):
        self.inSlide = 1
        s2html.MySGMLParser.start_slide(self, attrs)
    def end_slide(self):
        s2html.MySGMLParser.end_slide(self)
        self.inSlide = 0
---- end s2ps.py ----

The wierd thing is that when I use SlideParser, I get behavior as if
HTMLParser's functions (such as start_source) are using the HTMLParser
version of handle_data instead of the SlideParser version. Isn't that
incorrect? When I overload handle_data, shouldn't that apply to the
functions in HTMLParser who make calls to handle_data? I am befuddled
and would appreciate any clarification.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From david_ullrich at my-deja.com  Wed Jun 28 13:07:08 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 28 Jun 2000 17:07:08 GMT
Subject: unexpected behaviour in class variables
References: <8j83ml$2s7$1@slb3.atl.mindspring.net> <8F5FCA994gmcmhypernetcom@199.171.54.194> <8j8vjr$v63$1@nntp9.atl.mindspring.net> <8F60A1C6Bgmcmhypernetcom@199.171.54.154> <8jcj57$ec$1@nntp9.atl.mindspring.net>
Message-ID: <8jdbbc$int$1@nnrp1.deja.com>

In article <8jcj57$ec$1 at nntp9.atl.mindspring.net>,
  "Andrew Dalke" <dalke at acm.org> wrote:
> Gordon McMillan:
> >It ain't wrapped. Beat up Fred, he needs it <wink>.
>
> Beat him up?  I'm a software guy and try to stay away from that
> whole hardware concept.  How about I email him instead?

   Much better idea.


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


From wh at msgto.com  Mon Jun 26 16:14:07 2000
From: wh at msgto.com (Walter Hofmann)
Date: Mon, 26 Jun 2000 22:14:07 +0200
Subject: TCP packet size?
References: <39487D5B.FC96BE3F@rpgarchive.com>
Message-ID: <slrn8lfecf.23e.wh@frodo.uni-erlangen.de>

On Thu, 15 Jun 2000 01:53:15 -0500, chris <chris at rpgarchive.com> wrote:
>
>
>--------------6576511AB77059DEC203AB93
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Is there a way to ensure or predict packet size when using TCP?  I'm
>developing a Multi-user application that functions much like a
>muli-player game.  The applications are sharing xml data.  The xml data
>can get very big 3000-6000 bytes.  Unfortunately, when the size of the
>network message gets over about 2000, its to get broken up into smaller
>packets.  So, when I send a 3000 byte message it gets cut up into 2-3
>packets.  Anther problem I get is when I send two small messages back to
>back, they are sometimes received together.  Is there something I'm
>missing here?

No. That's the way TCP works. It gives you a "stream" of data. Use UDP
if you want to have packets (like most games do, AFAIK), but be prepared
to handle lost or duplicate packets. If you send UDP packets larger that
the MTU they will get fragmented, which will harm performance.

Using TCP you have to add message headers or an "end of record" string 
so that you know when to issue another "read" to get the rest of the
packet.

Walter


-- 
begin  LOVE-LETTER-FOR-YOU.txt.vbs
Hi! I'm a signature-virus. Distribute me!
end


From rsteelandt at bucodi.com  Wed Jun 21 05:27:17 2000
From: rsteelandt at bucodi.com (BuCoDI-Rony Steelandt)
Date: Wed, 21 Jun 2000 11:27:17 +0200
Subject: Tkinter and setting focus
References: <394F860E.EF6042D6@bucodi.com> <1_R35.5209$_55.108735@news2-win.server.ntlworld.com>
Message-ID: <39508A75.23DF0795@bucodi.com>

Thank you Richard, exactly what i needed. I didn't know you could that

Rony

richard_chamberlain a ?crit :

> Hi Rony,
>
> You could just generate your own tab event (I presume that's the behaviour
> you want)
>
> So:
>
> def validate(self,event):
>         # do some validation
>         self.field.event_generate('<Tab>')
>
> Richard
>
> BuCoDI-Rony Steelandt <rsteelandt at bucodi.com> wrote in message
> news:394F860E.EF6042D6 at bucodi.com...
> >
> > I have created a class wich expands the standard entry widget.
> > In that class i did a override of the return key to do some validation
> > of my entry
> >
> > self.entry.bind(<return>,self.validate)
> >
> > The validation routine works fine, but now comes my problem
> >
> > I want at the end of the validation, if validation is ok set the focus
> > to the next entry widget in the form.
> > I searched documentation but didn't find anything on this.
> > The problem is since my class is a general class, i don't know in the
> > class wich widget is the next to put the
> > focus on.
> > Here is some code to show what i'm doing:
> > class numentry:
> >  def
> >
> __init__(self,parent,rw,clmn,wdt,relf=SUNKEN,mask='$',min=0,max=999999999):
> >
> >   self.w = wdt
> >   self.m = mask
> >   self.min = min
> >   self.max = max
> >   self.field = Entry(parent,relief=SUNKEN,width=wdt)
> >   self.field.grid(row=rw,column=clmn)
> >   self.field.bind("<Return>",self.validate)
> >   self.returnval=''
> >   self.ind = 0
> >   self.pointindex = 0
> >
> >
> >  def validate(self,event):
> >     # do some validation
> > ===> here i want to set the focus to the next widget
> >
> >
> > #main
> > frame = Frame(root,width=100,height=100)
> > # frame.bind("<Up>",callback)
> > frame.pack()
> > f=numentry(frame,1,1,10,SUNKEN,'0000.00',5)
> > f2=numentry(frame,2,1,10,SUNKEN,'0000.00')
> > root.mainloop()
> >
> >
> > I'm realy sticked on that ?????
> > Anyone has a idea how to do that
> >
> > Rony Steelandt
> > --
> > ======================================================
> > Bureau de Conseil et de D?veloppement Informatique
> > 2, Rue Henri Avril
> > 22480 Saint - Nicolas du Pelem
> >
> > T?l : +33.2.96.29.79.08
> > Fax : +33.2.96.29.79.09
> >
> > e-mail : bucodi at worldnet.fr
> > ======================================================
> >
> >

--
======================================================
Bureau de Conseil et de D?veloppement Informatique
2, Rue Henri Avril
22480 Saint - Nicolas du Pelem

T?l : +33.2.96.29.79.08
Fax : +33.2.96.29.79.09

e-mail : bucodi at worldnet.fr
======================================================




From db3l at fitlinxx.com  Wed Jun 14 03:18:22 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 03:18:22 -0400
Subject: ANNOUNCE: wxPython 2.1.16
References: <003c01bfd5b5$c9cac810$0b01a8c0@ARES>
Message-ID: <uog547mtd.fsf@ctwd0143.fitlinxx.com>

"Robin Dunn" <robin at alldunn.com> writes:

> WARNING: A while back I asked what should be done about the Magic
> Method Names.  (Methods that are automatically turned into event
> handlers by virtue of their name.)  The consensus was that it is more
> confusing to have them than to try and expand them to have greater
> coverage.  I am finally getting around to removing the code that
> generates the event binding.  This means that if you are using any of
> the following method names without a EVT_* call that you need to
> modify your code to add the EVT_* to hook the event to the method.

One burp with this - it looks like this causes a small problem with
the wrappings for wxTaskBarIcon.  In windows2.py it still uses
wx._checkForCallback, but that function is now commented out in wx.py.

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


From olczyk at interaccess.com  Sun Jun 25 00:01:09 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sun, 25 Jun 2000 04:01:09 GMT
Subject: Python with braces.
Message-ID: <395683ae.299281703@nntp.interaccess.com>

Someone told me that there is an interpreter that can hanld Python
using braces instead of indentation for defining blocks. Can anyone
tell me where to find it?


From stephen at cerebralmaelstrom.com  Fri Jun  9 15:45:01 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Fri, 09 Jun 2000 19:45:01 GMT
Subject: Linewrapping in Pythonwin?
Message-ID: <1Pb05.47081$WS3.490567@typhoon.we.rr.com>

    I usually use PythonWin for my IDE, because IDLE has a couple quirks
which generally irritate the heck outta me. (Not demanding they get fixed!
*hide* Just feel like breaking a pencil every time I hit one :) heh.) Namely
the fact that right-clicking doesn't bring up a context-menu.. Its a natural
thing for me to select, rightclick, copy.. Anyways :)

    I do have one big um.. sugguestion? (hides more from Mark) in PythonWin.
Or, rather, a question.Is there a way to have it linewrap in the interactive
window? I don't see an option -- maybe i'm just being blind, but when i'm
doing some debugging and looking at various attributes which are possibly
quite long, it'd be much easier to have it all wrapped so I don't have to
scroll anywhere to see anything.

--Stephen




From dsavitsk at e-coli.net  Tue Jun 13 22:22:11 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Wed, 14 Jun 2000 02:22:11 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com>
Message-ID: <n%B15.521$dF.22239@news1.rdc1.il.home.com>

asp supports all sorts of scripting.  the easiest way to do this is with ...

<%@ LANGUAGE=VBScript %>
<%
' The default language is set above
' Place your VBScript stuff here
%>

<SCRIPT RUNAT = SERVER LANGUAGE=PythonScript>
# within this block PythonScript
# will be the language used at the
# server.
# you must have the ActiveX DLL (or
# something similar for iASP or
# chilisoft) installed.
# note that PerlScript is broken in
# NT 5.0 still (i think)
</SCRIPT>

<%
' this defaults to VBScript still
%>



"millja" <millja at wwc.edu> wrote in message
news:0a9f5150.911771a5 at usw-ex0109-068.remarq.com...
> I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m having a bit of
an issue.  My current situation is that
> I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m able to use
Python with ASP and
I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m able to use
> VBScript with ASP.  However,
I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m still trying to
figure out
> how to use the two together.
>
> What is occurring is that
I?ffffc3?ffffa2?ffffe2?ffff82?ffffac?ffffe2?ffff84?ffffa2m working on a
website, most of
> which is already written with VBScript.  I have a rather
> large function that I would like to integrate.  Is there
> some way to call this function from within the VBScript?
> Or, it would work if I can get numbers, generated by the
> VBScript, into the Python function and the results of this
> function back to the VBScript.
>
> Thanks for you time.
>
> jason milliron
>
>
>
> * Sent from AltaVista http://www.altavista.com Where you can also find
related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is
Beautiful




From paul.robinson at quantisci.co.uk  Thu Jun 29 07:20:40 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Thu, 29 Jun 2000 12:20:40 +0100
Subject: list.pop([i]) and list.remove(x) w/ for loops
References: <8jdfb6$ugh$1@netnews.upenn.edu>
Message-ID: <395B3108.AA2A4EB5@quantisci.co.uk>

David White wrote:
> 
> I was getting some unexpected behavior when using list.remove(x), wondering if
> someone can tell me what's going on.

All the replies so far seem fairly reasonable, but what about iterating
backwards:

for index in range(len(olist)-1, -1, -1):
    if condition: 
        del olist[index]

Are there any known problems with this idiom...?
Obviously it gets around the problem of duplicating the list.

	Paul.

-----------------------------------
Business Collaborator Team
Enviros Software Solutions
http://www.businesscollaborator.com
-----------------------------------



From thomas at xs4all.nl  Fri Jun 30 10:32:28 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 30 Jun 2000 14:32:28 GMT
Subject: Is this a leaker?
References: <GH075.17918$NP5.455672@newsread2.prod.itd.earthlink.net> <0S075.17928$NP5.457944@newsread2.prod.itd.earthlink.net> <slrn8lp6ul.u8u.thomas@centurion.xs4all.nl> <8ji745$1d5u$1@news2atm.raleigh.ibm.com>
Message-ID: <slrn8lpbra.v75.thomas@centurion.xs4all.nl>

On Fri, 30 Jun 2000 09:26:59 -0500, Arint? <n at n.com> wrote:
>> This depends on how you document your function :) If you document it as
>> returning a borrowed reference, the caller should not DECREF the returned
>> value. If you document it as returning a new reference, it should take
>care
>> of the DECREF itself. In this case, it doesn't make sense to return a
>> borrowed reference: If you do a DECREF on the fresly-created
>> Py_BuildValue()-object, it'll end up destroyed !

>I was speaking of returns to python scripts.  All of the 'return
>Py_BuildValue()' return to python scripts.  I don't have to decref in py
>script files do I?

Nope, you don't. Python scripts (or rather, the bytecode interpreter)
automatically DECREF everything they lose a reference to, DECREFing objects
left and right. Note that this means you should never return a borrowed
reference to a Python caller ! If you do, you might end up with a dangling
pointer yourself.

Thomas.


From mfletch at tpresence.com  Tue Jun 27 12:35:23 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 27 Jun 2000 12:35:23 -0400
Subject: determining an assigned IP address
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33F2@VRTMAIL>

def getLocalHostIP( remote = ("www.yahoo.com", 80), forceCompute=0):
	'''Get the "public" address of the local machine, i.e.
	that address which is connected to the general internet.
	
	Code by Donn Cave, posted to comp.lang.python'''
	try:
		if not forceCompute:
			return LOCALHOST_PUBLIC_IP
	except NameError:
		pass
	import socket
	s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	s.connect( remote )
	ip, localport = s.getsockname()
	s.close()
	global LOCALHOST_PUBLIC_IP
	LOCALHOST_PUBLIC_IP = ip
	return ip

HTH,
Mike

-----Original Message-----
From: Jonathan Epstein [mailto:Jonathan_Epstein at nih.gov]
Sent: Monday, June 26, 2000 10:10 AM
To: python-list at python.org
Subject: Re: determining an assigned IP address


The following Perl code works:

  use Sys::Hostname;
  use Socket;

  $ip=gethostbyname(hostname());
  $theip=inet_ntoa($ip);


It should be relatively straightforward to convert this to Python.

- Jonathan


Bill Tolbert wrote:
> 
> I've just set up Linux with my cable modem and I'd like to use Python to
> help me launch remote X apps. I need to know the ip address assigned by
> my isp, but so far I'm striking out. gethostbyname_ex in the socket
> module returns the address I supplied when I thought it was static.
> Right now the only way I know to get the right address is to log into
> another machine and use who.
> 
> All help is appreciated!
> 
> Bill
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
-- 
http://www.python.org/mailman/listinfo/python-list



From offer at sgi.com  Wed Jun 28 22:18:37 2000
From: offer at sgi.com (Nhi Vanye)
Date: 29 Jun 2000 02:18:37 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194> <395A9DFD.1C6C508A@roguewave.com>
Message-ID: <8jeblt$3ufed$1@fido.engr.sgi.com>

$ from bjorn at roguewave.com -#96744 | sed "1,$s/^/> /"
>
>
>Well, I get the problem too now (bang my head hard enough and I'll
>eventually understand ;-)

I agree, it suddenly clicked when I read an example in another thread.

>
>The relative import feature probably should be "shot in the head", but
>since it's in use (and presumably useful), how do we go about loading
>the global modules?
>
>The current suggestions are "import magicroot.os" and "import .os". I
>don't like either, but would probably prefer the first since it is more
>explicit...

Well I don't like the text 'magicroot', but the logic is valid. 

global would probably be reasonable magicmarker, as its a keyword it
can never be a valid module, and it fits with the concept.

So my vote goes for 

import global.os

:-)


>
>-- bjorn
>

richard.
-- 
Richard Offer           Widget FAQ --> http://reality.sgi.com/widgetFAQ
MTS-Core Design (Motif)
___________________________________________http://reality.sgi.com/offer



From gmcm at hypernet.com  Tue Jun 13 08:21:52 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 13 Jun 2000 12:21:52 GMT
Subject: How many times inner functions are compiled ?
References: <960897680.1890969865@news.libertysurf.fr>
Message-ID: <8F525B8C7gmcmhypernetcom@199.171.54.195>

Jerome Quelin wrote: 

>def outer():
>    def inner():
>        print "spam"
>    inner()
>outer()
>outer()

>In the snippet code above, I call outer() two times. I would like to
>know how many times python will compile the inner function ? 

Every time you call outer().


- Gordon


From porter at et.byu.edu  Thu Jun 22 11:55:23 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Thu, 22 Jun 2000 09:55:23 -0600 (MDT)
Subject: curses input with threads 
Message-ID: <Pine.GHP.4.21.0006220954150.6540-100000@don.et.byu.edu>

I have a multi-threaded program whose interface is done with curses.  It
pretty much just runs on its own, and displays its progress on the
screen.  I want to add a pause function.  I tried creating a separate
thread that runs stdscr.getch() to catch the keyboard entries.  But when I
do that, the entire program seems to pause to wait on that
getch() function.  

Is there any way to get input without freezing the other threads?

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------







From amitp at Xenon.Stanford.EDU  Sun Jun 25 22:39:38 2000
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 26 Jun 2000 02:39:38 GMT
Subject: timbot 1.1?
References: <20000619171244.A5561@bigskytel.com>
Message-ID: <8j6fpa$h4p$1@nntp.Stanford.EDU>

 David Porter  <jcm at bigskytel.com> wrote:
| 
| 
| http://24.2.96.79:8080/timmy/timbot.html
| 
| This is just by chance, right?
| 

There's a picture of timbot here:

http://www.hotendotey.com/fans.htm


	- Amit
-- 
--
Amit J Patel, Computer Science Department, Stanford University
http://www-cs-students.stanford.edu/~amitp/


From gmcm at hypernet.com  Tue Jun 13 16:09:55 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 13 Jun 2000 20:09:55 GMT
Subject: Client with select.select()
References: <#F2#YuI1$GA.106@net025s> <8i36rt$42mq$1@nntp6.u.washington.edu> <emt7cXW1$GA.180@net025s>
Message-ID: <8F52A2D93gmcmhypernetcom@199.171.54.155>

artaA[r]TA wrote:

[select]

>Then you can check with the write-set and send some data.
>Or am I missing something?

select will tell you a socket is writable as long as the outbound network 
buffers aren't full. It has nothing to do with whether the other end wants 
to read. That's all.

- Gordon


From thomas at xs4all.nl  Sat Jun 24 12:25:12 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 24 Jun 2000 16:25:12 GMT
Subject: Module of built-in wrappers
References: <GXk45.1367$9W1.102607@news-west.usenetserver.com>
Message-ID: <slrn8l9o6u.elg.thomas@centurion.xs4all.nl>

On Thu, 22 Jun 2000 02:20:22 -0800, Matthew Cline <matt at nightrealms.com> wrote:
>The Python FAQ gives an example of how to make a class wrapper around
>a builtin type, like a dictionary.  Is there any module that has
>class wrapers for every builtin type for which this can be done?

It can be done for all builtin types, but it's only logical for specific
types ;) (Subclassing a number isn't too functional, yet)

And yes, standard class wrappers are included with your Python. See the
standard UserList and UserDict modules, and the UserString module if you're
using something newer than Python 1.5.2 (like one of the 1.6 alphas.)

Thomas.


From edcjones at erols.com  Wed Jun 14 10:56:02 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Wed, 14 Jun 2000 10:56:02 -0400
Subject: Standalone http server, cgi software?
References: <3946EEA9.DFCBAB4D@erols.com> <394737F4.B3003678@stroeder.com>
Message-ID: <39479D02.42B5AD06@erols.com>

Michael Str?der wrote:

> Do you really need a stand-alone server? If yes, why? If you're
> looking for persistent web application programming you have to face
> several other problems like secure session managment, concurreny
> etc.

"standalone" turned out to be a good search keyword.

I am looking for a system that resides entirely on one PC. The idea is
to pass arguments to a Python program via a HTML form. The program
would look things up in a database, process them, etc. and return to
the browser a page of html containing the answers.

Thanks,
  Ed Jones




From embed at NOSPAM.geocities.com  Wed Jun 28 14:46:05 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 14:46:05 -0400
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu>
Message-ID: <aJr65.11287$227.233074@nnrp1.uunet.ca>

> I am *brand* new to python (as of Sunday), and I just got to the section
> in the tutorial about classes.  I was wondering why there really isn't
> such an idea as a "private" member of classes?  I understand that members
> can be hidden, but not really protected from the user.  I was wondering
> what the benefit of this would be?  Wouldn't it open the possibility of
> security holes?  Also, what benefit is gained by allowing a user to
> add/delete members from existing classes?  Wouldn't it be more desirable
> to create his/her own derived class?

The simplest way for me to put it is that it makes the Python language
simpler, easier to learn, and more "compact" in terms of remembering how to
program in it.  That freed up space can be used for thinking on higher
planes than I find I normally think on when programming in C. Just as there
is no way to declare a variable as an integer, or a function as returning an
integer, there is no way to make that member variable or member function
private, static, or a constant.  You might mourn that loss, but that shows
that you are transitioning still from thinking like a
statically-typed-language programmer to a dynamically-typed-language
programmer.

Now, sometimes an idiom replaces a keyword.   For example, if you want
something to be apparently private, one idiom is to append underscores to
the name, which in programmer-idiomspeak means "don't mess with this if you
know what's good for you". It could also mean (more open mindedly, mess with
this only if you really know what you're doing). In fact the latter is
actually a better way to build systems in a dynamically typed language like
Python.   Another way to make private members is to mess with __getattr__
and __setattr__.

The adjustment period from C or C++ involves a few "Do you really need X?"
type of  dilemmas. You'll find that there are ways to implement something
like X in Python, but that it is seldom worth bothering to make anything
hidden, in practice.

Warren





From mlh at idi.ntnu.no  Sat Jun 10 12:31:54 2000
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 10 Jun 2000 18:31:54 +0200
Subject: Stackless Python in Commercial Game Project!
References: <392E7C07.B456D9F@tismer.com>
Message-ID: <8htqhp$a0n$1@kopp.stud.ntnu.no>

Cool!

Just a question: What is the motivation for using Stackless Python as
opposed
to standard Python here? Threading? (Just curious...)

--

  Magnus       "It is a funny thing about life; if you accept to refuse
  Lie           anything but the best, you very often get it."
  Hetland                                            - Somerset Maugham





From kputland at servicemagic.com  Mon Jun 26 16:23:12 2000
From: kputland at servicemagic.com (Karl Putland)
Date: Mon, 26 Jun 2000 14:23:12 -0600
Subject: COM question
Message-ID: <004b01bfdfac$5a3672f0$8901010a@w3denver.com>

From: "Mark Hammond" <MarkH at ActiveState.com>
> "Klaus Doerfler" <klaus.doerfler at stud.fh-regensburg.de> wrote in message
> news:39535377.6E6E74FA at stud.fh-regensburg.de...
>
> > I'm writing a simple COM server and I want to use early-binding, so how
> > exactly does the server know about what the interface in the tlb-file is
> > like and how do I have to use it in the server? Are there any naming
> > conventions?
>
> Python doesnt directly support creating a TLB for a Python COM server - so
> your clients are forced to use late binding.
>
> Mark.
>

You could also search the pycom-dev archives for some experimental attempts
at automatically creating TLBs for Python COM servers.  Look in threads from
June-Aug of 1999.

http://mailman.pythonpros.com/mailman/listinfo/pycom-dev

--Karl Putland




From jkraska1 at san.rr.com  Fri Jun 16 02:23:01 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Fri, 16 Jun 2000 06:23:01 GMT
Subject: How does python invoke external C function?
References: <39498799.D28EDD46@esec.com.au>
Message-ID: <3949C868.F13EC826@san.rr.com>

Sam Wun wrote:
> 
> Hi,
> 
> I would like Python call an external C function ( ie. char
> *getString()),
> Does anyone knows how to do that?

You need to learn how to write a C external module for
python. There is information online at http://www.python.org,
but _Essential Python Reference_ is more thorough. Alternately,
if you're pretty good at C, you might get the python source
distribution and take a gander through a few of the objects.
The first one you should look at is xxobject.c, which is nothing
more than a skeletal implemenation of an external C module from
which you can start your efforts.

For one simple external C function, you should find it pretty
trivial, IMO.




C/


From mwh21 at cam.ac.uk  Tue Jun 20 10:12:47 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jun 2000 15:12:47 +0100
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl> <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu> <20000620143255.F26436@xs4all.nl>
Message-ID: <m3n1kg5tls.fsf@atrus.jesus.cam.ac.uk>

Thomas Wouters <thomas at xs4all.net> writes:

[snip] 
> def spam(x, y):
>     if x in range(1, y):
>         return 1
>     if x in range(y, 1):
>         return 1
>     return 0
> 
> Might not seem very useful, but it's just an example ;) This can
> easily be optimized, right ? We lookup 'range', it's a builtin, we
> store the address, and call it two times. Well, no. We dont *know*
> range a is a builtin, it could be a global that points elsewhere. If
> the 'spam' function contained a 'from module import *' or 'exec'
> statement, it could even be a local.
> 
> So, we have to check if range is still the same function. Which
> means we have to look it up and compare identity -- and we've
> already lost any advantage the cache could give us. Or we could
> place a marker on the __builtin__ module, and invalidate the cached
> entry of 'range' whenver 'range' is changed. But then we also have
> to add a marker to the global namespace, because something can *add*
> a range there.

One strategy occurs here is too assume that the global/builtin
namespace doesn't change all that often, so you compile `spam'
assuming that `range' refers to what it does at compile time, and then
if any function modifies builtins or the global namespace of `spam'
you throw away what you've compiled and start again (it's presumably
worthwhile to do this somewhat lazily, as the module namespace will
get repeatedly changed during mosule execution).

This strategy is used by MIT scheme (I think) and similar things are
done by most Common Lisps in their implementations of CLOS.

This would have the "advantage" of making functions which muck with
global namespaces howlingly expensive, which you might or might not
consider to be a good thing...

Hmm, implementation...

Cheers,
M.

-- 
  There are 'infinite' number of developed artifacts and one cannot
  develop appreciation  for them all.  It would be all right to not
  understand something, but it would be imbecilic to put judgements
  on things one don't understand.               -- Xah, comp.lang.lisp


From jeremy at beopen.com  Wed Jun 21 14:28:13 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 21 Jun 2000 14:28:13 -0400
Subject: "TypeError: read-only buffer, None" returned from file.write...
References: <MIS35.664$PQ4.13634@news1.online.no> <8ip2q3$2jrk$1@news.aros.net> <mu145.13877$R7.2335108@juliett.dax.net>
Message-ID: <6E1AD32AF87E515E.2A23DF4E6211A01A.00440780EBC1EDDA@lp.airnews.net>

It would be better, though, to test explicitly for None and not write
the result to a file when it is None.

sBuffer = m_XMLDoc.transformNode(xslDoc)

if sBudder is None:
	print "Error: transformNode returned None"
else:
	f = open( outFile, 'w')
	f.write(sBuffer)
	f.close()


Jeremy


From aahz at netcom.com  Thu Jun 22 12:41:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jun 2000 16:41:31 GMT
Subject: Classes: What is the standard trick for ...
References: <395235F9.94C0BBA1@erols.com>
Message-ID: <8itfjr$8de$1@nntp9.atl.mindspring.net>

In article <395235F9.94C0BBA1 at erols.com>,
Edward C. Jones <edcjones at erols.com> wrote:
>I would like to keep a collection of related functions together
>in a class:
>
>class functions:
>    def f1():
>        print 'f1'
>
>    def f2():
>        print 'f2'
>
>I would like a function call to look like "function.f1()". I
>would like to spare the user the need to create an instance of
>the class. Can I do this? How?

Use a module instead of a class.  I.e., create functions.py and require
the user to 'import functions'; then functions.f1() will work fine.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From wlav at hpatl26.cern.ch  Fri Jun  9 12:36:30 2000
From: wlav at hpatl26.cern.ch (Wim Lavrijsen)
Date: 9 Jun 2000 16:36:30 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <394113F2.8D6D0833@lemburg.com>
Message-ID: <8hr6ee$st8$1@sunnews.cern.ch>

"M.-A. Lemburg" <mal at lemburg.com> writes:

>reduces LOCs, scales well (from 5-100k LOCs
>if needed -- which is roughly equivalent to 25-250k LOCs
>in C++ from my own experience).

This is interesting. Could other people confirm this? Ie. that the
trade-off for Python is larger at small projects (factor ~5) than at
medium sized projects (factor ~2.5)?

And what is the cause of the difference?

Best regards,
      Wim Lavrijsen



From dsavitsk at e-coli.net  Thu Jun  1 12:41:25 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Thu, 01 Jun 2000 16:41:25 GMT
Subject: Calling Function in DLL on Windows
References: <dThZ4.101952$fF5.3223396@news1.rdc1.il.home.com><3935F38E.FF0061DC@sightreader.com> <xSu2lLAMsgN5Ewy0@jessikat.demon.co.uk>
Message-ID: <VmwZ4.102768$fF5.3267741@news1.rdc1.il.home.com>

> try Sam Rushing's windll/calldll stuff; should be findable at medusa
>
> Robin Becker

thanks for the suggestion

it looks like grabbing the calldll.pyd and windll.py files from sam rushing
at www.nightmare.com (both in dynwin.zip) will allow something like...

>>> from dynwin.windll import *
>>> # The DLL in qustion here is called
>>> # SFWLink.DLL. It is a connection to
>>> # a scanning engine for a NCS OMR Scanner
>>> mod1 = module('SFWLink') # Load's DLL
>>> mod1.LoadSFW('c:\\sfw\\', 'Param') # calls f(x)

This example is mostly from MHammond's Book where it is given about 1/2 of a
page.  He says to look at Sam's site.  Unfortunatly the documentation is not
really there for those of us who are new. MHammond also presents some
information on passing values ByRef.

As soon as i get to work with the hardware in question again (next week)
i'll put up a post on doing this, and how it works.




From eviltofu at rocketmail.com  Sat Jun 10 07:38:28 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 10 Jun 2000 11:38:28 GMT
Subject: Checking socket type?
Message-ID: <eviltofu-36614B.07382410062000@news.earthlink.net>

How would I check to see if an object is an instance of a socket?


From andyl at azaal.dircon.co.uk  Sun Jun  4 05:53:46 2000
From: andyl at azaal.dircon.co.uk (Andy Leighton)
Date: 4 Jun 2000 09:53:46 GMT
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net>
Message-ID: <slrn8jk9pa.mfl.andyl@azaal.dircon.co.uk>

On Sun, 4 Jun 2000 00:11:42 GMT, Ken McCracken <aa175 at torfree.net> wrote:
>Not that this exactly belongs in a python ng but I'm told the only usable
>onboard programming environment is quartus, an ansi forth system by Neal
>Bridges. 

Well LispMe! is pretty good too.  I would provide a URL but I don't have
it to hand.

-- 
Andy Leighton => andyl at azaal.dircon.co.uk
"... January is your third most common month for madness" - _Sarah Canary_


From jbauer at rubic.com  Sun Jun 11 09:40:27 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Sun, 11 Jun 2000 08:40:27 -0500
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21> <3942DC44.4A85EEA7@rubic.com> <39430A2E.99CE3FC@yahoo.com>
Message-ID: <394396CB.8D2D53F0@rubic.com>

Sam Wun wrote:
> I've tried to use command python build.py to build the 
> module, but it complains the _mysqlmodule.so is not found...
> 
> Do you know how to fix this problem?

As Dana mentioned, you may have to manually tweak some
of the paths in the Setup file.  I've got MySQLdb.py and
_mysqlmodule.so in my .../lib/python1.5/site-packages 
directory, which is where build.py is supposed to put
them.

Jeff Bauer
Rubicon Research



From jparkey at nospamrapidbs.com  Sat Jun 24 04:44:52 2000
From: jparkey at nospamrapidbs.com (John Parkey)
Date: Sat, 24 Jun 2000 09:44:52 +0100
Subject: HTML Parser? Sample Code?
References: <3952B827.5B96BE9C@mindspring.com>
Message-ID: <8j1sk2$b37$1@phys-ma.sol.co.uk>

Have you had a look at HTMLXtract? (search via Parnassus). It may save you
some work.

jp

"Nathaniel Leeds" <nmleeds at mindspring.com> wrote in message
news:3952B827.5B96BE9C at mindspring.com...
> Has anyone used the HTML parser.  I'd like to be able to skillfully
> disect html pages and put various types of information into lists and
> other objects using the tag structure of the HTML and not needing to
> write the stacking and pattern matching routines with regexp.  I'm new
> to python and am having trouble with the HTML parser documentation
> that's on line.  Does someone have some sample code I could look at?
>




From crystalc at interlog.com  Wed Jun 28 18:16:23 2000
From: crystalc at interlog.com (Brian Alexander)
Date: Wed, 28 Jun 2000 18:16:23 -0400
Subject: simple question ... exec in W32?
Message-ID: <Nxu65.20928$Ip.511269@cac1.rdr.news.psi.ca>

Hello;

How to I start a W32 application, like Word etc, from a Python script? I'm
actually trying to rewrite a backup script, which, additionally, does a lot
of non-backup processing overnight.

Many thanks,

Brian.




From shapr at uab.edu  Sun Jun 18 15:42:57 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sun, 18 Jun 2000 19:42:57 GMT
Subject: Python 3000 -> No, but there's 'R***'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <8iib7d$73f$1@news.jump.net>
Message-ID: <394D26D5.E69282A7@uab.edu>

Conrad Schneiker wrote:

> "Aahz Maruch" writes:

> > Y'know, those of you in the Ruby world who make a point of going around
> > trashing other languages are probably doing more to hurt the acceptance
> > of Ruby over the long haul than anything else. It's *rude*.

> Well, thanks anyway for your feedback, although it seems a little bit
> over-the-top to me.

I agree with Aahz, I don't think he's being over-the-top.

> > I was
> > almost interested enough in Ruby at one point to check it out, but I
> > have to say that the Attitude I've seen so far is even worse than the
> > Perlers.
> 
> Hmm. Even though I mentioned that I still use Perl from time to time, and
> made it clear that I thought Ruby was of value to some but by no means all
> people, and invited no response, one of the more rabid Perlers nevertheless
> accused me of trying to start a language war. I'm not sure if any Ruby
> poster has stooped comparably low--at least not yet. :-) But of course I
> would still never hold the posting of 1 or a few Perl hotheads against Larry
> Wall, or Perl, or the large number of otherwise civil Perlers.

You posted the first part of the Ruby FAQ to comp.lang.python at some
point in the last month or two. Would you be offended if I posted part
of the Python FAQ to comp.lang.ruby? However you answer, it doesn't seem
polite to me.

> If the attitude of 1 or 2 or 3 people keeps you from checking out Ruby, it
> seems to me that that sort of thinking amounts to sinking to the level of
> the people you loathe, plus it seems to overgeneralize quite a bit from a
> couple of recent data points. Obviously I don't want to encourage bad
> attitudes, but what I have seen doesn't seem comparable to Perl fanaticism.
> I happened to learn about Ruby on comp.lang.python last year in a series of
> what seemed to me at the time to be very civil discussions. None of the more
> recent postings has seemed notably less civil to me.

Possibly I'm being oversensitive as well, but I side with Aahz. The tone
of a user-community is important to me. I've found it harder to discuss
Perl civilly with people who use it than Python, my preferred language. 

> > It'd be real nice if the level-headed Ruby people restrained the
> > hotheads.
> 
> Well, I'm not sure we (or perhaps they, in your apparent opinion) would
> generally agree about who the hotheads are in this case :-) ... but in any
> case I still encourage anyone who thinks that Perl and Python have been
> unduly slighted to feel free to post their dissent on comp.lang.ruby. I'm
> sure at least _some_ people will be somewhat more careful about how they
> discuss Perl, Python, and Ruby as a result.

The point as I see it, is that comp.lang.python is not
comp.lang.recruiting. There are times when it's completely natural to
talk about Ruby on comp.lang.python. But, small irritations add up, and
repeated off-topic Ruby advocacy, even of a very small caliber, is not
what I want to read on comp.lang.python.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From pinard at iro.umontreal.ca  Tue Jun  6 14:52:57 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 14:52:57 -0400
Subject: Getting the module name from a method...
In-Reply-To: "Michael P. Reilly"'s message of "Tue, 17 Aug 1999 15:27:45 GMT"
References: <7pboln$n74$1@feed.teaser.fr> <Rbfu3.385$mV.141087@news.shore.net>
Message-ID: <oqwvk2bq0m.fsf@titan.progiciels-bpi.ca>

"Michael P. Reilly" <arcege at shore.net> writes:

> Olivier Deckmyn <olivier.deckmyn at mail.dotcom.fr> wrote:

> : I would like a method of my own to be able to print the module it is
> : declared in: ex: the text for the module : toto.py is :

> : class MyClass:
> :     def myMethod(self):
> :         print "I am in module", ????????

> : and then
> : MyClass().myMethod()
> : should produce :
> : 'toto'

> Unless the method was added as an instance attribute (self.method = ...),
> the global namespace should be the module where the method was defined,
> in this case "toto".  The less secure means would be globals()['__name__'],

In which way referring to `__name__' would not be secure?

> but you could also go with:

>   import os, sys
>   def mymodule():
>     try:
>       raise RuntimeError
>     except RuntimeError:
>       exc, val, tb = sys.exc_info()
>       code = tb.tb_frame.f_back.f_code
>       del exc, val, tb
>     if code.co_filename:
>       module = os.path.splitext(os.path.basename(code.co_filename))[0]
>     else:
>       module = '<string>'
>     return module

Why raise an exception?  Would not:

   import traceback
   frames = traceback.extract_stack()
   file_name, line_number, function_name, python_text = frames[-1]

give you enough information, while being of a much simpler writing?

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





From kaimar at ut.ee  Wed Jun 28 12:20:32 2000
From: kaimar at ut.ee (Kaimar Karu)
Date: Wed, 28 Jun 2000 18:20:32 +0200
Subject: Portal-like page using ZOPE
Message-ID: <395A25D0.1D603B9C@ut.ee>

Hi,

I'm working on a health education web page. The database for this
page is based on xml (and accessed through XMLDocument). I've created
the interface for displaying the content, and now I have to figure out
how to make the page portal-like.

The users must be able to log in and some of the pages content will
be dynamically created based on the user id. If the visitor has not
logged in, he can still see most of the content, but not all of it.

Besides the difference in content, the user must also be able to
create "bookmarks", which will be stored on the server and displayed
to the user when requested. And more, as there will be tests on the
web page (the idea of the page is to provide health education in
student's mother tongue (Estonian)), the results of these tests must
also be stored on the server.

There can be three types of users.

The first type is the 'ordinary user'. He can log in and have
bookmarks, but he cannot take part in any of the 'classes'.

The second type is the 'teacher'. He has the right to add users, and
these users will be placed 'under him'. He can manage their data and
see the results of their tests, and also the 'teacher' can create an
opening page for the students, where the topics and tests discussed
in the class will be listed. For the class, the teacher should be able
to disable all the other links on the page for the non-distraction.

The third type is the 'student'. This account can be created by the
maintainers of the site or by the 'teacher'. When the 'teacher' has
included the 'student' in a group for the moment, all of the content
which is seen by the 'student' can (but doesn't have to) be managed
by the 'teacher'. When the 'teacher' frees the 'student' from the
group, the content the 'student' sees is much alike the one the
'ordinary user' sees, except that the 'student' has the ability to
check his results and grades, if those are entered by the 'teacher'.

How can I achieve this using Zope?

TIA,
_______________________________________
Kaimar Karu | kaimar at ut.ee | 251 13 468
       = What a wonderful world =



From jean-louis.replacebylastname at ixiasoft.com  Wed Jun 14 10:34:54 2000
From: jean-louis.replacebylastname at ixiasoft.com (Jean-Louis Couturier)
Date: Wed, 14 Jun 2000 10:34:54 -0400
Subject: Problems with COM
References: <Wey15.3428$UM2.51434@weber.videotron.net> <M_B15.88$18.48718@pouncer.easynews.com>
Message-ID: <dJM15.4527$UM2.82286@weber.videotron.net>

> What does the SearchDocuments function look like ?
It is a function for our product called the TeXtML Server which stores and
indexes XML documents.  It has two arguments : a string (the query) and a
VARIANT (that should become a ResultSpace).  The  ResultSpace accesses a
twin object on the server.  This makes it improper to let a ResultSpace be
created out of the context of the SearchDocuments function.  The return
value of the function is a code indicating how the search went.

> Couldn't you just create the object in that function, and
> then return object.count to the calling function ?
>From what I understand, the object IS created in the function.  I just don't
know what parameter to give the function and get a resultSpace in return.

Jean-Louis Couturier
Developer
Ixiasoft

Roger Upole <rupole at compaq.net> wrote in message
news:M_B15.88$18.48718 at pouncer.easynews.com...
> What does the SearchDocuments function look like ?
> Couldn't you just create the object in that function, and
> then return object.count to the calling function ?
>
> BTW, resultSpace=win32com.client.Dispatch actually
> assigns the Dispatch function to resultspace, so you are
> passing a function rather than a COM object.
>
>                 Roger Upole




From akuchlin at mems-exchange.org  Wed Jun 28 17:36:54 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 28 Jun 2000 17:36:54 -0400
Subject: BXXP - Blocks eXtensible eXchange Protocol
References: <n1b65.7188$i5.105827@news1.frmt1.sfba.home.com>
Message-ID: <3d3dlxwkqx.fsf@kronos.cnri.reston.va.us>

"Alexander Gavrilov" <gavrilov at iname.com> writes:
> What do you think about BXXP? http://www.invisible.net

I was just reading the protocol spec this morning, and it looks
interesting.  I wish it didn't use XML for simple stuff like
exchanging profiles, which is overkill, and how to use the windowing
features in a protocol design is still unclear to me.  Started some
vague speculations about what a Pythonic API would look like, but I
should probably look at the Perl and Java APIs first.

--amk




From dana at mmi.oz.net  Thu Jun 15 11:12:42 2000
From: dana at mmi.oz.net (Dana Booth)
Date: 15 Jun 2000 15:12:42 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <8iab6a$idm$1@hunter.orchestream.com>
Message-ID: <8iarpa$gt6$2@216.39.141.36>

Tobe <tbradshaw at orchestream.com> wrote:

T: Lisp.. productive ??

Lisp is the default macro language of AutoCAD. The lead engineer at my shop
swears by it. She is very productive, hence her title as lead engineer. :)


-- 
-----
Dana Booth <dana at mmi.oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From wware at world.std.com  Thu Jun  1 13:37:07 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 1 Jun 2000 17:37:07 GMT
Subject: OT: too-tiny fonts in Netscape
References: <00de01bfcaf3$6236de90$0500a8c0@secret.pythonware.com> <39356936.7115883@news.telus.net> <39369663.495889@news.telus.net>
Message-ID: <FvHKxv.B5M@world.std.com>

This is nothing to do with Python, but since it came up...

David (invalid.address at 127.0.0.1) wrote:
> ... "far too small" is that the text is about
> four pixels high.

I actually run into this problem with Netscape pretty frequenly, on
my Solaris box at work. Is there any way to prevent this, perhaps to
set an option for "never use a font smaller than X, no matter what
the HTML says"? I think the version of Netscape is 4-something. Thanks.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From czellers at beopen.com  Tue Jun 27 10:28:27 2000
From: czellers at beopen.com (Clyde Zellers)
Date: Tue, 27 Jun 2000 10:28:27 -0400
Subject: Python  Article Announcement 
Message-ID: <002d01bfe043$f610f3c0$0101a8c0@one>

GUI Python: All This and Good Looks Too?
By Daniel Solin
June 27, 2000 

"Can I use Python for Web development?" "Can Python do serious programming?" "Is there anything Python can't do?". Since the answers to those three questions in 99% of the cases are "Yes, Yes and No", Python has probably earned its reputation as the coolest and most useful programming language today. In this series of articles, we'll show you how Python can be used for GUI development.

http://www.linuxdev.net/features/articles/ldpytk1_1.shtml 

from BeOpen.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000627/23855968/attachment.html>

From jamarijr at my-deja.com  Fri Jun  2 11:04:17 2000
From: jamarijr at my-deja.com (Arinté)
Date: Fri, 02 Jun 2000 15:04:17 GMT
Subject: Question on embedding. www.mcmillian-inc.com
Message-ID: <8h8id5$ooa$1@nnrp1.deja.com>

I just looked at that web page and saw a sample of embedding.  The
source is located here http://www.mcmillan-inc.com/embed2.html.  I am
not sure if I am reading this right, but it seems the c program runs
the script by calling "doit" in the *.py file.  How would you call a
*py file that didn't have any functions just the main thing.

some.py
----------------start file-------
print "junk"
----------------end file------------

How would you call that file using that source.

Thanks


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


From stephen at cerebralmaelstrom.com  Thu Jun  1 19:42:47 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 01 Jun 2000 23:42:47 GMT
Subject: Two questions on embedding :)
Message-ID: <XxCZ4.31420$WS3.364825@typhoon.we.rr.com>

    Last night I was fooling around and embedding Python into a little C++
Program I was working on, for no other reason then the fact that I had to do
some string manipulation that was a pain in C,...so, I spent God Only Knows
how long sticking Python into my program to make writing above said
manipulation easier. Somehow, I don't think the end effecti s Less Work at
all :)

    What I wanted to do was import two modules -- sys and languages -- the
latter of which was one that I wrote for the program. The latter provides
two functions: Load() and Translate(lang,string). Load returns nothing,
Translate returns the translated string.

Question#1:
    I ended up having to do PyRun_SimpleString("import sys, languages");
because I couldn't get the PyImport modules to work, at all :P

    PyImport_ImportModule("sys") didn't raise any exceptions but later calls
to PyRun_SimpleString("print sys.path") raised an NameError exception on
sys.

    I tried PyImport_AddModule("sys") afterwards, thinking from the docs
that mabye the above function didn't add it into sys.modules and this was
was needed.. still didn't work. I don't recall exactly what else I did, but
I tried about a dozen different combinations of commands and so on and so
forth until I just gave up and did PyRun_SimpleString("import sys");.

    Does anyone have an example program anywhere that incorporates importing
modules in a C++ level? Hrm. Actually, is there a need to bother? Is it any
better then just doing a PyRun_SimpleString?

    I ended up finding the solution to most of my problems by looking int he
'pysvr' demo but couldn't find a demo to show the above opperations.

Question#2:

    Later, I basicaly wanted to call: "languages.Translate('name','string')"
and have the result returned to the C program for processing. My code was
somewhat like the below:
----->
    PyObject *Result,*MainModule,*Globals;

    MainModule = PyImport_AddModule("__main__");
    Py_INCREF(MainModule); // since I'm storing it?

    Globals = PyModule_GetDict(MainModule);
    Py_INCREF(Globals);

    Result =
PyRun_String("languages.Translate('name','string')",Py_single_input,Globals,
Globals);
<-----
    Origionally, I was then moving on to do: if(PyString_Check(Result))
Output->Lines->SetText(PyString_AsString(Result));

    But I discovered that the if was never executing -- because
PyString_Check(Result) never returned true. I did a couple other debugging
checks, and it also didn't show up true as PyTuple_Check() or anything
else...
    The help files say that PyRun_String returns the result as a Python
object -- which it does.. but what KIND of python object? How do I get at
the result?

    To make it work, I ended up just appending a 'tmp =' in front of my
command, and for the assignment into Output->Lines I grabbed the text via
PyMapping_GetItemString(Globals,'tmp'); but that doesn't seem like the Right
Way To Do It. ;)


Sorry for the verbosity, but I wanted to include enough information that yuo
shuoldn't have to ask any questions.. but of course i'll forget that one
most important bit of information that you'll need so you'll have to ask
anyways.  Ah well. :)

--S




From pinard at iro.umontreal.ca  Sat Jun  3 21:32:23 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 03 Jun 2000 21:32:23 -0400
Subject: String.join revisited (URGENT for 1.6)
In-Reply-To: aahz@netcom.com's message of "3 Jun 2000 14:54:43 GMT"
References: <28052000.5@sanctum.jae.ddns.org>
	<39346FAC.F2797EA5@cosc.canterbury.ac.nz>
	<8h320n$6fo$1@newshost.accu.uu.nl> <03062000.1@sanctum.jae.ddns.org>
	<8hb67j$71h$1@slb6.atl.mindspring.net>
Message-ID: <oqvgzq44ew.fsf@titan.progiciels-bpi.ca>

aahz at netcom.com (Aahz Maruch) writes:

> So far, nobody has pointed out *ANY* disadvantages of retaining the
> current string.join() mechanism, with one exception: it destroys the
> "orthogonality" of moving all string functions to string methods.

I never really saw `string.join()' as a string method.  My guess why it
has been put there?  It was easy to explain `join' next to `split', given
they kind of complement each other, and have elegant argument similitude.

`string.split' was well decided, because it is clearly a method on strings
(on the first argument, _not_ the second).

`string.join' has been named and decided after its result, _not_ after
its arguments (and surely never after its second argument).  This was bad,
slightly confused, but bearable then.  Trying to propagate this old error
into a string class absolutely underlines it was a bad choice, and raises
horror in all souls having a bit of purity left! :-)

`string.join' is not, and has never been a string method.  That `join'
has been put into `string' for documentation convenience, nothing more.
We never, never decide how to sort methods by their result.  We always do
it by their most important argument.  The fact that the joining string
has been put second, and even optional, is another proof that it is not
the important argument (neither for `string.split', by the way).

Let's merely recognise the error, and correct it, instead of forcing ugly
intellectual stunts on everybody to justify it.  It dirties Python.

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





From gmcm at hypernet.com  Wed Jun 14 09:05:20 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 14 Jun 2000 13:05:20 GMT
Subject: [Q:] hash table performance!
References: <8i7kro$i04$1@nnrp1.deja.com>
Message-ID: <8F5352C34gmcmhypernetcom@199.171.54.194>

liwen_cao at my-deja.com wrote:

>I'm doing a project on large volume information processing. One of the
>tasks is to find out the duplicated files under a directory. I believe
>Python would be a good powerful tool for that(yes it is, I've
>implemented it in 40 lines of codes). However, performance IS a
>problem! Since I'm using directory as the hash table (hash_table={}...),
>I doubt the bottle neck is in the hash table: How can the generic hash
>table fits every cases?

Some of us have benchmarked Python's hash implementation against a bunch of 
other implementations. Haven't found anything that comes close.

>Is there a way of customize the length, hash algorithm of the hash
>table in python?  Or anyone can describe how the python hash table
>works.

Object/dictobject.c

>My way of doing it is simple: walk the directories and files, do MD5
>coding for every file, use the MD5 code as the hash key, insert the
>file name into the hash table. When two files has the same key, check
>the contents by bytes.

If you profiled the above (check out the profile module), you'd find that 
hashing is an insignificant part of the process.

- Gordon


From jelizondo at simpletel.com  Wed Jun 21 15:30:10 2000
From: jelizondo at simpletel.com (jelizondo at simpletel.com)
Date: Wed, 21 Jun 2000 19:30:10 GMT
Subject: Carrage Returns and Tabs in Windows Applications
Message-ID: <8ir53n$l5n$1@nnrp1.deja.com>

Does anyone know if you can code Python to automate Carrage Returns as
well as Tabs in Windows Applications?

Thanks....


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


From hinsen at cnrs-orleans.fr  Wed Jun  7 12:43:16 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 18:43:16 +0200
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain>
Message-ID: <m3hfb5h26z.fsf@chinon.cnrs-orleans.fr>

thor at localhost.localdomain (Manuel Gutierrez Algaba) writes:

> Python era un lenguaje casi perfecto, al menos hasta que empezaron
> a meterle tonterias sintacticas. Ahora estoy empezando a ver cual
> es el problema de python: los holandeses. Al parecer estan tan

No hay solamente holandeses qui hablan Holandes aqui!

> parida en Holandes. La verdad es que los holandeses son muy imaginativos:
> - Las cafeterias de drogas
> - El trafico de diamantes ( De Beers estara detras de Sierra Leona)
> - Tomates trangenicos. 
> ...

Pero tambien: Python, poffertjes, ...   ;-)
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From mhammond at skippinet.com.au  Sat Jun 10 00:05:37 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 04:05:37 GMT
Subject: windll
References: <3941ACC5.305B8830@videotron.ca> <3941B6B4.264FA7A1@san.rr.com>
Message-ID: <l8j05.833$cQ.10319@news-server.bigpond.net.au>

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:3941B6B4.264FA7A1 at san.rr.com...
> Jean-Yves Vinet wrote:
> >
> > Is there someting like to "windll" to use
> > dynamic librairy under Linux/Unix plateform.
>
> What does windll do? "Using" dynamic libraries
> under Unix is as easy as linking with them and
> then making certain that they are in your
> LD_LIBRARY_PATH.

Using them from Python is that easy?  Really?

windll tries to make using arbitrary DLLs, *from Python* that simple!

Mark.





From shecter at darmstadt.gmd.de  Thu Jun 15 11:50:14 2000
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 15 Jun 2000 17:50:14 +0200
Subject: Suggestion: "Completing" HTTP support in Python. (was Re: SSL and POST 
 with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com> <etdog5fcx7a.fsf@w20-575-109.mit.edu> <8hhttv$7r6$1@nnrp1.deja.com>
Message-ID: <3948FB36.416393C5@darmstadt.gmd.de>

Hi,

I was really suprised to see that HTTP client support in Python isn't
really quite complete.  What I mean is, I was hoping to find an
interface at a higher level, like this:

Wishful example #1
--------------
h = HTTP('my.host.name', '/loginscript', 8080)
h.add_parameter('username', 'Robb')
(code, msg, headers) = h.post()

Wishful example #2
--------------
h = HTTP('my.host.name', 8080)
h.add_header('X-local-hack', 'true')
(code, msg, headers) = h.get()

Notes: A black-box interface - no understanding of HTTP spec's details
(like how to quote parameters) is necessary.  A simplified and less
error-prone interface - no "endheaders" type call necessary. The
difference between GET and POST amounts to changing one word.

...without this kind of interface, I bet there's a lot of duplication
going on.  And, cookie support would of course also be great.  If
anything, as an option that could be enabled, where cookie information
would be managed by an HTTP object from invocation to invocation.

I've found bits and pieces of things that can do some of this (like
urlpost.py that I found on Parnassus, and cookie support from that
Python browser.)

I need this kind of interface because I'm working on an HTTP-based
distributed object project, and I've run screaming from Java...  I had
to implement vast parts of the HTTP library there, but there were
unending bugs that made the job a nightmare.

So - is anybody else interested in seeing a client interface that
allows the whole HTTP spec to be used, without getting down into the
mucky details?  I can see that I'll be working on this for sure.  I'd
be much more motivated if I knew that my work wasn't going to just sit
around.  I think this kind of interface should be in the core library.

Interested in comments,
- Robb



From gangli at msn.com  Tue Jun 13 10:28:08 2000
From: gangli at msn.com (gangli at msn.com)
Date: Tue, 13 Jun 2000 14:28:08 GMT
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no>
Message-ID: <8i5gd1$hm$1@nnrp1.deja.com>

In article <3946dc6b.193651816 at news.online.no>,
  thomas at cintra.no (Thomas Weholt) wrote:
> On Tue, 13 Jun 2000 16:39:23 +1000, "Jason" <it at onestopent.com.au>
> wrote:
>
>
> file = open("c:\\temp\\music.txt")
>
> And YES, the double backslashes are needed, or the \t in \temp is

You can use forwardlashes too, e.g.
file = open('c:/temp/music.txt')


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


From jbower at lims.com  Tue Jun 13 15:18:29 2000
From: jbower at lims.com (Jim Bower)
Date: Tue, 13 Jun 2000 19:18:29 GMT
Subject: Python and Windows Scripting Host
Message-ID: <9Ov15.1411$vM.56343@typhoon1.ba-dsg.net>

Just getting started with Python. Downloaded the basic distribution and the
Win32All distribution. Python appears to register OK (ran pyscript.py
separately just to verify).

However, I am getting "NameError: wscript" anywhere the wscript object is
mentioned. Have I missed something in the install?

Jim Bower jbower at lims.com





From cjc26 at nospam.cornell.edu  Sun Jun 11 20:27:37 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Mon, 12 Jun 2000 00:27:37 GMT
Subject: HELP with HTTPLIB/URLLIB
References: <sk7kt5ajh51169@corp.supernews.com>
Message-ID: <slrn8k8bmc.bdq.cjc26@localhost.mindriot.net>

* Alan Alper <al at findacorp.com> menulis:
| I need to duplicate the action of a browser POSTing form data to a server.
| GET is easy, but I don't know how to do POST. Any and all help, preferably
| with examples would be welcome.  Essentially, I'm trying to write a small
| agent to fetch classified ad data from several newspaper websites for me so
| that I don't need to visit the sites manually.  I have no clue how to do a
| POST.

urlopen() takes an optional second argument, which is the data to post.
For example:

urlopen("http://example.org/test.cgi", urlencode({name: "Cliff", id:
"4"})

Also, you might want to take a look at the library reference for urllib:
<http://www.python.org/doc/current/lib/module-urllib.html>


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From moshez at math.huji.ac.il  Mon Jun 12 13:07:46 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 12 Jun 2000 20:07:46 +0300 (IDT)
Subject: string.atoi() doesn't do negative base16 - why??
In-Reply-To: <newscache$pnp1wf$x97$1@jenpc07.jennic.com>
Message-ID: <Pine.GSO.4.10.10006122007070.15238-100000@sundial>

On Mon, 12 Jun 2000, Robert Cragie wrote:

> string.atoi('0xffffffec',0)

Or, just use longs and string.atol

Premature optimization and all that.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From scorder at incigna.com  Tue Jun 27 12:07:08 2000
From: scorder at incigna.com (Sam)
Date: Tue, 27 Jun 2000 12:07:08 -0400
Subject: Controlling Acrobat Reader with COM
References: <200006262159.AA3991339306@mail.mamey.com>
Message-ID: <s7khls0in7rrfbqnqdo4cha6chh9qqhstr@4ax.com>

On Mon, 26 Jun 2000 21:59:51 -0400, "Andres Corrada-Emmanuel"
<andres at mail.mamey.com> wrote:

>Hi,
>
>I'm trying to develop an application that will print (as in a paper copy) a PDF file automatically on Win32. I've used Mark Hammond's genpy utility to create a wrapper around "Acrobat Control for ActiveX". I'm able to get a handle on Acrobat but cannot execute any of its registered methods:
>
>>>> from win32com.client import Dispatch
>>>> acro = Dispatch('PDF.PdfCtrl.1')
>>>> acro.LoadFile(fileName='C:\Temp\ReadMe.pdf')
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>  File "C:\Program Files\Python\win32com\gen_py\CA8A9783-280D-11CF-A24D-444553540000x0x1x3.py", line 30, in LoadFile
>    return self._oleobj_.InvokeTypes(0x2, LCID, 1, (11, 0), ((8, 0),),fileName)
>com_error: (-2147418113, 'Unexpected failure', None, None)
>
>I'm very much a COM newbie, so this traceback means very little to me. Could someone enlighten me on what is going wrong in this case? I have no problem controlling Word with my setup so I'm stumped on this one.
>
>Andres Corrada
Try typing these strings directly into the interpreter.

>>> print 'c:\test\readme.pdf'
c:	est
eadme.pdf
>>> print "c:\test\readme.pdf"
c:	est
eadme.pdf
>>> print 'c:\\test\\readme.pdf'
c:\test\readme.pdf
>>> print 'c:/test/readme.pdf'
c:/test/readme.pdf


From aa8vb at yahoo.com  Tue Jun 27 13:12:59 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 27 Jun 2000 13:12:59 -0400
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
In-Reply-To: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>; from
 tuttledon@hotmail.com on Tue, Jun 27, 2000 at 01:10:46PM +0000
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <20000627131259.A119035@vislab.epa.gov>

Don Tuttle:
 |Here's the link
 |http://www.msdn.microsoft.com/vstudio/nextgen/technology/csharpintro.asp
 |
 |I look forward to you comments.

Thought I'd give it a read.  Their doc links:

     Introduction to C#
     C# Language Reference

are to DOS .EXE-wrapped .ZIP packages of MSWord DOC files.  Another
MSWin-only thing?  They're off to a good start.  Someone needs to introduce
Bill and MSResearch to PDF.  

Fortunately Sun StarOffice coupled with ps2pdf will get you there.

Up-top:
   |No part of this document may be reproduced, stored in or introduced into
   |a retrieval system, or transmitted in any form or by any means
   |(electronic, mechanical, photocopying, recording, or otherwise), or for
   |any purpose, without the express written permission of Microsoft
   |Corporation.
   |
   |Microsoft may have patents, patent applications, trademarks,
   |copyrights, or other intellectual property rights covering subject
   |matter in this document. Except as expressly provided in any written
   |license agreement from Microsoft, the furnishing of this document does
   |not give you any license to these patents, trademarks, copyrights, or
   |other intellectualproperty.

We're heading for a real open standard here.  They say I violated copyright
at least three times by downloading it to disk, unzipping it, and again by
printing it.

Enough time spent on this for now.

If there are any embrace-and-extend goodies in C#, I have faith that
someone in the Python community will find them and bring them to us.

-- 
Randall Hopper
aa8vb at yahoo.com



From shapr at uab.edu  Thu Jun  1 20:17:29 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 02 Jun 2000 00:17:29 GMT
Subject: A Mountain of Perl Books + Python Advocacy
References: <20000405053920.20579.qmail@web2102.mail.yahoo.com> <slrn8hdj3i.47b.tsummerfelt1@co146453-a.kico1.on.wave.home.com> <ur9bc7i3p.fsf@ctwd0143.fitlinxx.com> <slrn8hggff.lv.tsummerfelt1@co146453-a.kico1.on.wave.home.com> <39184F02.E6AD745D@basho.fc.hp.com> <c3qdjs8c1msmr6i8hjp7p5fo862r15nu7k@4ax.com>
Message-ID: <3937035C.EA99CB2B@uab.edu>

Chris Stromberger wrote:

> >I would invite everybody to read Eric Raymond's article: "Why Python?"
> >Honestly, I expected to see comments about it already in this news group
> >. . . but maybe I was just late, and missed the thread.

I've seen it in the special issue of the Linux Journal that is dedicated
to Python, I don't know if it's online.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From emile at fenx.com  Sat Jun 17 14:00:35 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 17 Jun 2000 11:00:35 -0700
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com>
Message-ID: <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com>

----- Original Message -----
From: "Stephen Hansen" <stephen at cerebralmaelstrom.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Saturday, June 17, 2000 8:51 AM
Subject: Re: Why whitespace denotation of blocks doesn't
work.


>     I discovered Python a bit over a year ago, and have
been using it
> relatively constantly in idle projects. Some are simple,
some are not.
> Currently i'm working on one that's not.
>
>     Initially, the indentation thing struck me as weird.
My main complaint
> about it is not that it's difficult to maintain proper
indentation, but that
> if you're scrolled far down in nested 'if', for instance,
it's sometimes
> hard to see which indentation-level is for which 'if'...
but then i'll just
> split-my-screen and scroll up.
>
<snipped>

Which is why I prefer an editor that shows whitespace
and use tabs.  That makes it easy to follow the column
of tab marks down the screen.

Emile van Sebille
emile at fenx.com
-------------------






From usenet at inter-slice.com  Sat Jun  3 11:33:38 2000
From: usenet at inter-slice.com (James Felix Black)
Date: Sat, 03 Jun 2000 10:33:38 -0500
Subject: Erm... MacOS X?
References: <eviltofu-1AEA18.02454303062000@news.earthlink.net>
Message-ID: <usenet-6DC18E.10333803062000@news.visi.com>

In article <eviltofu-1AEA18.02454303062000 at news.earthlink.net>, Jerome 
Chan <eviltofu at rocketmail.com> wrote:

> Has anyone compiled Python for MacOS X?

Yes.  It builds right out of the box.  The only trap is if you build on 
an HFS+ as opposed to UFS volume; then, you have to be careful about 
case (HFS+ is case-insensitive but preserving.)

(jfb)


From paul at prescod.net  Wed Jun  7 03:16:54 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 07 Jun 2000 02:16:54 -0500
Subject: Python <--> XML data binding.
References: <8hhlr6$eoh$1@slb3.atl.mindspring.net> <393C6944.2FBCE9B1@prescod.net> <8hkhge$j93$1@nntp9.atl.mindspring.net>
Message-ID: <393DF6E6.14B80B9C@prescod.net>

Werner Fouche wrote:
> 
> Paul,
> 
> I should have mentioned that Breeze reads the extended DTD specification
> called DT4DTDs developed by Extensibility -- their XML Authority Editor
> can convert many XML schema language dialects to DT4DTDs. DT4DTDs
> extend standard DTDs with the data types found in W3C XML Schema
> language specification.

I'm actually pretty familiar with DT4DTDs.  :)

http://www.w3.org/TR/dt4dtd

Let me just say in my defense that I didn't have that much control over
the design!!

> The biggest benefit of a tool such as what you are developing would be to
> quickly and reliable translate industry standard XML document specifications
> (say OFX 2.0) to crisp Python code! I am looking forward to the fruits of
> your current endevour.

I want you to keep looking forward to it but I also want to tell you
that I'm not particularly interested in generating Python code. In the
vast majority of cases code generators are hacks intended either to get
around a lack of dynamicity in a programming language (not a problem
with Python) or performance problems (which in Python's case, a code
generator would likely make worse, not better). 

Motto: Code generators are hacks. Sometimes necessary hacks, but hacks
nevertheless.

Let's say you take OFX 2.0 and want to map it into objects. It is
perfectly natural in the Java world to generate class definitions
because working with dynamically typed data in Java is so incredibly
painful that otherwise you would be tearing your hair -- nay -- your
skin off. But in Python, working with dynamically typed information is a
joy. So rather than generating code I'll just make the appropriate
object for you when you need it. No recompiling in advance required.

As far as keeping the "DTD and code" in sync, that can only go so far.
If you change your DTD whatever is your code depends on the DTD's
structures will break. In Java, it will break violently because of the
static type checking. My goal is to mitigate the breakage by providing a
mapping layer between the DTD and the objects so that when the DTD
changes you change the mapping, rather than the code. Something has to
break but fixing a mapping is easier than fixing code. It's like blowing
a fuse instead of burning out your computer!

Hopefully Breeze is intelligent enough to do that too, otherwise it will
be more headache than it is worth.

If the mapping process needs to be optimized for performance for some
particular application THEN it might be time to look at a code
generator. Optimization is an appropriate task for hacks.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Alabama's constitution is 100 years old, 300 pages long and has more
than 600 amendments. Highlights include "Amendment 393: Amendment of
Amendment No.  351", "Validation of Laws Regulating Court Costs in
Randolph County", "Miscegenation laws", "Bingo Games in Russell County",
"Suppression of dueling".  -
http://www.legislature.state.al.us/ALISHome.html



From jeffrey at digicool.com  Mon Jun 26 11:34:07 2000
From: jeffrey at digicool.com (Jeffrey P Shell)
Date: Mon, 26 Jun 2000 11:34:07 -0400
Subject: Darwin (Not Python, yet...)
References: <Pine.A32.3.90.1000624142526.18059A-100000@elvis.med.Virginia.EDU>
Message-ID: <8j7skg$1f$1@bob.news.rcn.net>

>   What's the state of Python on Darwin?
>   Any one do a build yet with X11, Tkinter, ( or WxWindows or Qt or ... )?
>   Any tips?
>
>   Darwin and OSX integration: Is this close enough to make one version
>     (with DLLs to support all of the options, for example, if you're
>      running Darwin, you may have X11, but you don't have Carbon, etc.)

Is there work being done on this?  Is Objective Everything running again?

There are so many possible API's in MacOS X for Python, that there could (in
theory) be three pythons targeted at different markets::

Darwin / BSD -- for command line folks (should work the same in Darwin and
MacSO X.  Should support threading && shared libraries easily so that Zope
can be more easily configured against it!).  BTW:::::::  As of Python 1.5.2,
building Python on an HFS+ partition is nearly impossible.  HFS+ is not case
sensitive, and the Python build directory conficts with the built 'python'.
'make install' subsequently fails.  I've also had odd experiences with fcntl
on my most recent attempt at building (I was running DP4 of MacOS X, but
building on the UFS (case sensitive file system) partition that I have MacOS
X Server installed on.).

Integration with Cocoa -- Actually, JPython could work here.  Using JPython,
i was able to import the com.apple.yellowbox.* packages which are the Cocoa
API's for Java and play around a little bit.  Everything covered in Apple's
Cocoa reference for Java is usable (naturally) in JPython.  Pretty cool.
Cocoa is much easier to use for scripting the MacOS X interface than
AppleEvents in Carbon.

Integration with Carbon -- This could be considered the MacOS X version of
MacPython, which has toolbox support, AppleEvents, and other MacOS (7-9)
specific features && IDE.


It would be nice to have one Python that could load and unload bundles to
support all of these features, but I imagine that would take a lot more work
than most people have time for.  I know I'm nowhere near smart enough to
even start trying to grokk it.

But a really good Python at the Darwin/BSD level would be great for Zope.
I've managed a couple of times to get it built in a way that Zope likes
(until today's 'fcntl' oddness.  hellIfIKnowWhatThatMeans()).

I have a document up on zope.org that talks about building Python 1.5.2 for
Zope usage.  I'm *begging* anyone smarter with Darwin and Python to try to
address the issues that i've had.  They should be documented on the page.

http://www.zope.org/Members/jshell/buildingZopeOnMacOSXServer

multigratzi, jshell




From richard_chamberlain at ntlworld.com  Fri Jun 30 17:21:45 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 30 Jun 2000 22:21:45 +0100
Subject: Are tk derived objects ever destroyed?
References: <962360149.1278823422@news.libertysurf.fr>
Message-ID: <5b875.1088$Tb2.20649@news6-win.server.ntlworld.com>

Hi Jerome,

This is a tricky question ;-)

Unless you explicitly call destroy against your class it doesn't seem to
call ___del__. You can get round it in the following manner:

#!/usr/bin/python
from Tkinter import *
class Frm(Toplevel):
    def __init__(self):
        Toplevel.__init__(self)
        Label(self, text='spam').pack()
        self.bind('<Destroy>',self.kill)
    def kill(self,event):
        self.destroy()
    def __del__(self):
        print 'Kai kai kai'
Button(text='create', command=Frm).pack()
mainloop()

Richard


Jerome Quelin <jerome.quelin at insalien.org> wrote in message
news:962360149.1278823422 at news.libertysurf.fr...
> Hi all,
>
> Look at the snippet code below:
>
> -- File toplevel_creator.py
> #!/usr/bin/python
> from Tkinter import *
> class Frm(Toplevel):
>     def __init__(self):
>         Toplevel.__init__(self)
>         Label(self, text='spam').pack()
>     def __del__(self):
>         print 'Kai kai kai'
> Button(text='create', command=Frm).pack()
> mainloop()
> -- End of file
>
> When running this code and creating some toplevel windows with the
> create button, the objects aren't destroyed (the __del__ function
> doesn't get called) when I destroy the window.
> I wonder why, since there aren't any other references to this toplevel
> window. Unless Tkinter keeps tracks of all widgets created? Or
> destroying the widgets does not clears all references to the widget?
> Then what must I do to destroy the object when the window gets
> destroyed?
>
> Thanks,
> Jerome
> --
> jerome.quelin at insalien.org
>




From MarkH at ActiveState.com  Thu Jun 22 19:23:31 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thr, 22 Jun 2000 23:23:31 GMT
Subject: Labels of remote volumes under NT
References: <2B1262E83448D211AE4B00A0C9D61B0301301218@msgeuro1.creo.be>
Message-ID: <TDX45.1837$SN2.5220@NEWS-SERVER.BIGPOND.NET.AU>

Almost all of your answers can be found with a quick deajnews search:

* List of processes: win32pdhutil

* List of devices: No Python functions, look in the registry

* VersionStamp: No - there was a rumour about one, but it never appeared.

* GetPrinter - Use the source, Luke.

* CurrentUser - win32wnet, win32api.GetCurrentUser, etc

* Labels mapped drives, something in win32api, otherwise win32net.

Mark.

--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!
"Pieter Claerhout" <PClaerhout at CREO.BE> wrote in message
news:2B1262E83448D211AE4B00A0C9D61B0301301218 at msgeuro1.creo.be...
> > Hello,
> >
> > is it possible with Python under NT, to find out the type and the
labels
> > of the drives of a remote computer?
> >
> > Kind regards,
> >
> >
> > Pieter
> >
> > Pieter Claerhout
> > Application Support - CreoScitex Europe
> >
>




From ax at infocentr.ru  Fri Jun 16 14:52:19 2000
From: ax at infocentr.ru (Aleksey Kishkin)
Date: Sat, 17 Jun 2000 00:52:19 +0600
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com>
Message-ID: <394A7763.DE2A771D@infocentr.ru>

Akira Kiyomiya wrote:
> 
> I wonder if Python runs faster than Perl.
> 
> Could someone give me some comparison and contrast with some languages such
> as Perl, Tcl, JavaScript, etc?
> 
> I am just curious.
> 
> Thanks
> 
> Akria

I did. I found the programm (sieve) for java and rewrote this
programm to all languages which I had at that moment.
It worked 10 seconds and printed how much this task was solved.

(it means bigger amount is better)

TCL  - 13 times / 10 sec 
perl - 64
python - 36
C without optimization - 6468
C with optimization  - 19111
Java without JIT (1.1.7) - 921
Java with JIT (kaffe)  - 8230
pascal (fpc)   - 4451
 
Well. that's all.


From sandysj at asme.org  Thu Jun 22 18:37:33 2000
From: sandysj at asme.org (Jeff Sandys)
Date: Thu, 22 Jun 2000 22:37:33 GMT
Subject: Countdown
Message-ID: <3952952D.88E3588D@asme.org>

Python is really elegant when counting up:

for i in range(len(thing)):
    print thing[i]

But when counting down looks ugly:

for j in range(len(thing)-1,-1,-1):
    print thing[i]

Advice so far:
Stop doing things backwards:
    (customer request)
Use reverse first:
    (destructive, need to unreverse after)
Live with it:
    (bytecode doesn't care, but it is still ugly)
Try range(len(thing)).reverse():
    (doesn't work but):
        rthing = range(len(thing))
        rthing.reverse()
        for k in rthing:
            print thing[i]
    (works, but also ugly)


From thomas at xs4all.net  Thu Jun 15 14:05:57 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 15 Jun 2000 20:05:57 +0200
Subject: Python Productivity over C++
In-Reply-To: <20000615195014.A12612@fuf.sh.cvut.cz>; from M.Vitecek@sh.cvut.cz on Thu, Jun 15, 2000 at 07:50:14PM +0200
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com> <20000615195014.A12612@fuf.sh.cvut.cz>
Message-ID: <20000615200557.K26436@xs4all.nl>

On Thu, Jun 15, 2000 at 07:50:14PM +0200, Michal Vitecek wrote:

[ access control ]

> >Look at the big picture.  Do encapsulation, access control and type
> >strictness really make C++ really less error prone than python?

> >I think not.

>  but what are the method names with '_' at the beginning in modules for?
>  aren't they there as a kind of hack to implement some way of access
>  control?

No, they are a form of reducing namespace clutter and name-clashing.

Variable names starting with a '_' do not get imported into the local
namespace when you use 'from <module> import *'. They are accessible just
fine, they just do not get bound to a *local* (to the 'from module import'
statement) name.

Class variables (including methods) starting with a double _ get 'mangled'
to make them private -- so you can store 'class-related' data without the
risk of a subclass accidentily shadowing it. The class-related data is still
very accessible, it's perfectly valid to say 'self._BaseClass__private_data'
or some such.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From erno at iki.fi  Tue Jun 27 08:07:02 2000
From: erno at iki.fi (Erno Kuusela)
Date: 27 Jun 2000 15:07:02 +0300
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk> <8imc1f$181$1@slb6.atl.mindspring.net> <394f2d55.0@katana.legend.co.uk> <8inrg6$mm2$1@slb7.atl.mindspring.net>
Message-ID: <kuem5juy3d.fsf@lasipalatsi.fi>

>>>>> "Aahz" == Aahz Maruch <aahz at netcom.com> writes:

    Aahz> But it's not performing any action on test or test2; it's
    Aahz> performing an action on the path /test/test2.  I believe it
    Aahz> actually uses the OS call stat(),

yep, strace says..

stat("/tmp/dir-with-mode-000/subdir", 0xbffff63c) = -1 EACCES (Permission denied)

ie the stat call is reporting a failure to get information on that
path.

    Aahz> but that doesn't really matter; the point is that whatever
    Aahz> internal function gets called on the *entire* path returns
    Aahz> an error, therefore the path cannot be a directory.

but it is indeed a directory - the os is just telling the you that 
you are not permitted to have any information on that filesystem object
(including whether it exists).

  -- erno


From digitig at cix.co.uk  Tue Jun  6 18:20:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Jun 2000 23:20 +0100 (BST)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <393CCEE6.E5C1C267@marconicomms.com>
Message-ID: <memo.20000606232055.11341E@digitig.compulink.co.uk>

In article <393CCEE6.E5C1C267 at marconicomms.com>, 
Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen) wrote:

> b) case-sensitivity with the awareness of same names. Just allow
> 
> >>> Orange = 1
> >>> Orange = 2
> 
> but throw an exception when
> >>> orange = 3
> or
> >>> ORANGE = 4 etc.

I consider that bad behaviour. I have said elsewhere that there are 
programming idioms I consider to be good practice that use case to 
distinguish objects. I'd have no objection to an editor that pointed out 
to me when I was doing it, but I would consider exceptions to be a serious 
penalty for no real advantage. Are we going to make C/C++ 
and Java case-insensitive, too? Or are these learners going to have to 
come to terms with the fact that "a" <> "A" at some point, and the sooner 
the better?

> a more difficult task is to recognize spelling mistakes:
> >>> ornge = 2
> >>> oragne =3
> >>> orange_colour = "orange"

The trouble with that is, having been forbidden the use of canvas after a 
from Tkinter import * (clashes with "Canvas"), now the confounded thing is 
going to assume aCanvas is a typo, amend it to Canvas and throw it out! 
We'd end up with /much/ too restricted a set of identifiers. As a Brit I 
have enough trouble with color/colour, but I emphatically do /not/ want 
the compiler to tell me I can't use my native spelling of "colour".

"Corrected in place" is a disaster in waiting, by the way. Did anyone here 
ever use Comal? That "corrected" typo's in place, but as it didn't know 
what the programmer intended it basically changed to random code. The 
result was that it replaced error messages with hard-to-find bugs.

> When programming or reading the input of a non-experienced programmer
> who might think  012 is decimal 12 the input will be interpreted as
> an octal number. I don't have any solution for this. Maybe there should 
> be
> a routine reading from stdin which doesn't interpret 012 as octal 
> number...

A Python-aware editor can help with this. I set my C++ editor to highlight 
octal in a different colour to decimal, which makes it easy to spot.


From embed at NOSPAM.geocities.com  Wed Jun 28 13:18:38 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 13:18:38 -0400
Subject: what is lambda?
References: <8jd8uu$gnh$1@nnrp1.deja.com> <8jdaht$i43$1@nnrp1.deja.com>
Message-ID: <erq65.11249$227.232310@nnrp1.uunet.ca>

>    You should note there's varying opinions on whether
> lambda is a good thing - if you're a beginner at programming
> as well as with Python it's not the first thing you should
> worry about.

It also (in my case) expanded the vocabulary I use internally (my own
pseudo-code-language) to include concepts that are not easily expressable in
less dynamic languages. In other words, lambda was a kind of
who's-programming-who keyword. Rather than expanding only the language, it
also opened up my brain a bit wider.

Warren




From loewis at informatik.hu-berlin.de  Tue Jun 27 04:57:22 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Jun 2000 10:57:22 +0200
Subject: Is this safe? (Making string comparisons ignore case)
References: <NqT55.6417$Gh6.421246@news-west.usenetserver.com>
Message-ID: <p6q4s6fzekt.fsf@informatik.hu-berlin.de>

Matthew Cline <matt at nightrealms.com> writes:

> I've done the following to make it so that all string comparisons
> are case insenstive.  Is this safe?  (I'm pleasently surprised that
> it actually works)

> def myCmp(self, other):
> 	return oldCmp(string.lower(self), string.lower(other))

That's quite incorrect, just try cmp(1,1). In general, replacing cmp
may have strange effects. Objects implementing __hash__ and __cmp__
typically implement them by delegating to hash() and cmp(). Such
objects would break when used as dictionary keys.

It is much better to use myCmp directly in the places where you need
it (and give it a different name, too!).

Regards,
Martin


From denalione at my-deja.com  Wed Jun 28 09:20:30 2000
From: denalione at my-deja.com (denalione at my-deja.com)
Date: Wed, 28 Jun 2000 13:20:30 GMT
Subject: Python Thread safety
Message-ID: <8jcu30$2nm$1@nnrp2.deja.com>

What is the current status of thread safety in Python?	I saw references to
Python not being thread safe but not what version was being discussed.


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


From akuchlin at mems-exchange.org  Mon Jun 26 10:19:25 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 26 Jun 2000 10:19:25 -0400
Subject: Encryption modules
References: <8j7lnq$9jq$1@nnrp1.deja.com>
Message-ID: <3dem5kwmmq.fsf@kronos.cnri.reston.va.us>

David Currie <davec999 at my-deja.com> writes:

> Hi.  Does anyone know where to get the cryptographic modules written by
> Andrew Kuchling that are referenced in the Python library
> documentation?  The link given for them is on Starship and seems to be

Try ftp://starship.python.net/pub/crew/amk/crypto .  The page was
moved, and the redirect to the new location isn't working for some
reason (Apache config, probably).

--amk


From ge at nowhere.none  Sat Jun 17 15:54:54 2000
From: ge at nowhere.none (Grant Edwards)
Date: Sat, 17 Jun 2000 19:54:54 GMT
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com> <slrn8kn881.sdl.nobody@isis.visi.com> <394BA666.E5F3B19B@uniserve.com> <15O25.1379$i06.308670@ptah.visi.com> <394BC4CD.6432CBBF@uniserve.com>
Message-ID: <iIQ25.1400$i06.313245@ptah.visi.com>

In article <394BC4CD.6432CBBF at uniserve.com>, Bob van der Poel wrote:

>> >>    Button(root, bitmap='queen')
>> >>
>> >> This is where the error occurs -- it doesn't matter if I use the name I
>> >> specified in the call to BitmapImage or let Tk generate one and retrieve it
>> >> with str().  What I can't figure out is how to use bobpic as the bitmap for
>> >> a Button.
>> >
>> >More from my snippet:
>> >
>> >       l = Label(w, image=bobpic, borderwidth=4, relief='sunken')
>> 
>> Yes, using it as the widget's image works, but I'm trying to
>> use it as the bitmap for the widget.
>
>Oh, ummm did you try:
>
>	 Button(root, bitmap=queen)

Yes, I did.  Does that work for you?

>without the quotes around the name? Remember, 'queen' is a string,

Yes, I know.  According to the documentation, the argument
value for the bitmap option is supposed to be a string
specifying a bitmap in a manner acceptible to Tk_Getbitmap.
Tk_Betbitmap accepts two types of strings:

    "@filename"   will read the data from a file.
    
    "bitmapname"  will use the previously defined bitmaped of
                  that name

>not the name of the variable where the data is stored. Also,
>don't let this bite you: if this is a function then the
>variable has to declared global otherwise when the function
>exits the variable (and the display) disappear.

Yes, I tried making it global and making it an instance
variable.  Neither one made a difference.

I am wondering if BitmapImage is somehow not adding the name of
the created bitmap to the global list of "known" bitmaps, or if
the name is somehow getting mangled so the name that Tk knows
about is different than the one Python presents to the user.

Does anybody know how to see the list of "known" bitmaps?

-- 
Grant Edwards                   grante             Yow!  .. Now KEN and BARBIE
                                  at               are PERMANENTLY ADDICTED to
                               visi.com            MIND-ALTERING DRUGS...


From msoulier at nortelnetworks.com  Wed Jun 28 11:36:53 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 28 Jun 2000 15:36:53 GMT
Subject: compilation
References: <8j84cf$6bj$1@bmerhc5e.ca.nortel.com> <slrn8lf95c.fu7.thomas@centurion.xs4all.nl> <8jahlv$rui$4@bmerhc5e.ca.nortel.com> <uya3q65n0.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8jd62l$bpb$1@bmerhc5e.ca.nortel.com>

In article <uya3q65n0.fsf at ctwd0143.fitlinxx.com>, David Bolen wrote:
>msoulier at nortelnetworks.com (Michael P. Soulier) writes:
>
>Ah, but the rub is how do you perform that check without actually executing
>the script, since its the execution of the script that may actually yield
>the availability of the function or variable :-)

	C seems to manage by finding such problems in either the compile or link
phases. I don't see why there can't be a compile switch for such checking.
Sure, you don't want it on all the time, that defeats dynamic binding. But for
the purposes of a compile pass, I don't see why it can't be done when it's
done all the time in C. 

	Mike

-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From adams_s at lab.eng.usyd.edu.au  Sun Jun  4 08:08:10 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Sun, 4 Jun 2000 22:08:10 +1000
Subject: User Input?
References: <cnj_4.3811$r7.919327@news1.rdc2.tx.home.com>
Message-ID: <8hdgge$ld9$1@news1.wire.net.au>

use raw_input

>>> def get_SPAM_n_EGGS():
                    spam = raw_input('Spam? ')
                    eggs = raw_input('and Eggs? ')
                    print spam, eggs


>>> get_SPAM_n_EGGS()
Spam? loads
and Eggs? yessiree
loads yessiree
>>>

hope that helps

Steven

Michael Morrison <borlak at home.com> wrote in message
news:cnj_4.3811$r7.919327 at news1.rdc2.tx.home.com...
> My friend was talking about a programming asignment and I was telling him
> about python, so we decided to make programs in our different languages
(his
> being QBasic :P) ...
>
> Anyway, it dealt with user input, and from only having read Learning
Python,
> I didn't know how to do it (and still don't).
> Is there an easy way to do user input in python?  I found a non-standard
> module, but it looked too complicated (not something I'd expect from the
> ease-of-use I've had so far).
>
> Example: python blah.py
> Name? (type in name)
> Address? (type in address)
> etc....
>
> Any help would be appreciated, thanks!
>
>
>
>




From bill_tolbert at bigfoot.com  Fri Jun 16 11:04:38 2000
From: bill_tolbert at bigfoot.com (CB)
Date: Fri, 16 Jun 2000 15:04:38 GMT
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <Pine.BSF.4.10.10006152355270.4382-100000@chi.pair.com> <8id9bm$ko3$1@nnrp1.deja.com>
Message-ID: <8idflo$poj$1@nnrp1.deja.com>

As a Python newbie and pair programming enthusiast, I'd love a chance to
try some virtual pair programming. I'd be open to something like VNC or
a shared editor. Hopefully I'll have a cable modem running by this
afternoon. That should help!

If you're interested let me know.

Bill in Durham



> Try VNC (http://www.uk.research.att.com/vnc)  The idea is that you
> share desktop between two computers.  This way you can edit,
> run/compile, and see program behaviour *on both screens* at the same
> time.  Computers can be different platforms (Mac/X/MSWin).
>
> VNC is free, GNU GPL, multi platform and can work even with modem
> connections (although that can be a slow and painfull experience if
> speed < 56k).
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From andyl at azaal.dircon.co.uk  Wed Jun 21 16:08:18 2000
From: andyl at azaal.dircon.co.uk (Andy Leighton)
Date: 21 Jun 2000 20:08:18 GMT
Subject: Cold Fusion and IBM Websphere using Python as CGI ?
References: <00ad01bfdb88$1a5e8320$8235a8c0@pc-p2457.patni.com>
Message-ID: <slrn8l285i.7mp.andyl@azaal.dircon.co.uk>

On Wed, 21 Jun 2000 19:23:39 +0530, Anil Singh <anil.singh at patni.com> wrote:
>This is a multi-part message in MIME format.

Yuck

>We have developed a Web application using Python as CGI on IIS on =
>Windows NT 4.0.
>The Application layer of our website uses C exe for various services =
>like=20
>database connectivity etc. We are trying to move to other webservers =
>specifically to
>Cold Fusion Server 4.0 or IBM Web Server 1.0 in near future. But we =
>don't want to=20
>change the CGI script language from Python. We would like to know =
>whether Python Scripts=20
>are compatible to Cold Fusion Server 4.0 & IBM IBM Web Server. If it =
>require changes=20
>what will be the nature of changes ?

Well Cold Fusion is not a web server, it is an application server.
It will only deal with CFML pages.  It is provided with an ISAPI
module which means that you will be able to use it with your
current web server (IIS).  I am not sure what you mean by IBM
Web Server 1.0 so I cannot comment.

Generally I am not sure why you want to move if your application
works OK now.  If you are going the CF route, you will have to
rewrite everything or you are wasting your money.

BTW pls fix your newsreader so that it doesn't add an html copy
of your message to the bottom of your posts.

-- 
Andy Leighton => andyl at azaal.dircon.co.uk
"... January is your third most common month for madness" - _Sarah Canary_


From herzog at online.de  Wed Jun 21 14:00:51 2000
From: herzog at online.de (Bernhard Herzog)
Date: 21 Jun 2000 20:00:51 +0200
Subject: apply for objects
References: <8icpdd$8c7@cs.vu.nl>
Message-ID: <m3lmzyykvg.fsf@greebo.nodomain.de>

Hahn S <shahn at kits.cs.vu.nl> writes:

> Ok, i have figured out how to use apply on methods. You just have to get
> the method from the class and use the im_func in the apply. The first
> argument must be the one which should be the self object.
> 
> Example:
> >>> class D:
> 	def start(self, hoi, daar=100):
> 		print self.aap
> 		print hoi
> 		print daar
> 
> 
> >>> d = D()
> >>> d.aap = 'Artis is leuk'
> >>> apply(D.start.im_func, (d, 'Hello World'))
> Artis is leuk
> Hello World
> 100

>>> apply(d.start, ('Hello World',))
Artis is leuk
Hello World
100

Works just as well. There's no need to mess around with im_func.

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


From michael at stroeder.com  Sun Jun 11 09:02:18 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 11 Jun 2000 15:02:18 +0200
Subject: Will logging system be useful for python programmers ?
References: <39437BAA.2DC4CE90@mail.ru>
Message-ID: <39438DDA.4DD32FCF@stroeder.com>

"Igor E. Poteryaev" wrote:
> 
> Will logging system be useful for python programmers ?

Yes.

Best would be if it's added to the standard Python distribution in
the future.

Ciao, Michael.


From petersc at stud.ntnu.no  Fri Jun 16 02:29:44 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 16 Jun 2000 08:29:44 +0200
Subject: Python segfault
References: <slrn8kits2.ggk.perly@typhoon.xnet.com>
Message-ID: <3949C958.D3AA1F4@stud.ntnu.no>

Pat Szuta wrote:
> 
> I don't know if anyone has seen this,  but python segfaults when given a
> rather large input.
> 
> [pat at perly pat]$ perl -e "print '2+2+' x 10000" > input
> [pat at perly pat]$ python < input

Strange. Looks like the limit is exactly at 8192. That makes 16384
operations the parser (or whatever crashes) can hold.

Not that the problem is not the length of the input itself:

$ perl -e "print '22222222 ' x 200000 > input
$ python < input

works just fine. I think the problem is the add (or mul) operation.
So the problem might be in the expression parser.

Is there anyone with some more insight around?

else-I-will-have-to-check-it-myself-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From digitig at cix.co.uk  Fri Jun 16 19:34:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 17 Jun 2000 00:34 +0100 (BST)
Subject: Python Productivity over C++
References: <3949A5DB.5D4ECFCA@sightreader.com>
Message-ID: <memo.20000617003446.42607D@digitig.compulink.co.uk>

In article <3949A5DB.5D4ECFCA at sightreader.com>, kens at sightreader.com (Ken 
Seehof) wrote:

> This is a multi-part message in MIME format.
> --------------241ADB457183291FAD22762F
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
> 
> Actually C++ doesn't have true access control either.  Just insert 
> before your
> includes:
> 
> #define private public
> 
> . . . well, almost ;-)

More effectively, declare a class with the same structure as the one you 
want to access, but with everything public, then cast the instance of the 
original class to your type. It will almost certainly work, and Stroustup 
says something to the effect that you deserve everything you'll get...


From rotten at colltech.com  Fri Jun 16 15:00:05 2000
From: rotten at colltech.com (Richard Otten)
Date: Fri, 16 Jun 2000 14:00:05 -0500 (CDT)
Subject: install gone awry
Message-ID: <200006161900.OAA17626@psasolar.colltech.com>

Hello Python gang,

  I have installed Python literally dozens of times.
  For some reason my most recent intall has had something go wrong with it.
I noticed the problem when compileall.py got stuck in an infinite loop.

  It looks like the "os" function has gagged.

  Here is screenshot illustrating the problem:
------------------------------------------------------------------------------
zebra:/export/home/rotten/Python-1.5.2 $ uname -a
SunOS zebra 5.7 Generic_106541-09 sun4u sparc SUNW,Ultra-4

zebra:/export/home/rotten/Python-1.5.2 $ python
Python 1.5.2 (#2, Jun 16 2000, 14:39:17)  [GCC 2.95.2 19991024 (release)] on sun
os5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.listdir('.')
['', '', '', '', '', '', '', '', '', 'ammar', 'e', '', 'dules', 's', '', 'rser',
 '', '', '', '', 'tus', 'hon1.5.a', 'kefile', 'ildno', '', '', '', 'd', '']
>>> ^D

zebra:/export/home/rotten/Python-1.5.2 $ ls
./                    Modules/              config.h
../                   Objects/              config.h.in
BeOS/                 PC/                   config.log
Demo/                 PCbuild/              config.status*
Grammar/              Parser/               configure*
Include/              Python/               configure.in
Lib/                  README                getbuildinfo.o
Makefile              Tools/                install-sh*
Makefile.almostgood   acconfig.h            libpython1.5.a
Makefile.in           buildno               python*
Misc/                 config.cache

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

Any ideas as to what might have gone wrong?

-- 
Rick Otten
rotten at colltech.com
O=='=++



From godzilla at netmeg.net  Tue Jun  6 10:13:53 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Tue, 06 Jun 2000 14:13:53 GMT
Subject: author of htmllib/formatter?
Message-ID: <m2og5ec2q9.fsf@netmeg.net>

Does anyone know who are the authors of the htmllib and formatter
modules in the standard python distribution?

i am having some trouble with htmllib.HTMLParser and am trying to
track down the problem in the code to no avail.

thanks

les schaffer


From alan.gauld at gssec.bt.co.uk  Tue Jun 13 09:56:16 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Tue, 13 Jun 2000 14:56:16 +0100
Subject: tkinter question
References: <39454A0E.1A630514@uniserve.com>
Message-ID: <39463D7F.C05A8969@gssec.bt.co.uk>

Bob van der Poel wrote:
> Third, how can I define a widget command with an arg. For example:
> 
>         q = Button( text="Quit", command = quitit("sure?") )

Use a lambda:

foo = lambda s="sure?" : # define it here
q = Button(text="Quit"), command=foo)

Alternatively you can use the bind mechanism which passes an 
event object to your handler - but the event might not hold 
what you need...

Thirdly you can make your handler a method of a class and 
have the class store the 'parameters'.

Alan G.

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


From noone at do.not.use  Sat Jun  3 15:39:21 2000
From: noone at do.not.use (Russell Turpin)
Date: Sat, 03 Jun 2000 14:39:21 -0500
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com>
Message-ID: <39395EE9.DFF9603B@do.not.use>

Courageous wrote:
> For what it's worth, I'm implementing a dynarray which
> leaves buffer room on both the head and tail as we speak.

This sounds good! It is the vector implementation, just
improved. 

Russell


From redaelli at inc.it  Sat Jun  3 17:59:48 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sat, 03 Jun 2000 23:59:48 +0200
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <04ce01bfcdbd$1fa80000$6401a8c0@home>
Message-ID: <39397FD4.67C7FD73@inc.it>

Oh, well. It seems I'm missing an obvious thing.
I post here my program and the data that caused the problem

-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/
-------------- next part --------------
 1
 4 3
      1  0.000000E+00  0.000000E+00
      2  0.000000E+00  0.300000E+03
      3  0.500000E+03  0.300000E+03
      4  0.500000E+03  0.000000E+00
      1     1     2
      2     2     3
      3     3     4
      1  0.300000E+03  0.000000E+00  0.100000E+01
  0.000000E+00 0.000000E+00 0.000000E+00 0.250000E+04 -.146922E+03 0.652654E+04
  0.150000E+02 0.478073E-01 -.395235E-04 0.250000E+04 -.146922E+03 0.668798E+04
  0.300000E+02 0.956146E-01 -.174419E-03 0.250000E+04 -.146922E+03 0.720229E+04
  0.450000E+02 0.143422E+00 -.429172E-03 0.250000E+04 -.146922E+03 0.806948E+04
  0.600000E+02 0.191229E+00 -.828271E-03 0.250000E+04 -.146922E+03 0.928954E+04
  0.750000E+02 0.239036E+00 -.139620E-02 0.250000E+04 -.146922E+03 0.108625E+05
  0.900000E+02 0.286844E+00 -.215745E-02 0.250000E+04 -.146922E+03 0.127883E+05
  0.105000E+03 0.334651E+00 -.313651E-02 0.250000E+04 -.146922E+03 0.150670E+05
  0.120000E+03 0.382458E+00 -.435786E-02 0.250000E+04 -.146922E+03 0.176985E+05
  0.135000E+03 0.430266E+00 -.584599E-02 0.250000E+04 -.146922E+03 0.206830E+05
  0.150000E+03 0.478073E+00 -.762539E-02 0.250000E+04 -.146922E+03 0.240203E+05
  0.165000E+03 0.525880E+00 -.972055E-02 0.250000E+04 -.146922E+03 0.277105E+05
  0.180000E+03 0.573687E+00 -.121559E-01 0.250000E+04 -.146922E+03 0.317535E+05
  0.195000E+03 0.621495E+00 -.149561E-01 0.250000E+04 -.146922E+03 0.361495E+05
  0.210000E+03 0.669302E+00 -.181454E-01 0.250000E+04 -.146922E+03 0.408983E+05
  0.225000E+03 0.717109E+00 -.217484E-01 0.250000E+04 -.146922E+03 0.460000E+05
  0.240000E+03 0.764916E+00 -.257897E-01 0.250000E+04 -.146922E+03 0.514546E+05
  0.255000E+03 0.812724E+00 -.302936E-01 0.250000E+04 -.146922E+03 0.572620E+05
  0.270000E+03 0.860531E+00 -.352846E-01 0.250000E+04 -.146922E+03 0.634223E+05
  0.285000E+03 0.908338E+00 -.407874E-01 0.250000E+04 -.146922E+03 0.699355E+05
  0.300000E+03 0.956146E+00 -.468262E-01 0.250000E+04 -.146922E+03 0.768016E+05
      2  0.500000E+03  0.100000E+01  0.000000E+00
  0.000000E+00 -.956146E+00 0.105359E+06 0.000000E+00 -.810241E-12 0.118531E+15
  0.250000E+02 -.956146E+00 0.354593E+09 0.000000E+00 -.250000E+03 0.118531E+15
  0.500000E+02 -.956146E+00 0.553234E+09 0.000000E+00 -.500000E+03 0.118531E+15
  0.750000E+02 -.956146E+00 0.615784E+09 0.000000E+00 -.750000E+03 0.118531E+15
  0.100000E+03 -.956146E+00 0.561998E+09 0.000000E+00 -.100000E+04 0.118531E+15
  0.125000E+03 -.956146E+00 0.411631E+09 0.000000E+00 -.125000E+04 0.118531E+15
  0.150000E+03 -.956146E+00 0.184438E+09 0.000000E+00 -.150000E+04 0.118531E+15
  0.175000E+03 -.956146E+00 -.998256E+08 0.000000E+00 -.175000E+04 0.118531E+15
  0.200000E+03 -.956146E+00 -.421405E+09 0.000000E+00 -.200000E+04 0.118531E+15
  0.225000E+03 -.956146E+00 -.760544E+09 0.000000E+00 -.225000E+04 0.118531E+15
  0.250000E+03 -.956146E+00 -.109749E+10 0.000000E+00 -.250000E+04 0.118531E+15
  0.275000E+03 -.956146E+00 -.141248E+10 0.000000E+00 -.275000E+04 0.118531E+15
  0.300000E+03 -.956146E+00 -.168578E+10 0.000000E+00 -.300000E+04 0.118531E+15
  0.325000E+03 -.956146E+00 -.189761E+10 0.000000E+00 -.325000E+04 0.118531E+15
  0.350000E+03 -.956146E+00 -.202822E+10 0.000000E+00 -.350000E+04 0.118531E+15
  0.375000E+03 -.956146E+00 -.205787E+10 0.000000E+00 -.375000E+04 0.118531E+15
  0.400000E+03 -.956146E+00 -.196678E+10 0.000000E+00 -.400000E+04 0.118531E+15
  0.425000E+03 -.956146E+00 -.173522E+10 0.000000E+00 -.425000E+04 0.118531E+15
  0.450000E+03 -.956146E+00 -.134343E+10 0.000000E+00 -.450000E+04 0.118531E+15
  0.475000E+03 -.956146E+00 -.771639E+09 0.000000E+00 -.475000E+04 0.118531E+15
  0.500000E+03 -.956146E+00 -.105359E+06 0.000000E+00 -.500000E+04 0.118531E+15
      3  0.300000E+03  0.000000E+00  -.100000E+01
  0.000000E+00 0.000000E+00 -.129038E+07 0.000000E+00 0.000000E+00 0.000000E+00
  0.150000E+02 0.000000E+00 -.156394E+10 0.000000E+00 0.000000E+00 0.161438E+03
  0.300000E+02 0.000000E+00 -.243946E+10 0.000000E+00 0.000000E+00 0.675750E+03
  0.450000E+02 0.000000E+00 -.271495E+10 0.000000E+00 0.000000E+00 0.154294E+04
  0.600000E+02 0.000000E+00 -.247753E+10 0.000000E+00 0.000000E+00 0.276300E+04
  0.750000E+02 0.000000E+00 -.181429E+10 0.000000E+00 0.000000E+00 0.433594E+04
  0.900000E+02 0.000000E+00 -.812342E+09 0.000000E+00 0.000000E+00 0.626175E+04
  0.105000E+03 0.000000E+00 0.441215E+09 0.000000E+00 0.000000E+00 0.854044E+04
  0.120000E+03 0.000000E+00 0.185928E+10 0.000000E+00 0.000000E+00 0.111720E+05
  0.135000E+03 0.000000E+00 0.335473E+10 0.000000E+00 0.000000E+00 0.141564E+05
  0.150000E+03 0.000000E+00 0.484049E+10 0.000000E+00 0.000000E+00 0.174938E+05
  0.165000E+03 0.000000E+00 0.622943E+10 0.000000E+00 0.000000E+00 0.211839E+05
  0.180000E+03 0.000000E+00 0.743446E+10 0.000000E+00 0.000000E+00 0.252270E+05
  0.195000E+03 0.000000E+00 0.836847E+10 0.000000E+00 0.000000E+00 0.296229E+05
  0.210000E+03 0.000000E+00 0.894436E+10 0.000000E+00 0.000000E+00 0.343718E+05
  0.225000E+03 0.000000E+00 0.907502E+10 0.000000E+00 0.000000E+00 0.394735E+05
  0.240000E+03 0.000000E+00 0.867335E+10 0.000000E+00 0.000000E+00 0.449280E+05
  0.255000E+03 0.000000E+00 0.765225E+10 0.000000E+00 0.000000E+00 0.507355E+05
  0.270000E+03 0.000000E+00 0.592461E+10 0.000000E+00 0.000000E+00 0.568958E+05
  0.285000E+03 0.000000E+00 0.340332E+10 0.000000E+00 0.000000E+00 0.634090E+05
  0.300000E+03 0.000000E+00 0.129038E+07 0.000000E+00 0.000000E+00 0.702750E+05
-------------- next part --------------
#!/usr/bin/env python

## pyplot
## program to plot'n'show the results of "telaio", a fortran program
## 
# Author: Jesper Skov <jskov at cygnus.co.uk>
# A rewite of the C canvas example in the GNOME Developer's Information

import GDK
from gtk import *
from gnome.ui import *
import GdkImlib

import glob # to look for all LCASE.* files
from string import * ## to process informations
from UserList import UserList ## to create lists that starts counting from 1
from math import * ## we need almost everything from math....

def fortran_float(string):
    data = [mantissa,exponent] = split(string,"E",2)
    print data
    map(float,data)
    return mantissa*pow(10,exponent)

class CanvasExample:
    def __init__(self):

        self.width = 600
        self.height = 400

        self.all = []

        self.colors = ("red",
                       "yellow",
                       "green",
                       "cyan",
                       "blue",
                       "magenta")

    def show_warped(self, widget, event=None):
        return
    
    def show_N(self, widget, event=None):
        return
    
    def show_T(self, widget, event=None):
        return
    
    def show_M(self, widget, event=None):
        return
    
    def main(self):
        # Open window to hold canvas.
        win = GtkWindow()
        win.connect('destroy', mainquit)
        win.set_title('Telaio')

        # Create VBox to hold canvas and buttons.
        vbox = GtkVBox()
        win.add(vbox)
        vbox.show()

        # Create canvas.
        self.canvas = GnomeCanvas()
        self.canvas.set_usize(self.width, self.height)
        self.canvas.set_scroll_region(0,0, self.width, self.height)
        vbox.pack_start(self.canvas)
        self.canvas.show()

        # Create buttons.
        hbox = GtkHBox()
        vbox.pack_start(hbox, expand=FALSE)
        hbox.show()

        label = GtkLabel('Condizione di carico:')
        label.set_alignment(0, 0.5)
        hbox.pack_start(label, expand=FALSE)
        label.show()
        
        spinner = GtkSpinButton(GtkAdjustment(0,0,10,1,2,0), 0, 0)
        hbox.pack_start(spinner)
        spinner.show()

        b = GtkButton("Deformata")
        b.connect("clicked", self.show_warped)
        hbox.pack_start(b)
        b.show()

        b = GtkButton("N")
        b.connect("clicked", self.show_N)
        hbox.pack_start(b)
        b.show()

        b = GtkButton("T")
        b.connect("clicked", self.show_T)
        hbox.pack_start(b)
        b.show()

        b = GtkButton("M")
        b.connect("clicked", self.show_M)
        hbox.pack_start(b)
        b.show()

        b = GtkButton("Quit")
        b.connect("clicked", mainquit)
        hbox.pack_start(b)
        b.show()

        win.show()

class Node:
    def __init__(self,number,x,y):
        self.number = number
        self.x = x
        self.y = y
        
class Rod:
    def __init__(self, number, node1, node2):
        self.number = number
        self.node1 = node1
        self.node2 = node2
        self.sections = range(0,21) ## list of sections. Each section will be a list containg [position, u displacement, v displacement, n, t, m]. From index 0 to index 20
        
## Lista is a list with preallocated length that starts counting from 1. This is useful to avoid unnecessary mess within file processing expecially regarding the rods and nodes lists
class Lista(UserList):
    def __init__(self):
        UserList.__init__(self)
        self.data = []
    def __setattr__(self, attr, value):
        if attr == "size" and value>0:
            self.data = [None]*value
        else:
            self.__dict__[attr] = value
    def __getitem__(self,index):
        if index > 0:
            return self.data[index-1]
        else:
            raise IndexError
    def __setitem__(self,index,value):
        self.data[index-1] = value
    
class LoadCase:
    def __init__(self):
        self.nodes   = Lista() ## the list of nodes
        self.rods    = Lista() ## the list of rods

class Telaio:
    def __init__(self):
        ## Loading all the load cases 
        ## Looking for all LCASE.xxx files
        self.filelist = glob.glob("LCASE*")
        self.load_cases = Lista()
        self.load_cases.size = len(self.filelist)  ## Load cases are a list of cases
        
        for filename in self.filelist:
            file = open(filename)
            ## Obtaing case number 
            case_n = int( split(filename,".")[1] ) ## split filename into fields separated by "."; take the second (0 is the first), translate it in a number
            ## Creating Load case
            loadcase = LoadCase()
            self.load_cases[case_n] = loadcase
            ## Reading the trailing 1
            file.readline()
            ## read nodes and rods numbers
            number_list = [nodes_n, rods_n] = split( file.readline() )
            [loadcase.nodes_n, loadcase.rods_n] = map(int, number_list) ## Convert them to numbers and assign 'em
            ## Reading nodes coordinates
            for node_i in range(0,loadcase.nodes_n):
                [node, x, y] = split( file.readline() )
                node = int(node);
                print "x(%s)='%s' y(%s)='%s'" % (type(x),x,type(y),y)
                x = float(x)
                y = float(y) ## Converting in numeric form (could be done in during node creation, but it would generate beastly thing like "loadcase.nodes[int(node)] = Node( float(x), float(y) )" )
                loadcase.nodes[node] = Node(node,xx,yy)
                ## Note: node_i is NOT used!
            ## Reading rod's end nodes
            for rod_i in range(0, loadcase.rods_n):
                [rod_n, node1, node2] = split( file.readline() )
                map(int, [rod_n, node1, node2]) ## conveting into numeric form applying int to every element of the list.
                loadcase.rods[rod_n] =  Rod(rod_n,node1,node2)
                ## Note rod_i is NOT used!
            ## for each rod, read it's number, length cosine and sine
            for rod_i in range(0, loadcase.rods_n):
                [rod_n, lenght, cosine, sine] = split( file.readline() )
                rod_n = int(rod_n); map(float, [lenght, cosine, sine])  ## Converting into numeric form
                current_rod = loadcase.rods[rod_n]
                [current_rod.lenght, current_rod.cosine, current_rod.sine]= [lenght, cosine, sine]
                ## Reading position, displacements and stresses
                for i in range(0,21): ## from 0 to 20
                    section = [x,u,v,n,t,m] = split( file.readline() )
                    map( float, section) ## Converting list's element into float
                    current_rod.sections[i] = section
                ## End reading rod's sections
            ## End reading rods
        ## End reading files.
        
if __name__ == '__main__':
    t = Telaio()
    c = CanvasExample()
    c.main()
    mainloop()


From shapr at uab.edu  Sat Jun  3 17:39:21 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sat, 03 Jun 2000 21:39:21 GMT
Subject: Python GUI requires tcl?
References: <8hbeol$obo$1@samba.rahul.net> <8hbjl0$611$1@news1.xs4all.nl> <393968C8.CBFAAF1@bogusaddress.com>
Message-ID: <3939817F.3FDBD89B@uab.edu>

Joe Smith wrote:
> 
> Could one use jpython and use java's graphics library?  How stable is
> that?

JPython is as stable as the JVM you run it on in my experience. This
means you could happily access any java based editor such as jEdit from
jpython. I don't think there are any Python specific java-based editors
though.

There is a very cool jTkinter module written by Finn Bock on
sourceforge:
https://sourceforge.net/project/?group_id=1133
The cool part about this is that it lets you run IDLE and other Tkinter
apps in Jpython.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From cjc26 at nospam.cornell.edu  Fri Jun  9 14:25:23 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 09 Jun 2000 18:25:23 GMT
Subject: Erm... MacOS X?
References: <eviltofu-1AEA18.02454303062000@news.earthlink.net> <usenet-6DC18E.10333803062000@news.visi.com> <slrn8ji7hc.5b1.cjc26@localhost.mindriot.net> <eviltofu-86F5A2.13584703062000@news.earthlink.net>
Message-ID: <slrn8k2dot.q7v.cjc26@localhost.mindriot.net>

* Jerome Chan <eviltofu at rocketmail.com> menulis:
| 
| There is a man Page for Zlib and Gzip but I can't seem to find the 
| location of the zlib.h file and the zlib library. Do you know where it 
| is or do I have to download and compile it?

I found it in three places:

/System/Library/Frameworks/Kernel.framework/Versions/A/Headers/net/zlib.h
/System/Library/Frameworks/System.framework/Versions/B/Headers/bsd/net/zlib.h
/System/Library/Frameworks/Zip.framework/Versions/A/Headers/zlib.h


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From pinard at iro.umontreal.ca  Wed Jun 21 12:02:36 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 21 Jun 2000 12:02:36 -0400
Subject: French Python documentation
Message-ID: <oqwvjj58f7.fsf@titan.progiciels-bpi.ca>

Hello, my friends.  Robert, Cc:ed, is asking me what exists, in French,
to document Python.  He would like to get as much as possible for the
unilingual members of his team.  Please reply to the list, or to me at
least, so I can save the collective knowledge and wisdom on this point! :-)

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





From dnotspam-jen at personic.com  Fri Jun 30 21:39:21 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Fri, 30 Jun 2000 18:39:21 -0700
Subject: Communicating between two computers
Message-ID: <395d48b3$0$21842@wodc7nh0.news.uu.net>

FYI, I'm doing all of this on Windows NT, in case that makes a difference
...

Is there a good way to get a script on one computer to talk to a script on
another computer?  The process I'm thinking of is pretty simple ... I want a
script on a Web server to be able to send a command to a build server to
start a build process.  I'd also like the Web server to be able to query the
build servers to find out their status (so that two builds do not run at
once on the same machine, for example).

I've read a little about SOAP, but I think that my needs are probably a lot
simpler than that.  The easiest way to accomplish this is to setup a Web
server on each of the build machines, too, I guess ... but I was hoping that
I could just have the scripts that run on the build servers run as NT
services.

Thanks!

Jen






From alex at magenta.com  Mon Jun 12 04:57:10 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 12 Jun 2000 10:57:10 +0200
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au>
Message-ID: <8i28pu01ocd@news2.newsguy.com>

Neil Hodgson <neilh at scintilla.org> wrote in message
news:Cl%05.2665$cQ.22917 at news-server.bigpond.net.au...
> > In many Perl vs. "other scripting language" discussions, I have often
> > seen Perlers state that those who dislike Perl's syntax and structure
> > are often non-Unix types, and thus are not used to common Unix
> > concepts. And, since Perl was born as a Unix admin tool, us "Windows
> > types" will likely be lost.
>
>    I think that was part of my problem with Perl. While I can use Unix,
its
> at a low level of competence - I count myself lucky to survive a session
> with vi without pulling the power cord to stop the torture. Coming from an

I still use and program Unix at a very high level of competence, and vi
is by far my favourite editor (gvim, of course, these days; under Windows
too, but, basically, everywhere and anywhere).

> RSX/11M+, Mac, OS/2, Windows background I found the syntax choices in
Perl,
> which probably appear obvious to native Unixers, bizarre and difficult to

I find (and always have found) Perl's syntax bizarre and difficult, despite
working for many years with Perl.  And I'm as close to a "native Unixer" as
you can get around here, though I also have mainframes and VMS backgrounds.
Unix's various shells' syntax is *NOT* Unix's strong point!-)

> reason with. Python does not try to use much 'clever syntax' at all making
> it much easier for me.

Exactly.  Here too.


>    This leads to not liking the 'augmented assignment' addition. The only
> examples of augmented assignment that are used often enough to have their
> own special syntax are addition and subtraction. Even for these, I'd
prefer
> to see syntax that looks like function calls similar to Modula 2.

Here, I deeply disagree.  Augmented-assignment is the only substantial
contribution of the C syntax: a truly excellent, outstanding idea.  The
point is that "multiply X by 7" is *more fundamental* in some sense than
"compute the expression X times 7, and set a new value of X to that", as
also reflected by the much more concise English (and other natural lang)
syntax expressing it; augmented-assignment gives concise syntax to match
the fundamental semantics.  Languages with no interest in mimicking C per
se, such as Icon, have long been adopting that syntactic idea.  Python
would also benefit from it.


Alex





From holger at fuessler-net.de  Tue Jun 27 12:38:00 2000
From: holger at fuessler-net.de (holger at fuessler-net.de)
Date: Tue, 27 Jun 2000 18:38:00 +0200
Subject: Problems Compiling MySQLmodule.c
Message-ID: <89laj8.1re.ln@odin.hmf.de>

Hi!

I'm not sure I'm right here, but I'll give it a try.

I'm trying to compile MySQLmodule.c. The compiling and installation of
python works fine. When trying to compile MySQLmodule.c (whether as a shared
module or with access to the source tree doesn't matter) I get the following
error.

--schnipp--

gcc -shared -I/opt/include/python1.5 -I/opt/include/mysql MySQLmodule.c
-lmysqlclient -L/opt/lib/python1.5/config/ -lpython1.5 -o MySQLmodule.so
MySQLmodule.c: In function pythonify_row':
MySQLmodule.c:238: warning: assignment from incompatible pointer type
MySQLmodule.c: In function pythonify_res_fields':
MySQLmodule.c:384: invalid lvalue in unary &'
MySQLmodule.c: In function STH_fetchdict':
MySQLmodule.c:1125: invalid lvalue in unary &'
MySQLmodule.c:1147: invalid lvalue in unary &' 

--schnipp

Maybe you know what's wrong.

Thanks in advance

Holger

-- 
Holger M. Fuessler
holger at fuessler-net.de
Tel.: +49 621 1221292


From DavidA at ActiveState.com  Thu Jun  8 13:29:14 2000
From: DavidA at ActiveState.com (David Ascher)
Date: Thu, 8 Jun 2000 10:29:14 -0700
Subject: Rich comparisons (was:  My wishlist for Python3K)
In-Reply-To: <m3wvk06s29.fsf@brant.geog.ubc.ca>
Message-ID: <PLEJJNOHDIGGLDPOGPJJKEDECHAA.DavidA@ActiveState.com>

> but I haven't heard it mentioned as part of 1.6/1.7 -- anyone know
> the status of the rich comparison proposal?

Still in my TODO list, still in development.  Alas, it's less simple than I
thought, requiring messing around with the bytecode compilation stage in
order to be backwards compatible.

It won't make 1.6.

--david




From alex at magenta.com  Mon Jun 26 10:08:59 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 26 Jun 2000 16:08:59 +0200
Subject: Differentiating between WinNT and Win32
References: <eviltofu-F9E289.07322926062000@news.earthlink.net>
Message-ID: <8j7pjl01vq7@news2.newsguy.com>

Jerome Chan <eviltofu at rocketmail.com> wrote in message
news:eviltofu-F9E289.07322926062000 at news.earthlink.net...
> Is there a way to do this in Python? I went DejaDiving and found
> something called platform.py but the link to it on starship is dead.
> Anyone has any clues?

import win32api
win32api.GetVersionEx()

will return, e.g., the tuple:

(4, 0, 1381, 2, 'Service Pack 5')


The 4th element in the tuple ("2" here) has the specific info
you require; it's 1 for Win/95 and Win/98, 2 for Win/NT (note
that all of these are "Win32"...).


Alex







From jcm at bigskytel.com  Tue Jun 27 18:08:57 2000
From: jcm at bigskytel.com (David Porter)
Date: Tue, 27 Jun 2000 16:08:57 -0600
Subject: "No module named _locale" (1.5.2 w/ RH 6.2)
In-Reply-To: <16600709@oberon.noris.de>; from hartmut@oberon.noris.de on Tue, Jun 27, 2000 at 10:28:22PM +0100
References: <16600709@oberon.noris.de>
Message-ID: <20000627160857.A32688@novara.avenue>

* hartmut Goebel <hartmut at oberon.noris.de>:

> >>> import locale
> Import Error: No module named _locale
> 
> Well, this is true, since there is no _locale.* on my system.
> I'm running RedHat 6.2 with Python 1.5.2 (from ReadHat's 6.2 rpm)

Do you have _localemodule.so ? I don't have any _locale.* on my system either,
but 'import locale' works fine.





From boud at rempt.xs4all.nl  Thu Jun 29 15:16:33 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: Thu, 29 Jun 2000 21:16:33 +0200 (CEST)
Subject: Following a file, or cloning tail
In-Reply-To: <0b4201bfe16b$78339560$6401a8c0@home>
Message-ID: <Pine.LNX.4.21.0006292102260.25464-100000@calcifer.valdyas.org>

On Wed, 28 Jun 2000, Darrell Gallion wrote:

> If your on Windows and can wait for the file to complete writing, check out
> http://www.dorb.com/darrell/win32WorkSvr/
> 

I have to be portable, worse luck!

> The best solution might be to use popen if you can.

You mean simply a os.popen("tail -f myfile","r").readline() should
work? It sounds to easy to be true ;-)

> 
> Trail might be a problem if the other application doesn't flush() the file
> once in a while.

It calls flush() allright, that's not the problem.

> If it does you should be able to just call read() on an interval or use
> select to avoid polling.
> Select won't help you on Windows though, but Win32 calls will.
> 

Thanks for all the pointers!

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




From moshez at math.huji.ac.il  Sat Jun 10 16:24:13 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 23:24:13 +0300 (IDT)
Subject: No strptime in Windows?
In-Reply-To: <8hu4tc$6m0$1@nnrp1.deja.com>
Message-ID: <Pine.GSO.4.10.10006102322220.9354-100000@sundial>

On Sat, 10 Jun 2000 david_ullrich at my-deja.com wrote:

>     The docs say that time.strptime may not be available
> on all platforms. It doesn't seem to be available under MS
> Windows. I don't see why not - there's nothing Windological
> about strptime but the same applies to the other time
> functions that _are_ included.

The thing is, Python doesn't invent those functions: it merely wraps them.
If the platform doesn't have strptime, then Python won't either. Choose an
strptime-less platform, you're strptime-ly screwed.

I think there are long terms plan to supply a pure Python substitute, but
it hasn't happened yet.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From gmcm at hypernet.com  Wed Jun 28 18:59:03 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 28 Jun 2000 22:59:03 GMT
Subject: Module Name Problem
References: <eviltofu-C678D0.18293328062000@news.earthlink.net>
Message-ID: <8F61CAE59gmcmhypernetcom@199.171.54.194>

Jerome Chan wrote:

>If I have a module called
>
>maxvu.os
>
>and I use
>
>import os
>
>Somehow, when I try to use the os calls or maxvu.os calls it gets 
>confused. Is there a workaround besides renaming my module?


Not if you want access to os within maxvu.

- Gordon


From scarblac-spamtrap at pino.selwerd.nl  Wed Jun  7 17:44:59 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Jun 2000 21:44:59 GMT
Subject: Why can't I restore my redirected stdout?
References: <8hm5od$lki$1@news5.jaring.my>
Message-ID: <slrn8jtgg8.oe2.scarblac-spamtrap@flits104-37.flits.rug.nl>

Raymond Ng Tong Leng wrote in comp.lang.python:
> I am trying to implement a verbose and quiet mode in my Python script. So I
> wrote the following class:
> 
> class Quiet:
>     def write(self, data):
>         pass
> 
> Then I set sys.stdout to Quiet by:
> 
> >>> sys.stdout = Quiet()
> >>> print 'hello'
> >>>
> 
> 'hello' doesn't come out which is what I want. But, when I tried to restore
> sys.stdout's original settings:
> 
> >>> sys.stdout = sys.__stdout__
> >>> print 'hello'
> >>>
> 
> 'hello' is not printed either! How do you restore sys.stdout to its original
> settings?

I just typed this in, and it worked fine! Something else must be happening.
Are you in idle or so? Does it work inside a normal python interpreter?
What's your platform?

> Also, I was wondering, are there other ways to suppress
> sys.stdout? I'm just curious whether I could suppress sys.stdout without
> redirecting to a class. It seems to me that I should be able to redirect
> sys.stdout to a null device or something. Thanks!

Seems to me your solution is the easiest way to get a cross platform null
device, but maybe there's something like it in the library already, I don't
know.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From thomas at xs4all.net  Mon Jun  5 08:22:11 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Jun 2000 14:22:11 +0200
Subject: Starship ever gonna fly?
In-Reply-To: <393BA4CB.C4DC60D7@sightreader.com>; from 12klat@sightreader.com on Mon, Jun 05, 2000 at 06:02:04AM -0700
References: <f9oljso6hl7r7blv64an77h8lfj6dhr6f9@4ax.com> <20000605012821.V469@xs4all.nl> <393BA4CB.C4DC60D7@sightreader.com>
Message-ID: <20000605142211.W469@xs4all.nl>

On Mon, Jun 05, 2000 at 06:02:04AM -0700, Ken Seehof wrote:
> > > Whoever killed Starship did real and lasting damage
> > > it seems.  Did anyone call the FBI?

> > I was under the impression that the Big Crash was hardware-related, not a
> > malicious attack.

> The Perl people did it.

Damnit. That does it. I'm gonna kick our butts tomorrow evening, at the
Amsterdam Perl Mongers meeting. Damned Perl Secret Underground.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From digitig at cix.co.uk  Sat Jun  3 15:12:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 3 Jun 2000 20:12 +0100 (BST)
Subject: A Mountain of Perl Books + Python Advocacy
References: <3937035C.EA99CB2B@uab.edu>
Message-ID: <memo.20000603201240.52825A@digitig.compulink.co.uk>

In article <3937035C.EA99CB2B at uab.edu>, shapr at uab.edu (Shae Erisson) 
wrote:

> Chris Stromberger wrote:
> 
> > >I would invite everybody to read Eric Raymond's article: "Why 
> > Python?"
> > >Honestly, I expected to see comments about it already in this news 
> > group
> > >. . . but maybe I was just late, and missed the thread.
> 
> I've seen it in the special issue of the Linux Journal that is dedicated
> to Python, I don't know if it's online.

There's a link from 
http://www2.linuxjournal.com/cgi-bin/frames.pl/index.html that says it 
includes May's Python supplement, but I've not downloaded it.

I'm hesitant to show that article to friends. I keep thinking that after 
such praise, they're sure to be disappointed in the reality. Then I think 
"but what's to disappoint?"


From petersc at stud.ntnu.no  Tue Jun 20 02:48:09 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 20 Jun 2000 08:48:09 +0200
Subject: serial I/O
References: <394F01DD.E0CF3EB@stud.ntnu.no> <cMD35.4116$fw6.82744@news6-win.server.ntlworld.com>
Message-ID: <394F13A9.282D3431@stud.ntnu.no>

richard_chamberlain wrote:
> 
> Hi Peter,
> 
> Have a look at ftp.python.org/pub/python/contrib/sio-151.zip

Thanks! Don't know how I could miss that.

just-have-to-get-the-protocol-right-then-ly y'rs
peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From tim_one at email.msn.com  Sun Jun 11 15:55:01 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 11 Jun 2000 15:55:01 -0400
Subject: No strptime in Windows?
In-Reply-To: <8i0lhu$ped$1@nnrp1.deja.com>
Message-ID: <000901bfd3de$ee54d280$27a2143f@tim>

[David Ullrich]
> ...
>        Maybe I'm not clear on exactly what "the
> platform" is. I've taken that to mean "Windows".
> But maybe when people say I'm using the Windows
> platform that's shorthand for saying the platform
> I'm using is "Windows C"?  (No, that can't be it,
> because surely there's no such thing as "Windows
> C".) Or the "platform" is the specific C compiler
> that was used? (No, that can't be it either...)

"The platform" is the combination of hardware, compiler, operating system, C
library, and (less so) Python build-time configuration options.

>        Both replies explained that strptime must not
> be available on the Windows platform. I surmised as
> much. But the problem is I'm not aware that there's
> anything like strftime() in Windows either.

In this particular case, Microsoft's runtime C library does support strftime
directly, but does not support strptime.

> ...
> exactly what is it that's missing strptime, so that my Python
> doesn't have one available to wrap?

OK, but the answer isn't going to help you <wink>:  msvcrt.dll (that's the
MicroSoft Visual C RunTime DLL).

>     (When I asked the question I was assuming that
> most of the time.this/that functions were done "by
> hand" in the Windows version of Python.)

They're not -- most of them are standard ANSI C library functions (for which
see the C standard), and "even Microsoft" <wink> takes a stab at supporting
those.  Since MS didn't see fit to support strptime (which is not std ANSI
C), the only way it will show up in an MS Windows version of Python is if
someone contributes implementation code for it using straight C.

note-that-strptime-is-missing-under-some-unices-too-ly y'rs  - tim





From moshez at math.huji.ac.il  Tue Jun 13 00:50:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 07:50:05 +0300 (IDT)
Subject: tkinter question
In-Reply-To: <yad15.25$x_6.8854@news1.rdc2.tx.home.com>
Message-ID: <Pine.GSO.4.10.10006130746350.20303-100000@sundial>

On Mon, 12 Jun 2000, Rainer Deyke wrote:

> Bob van der Poel <bvdpoel at uniserve.com> wrote in message
> news:39454A0E.1A630514 at uniserve.com...
> 
> > Third, how can I define a widget command with an arg. For example:
> >
> > q = Button( text="Quit", command = quitit("sure?") )
> 
> One word: lambda.

There's this joke, about a person who's asked "In one word, how's it
going?" "Good" "And in two words?" "Not good"

In three words: don't use lambda <wink>

There are lots of ways to write classes to help this work smoothly:

class Curry:

	def __init__(self, func, *args):
		self.args = args
		self.func = func

	def __call__(self, *args):
		return apply(self.func, self.args+args)

Use like

f = Curry(operation.add, 1)
f(2) ==> 3

Much stabler them lambdas, because it doesn't depend on scoping
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From barryhart at my-deja.com  Mon Jun 12 19:10:23 2000
From: barryhart at my-deja.com (barryhart at my-deja.com)
Date: Mon, 12 Jun 2000 23:10:23 GMT
Subject: PythonWin and collections
References: <8hv12v$ol1$1@nnrp1.deja.com> <8i3g8d$kev$1@nnrp1.deja.com>
Message-ID: <8i3qkm$rsh$1@nnrp1.deja.com>

Mark,

> You could try win32all build 132, where there is specific ADSI
> support.  There is a demo there which should point you in the right
> direction.  This does support ADSI enumerations correctly.

Thanks for the tip. Is this build publicly available anywhere? I was
looking at
http://starship.python.net/crew/mhammond/win32/Downloads.html and build
131 seemed to be the latest one there.

Thanks,
Barry


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


From ivanlan at home.com  Wed Jun  7 11:18:01 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Wed, 07 Jun 2000 09:18:01 -0600
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <393f5114.72916406@nntp.interaccess.com>
Message-ID: <393E67A9.772BC631@home.com>

Hi All--

"Thaddeus L. Olczyk" wrote:
> 
[bobbit]
> 
> Could be good are bad news. Technologies tend to get perverted when
> Microsoft starts integrating them into their products. Isn't
> there some sort of lawsuit about that ;)
> 

The PSU is handling the details.

<hard-at-work-preventing-perversion>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From sean at digitome.com  Mon Jun 19 17:23:05 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Mon, 19 Jun 2000 21:23:05 GMT
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee>
Message-ID: <394e8973.14164217@news.iol.ie>

On Fri, 02 Jun 2000 18:01:12 +0300, Kaimar Karu <kaimar at ut.ee> wrote:

>Hi,
>
>Where can I find some kind of helping material (tutorials etc) about how
>to use xml in python?
>I need some examples, just pure specification is not enough.
>
My book "XML Processing with Python", ISBN 0 13 021119 2,
has just been published by Prentice Hall. This should
help.

regards,



From jflores at codeit.com  Thu Jun  8 19:32:14 2000
From: jflores at codeit.com (Grakka)
Date: Thu, 08 Jun 2000 16:32:14 -0700
Subject: Pythonwin is terrible!
In-Reply-To: <20000608155108.A8398@reepicheep.avalongroup.net>
References: <8hp2ku$a25@news.or.intel.com>
 <8hp2ku$a25@news.or.intel.com>
Message-ID: <4.3.2.7.0.20000608162836.00a9e100@mail.codeit.com>

It is not Python, my friend, actually, extract the word "Python" from the 
subject of this thread, and.. voila.

At 03:51 PM 6/8/00 -0700, Timothy Grant wrote:
>On Thu, Jun 08, 2000 at 02:19:23PM -0700, Jeff wrote:
> >
> > If there are no answers to these questions, someone needs to work on
> > pythonwin's "development" capabilities.
>
>I vote for you to do it.
>
>--
>Stand Fast,
>     tjg.
>
>Timothy Grant                         tjg at exceptionalminds.com
>Chief Technology Officer              www.exceptionalminds.com
>Red Hat Certified Engineer                           MIG #1433
>Avalon Technology Group, Inc.
> >>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<
>
>--
>http://www.python.org/mailman/listinfo/python-list




From dgrisby at uk.research.att.com  Thu Jun  1 11:54:44 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 1 Jun 2000 15:54:44 GMT
Subject: Interprocess comunication
References: <8F456BAC0lminutieidit@192.6.1.10> <3934f634.704981920@news.u-psud.fr> <8F45B7D7Clminutieidit@192.6.1.10> <39366A38.262E0E07@earth.ox.ac.uk>
Message-ID: <8h6104$cam$1@pegasus.csx.cam.ac.uk>

In article <39366A38.262E0E07 at earth.ox.ac.uk>,
 Nick Belshaw  <nickb at earth.ox.ac.uk> wrote:

>You might want to try the Medusa stuff  - asyncore/asynchat.py - in the
>standard library.

Alternatively, you might find it easier to use one of the distributed
object systems.

  http://www.vex.net/parnassus/apyllo.py?i=549099265

lists a bunch of them. (I hope that URL stays valid...) Some are
Python-only, others are CORBA-based so you can communicate between
Python and other languages. Even if you don't want to do that now, it
might be a good insurance policy for the future.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From arta at NOSPAMx-stream.nl  Thu Jun 15 14:26:14 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Thu, 15 Jun 2000 20:26:14 +0200
Subject: locate
Message-ID: <eN#Favv1$GA.319@net003s>

Hi,

I'm wondering if there's a method to do something like LOCATE in BASIC.
Example:

LOCATE  4,13
PRINT "COOL!"

This prints the message COOL! on the screen on y, x location 4,13.

Can somebody tell me, if this is possible in Python?

A[r]TA

--
-Life can kick your ass, but all you got to do is stand up and move on...





From nmleeds at mindspring.com  Thu Jun 22 21:06:47 2000
From: nmleeds at mindspring.com (Nathaniel Leeds)
Date: Thu, 22 Jun 2000 18:06:47 -0700
Subject: HTML Parser? Sample Code?
Message-ID: <3952B827.5B96BE9C@mindspring.com>

Has anyone used the HTML parser.  I'd like to be able to skillfully
disect html pages and put various types of information into lists and
other objects using the tag structure of the HTML and not needing to
write the stacking and pattern matching routines with regexp.  I'm new
to python and am having trouble with the HTML parser documentation
that's on line.  Does someone have some sample code I could look at?



From yawyi at gmx.de  Mon Jun 26 06:47:40 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Mon, 26 Jun 2000 12:47:40 +0200
Subject: smtplib
References: <000c01bfdec7$30f84b60$bdcb45c6@timaratz>
Message-ID: <3957535f$0$1860@personalnews.de.uu.net>

Peter Timaratz <timaratz at lisco.com> wrote:
>I'm using smtplib with a CGI script. I'm puzzled as to why this module
>doesn't support sending a subject for the e-mail.
>
>And the e-mail I receive doesn't show the to address in my e-mail client.

smtplib implements SMTP as per RFC 821, which defines the transport
protocol only, not the format of the message content. See RFC 822
for that.

Subject and From headers are part of the message body.
To and From adresses in SMTP are sometimes called "envelope"
adresses, and can be different from the adresses on the message
inside the envelope. SMTP servers usually put Received headers in
front of the messages they receive to capture the envelope
adresses, but they are not normally displayed in email clients.
Most email clients have an option to display all headers or the
raw message ("view" -> "page source" in Netscape).

In short: Place any headers like From:, Subject: and Date: in
front of your message. Header and Body are separated by a single
blank line.

Johannes



From jhauser at ifm.uni-kiel.de  Sun Jun 18 16:13:20 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 18 Jun 2000 22:13:20 +0200
Subject: Getting Password from User
References: <8ij9kr$qa5$1@bob.news.rcn.net>
Message-ID: <87ln027nof.fsf@ifm.uni-kiel.de>

Look for the module getpass in the library reference. It also supports
windows according to the document string.

HTH,
__Janko
-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From jerome.quelin at insalien.org  Mon Jun 26 14:50:46 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Mon, 26 Jun 2000 20:50:46 +0200
Subject: Final garbage collection?
References: <962028703.1248416950@news.libertysurf.fr> <FwrnM8.MMA@world.std.com>
Message-ID: <962045446.1993274611@news.libertysurf.fr>

wware at world.std.com (Will Ware) wrote:
>That's correct. Garbage collection in Python is accomplished by
>reference counting. When a gets a pointer to b, b's reference count
>is incremented, and vice versa.
I know that, but I expected Python to do a final check-up when exiting,
just like Perl does. I know this is a Python newsgroup, but here is a sample
snippet of Perl code:

$ cat test.pl
#!/usr/bin/perl
package Test;
sub new {
    my $self = bless ({}, 'Test');
    $self->{name} = $_[1];
    return $self;
}
sub DESTROY {
    my $self = shift;
    print "Ouch! ($self->{name})\n";
}
package main;
$a = new Test('a');
$b = new Test('b');
$c = new Test('c');
$a->{cross_ref} = $b;
$b->{cross_ref} = $a;
$ ./test.pl
Ouch! (c)
Ouch! (b)
Ouch! (a)
$

Perl implement its garbage collection by counting references too. And circular
references aren't collected since their reference count does not reach zero,
just like in Python.
But, (and the following is taken from perl documentation -man perlobj):
       When an interpreter thread finally shuts down (usually
       when your program exits), then a rather costly but
       complete mark-and-sweep style of garbage collection is
       performed, and everything allocated by that thread gets
       destroyed.  This is essential to support Perl as an
       embedded or a multithreadable language.

I was wondering why Python does not implement such a mechanism, that
seems to me a Good Thing (tm), because _in final_, every object _is_
destroyed, either if their reference count is nul or not, since the memory is
reclaimed by the system.

Sorry for the above Perl code for Python integrists.

Jerome
--
jerome.quelin at insalien.org


From m.faassen at vet.uu.nl  Tue Jun 13 09:37:54 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 13 Jun 2000 13:37:54 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i1uv0$hki$0@216.39.141.21> <1d09kssdshm96qgilpm7rabmibt2blmv5j@4ax.com>
Message-ID: <8i5dfi$gac$1@newshost.accu.uu.nl>

mkx at excite.com wrote:
> On 12 Jun 2000 06:11:44 GMT, dana_booth <dana at oz.net> wrote:

>>mec: So my question is this: Is Python a more "Windows-friendly" language?
>>
>>Okay, I couldn't leave this alone. :)
>>
>>I think "Windows-friendly" is relative. I know Windows only programmers who
>>can out-Perl just about anyone. (FWIW, I'm in the UNIX crowd. :)

> I should have replaced "Windows-friendly" with "less painful to
> non-Unix users"...

Um, Perl can be painful to Unix users too. I am a Unix user and I never
really got into Perl; didn't have to as I discovered Python and didn't
look back.

Regards,

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


From hinsen at cnrs-orleans.fr  Wed Jun  7 05:19:22 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 11:19:22 +0200
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl>
Message-ID: <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr>

m.faassen at vet.uu.nl (Martijn Faassen) writes:

> Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
> nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?

*Blijken* te spreken. Misschien zijn we allemal bots die van een
enkele (nederlands) programmeur komen. Of misschien is er maar een
enkele mens in deze nieuwsgroep, die honderde adressen gebruikt om
de indruk te wekken dat Python heel populair is. Of misschien is
er een vertalingsprogramma dat alle teksten gewoon naar het Engels
vertaald, maar af en toe naar het Nederlands.

Er zijn dus merdere rationale verklaringen, dus geen behoefte om
aan kosmische krachten te geloven!
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From wlfraed at ix.netcom.com  Wed Jun 28 00:23:52 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 27 Jun 2000 21:23:52 -0700
Subject: Books
References: <8jb1ec$9mh2@news.qualitynet.net>
Message-ID: <autilso0l591oo924n3mogcva7r2892p5b@4ax.com>

On Tue, 27 Jun 2000 17:09:09 +0300, "Helmit" <helmit_world at hotmail.com>
declaimed the following in comp.lang.python:

> Does anyone know any good python books for teaching beginers?
> 
	Wide-open question...

	Beginners in programming;
	beginners in Python;
	beginners in OO...

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From bjorn at roguewave.com  Tue Jun 20 19:11:22 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 20 Jun 2000 17:11:22 -0600
Subject: List without duplicates?
References: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <394FFA1A.FBF8F81D@roguewave.com>

William Dandreta wrote:
> 
> Is there a built in Python function that eliminates duplicates in a list?
> 
> I have been sorting the list and compairing adjacent elements and if equal
> deleting one:
> -------------------------------
> mylist.sort()
>     j = len(mylist)-1
>     while j > 0:
>       if mylist[j] == mylist[j-1]:
>         del mylist[j]
>       j = j - 1
> ------------------------------------
> I recently tried using a dictionary. The list is the keys and I put and
> empty string for the entry
> --------------------
> mydict = {}
> for e in mylist:
>   if mydict.has_key(e): continue
>   else: mydict[e] = ''
> ------------------------------------
> Is ther a better way?
> 
> Bill

This should probably be a faq by now...:

def unique(lst):
    d = {}
    for item in lst:
        d[item] = None
    return d.keys()

-- bjorn



From bvdpoel at uniserve.com  Sun Jun 25 13:33:09 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sun, 25 Jun 2000 10:33:09 -0700
Subject: tkinter font selector?
Message-ID: <39564255.65F7C155@uniserve.com>

Has anyone done a font selector widget for tkinter? Be nice if I could
have a font option which let a user select (from the available fonts on
the system) the font/size/etc to use for a text display.


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



From jeremy at beopen.com  Wed Jun 14 09:13:52 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: Wed, 14 Jun 2000 09:13:52 -0400 (EDT)
Subject: tuple creation in C extensions
In-Reply-To: <l03102800b56cf504d1d6@[193.78.237.168]>
References: <l03102801b569016fbd23@[193.78.237.174]>
	<l03102800b56cf504d1d6@[193.78.237.168]>
Message-ID: <14663.34064.737437.53886@localhost.localdomain>

>>>>> "JvR" == Just van Rossum <just at letterror.com> writes:

  JvR> At 6:10 PM -0400 13-06-2000, Jeremy Hylton wrote:
  [two ways to build a tuple omitted]
  >> This is a shorter-to-write, slower-to-run version.  If you use
  >> Py_BuildValue, the code has to parse the format string in order
  >> to build the object.  If you are coding in C because you're
  >> trying to speed up a speed-critical part of your program, the
  >> code you posted is the better way.

  JvR> On the other hand, creating tuples is probably not often
  JvR> something one does inside an inner loop...

Yes, in which case Py_BuildValue is fine.

Jeremy





From moshez at math.huji.ac.il  Sat Jun  3 06:38:33 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 3 Jun 2000 13:38:33 +0300 (IDT)
Subject: compiler_question->diret to whom(?)
In-Reply-To: <3938ACF7.9977D53A@mindspring.com>
Message-ID: <Pine.GSO.4.10.10006031337540.17165-100000@sundial>

On Sat, 3 Jun 2000, David Broadwell wrote:

> Ok, i know that the optimizing compiler is being built. But where can I
> direct a question-suggestion about it?
> 
> Ie: "hm, i've got an idea fo an optimization and i want to see if it
> will be of use."

Try the compiler-sig.
You'll find more details about it if you follow the "SIGs" link from
http://www.python.org

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From olczyk at interaccess.com  Wed Jun 28 14:01:36 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 28 Jun 2000 18:01:36 GMT
Subject: Problem with redirection, buffering and chaining scripts.
Message-ID: <395c3b52.183760656@nntp.interaccess.com>

I'm having this problem chaining two scripts together 
( Windows NT ), demonstrated by the demo below.

When script1.py calls script2.py nothing gets printed
until the script finishes. Then the whole output of script2.py
is flushed. Taking one of the command and executing it by hand,
the output comes one line at a time ( as I want it ).

Any idea of how to get script1.py to print output from
script2.py as it comes?

----------------------script1.py--------------------------------
import os,stat,time

def scan_dir(path,func):
    filenames=os.listdir(path)
    for filename in filenames:
        st=os.stat(path+'/'+filename)
        if stat.S_ISDIR(st[0]):
            scan_dir(path+'/'+filename,func)
            func(path+'/'+filename)       

def print_dir(x):
    command="e:/python/python -u script2.py "+x
    print command
    for line in os.popen(command+' 2>&1','r').readlines():
        print ">>",line,

scan_dir("l:/",print_dir)
------------------script2.py--------------------------------------

import os,sys,time


fname=sys.argv[1]

command="e:/bin/ls -l "
command=command+fname+' 2>&1'
print command
for line in os.popen(command,'r').readlines():
   print ">>",line,
   time.sleep(1)

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


From thor at localhost.localdomain  Thu Jun 15 13:30:27 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 15 Jun 2000 17:30:27 GMT
Subject: Python is not for the desert [ Re: How Fast Does Python Run?]
References: <8ib1bs$1pv@autodesk.autodesk.com>
Message-ID: <slrn8kibsg.26i.thor@localhost.localdomain>

On Thu, 15 Jun 2000 09:47:45 -0700, Akira Kiyomiya  wrote:
>I wonder if Python runs faster than Perl.

Perl is sure much faster, but that's only because they use a 
secret weapon:

The crazy Camel of the desert. 

This got some advantages and some disadvantages:

Advantages:

- If you're lucky and you own a female Camel you can drink some milk.

Disadvantages:

- Camel may bite you, or hit you. You know, they're quite illiterate
and rude.

- While going thru the desert you can be violently anally raped by
some arabs, or much worse, you may have to understand other's people
Perl's code. 


-- 
MGA


From rajarshi at home.com  Thu Jun  8 10:45:59 2000
From: rajarshi at home.com (Rajarshi Ray)
Date: Thu, 08 Jun 2000 14:45:59 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl>
Message-ID: <393FB13F.5F60E657@home.com>

"Dik T. Winter" wrote:
> 
> In article <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta at 4ax.com> urner at alumni.princeton.edu writes:
>  > Recent number theory breakthrough:  you can get the
>  > nth digit of PI without doing all the work needed
>  > to compute all the digits up to the nth digit (same
>  > with some other numbers -- not sure if we know all
>  > the relevant criteria as to which).
> 
> I checked the pages and have to make a correction here.  There is no such
> claim.  The only claim is that it can be done without using much memory.
> As one of the referred pages says:
>   "we have not found a faster algorithm, nor have we proven that one does
>   not exist."
> Se essentially the amount of work is still the same, but the memory
> requirents are smaller.

Aw, poor Ullrich. You've shattered his dream of making a fast parallel
calculation of Pi! How dare you!!


> --
> dik t. winter, cwi, kruislaan 413, 1098 sj  amsterdam, nederland, +31205924131
> home: bovenover 215, 1025 jn  amsterdam, nederland; http://www.cwi.nl/~dik/

-- 
Sir, I have found you an argument. I am not obliged to find you
an understanding.

                                               - Samuel Johnson


From digitig at cix.co.uk  Tue Jun 20 14:08:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 20 Jun 2000 19:08 +0100 (BST)
Subject: Python 3000 -> No, but there's 'Ruby'
References: <394BB8DE.AE1D6A7A@webtv.com>
Message-ID: <memo.20000620190809.43609A@digitig.compulink.co.uk>

In article <394BB8DE.AE1D6A7A at webtv.com>, balrog at webtv.com (balrog) wrote:

> > Is there a link where I can look up what's being considered for Python
> > 3000?
> 
> Half-seriously, Ruby is nicer looking & acting;  

Nicer looking to whom? It looks dreadful to me.


From joerg.wegener at sap.com  Fri Jun 23 10:39:13 2000
From: joerg.wegener at sap.com (Joerg Wegener)
Date: Fri, 23 Jun 2000 16:39:13 +0200
Subject: SAP R/3 and Python ?
References: <3953212C.84E3127F@lemburg.com> <8ivqtr$oo8$1@nntp9.atl.mindspring.net>
Message-ID: <8ivsoe$7gu$1@news1.wdf.sap-ag.de>

Hi,


"Aahz Maruch" <aahz at netcom.com> wrote in message
news:8ivqtr$oo8$1 at nntp9.atl.mindspring.net...
> In article <3953212C.84E3127F at lemburg.com>,
> M.-A. Lemburg <mal at lemburg.com> wrote:
> >
> >does anyone have an idea of how to interface to SAP R/3 from Python ?
> >
> >If there are no such Python interfaces, how much effort would it take
> >to add orders from a web system into an R/3 system ?
> >
> >[I know, my ignorance shines through here: I've never used R/3, but a
> >customer has confronted me with that need.]
>
> Disclaimer: I know little about R/3, too.
>
> IIRC, R/3 uses a database backend.  At worst, it should be possible for
> you to dump data directly into R/3 as long as you're careful to follow
> the virtual schema.

By all means, do NOT attempt this! SAP introduced an additional layer of
indirection between its applications and the database. The logical structure
is stored an the ABAP Dictionary, a data dictionary. Applications access the
database using the definitions found there.

BUT: There is NO 1:1-mapping from the ABAP Dictionary to the actual tables!
The R/3 Database Server performs various optimizations. The reason is that
R/3 is independent of the underlying RDBMS, therefore the database server
encapsulates the application layer from the database. Example: if your ABAP
table has 140 attributes, but the database only allows 128 attributes per
table, the database server will create two tables and distribute the
attributes. In other words, the physical implementation of the tables
depends upon the vendor and version of:
- the operating system
- the database system
used in the individual R/3 installation.

Another good reason not to access the database tables directly: by doing so,
you bypass the business logic in the application layer, and you could well
leave the database in an inconsistent state.

I think the best way to access R/3 is via a protocol called RFC (Remote
Function Call) or via BAPIs (Business APIs). IIRC there is something called
RFC-SDK which allows to write or generate code (I'm not sure here) which
gives access to RFCs via DDE/OLE. Use this in combination with WIN32-Python.
The downside is that you need much more knowledge about R/3 functions, but
this is the best and most reliable way.

Hope this helps

Joerg


> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> Why don't they make poppasmarts or didjeridon'ts?  --Aahz




From ge at nowhere.none  Wed Jun 14 17:51:17 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 14 Jun 2000 17:51:17 -0400
Subject: TCP packet size?
References: <39487D5B.FC96BE3F@rpgarchive.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296AD4@mail.sauder.com>

In article <39487D5B.FC96BE3F at rpgarchive.com>, chris wrote:

> Is there a way to ensure or predict packet size when using TCP?

Nope.

TCP/IP guarantees that the bytes get there in the right order.
It neither knows nor cares about any sort of boundaries between
bytes.

> I'm developing a Multi-user application that functions much
> like a muli-player game.  The applications are sharing xml
> data.  The xml data can get very big 3000-6000 bytes.
> Unfortunately, when the size of the network message gets over
> about 2000, its to get broken up into smaller packets.  So,
> when I send a 3000 byte message it gets cut up into 2-3
> packets.  Anther problem I get is when I send two small
> messages back to back, they are sometimes received together.
> Is there something I'm missing here?

That's the way TCP/IP works.  You're using a stream protocol to
try to send datagrams.  That's like trying to juggle tennis
balls while wearing shoes on your hands.  Shoes are a damn fine
invention, but you've got to use them for the right thing.

You want to use a datagram service instead of a stream service.

There's UDP, but it's unreliable, so you've got to add your own
ack/retry scheme -- bit of pain.  On Linux "man socket" claims
to support reliable datagrams:

   SOCK_SEQPACKET

       Provides a sequenced, reliable, two-way connection-
       based data transmission path for datagrams of fixed
       maximum length; a consumer is required to read an entire
       packet with each read system call.

That sounds like what you want, except it's not implemented for
address type AF_INET (which is probably what you're using).
Bummer.

There's also plain old unreliable datagrams (SOCK_DGRAM), and
reliable, unordered datagrams (SOCK_RDM). Probably the only
ones you can depend on being available are SOCK_STREAM (TCP/IP:
which is what you're using) and SOCK_DGRAM (UDP/IP: unreliable
datagrams).

>I tried using headers and breaking up messages into smaller chunks, but
>the size of the receiving packets doesn't always match what I sent.  I'd
>appreciate any help or suggestions on how I can tackle this problem.
>I'd be happy to send someone my network code.

Maybe you can find somebody who's already written a reliable
datagrram layer that runs on top of UDP/IP?

-- 
Grant Edwards                   grante             Yow!  Yow! We're going to
                                  at               a new disco!
                               visi.com            



From rayleyva at my-deja.com  Sun Jun  4 17:24:40 2000
From: rayleyva at my-deja.com (rayleyva at my-deja.com)
Date: Sun, 04 Jun 2000 21:24:40 GMT
Subject: Found it with ( find /usr | grep config/Makefile )
References: <8h74jt$d9r$1@news2.isis.unc.edu> <m3ya4nzpjj.fsf@h0050047c4794.ne.mediaone.net> <8h9kdo$is4$1@nnrp1.deja.com>
Message-ID: <8heheb$oqr$1@nnrp1.deja.com>

In article <8h9kdo$is4$1 at nnrp1.deja.com>,
  rayleyva at my-deja.com wrote:
> unfortunately the make command for PIL still doesn't work :
>
> make -f Makefile.pre.in boot
>
> Still looks for the file under :
>
> /usr/lib/python/config/Makefile
>
> instead of going to the appropriate directory
>
> /usr/local/lib/python1.5/config/Makefile
>
> I've been playing around with the Makefile.pre.in ... I'm a total
novice
> when it comes to shell-scripts ( wish I had payed more attention to
them
> now! ).
>
> Any ideas would be really, really welcome.
>
> Thanks
> Ray
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

ok ... took a longer look at it, and got it to compile no problems with
the following :

make -f Makefile.pre.in VERSION=1.5 installdir=/usr/local
exec_installdir=/usr/local Makefile

w/o the Makefile at the end it wasn't working.  Just putting it up in
case anyone else can use this.

Ray


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


From rob at hooft.net  Fri Jun 23 04:53:44 2000
From: rob at hooft.net (Rob Hooft)
Date: 23 Jun 2000 10:53:44 +0200
Subject: Countdown
References: <3952952D.88E3588D@asme.org> <3952B3E1.DC36612B@caltech.edu>
Message-ID: <m21z1ook13.fsf@hooft.net>

 RK> So stick it in a function.

 RK> def revrange(*args, **kw):
 RK>    l = apply(range, args, kw)
 RK>    l.reverse()
 RK>    return l

But then why not:

def revxrange(n):
    return xrange(n-1,-1,-1)

....Or more complicated if you need more complicated cases.

Rob

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


From gmcm at hypernet.com  Wed Jun 21 18:34:08 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 21 Jun 2000 22:34:08 GMT
Subject: Python sockets/select
References: <8ip6kt$1lc$1@bob.news.rcn.net>
Message-ID: <8F5ABD5B2gmcmhypernetcom@199.171.54.154>

[posted and mailed]

David Allen wrote:

>When I select() sockets for reading, writing, and error conditions,
>there are sockets (usually connected to clients who have already
>quit or died unexpectedly) that don't show up in the error queue,
>but continually flag select() as being readable.  When I read them,
>of course, '' comes back.

Using Windows, huh? Donn gave you the correct answer for real OSes, but 
Windows does some things, um, differently. AFAICT, the error list is 
useless on Windows.

>For some reason, sometimes '' comes in for valid clients, so I 
>can't just disconnect a client because I get that from them.  

Other than protocol-eating nano-viruses, this shouldn't happen.

Note that, on Windows, select can return a socket as writable, but when you 
go to write it, you get an EWOULDBLOCK (10035 in the Gates scheme). But in 
my experience, getting 0 bytes on read means that sucker's dead.

(Well, a socket ready to accept() also returns as readable, but of course, 
can't be read yet.)

- Gordon


From jkraska1 at san.rr.com  Tue Jun 13 22:00:21 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 14 Jun 2000 02:00:21 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <39468318.9EA637E0@interet.com>
Message-ID: <3946E7DF.9672FFE9@san.rr.com>

> > 2. How long did you use Python before you achieved increased
> > productivity ?

2-4 weeks for me.

> Maybe six months.

> Relative to C++, Python is a simpler OO language to learn
> and use.

It's *FAR FAR FAR* easier to learn. Furthermore, given
the deliberate paucity of primitives and forms, you're
very unlikely to run across unusual python idioms or
forms in other people's code, so being able to read and
quickly understand other people's code is much easier
in python, IMO.


C/


From jpb at mycroft.communiweb.net  Sat Jun  3 10:39:07 2000
From: jpb at mycroft.communiweb.net (jpb)
Date: Sat, 03 Jun 2000 14:39:07 GMT
Subject: Erm... MacOS X?
References: <eviltofu-1AEA18.02454303062000@news.earthlink.net>
Message-ID: <jpb-4E0BEF.10391003062000@news-server.cfl.rr.com>

In article <eviltofu-1AEA18.02454303062000 at news.earthlink.net>, Jerome 
Chan <eviltofu at rocketmail.com> wrote:

>Has anyone compiled Python for MacOS X?

You can download it from softrak.stepwise.com

Just search for Python and MacOS X

-- 
Joseph Block <jpb at communiweb.net>
"Any government that is powerful enough to give you everything you want,
 is powerful enough to take away everything you have."
 - Barry Goldwater


From sholden at bellatlantic.net  Tue Jun 27 00:14:26 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Tue, 27 Jun 2000 04:14:26 GMT
Subject: Pick or Universe Module
References: <3957612D.B631299D@laddservices.net>
Message-ID: <39582A17.6B772CFA@bellatlantic.net>

There is a product called Redback (?) which bring Pick/Universe
data into the COM world, and you could use PythonCOM to access
this.  Be aware it is payware, though.

regards
Steve Holden

Brad Losh wrote:
> 
> has anyone seen a python way to talk to a Pick or Unverse DBMS.  if
> there is nothing I would like to start something, just not sure how.
> 
> Brad Losh

-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From akira.kiyomiya at autodesk.com  Thu Jun 15 12:47:45 2000
From: akira.kiyomiya at autodesk.com (Akira Kiyomiya)
Date: Thu, 15 Jun 2000 09:47:45 -0700
Subject: How Fast Does Python Run?
Message-ID: <8ib1bs$1pv@autodesk.autodesk.com>

I wonder if Python runs faster than Perl.

Could someone give me some comparison and contrast with some languages such
as Perl, Tcl, JavaScript, etc?

I am just curious.

Thanks

Akria








From dfan at harmonixmusic.com  Wed Jun  7 10:22:05 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 07 Jun 2000 10:22:05 -0400
Subject: Embedding + extending without DLLs
Message-ID: <wkbt1da7w2.fsf@turangalila.harmonixmusic.com>

I would like to simultaneously embed and extend Python in a Win32
project.

That is, I'm writing a C++ application which uses Python as a
scripting language, and the Python code can call back into the C++
code to perform actions or queries.

I actually already have a little toy program that does this, and it
works fine.  However, it works by having a main .exe and then a
separate foo.dll, which Python then finds with an 'import foo'
statement.

What I'd prefer is to have Python be able to call directly into my
.exe, since the project already exists as a monolithic executable and
isn't split up into DLLs.  I suppose I can turn the whole thing into a
DLL and write a tiny .exe to load it up, but I wanted to check if
there's a natural way to let Python call my code without changing the
structure of my program.  Is there?

Thanks,
Dan

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


From robin at jessikat.demon.co.uk  Thu Jun 22 14:36:47 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 22 Jun 2000 19:36:47 +0100
Subject: Roman Number Representation
References: <200006221810.AA05499@ch1d2833nwk>
Message-ID: <KoAJ5WA$ylU5Ewvu@jessikat.demon.co.uk>

In article <200006221810.AA05499 at ch1d2833nwk>, Ted Horst
<Ted.Horst at ubsw.com> writes
>
>Ok, here is my take on how to do roman numerals in python.  rn and an are  
>simple, basic routines that are only valid for positive numbers.  rn2 and  
>an2 expand into negative numbers and use a goofy notation that I made up for  
>large numbers.  Their range is limited only by your memory and patience.
>
...
thanks for all the reponses + at least 3 lots of code
-- 
Robin Becker


From jmccarty at sun1307.ssd.usa.alcatel.com  Fri Jun 16 15:36:48 2000
From: jmccarty at sun1307.ssd.usa.alcatel.com (Mike Mccarty Sr)
Date: 16 Jun 2000 19:36:48 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <394a5109.80394991@10.1.1.28> <394A5795.F4DE98FB@san.rr.com> <394a5ed7.83928743@10.1.1.28>
Message-ID: <8idvkg$qm8$1@relay1.dsccc.com>

In article <394a5ed7.83928743 at 10.1.1.28>,
Tim Dixon <tdixon.no at spam.fwi.com> wrote:
)On Fri, 16 Jun 2000 16:33:59 GMT, Courageous <jkraska1 at san.rr.com>
)wrote:
)
)>
)>> Suppose, however, that you could get digit 'n' of pi directly (or e,
)>> or any other inifinte, nonrepeating sequence).  All I have to do is
)>> communicate which digit of pi the message key starts at, and I can
)>> generate the rest of the key.
)>
)>This is a collosally bad idea, IMO.
)>
)>
)
)Just for curiosity, why?

For one thing, it doesn't implement a one-time-pad. If I guess that you
are using PI as your key, then I can decrypt all your messages.

Mike


-- 
----
char *p="char *p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
I don't speak for Alcatel      <- They make me say that.


From richard at insight20.freeserve.co.uk  Wed Jun 14 14:40:48 2000
From: richard at insight20.freeserve.co.uk (richard at insight20.freeserve.co.uk)
Date: 14 Jun 2000 19:40:48 +0100
Subject: python <--> MySQL w.r.t. binary files (bitmap images like .jpg's)
Message-ID: <86ya48qf67.fsf@insight20.freeserve.co.uk>

Hi

fairly new to python - hope this a helpful post

set context - here's a message I sent - 

"Hi all

 Got the ans to the issue of loading binaries, which includes bitmap
 image files like .jpg's and .png's, into the MySQL database.

 Proves that the Python/MySQL pairing makes it simple (touch wood).

 For those of you less familiar, MySQL is a relational database.  Ok -
 slight qualification - the relational database is the data, organised
 in "tables" which relate to eachother, all regarded as a single entity
 known as "the database".  (on a unix file system (and other current
 platforms) the "database" is represented by a directory and the
 "tables" by individual files within that directory - there are also
 other files which are maintained by the database *program* to
 facilitate indexing etc)
 Have slightly "jumped the gun" in making the last distinction - MySQL
 is a *database program* which manages and operates upon databases and
 provides access into the databases for the user (the program can
 oversee many databases).

 So, loading binaries to the database... given that there is a datatype
 "BLOB" (TINYBLOB (256Bytes?) up to and including LONGBLOB for binary
 data up to 4GB)...

 The interface to the MySQL database program is the "mysql" monitor
 (that's what you start-up when you type "mysql" at the command line) -
 it starts-up and you identify yourself to the database program.  You
 can also reach into the database from programming languages via
 modules written for the language to interface with database program
 through "calls" provided for this purpose by the database program.

 The problem I ran into was that binary files can "accidentally appear
 to have" control characters which trigger events (unwanted) in the db
 interface.  For instance in a .jpg, you are going to at some point
 have a sequence of bits at some multiple of 8bit (1byte) offset from
 the beginning of the file which reads 00001010, which has no relation
 at all to the ASCII "newline" character, which in the ASCII character
 set has this bit pattern (it's represented by the hexadecimal number
 0x0A, which if you convert that to binary is the bit sequence I have
 just quoted).  There are other bit patterns which "appear" the same as
 other ASCII control characters.  The single-quote and double-quote
 characters also cause problems to the "mysql" MySQL monitor and/or the
 interface.

 What you need to do is to "escape" these bit patterns in the binary
 file by inserting the slash character before these "problem"
 bit-patterns (they are not seen when the binary file is recovered from
 MySQL).

 Jumping straight to the answer - (correct me if I am wrong on the
 location of this function) the MySQL API (Application Programming
 Interface) provides the function "mysql_escape_string()" which
 performs the "escaping" of the characters which would otherwise cause
 the interface problems during reading-in.  This has been "wrappered"
 in the MySQLdb-0.2.1 module which abstracts the API as a set of Python
 objects.  It is in the _mysql module, which you would not normally use
 directly because it presents a fundamental API which is in general
 abstracted into higher level API's so that your SQL is transportable
 between databases.  However, this is an application-specific problem,
 and the answer is application specific.  So, let's rock...

 In patchy pieces of Python (complete listing of an example follows)

 import _mysql
 import MySQLdb
 (connect to database)
 (create a cursor object)
 ("open" your binary (eg. (bitmap) image) file and get a file handle
 object)
 (build a string of the SQL you want to "execute" which looks like this
 sqlstr="update ... set ...='" + _mysql.escape_string(imgfile.read())
 +"' where ...")
 cursor_obj.execute(sqlstr)

 That's it - the binary is in the database.

 When you "select" the thing, you get it back as the binary pre- the
 "escaping" operation.

 Have fun.

 Here's the example code listing...

 Richard Smith

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

 #!/usr/local/bin/python

 import _mysql
 # MySQL (ie. implementation) specific - but fairly
 # implementation-specific problem

 import MySQLdb
 db=MySQLdb.connect(db="cheeses",user="a_user",passwd="a_passwd")
 c=db.cursor()

 imgfile=open("a_cheese.png",'rb')
 ## imgvar=imgfile.read() # avoid this

 # put SQL strings in place
 # "examine_str"
 sql_db_view="select * from cheese_name where cheese_name_pk=1"
 # reset
 sql_reset_img1="update cheese_name set picture_img=NULL where cheese_name_pk=1"
 # specifically retrieve the image field
 sql_db_image_view="select picture_img from cheese_name where cheese_name_pk=1"
 # "do it" SQL command - the "update" loading of the image file into the database
 sqlstr="update cheese_name set picture_img='" + _mysql.escape_string(imgfile.read()) +"' where cheese_name_pk=1"

 # do reset...
 c.execute(sql_reset_img1)

 # do the "update" load...
 c.execute(sqlstr)

 # look at what the result is...
 c.execute(sql_db_image_view)

 # do file write
 f_out=open('tmp.img.png','wb')
 f_out.write(c.fetchone()[0])
 # flush the cache / buffer whatever now
 f_out.close()

 # other tidy-ups
 imgfile.close()"

To which I got one answer...

"> import _mysql
 > import MySQLdb
 > (connect to database)
 > (create a cursor object)
 > ("open" your binary (eg. (bitmap) image) file and get a file handle
 > object)
 > (build a string of the SQL you want to "execute" which looks like this
 > sqlstr="update ... set ...='" + _mysql.escape_string(imgfile.read())
 > +"' where ...")
 > cursor_obj.execute(sqlstr)

 If you were using Oracle and Pro*C, you would do something like:

 EXEC SQL UPDATE ... set column=:value

 where ':value' represents a C variable containing a value.  The value in the
 C variable isn't parsed by the SQL parser and needn't be escaped.

 Perl's DBI and DBD interface has something similar, though I can't remember
 exactly how it's done.

 I would be amazed if there wasn't a similar binding step in Python's
 database interface (though, I admit, it does sound less well thought out
 than the Perl one, which has a standard front-end and a variety of back-ends
 for different databases).  You should probably use this rather than taking a
 large binary file, converting it (effectively) to text, forcing the parser
 to parse it and convert it back to binary."

(from a very helpful much appreciated character)

So - is there a neater way of doing this - such as in line with the
hint of the (very much appreciated) respondent?

Richard Smith


From petersc at stud.ntnu.no  Sun Jun  4 07:40:00 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 13:40:00 +0200
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk> <393A3A9A.FB9F2E37@stud.ntnu.no> <393A4C60.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <393A4010.662097AD@stud.ntnu.no>

Bablos wrote:
> 
> This is buffered on the Amiga, I assume.  I see very infrequent
> updates, so I assume it saves up a large number of print statements
> and then flushes them when it reaches a certain size.  Irmen?

import sys

and then just put a

sys.stdout.flush()

after the print statement. Your \v does NOT work on my linux.
> But hey, at least '\v' works for me.  If it's all so very platform
> dependant, surely it would be wise to standardise it somehow to
> maintain some form of standard method?

Good question, I've been thinking about it. But then you
would have to have a very platform dependent module which
uses different methods according to the needs of the platform.
Not very nice to maintain.

not-too-bad-an-idea-though-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From tim_one at email.msn.com  Sat Jun 10 22:52:16 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 10 Jun 2000 22:52:16 -0400
Subject: spaces in filenames
In-Reply-To: <slrn8k2o30.3r4.nascheme@cranky.arctrix.com>
Message-ID: <000201bfd350$0d818660$f6a2143f@tim>

[Tim, suggests forcing a Windows Python 1.5.2 install to use a
 path without embedded spaces]

[Neil Schemenauer]
> This is something that pissed me off about Windows 95.  Microsoft
> just had to show off their new feature allowing spaces in
> filenames by putting a space in one of the most important
> directory names on the system.  Unix has allowed weird characters
> in filenames for ages.  Just because you can do it doesn't mean
> its a good idea.  I wonder how much time has been wasted
> worldwide because Microsoft did not call "Program Files"
> "Programs".

Probably lots, but probably not by end-users.  This is something only
developers give a rip about (but anyone who downloads Python with the intent
of programming falls into this category!).

"The typical end-user" is exemplified for me by a particular sister, an
artist by training and inclination but bureaucrat by twist of fate.  I
bought her her first computer a couple years ago, and it was a Win95 box
(had I bought her a Linux box, she'd still be trying to boot it <0.3 wink>).
Spaces in file names are as natural as breathing to her -- that's the way
*everything* in her written life works.  I don't think ehe's ever seen a DOS
box.  And if you stick-- as I suspect the overwhelming majority of Windows
users do --purely to GUI apps, spaces in paths simply don't bite you.

or-when-they-do-you-blame-something-else-like-a-mean-brother-ly y'rs  - tim





From darrell at dorb.com  Tue Jun 27 00:31:35 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 26 Jun 2000 21:31:35 -0700
Subject: Script which only runs in Windows
References: <8j8d2n+qbe9@eGroups.com>
Message-ID: <03d801bfdff0$943d8cb0$6401a8c0@home>

----- Original Message ----- 
From: <al at servana.com>
To: <python-list at cwi.nl>
Sent: Monday, June 26, 2000 1:05 PM
Subject: Script which only runs in Windows


> Hello.
> 
> We have an odd problem with Python which we cannot resolve.
> 
> Below, is a script we have been running in Win32 for months.
> 
> We moved our scripts from Win32 to Slackware Linux and IF statements 
> which
> are not in functions or classes will cause errors when you
> try to run the script.
> 
> I have had retyped the script thinking it was the tab/space
> issue, but nothing changes - in Linux we cannot use IF outside
> a function or CLASS for some reason.
> 
> We are posting this here as it seems very serious and we were hoping 
> someone else had seen the problem.
> 
> #
> # here is where we do some work
> #
> try:
>     sResult = os.system('python ftpasget.py -q -g * -d ' + sFileTime 
> + ' -l ' + sSugLogin + ' -p ' + sSugPWord + ' ' + sSugSite + ' ' + 
> sBillDir + ' ' + sOutputDir + sDirTime)
> except:
>     sGetMinutes = 0
> 

I'd say the os.system call is failing.
Try printing it out and testing it manually.
This causes sResult to not exist.

--Darrell




From olczyk at interaccess.com  Wed Jun 14 17:32:34 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 14 Jun 2000 21:32:34 GMT
Subject: boolean true and false values.
Message-ID: <3947f922.425016156@nntp.interaccess.com>

Does python have true and false values.
I would like to do something like

class A:
     def _init_(self):
           self.processed=false
     def process(self):
	self.processed=true

at the moment I am using
false=(1==0)
true=(1==1)
but would like something more elegant, or at least less sloppy.


From cut_me_out at hotmail.com  Sat Jun  3 19:26:06 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Jun 2000 19:26:06 -0400
Subject: Automatic reload()?
References: <8hbnt1$62i$1@news2.isis.unc.edu> <m3r9aexydb.fsf@atrus.jesus.cam.ac.uk> <etdd7lyo1pp.fsf@w20-575-109.mit.edu>
Message-ID: <etdbt1iny7l.fsf@w20-575-109.mit.edu>

import sys, string
from types import ModuleType

class Reloader:

    ''' 
    This is so complex that I've probably added more problems than I've
    solved.  But anyway, here is my attempt at something that figures
    out an appropriate reloading order for you.  You create an instance
    of the class, call it 'rr', before you have imported any of the
    modules you'll be wanting to reload (the .pythonrc file would be a
    good place.)  Then when you wish to reload the modules in a given
    namespace, you run the command

    exec (rr (globals ())) 

    In my case, the namespace would usually be __main__, so this would
    just go at the top of the script I'm currently working on in
    emacs.

    Alex.
    '''

    def __init__ (self):
        self.keepem = sys.modules.copy ()

    def get_new_modules (self):
        modules = {}
        for i, m in sys.modules.items():
            if (not self.keepem.has_key (i)) and \
               (type(m) == ModuleType):
                modules[i] = m
        return modules

    def clear_sys_modules (self):
        for i in self.modules.keys ():
            if sys.modules.has_key (i):
                del sys.modules[i]

    def get_dependancies (self, module_name):
        self.clear_sys_modules ()
        exec 'import ' + module_name
        new_modules = self.get_new_modules ()
        del new_modules[module_name]
        return new_modules

    def determine_reloads (self):
        self.return_commands = []
        while self.dependancies:
            for i, deps in self.dependancies.items ():
                if not deps:
                    break
            else:
                raise 'Cyclic dependancy(?)'
            exec 'import ' + i
            reload (eval (i))
            if self.namespace.has_key (i) and \
               (self.namespace[i] == self.modules[i]):
                self.return_commands.append (
                    'import %s; reload (%s)' % (i, i))
            del self.dependancies[i]
            for deps in self.dependancies.values ():
                while deps.count (i):
                    deps.remove (i)

    def __call__ (self, namespace):
        self.namespace = namespace
        self.modules = self.get_new_modules ()
        self.dependancies = {}
        for module_name in self.modules.keys ():
            deps = self.get_dependancies (module_name).keys ()
            self.dependancies[module_name] = deps
        self.determine_reloads ()
        return string.join (self.return_commands, ';')

rr = Reloader ()

if __name__ == '__main__':
    '''This is a test script, intended to be run with two modules, B.py
    and C.py somewhere in the directories in sys.path.  Their contents
    should be

    B.py:
    ######################################################################
    import C

    class B (C.C):
    
        def __init__ (self):
            C.C.__init__ (self)
        
    C.py:
    ######################################################################
    class C:

        def __init__ (self):
            pass
    '''
    import B, C

    bid = id (B.B)
    cid = id (C.C)
    exec (rr (globals ()))
    assert (bid != id (B.B)) and (cid != id (C.C))

    t = B.B ()


From m.faassen at vet.uu.nl  Thu Jun 15 09:44:45 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jun 2000 13:44:45 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net> <uitvbluqx.fsf@ctwd0143.fitlinxx.com> <2RZ15.3850$Gp6.258234@dfiatx1-snr1.gtei.net>
Message-ID: <8iamkd$dro$1@newshost.accu.uu.nl>

jin choung <jin.choung at gte.net> wrote:
[snip]
> yes, it was my assumption that C++ was the OO god but it does not surprise
> me at all that i am way off on it :)  but for the c++ book at least, it does
> not assume any kind of knowledge of any abstract concepts and it starts from
> the foundation of absolute ignorance.

Right, as I guessed earlier in the thread; the difference is in the books,
not the languages. Python is easier to learn than C++. Your C++ book is
however better for beginners than your Python book. My friends experience with
"Learning Python" is similar. I should get her one of the other Python
books that are better for absolute newbies, I guess.

Regards,

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


From hzhu at rocket.knowledgetrack.com  Fri Jun  9 18:57:55 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 09 Jun 2000 22:57:55 GMT
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net> <slrn8k2gq6.pi.hzhu@rocket.knowledgetrack.com> <20000609220152.E20700@xs4all.nl>
Message-ID: <slrn8k2tog.10m.hzhu@rocket.knowledgetrack.com>

>But Guido suggested __add_ad__ in his latest posting on this subject, and it
>is ultimately his choice. If he comes up with a better name, or changes his
>mind in favor of whatever suggestion whoever made, it's a simple case of
>search-and-replace to fix it, anyway ;)

Before he makes final decision I'd make another suggestion

__add_by__

Same length.  A little less cryptical.  I've tried a few stituations and it
makes some sense.  Another possibility is __incr__ but this can't be done
systematically for all such operators.

Huaiyu


From kens at sightreader.com  Tue Jun 13 05:15:29 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 13 Jun 2000 02:15:29 -0700
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <3945FBB1.9CE9B636@sightreader.com>

Steve Mullarkey wrote:

> <snip>

> In "C++" I have :-
>
> 1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
> Profilers, Programmers Editor including Class Browser, etc..

I look forward to some nice tools under development.  Right now, I'd say
that overall C++ tools are better than the python tools (aat least in the
Windows environment).  On the other hand, imho, python doesn't need the
tools as much.

GUI building tools - C++ wins.  Until Boa Constructor arrives.  There's also
PythonWorks, which I haven't looked at.

Memory leak tools - Python doesn't need them (except for circular refs which
aren't that tough anyway).  Python wins here.  The next major release of
python will probably have true garbage collection, which will remove orphan
circular refs.

Profilers - Never looked, so I don't know if python has any good ones.  If
I'm concerned with saving a few milliseconds, I code an extension.

> 2. I use the STL (Standard Template Library) which gives powerful string
> and container classes.

Having strings and container classes as built-in types with language syntax
is a big advantage for python.  I find that while STL is perhaps as
/powerful/ as python's set built-in types, it still takes me longer to code
with it, and my code is much less readable than it would be in python.

> 3. I use incremental compiling and linking. This reduces wait times to
> very small amounts.

True, but to me this isn't as nice as having an interactive interpreter.
Suppose you write a function to manipulate a list in some way.  Now you want
to test it.  In python, you call the function from the interpreter, perhaps
expressing the list as a literal.  In C++ you write a test function, maybe
complete with some user interface code, debug your test function, compile
(very quickly with your incremental compiler) and run your test function,
and then try to figure out whether your bug is in the function being tested
or your test function.  Anyway, I find that the typical write-test-debug
cycle goes much faster in an interpreter, and with less distraction.

BTW, seems to me that in theory one could write a C++ interpreter.  The idea
would be to make the compiler available through a library function that
would compile the one-liner into a (very small) shared library, link to it,
and call it.  Does anyone know it that has been done?

> 4. I can also use background compiling and linking.
>
> 5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
> etc..
>
> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated. I am genuinely interested
> to ask "C++" programmers :-
>
> 1. What productivity increase do you achieve ?

Very hard to measure.  I think maybe 4 to 1, but I can't prove it.

> 2. How long did you use Python before you achieved increased
> productivity ?

I could match my C++ within a couple weeks.  Maybe a couple months to get to
4-1, mostly because I had to look for new libraries.

> Best regards,
>
> Steve Mullarkey.
> Tascom Pty Ltd.
> Berry, NSW 2535, Australia.
> Tel : (61)-(2)-4464-3328

Where does it come from?

The interpreter certainly helps, but that's not really the main thing.

When I code in C++ or python, I spend most of my time testing and debugging,
so the question is: How does python save time on debugging?

In C++ most persistent bugs are related to memory and type issues.  Python
handles memory for you, and gives less potential for type related problems
(despite the nonsense about C++ being safer because of static type checking
(oops, I hope I didn't just start a flame war :-) ).  So most of these bugs
don't appear in the first place.

I find that the Visual C++ debugger is better than the python debugger
(though eventually someone will write a really awesome python debugger,
which should be relatively easy given python's introspective capabilities).
However, I find that in the two years I've been programming with python, I
use the debugger less often than when I'm programming with C++ (which I've
been doing for about twelve years).

Python bugs are usually much easier than C++ bugs.  Write down a brief
description of the next ten significant bugs you encounter in C++ and you
will see what I mean.

Dealing with types in C++ is often very time consuming.  C++ has a vast
number of redundant basic types that various people have wrapped and aliased
in various ways.  By basic types I mean integers, floats, lists, strings,
and mappings (ok, I'm being python-centered here).  In C++ I find myself
spending more time making sure I have the right kind of integer, or
converting from one string type to another, or figuring out how to iterate a
particular kind of list, etc.  In python, if someone invents a new sequence
type, or implements a special kind of mapping object, I immediately know how
to use it with no guessing or reading manuals.  The STL certainly helps C++
overcome this, but STL was invented long after C++, so you still have to
deal with the multitude of types.  Python's containers were there from day
1.

The learning curve for third party libraries is a major factor in
productivity.  In python it is often said that there is one obvious way to
do a thing.  As a result, it is usually easy to remember (or even guess)
correct usage.  You never have to ask, "Is that supposed to be an unsigned
short int or a long int?", or "Is that a const char *, a char const *, or a
MegaWidgetString& ?", or "Am I supposed to delete that?".

I like to write my own custom programming tools (whether in C++ or python),
especially for large projects.  Python programming tools are easier to write
because of it's dynamic nature, and because string processing is easier.

Why is string processing easier in python?  Python doesn't have more string
processing capability but python can do the same things with much less
syntactical messiness.  There is no need for a string class to wrap the
basic type.  And then there's the handy slice operators that are
syntactically cleaner than anything you can get in C++.  A regex library is
a regex library, but it sure is nice to test your regular expression on the
command line (perhaps 10 seconds per test rather than 20-30 seconds with a
C++ compiler).

I don't have anything against C++, but for me, the role of C++ is changing.
It is becoming what assembly used to be when I was programming with C: a way
to code that inner loop that has to go really fast.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000613/8f553a61/attachment.vcf>

From dworkin at ccs.neu.edu  Tue Jun 20 22:26:20 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 20 Jun 2000 22:26:20 -0400
Subject: Python vs Perl
In-Reply-To: andyl@azaal.dircon.co.uk's message of "19 Jun 2000 09:18:39 GMT"
References: <8ikai5$v3d$1@nnrp1.deja.com> <394ebe95.255648282@news.online.no>
	<slrn8krpbf.gc.andyl@azaal.dircon.co.uk>
Message-ID: <vndvgz3vkfn.fsf@camelot-new.ccs.neu.edu>

andyl at azaal.dircon.co.uk (Andy Leighton) writes:

> >* Perl is harder to read than Python. This makes debugging and re-use
> >of other peoples code harder.
> 
> Of course if you only know Perl, I would think Perl would be the easiest
> to read.

I have seen quite a bit of evidence to the contrary recently.

-Justin

 



From tanzer at swing.co.at  Fri Jun  9 02:45:21 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 09 Jun 2000 08:45:21 +0200
Subject: Python Productivity over C++ 
In-Reply-To: Your message of "Fri, 09 Jun 2000 05:12:56 GMT."
             <8F4E9C028stevemulozemailcomau@139.130.250.4> 
Message-ID: <m130IXh-000wcEC@swing.co.at>

stevemul at ozemail.com.au (Steve Mullarkey) wrote:

> I have read in several places productivity claims of 5 to 10 times over
> 'C' and "C++". I would like to ask for some feedback from "C++"
> programmers who have moved to Python as to whether these estimates are
> realistic.
> 
> In "C++" I have :-
> 
> 1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
> Profilers, Programmers Editor including Class Browser, etc..
> 
> 2. I use the STL (Standard Template Library) which gives powerful string
> and container classes.
> 
> 3. I use incremental compiling and linking. This reduces wait times to
> very small amounts.
> 
> 4. I can also use background compiling and linking.
> 
> 5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
> etc..
> 
> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated.

There are lots of reasons. The ones first springing to my mind are:

- The tools you mention notwithstanding, C++ still forces you to write
  lots of low-level code by hand (e.g., memory management, the
  `orthodox canonical class form').

- C++ doesn't offer any means of meta-programming. If you want to add
  some mechanism (like serialization, introspection, ...) to a class
  hierarchy, you'll have to add member functions to every class in the
  hierarchy. In Python you can implement the complete mechanism in the
  root class of the hierarchy.

- Incremental compiling/linking doesn't help when you need to refactor
  your product or want to add a mechanism to a class hierarchy. In
  that case, you might have to wait a long time for the product to be
  recompiled and linked. In Python, you might be able to pull the
  change into the running application.

- I could go on, but I have to leave for work...
 
> I am genuinely interested to ask "C++" programmers :-
> 
> 1. What productivity increase do you achieve ?

At least a factor of five if comparing C++ and Python.
 
> 2. How long did you use Python before you achieved increased
> productivity ?

Less than two weeks.

Regards,
Christian

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From aahz at netcom.com  Wed Jun  7 09:50:50 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 13:50:50 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr>
Message-ID: <8hljvq$mun$1@slb7.atl.mindspring.net>

In article <m33dmpj1b9.fsf at chinon.cnrs-orleans.fr>,
Konrad Hinsen  <hinsen at cnrs-orleans.fr> wrote:
>m.faassen at vet.uu.nl (Martijn Faassen) writes:
>>
>> Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
>> nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?
>
>*Blijken* te spreken. Misschien zijn we allemal bots die van een
>enkele (nederlands) programmeur komen. Of misschien is er maar een
>enkele mens in deze nieuwsgroep, die honderde adressen gebruikt om
>de indruk te wekken dat Python heel populair is. Of misschien is
>er een vertalingsprogramma dat alle teksten gewoon naar het Engels
>vertaald, maar af en toe naar het Nederlands.
>
>Er zijn dus merdere rationale verklaringen, dus geen behoefte om
>aan kosmische krachten te geloven!

Okay, okay, a joke's a joke, but if people are actually going to write
full paragraphs in Dutch, it'd be nice if someone wrote translations for
the small minority of us who don't know Dutch.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From bvdpoel at uniserve.com  Wed Jun 14 21:17:10 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Wed, 14 Jun 2000 18:17:10 -0700
Subject: tkinter: wait_variable
Message-ID: <39482E96.350D115E@uniserve.com>

How does wait_variable() work? I've got the following in a class:

	__marked = []

	....

	def mark(self):
		a callback to mark certain items
		self.__marked.append(x)

	def wait(self, root):
		root.wait_variable(self.__marked)
		return self.__marked


and I call it, after setting up,etc:

	print myclass.wait(root)

However, it never seems to return. With test print statements I have
determined that my __marked variable is being updated...but
wait_variable() doesn't seem to sense it.


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



From mwh21 at cam.ac.uk  Wed Jun  7 18:53:51 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 07 Jun 2000 23:53:51 +0100
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com>
Message-ID: <m3em69w1a8.fsf@atrus.jesus.cam.ac.uk>

Before we start: you're a sick man.

Ben Wolfson <rumjuggler at cryptarchy.org> writes:

> I'm attempting to abuse Python's lambdas, and I've come across a
> problem I can't figure out.  Well, the problem is actually that I'm
> attempting to abuse lambdas, but aside from that I don't know why the
> abuse is failing.  Here's what I've got:
> 
> def __everyother(): #factory to return lambdas.
> 	return lambda l, evens, odds, app=lambda a,e: lambda
> a=a,e=e:a.append(e): Compose(apply(map, (lambda e, evens=evens,
> odds=odds, l=l,app=app: If(l.index(e)%2==0, app(evens,e), app(odds,e))
> ,l)), Return(evens, odds))
> 
> The functions it calls are:
> 
> def Return(*args):
> 	return lambda args=args: args
> 
> def Compose(*args):
> 	return lambda fs=args: For(len(fs), lambda c, fs=fs:
> If(lambda c=c,fs=fs: c==len(fs)-1, Return(fs[-1]()), fs[c]))()
> 
> def For(count, fct, *args, **kwargs):
> 	for i in range(count):
> 		fct(i,*args,**kwargs)

You don't post the code for If, and I'm a bit confused, because in one
place:

> odds=odds, l=l,app=app: If(l.index(e)%2==0, app(evens,e), app(odds,e))

you pass a value as the first parameter, and in another:

> If(lambda c=c,fs=fs: c==len(fs)-1, Return(fs[-1]()), fs[c]))()

you pass a thunk.  Which do you want?  I suspect this may be giving
the error you're suffering, though I can't get it to work at all.

Also:

> a=a,e=e:a.append(e): Compose(apply(map, (lambda e, evens=evens,
                               ^^^^^
what's this apply for?

i'm-sure-this-is-good-practice-for-a-maths-exam-ly y'rs
Michael

-- 
  I also fondly recall Paris because that's where I learned to
  debug Zetalisp while drunk.                          -- Olin Shivers


From m.faassen at vet.uu.nl  Wed Jun  7 20:25:54 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Jun 2000 00:25:54 GMT
Subject: TypeError: unsubscriptable object
References: <393ECFC3.879BC357@mordor.ch>
Message-ID: <8hmp6i$ets$6@newshost.accu.uu.nl>

Cyrille Artho <cartho at mordor.ch> wrote:
[snip]
> I tried it with simpler examples, where it worked (W was also a
> member of an object, where I accessed it later - no problems). Why
> doesn't it work now? I can send someone the entire script I have written
> so far if it helps. Help would be greatly appreciated!

What's possible (but it's not possible to determine this from your snippets)
is that you accidentally assign something else to W elsewhere. Try putting
in a 'print self.W' just before you try to change it, and see what it
says. If it's an unsubscriptable object (something isn't listlike in its
behavior), then you know something's wrong. :)

Good luck!

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


From mfletch at tpresence.com  Mon Jun 12 10:40:33 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 12 Jun 2000 10:40:33 -0400
Subject: Newbie: displaying binary numbers
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33B0@VRTMAIL>

Here are some _very_ inefficient mechanisms for going to/from arbitrary-size
integers... note that unbits will crash and burn if you pass it something
too large for an integer representation.

Enjoy,
Mike

>>> def unbits( value ):
... 	basevalue = string.atol( value, 2 )
... 	if basevalue & 2L**(len(value)-1):
... 		basevalue = basevalue-2L**len(value)
... 	return int( basevalue )
... 
>>> def bits( value, size=32 ):
... 	bits = ['0']*size
... 	for bit in range( size-1 ):
... 		if value & 2L**bit: bits[bit] = '1'
... 	bits.reverse()
... 	if value < 0:
... 		bits[0] = '1'
... 	return string.join( bits, '')
... 



From fiona at sitegnome.com  Fri Jun 16 07:10:56 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 16 Jun 2000 11:10:56 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 16th, 2000
Message-ID: <20000616111056.18561.qmail@synop.com>

Hi!

My arrangement with FAQTS was to get 225 entries into 
http://python.faqts.com by June 16.  I've done it! With a lot of help 
from you guys.  Thanks everyone.

This doesn't mean my daily entries will cease, I'm going to continue 
working on the FAQTS Python Knowledge base for a while to come.  Its fun 
and educational :-)

Regards,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
How can I map a drive using pythonwin, and see if it was successful or not?
http://www.faqts.com/knowledge-base/view.phtml/aid/3791
-------------------------------------------------------------
Fiona Czuczman
Fred Gansevles, Roger Upole

This code written using the 'dynwin' package.

# wnet.py -- wnet module

from dynwin import winwin, winclass, msgloop, windll
from dynwin.structob import struct_object
from dynwin.oracle import Oracle

mpr = windll.module ('mpr')

RESOURCETYPE_DISK       = 0x00000001

class NETRESOURCE (struct_object):
        oracle = Oracle (
                'NETRESOURCE',
                'Lllllllll',
                ('dwScope',
                 'dwType',
                 'dwDisplayType',
                 'dwUsage',
                 'lpLocalName',
                 'lpRemoteName',
                 'lpComment',
                 'lpProvider')
                )

def use(drive, host, user, passwd):
    nr = NETRESOURCE()
    nr.dwType = RESOURCETYPE_DISK
    lpLocalName = windll.cstring(drive)
    lpRemoteName = windll.cstring(r'\\%s\%s' % (host, user))
    lpProvider = windll.cstring('Microsoft Windows Network')
    nr.lpLocalName = lpLocalName.address()
    nr.lpRemoteName = lpRemoteName.address()
    nr.lpProvider = lpProvider.address()
    return mpr.WNetAddConnection2(nr,
                windll.cstring(passwd), windll.cstring(user), 0)

def unuse(drive):
    return mpr.WNetCancelConnection2(windll.cstring(drive), 1, 1)

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

You should be able to use the COM interface for Windows Scripting host.
import win32com.client
wnt=win32com.client.Dispatch('Wscript.Network')
wnt.MapNetworkDrive('z:','//someserver/someshare')
To test it, just try to access the drive:
os.listdir('z:') or something similar.


-------------------------------------------------------------
What is a tuple and why is it used (as opposed to an array, etc)?
http://www.faqts.com/knowledge-base/view.phtml/aid/3795
-------------------------------------------------------------
Fiona Czuczman
Stephen Hansen, Aahz Maruch, Warren Postma

A Tuple is of a fixed length, the length it was when it was created. A
list is of a dynamic length -- its actual size, I believe, from a 
memory-ish standpoint, that lists are actually larger then they appear 
to be in the Python environment. On the C level, my understanding from 
reading various posts is that the list allocates additinoal memory for 
elements which may be added in later, so that it doesn't have to 
allocate One Piece Of Memory every time an element is added. A Tuple is 
immutable. Once created, with its given values, it can not be changed. a 
= (1,2,3) can not have a[1] = 5 done. A list is mutable. You can change 
the individual elements at any time.

My unknowing guess? The overhead to create the truly 'dynamic' lists is 
completely wasted in enough circumstances that it warrented making an 
array which was fixed, in both size, and content.

----------

>From my POV, the primary purpose of tuples is for dictionaries: they
allow you to do things like

foo = {}
foo[(1,2)] = 3

This essentially permits the use of multi-column primary keys (to use
database terminology).  In addition, as Stephen Hansen pointed out,
tuples are a fair bit more efficient than lists.

---------

I first came across the term tuple in relation to Relational Algebra and 
in particular, in the realm of SQL and Databases.

A row of data in a database is often called a tuple. Really, it's a way 
of creating a Structure like you would do in C, very efficiently, which 
once created, is immutable. This means you can create a hash on it, and 
then use the hash to get the piece of data out of a database table. This 
is exactly how Tuples work, and they are hashed and then placed into an 
associative array (Python calls associative arrays Dictionaries).

So whereas there is one native Array type in C, the List type in Python 
is a mutable dynamically sized array, and a Tuple is a fixed immutable 
grouping of items, and a Dictionary is like an array where the key 
doesn't have to be a number.  So, three kinds of things which have 
something in common with a C array is better than just having a return 
to the Bad Old Days where C arrays were fixed in size, mutable to the 
point that you could easily crash your system, and totally 
non-polymorphic.


-------------------------------------------------------------
Where should I put my HTMLgen directory?
How do I add a path to sys.path?
http://www.faqts.com/knowledge-base/view.phtml/aid/3796
-------------------------------------------------------------
Fiona Czuczman
Jeff Kunce

When you do an "import" python looks for the module in all the 
directories in sys.path

Below are four ways to let python know where you have installed HTMLgen. 
I use #4 myself.

1) Add the directory to sys.path in your program before you import the 
module:
    import sys
    sys.path.append('C:\\Program Files\\Python\\Lib\\HTMLgen')

2) Add the directory to your systems "PYTHONPATH" environment
variable before you start your program. For windows:
    SET PYTHONPATH=C:\Program Files\Python\Lib\HTMLgen
or if PYTHONPATH is already defined:
    SET PYTHONPATH=C:\Program Files\Python\Lib\HTMLgen;%PYTHONPATH%

3) Create a file 'C:\\Program Files\\Python\\HTMLgen.pth' containing the
single line: Lib/HTMLgen

4) Import HTMLgen as a package
    a) create an empty file C:\\Program
Files\\Python\\Lib\\HTMLgen\__init__.py
    b) in your program, import HTMlgen like this:
        from HTMLgen import HTMLgen

All these procedures are explained in the documentation, and in the 
various python books. None is specific to HTMLgen.

And, personally, I would not put HTMLgen in the 'Lib' directory - I 
think of that as modules that are distributed with python, and may get 
overwritten with a new release. I'd suggest installing it as
   C:\Program Files\Python\local\HTMLgen
or just
   C:\Program Files\Python\HTMLgen


-------------------------------------------------------------
I have an ascii string like 0C44DH. Is there a Python function to convert it to decimal?
http://www.faqts.com/knowledge-base/view.phtml/aid/3799
-------------------------------------------------------------
Fiona Czuczman
Donn Cave

Yes, everything except that trailing H anyway.  I'd use 
string.atoi('c44d', 16), and I guess in the end that is about the same 
thing as int('c44d', 16)


-------------------------------------------------------------
How do I get the current index of the insertion character in an Entry widget?
http://www.faqts.com/knowledge-base/view.phtml/aid/3789
-------------------------------------------------------------
Fiona Czuczman
Hans-Joachim Widmaier

After digging in Tk docs for some time I've finally found how to get the 
current index of the insertion character in an Entry widget:

  self.index("insert")

will do the trick. Fair enough, "insert" isn't listed in the INDICES 
section, just in a note (in the Text widget man page).


## Edited Entries ##############################################


-------------------------------------------------------------
What is the Python code that will print primes below 1000?
http://www.faqts.com/knowledge-base/view.phtml/aid/2823
-------------------------------------------------------------
Fiona Czuczman
Fran?ois Pinard,Emile van Sebille

initial answer:

import re
for n in range(2, 1000):
    if not re.match('(11+)\\1+$', '1'*n):
        print n

and then again:

While the above submission is an interesting
application of regular expressions, it's not really
the way to get a list of primes.  Below is primes.py,
taken from the standard distribution that is faster
by a factor of 10-20:

primesuspects = range(3,1000,2)
ptr = 0
while ptr < len(primesuspects):
  val = primesuspects[ptr]
  for i in primesuspects[ptr+1:]:
    if not i % val:
      primesuspects.remove(i)
  ptr = ptr + 1
print primesuspects






From claird at starbase.neosoft.com  Sun Jun 11 12:18:18 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 11 Jun 2000 11:18:18 -0500
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hqsmc$eek$1@newshost.accu.uu.nl> <39436762@news.xtml.co.uk>
Message-ID: <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net>

In article <39436762 at news.xtml.co.uk>,
Nick <nick at spam_off_videosystem.co.uk> wrote:
			.
		[personal testimony]
			.
			.
>sys,os,os.path,string,re etc. You can write much more reusable, unit testable
>code in Python more easily. I think the lack of typing really aids reusablity
>and code productivity.
			.
			.
			.
I like to repeat the points Nick summarizes here.

C++ advocates (and, in similar ways, those for C
and Java also) often talk about the availability
of tools for those other languages, and about the
superior hygiene static typing affords.  The former
proposition is easy to answer:  Python doesn't 
*need* memory-management tools (for examples), be-
cause, by design, Python enlists the computer to do
that work, rather than forcing it on the developer.

On the other hand, I'm still searching for the best
way to speak about the "well you Python people only
find those errors at run-time" objection.  I find
it a challenge to communicate effectively to
newcomers how much better it is to have a language
that encourages unit testing as Python does. 
Python's developmental productivity is so much
greater, in my experience, as also to defy easy
explanation for those who haven't already used it.
-- 

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


From redaelli at inc.it  Sat Jun  3 17:56:15 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Sat, 03 Jun 2000 23:56:15 +0200
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it> <39397BC1.570AF5C8@stud.ntnu.no>
Message-ID: <39397EFF.E1B76E6E@inc.it>

Peter Schneider-Kamp wrote:
> 
> Paolo Redaelli wrote:
> >
> > but if I run a program that does x=float(x) where x = "0.000000E+00 "
> > (being a string) here's the output.
> 
> Can you give some code? I tried the following:
> 
> === BEGIN dumm.py ===
> x = "0.000000E+00 "
> x = float(x)
> === END dumm.py ===
> 
> it worked just fine. What version of Python are you using?
1.5.2 from RedHat 6.2
It works for me if I use the interactive version.
I took that very code, placed into a program and whooops... it hangs!
That's driving me crazy..... 
-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/


From mogul at primus.ca  Wed Jun  7 07:28:25 2000
From: mogul at primus.ca (Robin Porter)
Date: Wed, 7 Jun 2000 07:28:25 -0400
Subject: Another Newbie question....
Message-ID: <Xhq%4.864$0l2.6976@news2.tor.primus.ca>

When I try to run a program I have inputed (PAYROLL from an earlier
question),  I am prompted with the first line of input (HOURS WORKED) but
after I press enter - nothing happens.  I am not prompted for the rate of
pay.  Could you let me know why this is happening.

    Thank you very much for you help.

Robin




From bobl at tricity.wsu.edu  Mon Jun 12 21:14:36 2000
From: bobl at tricity.wsu.edu (Bob Lewis)
Date: Mon, 12 Jun 2000 18:14:36 -0700
Subject: unsettling tkinter bug -- or is it just me?
Message-ID: <39458AFC.8AD5B850@tricity.wsu.edu>

Consider the following code:
--------
from Tkinter import *

root = Tk()
img = PhotoImage(file='bobl.gif')

def func():
    cnvs = Canvas(root, width=img.width(), height=img.height())
    cnvs.create_image(0,0, image=img, anchor=NW)
    cnvs.pack(side=TOP)
    Button(root, text='Quit', command=root.quit).pack()

func()
root.mainloop()
--------
This runs just fine: It creates a root toplevel with "bobl.gif" displayed
in it and a "Quit" button.

Now we make one wafer-thin change: Move the the "img =" inside "func()":
--------
from Tkinter import *

root = Tk()

def func():
    img = PhotoImage(file='bobl.gif')
    cnvs = Canvas(root, width=img.width(), height=img.height())
    cnvs.create_image(0,0, image=img, anchor=NW)
    cnvs.pack(side=TOP)
    Button(root, text='Quit', command=root.quit).pack()

func()
root.mainloop()
--------
And now, while the canvas has the right dimensions, the image does
not appear.  Okay, so "img" went from a global to local scope, but
that shouldn't have affected execution, right?  Have I missed something?

This is on a Red Hat Linux system running versions 1.5.2-13 of both
python and tkinter.

	- Bob Lewis
	  bobl at tricity.wsu.edu


From moshez at math.huji.ac.il  Thu Jun  8 15:56:46 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 8 Jun 2000 22:56:46 +0300 (IDT)
Subject: wishlist Python3K: subclass ints
In-Reply-To: <8hoof0$r9c$1@slb1.atl.mindspring.net>
Message-ID: <Pine.GSO.4.10.10006082253450.17560-100000@sundial>

On 8 Jun 2000, Aahz Maruch wrote:

> In article <393FD0C7.D70479A2 at visionart.com>,
> Pete Shinners  <pete at visionart.com> wrote:
> >
> > - Also gone are all the nit-picking type checking routines.
> >   instead of first checking if a variable is a "class" type,
> >   then checking if the object is a certain class, all variables
> >   hold objects of the same "rank". the best move would be to
> >   have "int" and "float" be subclassed from a base class like
> >   "realnumber".
> 
> Close, but no cigar.  Start with the base class "number".  Then
> everything else follows: int, float, rational, complex.

Actually, in the number case specifically, I'd go for the
no-hierarchy-have-a-good-model-optimize-behind-the-covers strategy:

an int is a rational is a complex, any number might be exact or inexact.
Inexact is implemented internally as floating-point, of course, but it's
quite a good warning to users.

Note that a really crazy extension might have non-rational exact real
numbers (e.g., constructive reals), and it's a difficult question of how
to integrate it. Lucky for us, Python is a New Jersey language, so we can
just ignore the question.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From boud at rempt.xs4all.nl  Fri Jun  9 16:19:15 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 9 Jun 2000 20:19:15 GMT
Subject: Reloading on the fly: was: Re: Python Productivity over C++
References: <20000609185450.91797.qmail@hotmail.com>
Message-ID: <8hrjg3$dn2$2@news1.xs4all.nl>

Hung Jung Lu <hungjunglu at hotmail.com> wrote:

> I really have to raise my voice here: in Python, running means also 
> compiling. Steve: sorry to point this out to you, but you really ought to 
> take a look at dynamical reloading of modern script languages like Python. 
> To me, that's the biggest advantage of modern script languages. In Python, 
> there are no separate commands for compiling and running. And you can reload 
> a Python module on the fly. Do you understand that? For many applications 
> (like webservers), it is absolutely crucial that the main application keeps 
> running while you are adding/modifying part of the program. That's an area 
> where C++ can not do well. In Python, I often have programs running, and 
> then I need to modify part of the program. I go ahead and modify the module, 
> and since I have designed the main program to reload the submodules, I never 
> have to stop the main application, this even if there are compiling/runtime 
> errors from the submodule. Doing that in C++ is an absolute nightmare.

Could you post an example of designing the main program to reload
submodules? I'd like to get that working, too, and coming from
languages where that's impossible, I've a bit of trouble coming up
with a solution...

-- 

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


From claird at starbase.neosoft.com  Thu Jun 29 11:21:03 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 29 Jun 2000 10:21:03 -0500
Subject: Yours in namelessness (was: How to concatenate external files ..?)
References: <slm1odapjev11@corp.supernews.com> <8jf6lf$v2j$1@nnrp1.deja.com> <slrn8lmb4t.lpi.thantos@brimstone.mecha>
Message-ID: <B28AE9FD56BBFCF2.E102567A6FD85989.73C3BA8E8149CDA9@lp.airnews.net>

In article <slrn8lmb4t.lpi.thantos at brimstone.mecha>,
Alexander Williams <thantos at gw.total-web.net> wrote:
>On Thu, 29 Jun 2000 09:59:15 GMT, richard_chamberlain at my-deja.com
><richard_chamberlain at my-deja.com> wrote:
>
>>               file=open(x,'r')
>
>Teach me to post while still mostly asleep here at the office.  I
>forgot entirely about opening the iterated filename for reading.
>Bugger.
>
>To try and make up for it, how about this wee gem?
>
>> # "fls" is the globbed list of filenames
>> # "outfile" is the file object already opened to write to
>> flsHandles = map(lambda f: open(f, 'r').read, fls)
>> for fH in flsHandles:
>>     outfile.write(fH())
>
>Note that you never explicitly give each opened file its own name;
>instead, you create a list of all the read methods of all the file
>objects created dynamically from the list of filenames.  (That refered
>to by "open(FILENAME, 'r').open" is just a function/method reference,
>it can still be invoked.)
			.
			.
			.
Well, let's get even more anonymous, and suggest
  for fH in = map(lambda f: open(f, 'r').read, fls):
      outfile.write(fH())
-- 

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


From aahz at netcom.com  Sun Jun 18 17:53:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jun 2000 21:53:57 GMT
Subject: Function that knows the names of its actual parameters
References: <394a6395.7217524@news.btx.dtag.de>
Message-ID: <8ijgdl$oh8$1@slb6.atl.mindspring.net>

In article <394a6395.7217524 at news.btx.dtag.de>,
Stefan Franke <spamfranke at bigfoot.de> wrote:
>
>I found myself in need for the following functionality a couple of times:
>
>>>> a=1
>>>> b="foo"
>>>> c=b
>>>> magic(c, b, a)
>{'a': 1, 'b': 'foo', 'c': 'foo'}
>
>that is, a function that goes one step further than **dict arguments
>and returns a dictionary with names and values of its actual
>parameters.  It's useful every time one tries to write generators for
>Python or other code (right now I'm writing some SQL code emitting
>functions).

Why not simply write a function that takes a dict as its parameter?
Maybe if you gave a better description of an occasion you think you need
this feature we could give you a better answer.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From eddthompson at ssi.parlorcity.com  Fri Jun  9 21:26:25 2000
From: eddthompson at ssi.parlorcity.com (Blu Dragoon)
Date: Sat, 10 Jun 2000 01:26:25 -0000
Subject: function and library listing
References: <393f4787$0$25327@wodc7nh0.news.uu.net> <slrn8jumb2.qlt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <3941e0b4$0$9640@wodc7nh7.news.uu.net>

actually I have seen that and bought the book Iwas looking for something
better
Remco Gerlich wrote in message ...
>Blu Dragoon wrote in comp.lang.python:
>> Is there a site that list what functions and or libraries  are availible.
>> Not nesseccarily all of them just most of them.
>
>Go to http://www.python.org
>Click on 'Documentation'
>Click on 'Library Reference' (or 'Module Index')
>
>Was that really so hard to find?
>
>--
>Remco Gerlich,  scarblac at pino.selwerd.nl
>"This gubblick contains many nonsklarkish English flutzpahs, but the
> overall pluggandisp can be glorked from context"  (David Moser)




From jkraska1 at san.rr.com  Sat Jun  3 18:56:25 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 03 Jun 2000 22:56:25 GMT
Subject: PERFORMANCE: list versus a "dlist"
Message-ID: <39398DC0.C04E9784@san.rr.com>

Python lists, which are in actuality tail-preallocated dynamic
arrays, are compared to the performance of a similar dynamic
array using a head and tail preallocation strategy. This
construct is referred to as a "dlist" -- a double-ended list.

The tail preallocation strategy for the dlist involved
keeping actual length and allocated length, and doubling the
allocated space whenever an addition would involve appending
past the allocated region.

The head preallocation strategy involved keeping a "virtual
index" into the array which is initially set to a position
into the allocated region of memory at a position corresponding
to 1/3 of the way through. Certain critical conditions are
monitored which will trigger a reset to 1/3 of the way through
via either a reallocate/copy or a simple move, depending on
circumstance.

The performance of this construct is compared to python's
list as a FIFO (insert at head, pop at tail). Results are
reported as follows:

#insert/delete       DLIST            LIST

100                  .006             .006
1000                 .006             .018
10000                .065             .365

As before, on containers containing < 100 items or so, the
use of this additional datatype made no measurable difference
in performance of the tests. However, at list sizes at or near
1000 elements, the DLIST implementation broke forth as a clear
winner, where scaleability of the dlist imeplementation seemed
to indicate at or near O(1)+k cost for insertion at the head.

This is an early implemenation of this datstructure, and none
of the typical sequence manipulation code is available for it
as of yet.





C/


From shouldbe at message.com  Mon Jun 12 10:25:47 2000
From: shouldbe at message.com (Arinté)
Date: Mon, 12 Jun 2000 09:25:47 -0500
Subject: Python Issues.
Message-ID: <8i2o4s$1d3e$1@news2atm.raleigh.ibm.com>

What is the webpage that shows all the issues python has with certain OS'es?
Someone posted it a while ago when I had trouble with python and NT, but I
couldn't find it on deja.  Post the link again please...


thanx




From dgoodger at bigfoot.com  Tue Jun 27 23:11:45 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Tue, 27 Jun 2000 23:11:45 -0400
Subject: multiple replacements
In-Reply-To: <8ja7ok$6be$1@nnrp1.deja.com>
References: <8ja7ok$6be$1@nnrp1.deja.com>
Message-ID: <B57EE530.5823%dgoodger@bigfoot.com>

on 2000-06-27 08:47, siva1311 at my-deja.com (siva1311 at my-deja.com) wrote:
> I am a newbie Python programmer and was wondering the best way to make
> multiple replacements of text strings in a file.

Perhaps a bit complex for a newbie, but the most definitive answer I've seen
so far has been Fredrik Lundh's 2000-04-12 reply to the "Python idiom:
Multiple search-and-replace" thread:

=============== forward ================

Randall Hopper <aa8vb at yahoo.com> wrote:
> Is there a Python feature or standard library API that will get me less
> Python code spinning inside this loop?   re.multisub or equivalent? :-)

haven't benchmarked it, but I suspect that this approach
is more efficient:

...

# based on re-example-5.py

import re
import string

symbol_map = { "foo": "FOO", "bar": "BAR" }

def symbol_replace(match, get=symbol_map.get):
    return get(match.group(1), "")

symbol_pattern = re.compile(
    "(" + string.join(map(re.escape, symbol_map.keys()), "|") + ")"
    )

print symbol_pattern.sub(symbol_replace, "foobarfiebarfoo")

...

</F>

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



From android at micron.net  Thu Jun 15 08:53:07 2000
From: android at micron.net (Peter Immarco)
Date: Thu, 15 Jun 2000 08:53:07 -0400
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my> <20000614200609.A323@activestate.com> <39484ED4.2D3CBCD7@micron.net> <ld325.16426$N4.617712@ozemail.com.au>
Message-ID: <3948D1B2.44449B90@micron.net>

Chris,

Excellent point!  I learned 8-Queens from "The Art Of Prolog" and they show a
wonderful implementation of N-Queens.  I had been programming in Prolog for 3
years when I got that awesome book, and it showed me aspects of Prolog I would
never have dreamed of.  Sometimes being a self-taught programmer has its
pitfalls, but nothing that can't be cleared up by a great book.

Have you seen any "new classics"?  Most of the books I have that I like on
Prolog are from 5 - 15 years old.  I'm using LPA Prolog, how about you?

Peter.

the_wrights wrote:

>
> Someone not experienced in prolog might be seduced by the backtracking, only
> to find that the default control structures in the naive implementation of 8
> queens led to hopeless inefficiencies.
>
> Prolog is indeed a good language for this type of problem, but those new to
> the language should get O'Keefe's book (The Craft of Prolog) and Sterling
> and Shapiro (The Art of Prolog) so that the folk lore is available to them.
>
> cheers
>
> chris



From michael.husmann at teleatlas.com  Fri Jun  9 01:20:25 2000
From: michael.husmann at teleatlas.com (Michael Husmann)
Date: Fri, 09 Jun 2000 07:20:25 +0200
Subject: PLEASE HELP ON SOLARIS!!!
References: <393FBD20.E4664BD4@usa.alcatel.com>
Message-ID: <39407E99.A311022A@teleatlas.com>

> Is ANYBODY out there trying to run Python on Solaris?  If so, how did
> you get it to build?

I am running python 1.5.2. on solaris 2.5.1 and it works
perfectly. I have compiled everything here on my machine and
I do not remember that I have had any trouble.

If you need help so please let me know

-
Michael


From h__l at my-deja.com  Thu Jun 22 18:53:55 2000
From: h__l at my-deja.com (h__l at my-deja.com)
Date: Thu, 22 Jun 2000 22:53:55 GMT
Subject: python optimization
Message-ID: <8iu5dt$s6b$1@nnrp1.deja.com>

This may have already been covered but I have not seen this particular
question answered anywhere.  I am sure I missed this in the docs (or
read the wrong ones) but what are the OPT options for optimization?

I know of -O0  <turns it off?>
and also -O2 <default it seems for the RPM I d/l for python 1.5>

Is there a higher level? would this generate nonportable code (despite
the portability python enjoys)? error
prone code? How should I test this (I have a need to test it on both
win32 and also linux (RH 6.1).

Thanks in advance


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


From bmcd at es.co.nz  Fri Jun 16 22:06:29 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 17 Jun 2000 02:06:29 GMT
Subject: Question?
References: <394AD516.C150F0AB@yahoo.com>
Message-ID: <slrn8klnec.120.bmcd@ripley.sulaco.net>

On Sat, 17 Jun 2000 03:32:06 +0200, Charles W.W. <kingwang98 at yahoo.com> wrote:
> 

Answer!

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From npnigro at my-deja.com  Thu Jun 29 13:46:29 2000
From: npnigro at my-deja.com (npnigro at my-deja.com)
Date: Thu, 29 Jun 2000 17:46:29 GMT
Subject: ExitWindowsEx()
Message-ID: <8jg21f$kd6$1@nnrp1.deja.com>

I'm having problems with ExitWindowsEx function (a Win32api import) in
Win95 and Win98.  Unfortunately when I run the function, it closes the
active dos window that I'm running the script on and it appears to
close other active programs on the taskbar (they disappear).  However,
other active programs are not closed for they remain on open on my
desktop.  I'm sending the flags to force a reboot (2 or 4) yet it
neither forces nor reboots anything.  Any suggestions?  I would
definitely be willing to execute any sort of similar DOS prompt command
(such as rundll user,ExitWindowsExec) if it were to close all programs
forcefully.  I have yet to find such a gem... any help would be greatly
appreciated.

Nick


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


From rvprasad at cis.ksu.edu  Fri Jun 23 01:01:45 2000
From: rvprasad at cis.ksu.edu (Venkatesh Prasad Ranganath)
Date: 23 Jun 2000 00:01:45 -0500
Subject: what is a .pyo file?
Message-ID: <m3em5pyoqu.fsf@boss.ksu.edu>

Hi,

what are the contents of .pyo file?  When is it generated?  What is it used for?

Thank you,

-- 
Venkatesh Prasad Ranganath


From scherbi at bam.com  Wed Jun  7 06:57:49 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 07 Jun 2000 10:57:49 +0000
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <001101bfcfde$0a6a6f20$48a2143f@tim> <393D3993.84B507C9@bam.com> <m3ln0ivtzc.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <393E2AAD.641B9D38@bam.com>

Michael Hudson wrote:

> Bill Scherer <scherbi at bam.com> writes:
>
> > > [Michael Hudson]
> > > > ...
> > > > Tangentially, does anybody know of any implementations of JavaScript
> > > > (not necessarily connected to a browser) that let you type interactive
> > > > commands into a listener and genereally muck about with the language?
> >
> > There's the NJS Javascript Interpreter:
> >      http://www.bbassett.net/njs/
>
> Ta.  It doesn't actually have a listener though does it?  I can't find
> mention of one in the docs or command line help though...  still, beats
> typing javascript:... urls, I think.

I haven't actually used it for anything. Ijust knew of it.  I also came across
this: type in 'mocha:' for a url in Navigator 4+ and you get a JavaScript
console.  That may be more to your liking...

-Bill




From ng at hardlight.couk.com  Tue Jun  6 12:08:43 2000
From: ng at hardlight.couk.com (.:|:.)
Date: Tue, 6 Jun 2000 17:08:43 +0100
Subject: What's in a name?
References: <Pine.SOL.4.05.10005241201170.14111-100000@spgsun1.arlut.utexas.edu> <392c17d3$0$17863@wodc7nh7.news.uu.net> <Pine.SOL.4.05.10005241328080.14321-100000@spgsun1.arlut.utexas.edu> <392C3738.58C41202@muc.das-werk.de> <392CB840.1E5E5B06@teleatlas.com> <392CBAB5.37BFCE21@san.rr.com>
Message-ID: <393d24ee.0@news.proweb.co.uk>

"Courageous" <jkraska1 at san.rr.com> wrote in message
news:392CBAB5.37BFCE21 at san.rr.com...
> Can you name some situations
> in reading and writing english (heh) where case is essential to being
> understood?

river riding

River Riding


Matt





From s2mdalle at titan.vcu.edu  Tue Jun 20 21:37:40 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Tue, 20 Jun 2000 20:37:40 -0500
Subject: Overriding 'select'
References: <8ihgad$2qj$1@bob.news.rcn.net> <20000618134928.V26436@xs4all.nl>
Message-ID: <8ip6ei$s5t$1@bob.news.rcn.net>

> So, add this to your Connection object:
> 
>     def fileno(self):
>         return self.socket.fileno()
> 
> and all should be well. The select.select() returns the objects passed
> in, not their filedescriptor, so you do not need to do your own
> bookkeeping.

This absolutely works like a charm.  Thanks a lot.  Much faster than
what I was using too.  :)

-- 
David Allen
http://opop.nols.com/
----------------------------------------
It's that time of year again, Defcon has descended upon us. Time to venture
to SYN City to see the lights, meet strange new people, and create new
blurs we later dub memories.
	-- Attrition.org



From olczyk at interaccess.com  Fri Jun  9 14:12:04 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Fri, 09 Jun 2000 18:12:04 GMT
Subject: Problem with script and typing
References: <394123ad.89325968@nntp.interaccess.com> <8hr9jb$14g$1@sunnews.cern.ch>
Message-ID: <394430d0.92688156@nntp.interaccess.com>

On 9 Jun 2000 17:30:19 GMT, wlav at atlas03.cern.ch (Wim Lavrijsen)
wrote:

>>def is_project(s):
>>    if s[len(s)-3:] == 'prj':
>>        project=FindProject(s)
>>## error here
>>        if project.IsNull():
>>            project=Project(s)
>>        project.Scan()
>
>Again ... 'global project'. Also, should this be a member of Project?
Actually ( now that I think of it ) since the list was supposed to be
called projects. But I don't access projects from here. Is global ( in
the other spot you suggest ) really needed?
Here is what is supposed to happen:
is_project is sent a filename. It checks whether the extension
is 'prj'. If it is it looks for it in the projects list ( FindProject
does this ). If it fails then it creates a new project ( which install
itself in the projects list). Then it calls Scan on the project.

>>2) Is there a better way of handling null objects? ( Most languages I 
>>    know have objects which are explicitly null )
>
>Yep, use None:
>
>>>> project = None
>>>> if project: print "there's a project!"
>...
>>>> if not project: print "no project!"
>...
OK. That solves the specific problem, but what about the typing
problem. Since projects only contains instances of Projects,
FindProject returns a Project. The compiler seems to complain
that it doesn't know that what FindProject returns is a project.
So I can't call specific methods on the return value.


From swun at esec.com.au  Mon Jun 12 10:13:36 2000
From: swun at esec.com.au (Sam Wun)
Date: Tue, 13 Jun 2000 00:13:36 +1000
Subject: python and php??
Message-ID: <3944F010.9F7A6DD2@esec.com.au>

Can I invoke python in php script?

Sam.



From Jonathan_Epstein at nih.gov  Tue Jun 27 11:58:18 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Tue, 27 Jun 2000 11:58:18 -0400
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <3958CF1A.EE90DD44@nih.gov>

Also note The Register's commentary on this topic, including mentioning
Python:
  http://www.theregister.Co.uk/content/1/11606.html

- Jonathan


Don Tuttle wrote:
> 
> Consider this a heads up NOT a promotion!!
> 
> In case you haven't heard Microsoft is coming out with a massive new Web
> stratagy that, in part, uses a new language called C# (That sharp, like in
> music).  Here's a part of what they have to say:
> 
> "Because of its elegant object-oriented design, C# is a great choice for
> architecting a wide range of components-from high-level business objects to
> system-level applications. Using simple C# language constructs, these
> components can be converted into Web services, allowing them to be invoked
> across the Internet, from any language running on any operating system."
> 
> "More than anything else, C# is designed to bring rapid development to the
> C++ programmer without sacrificing the power and control that have been a
> hallmark of C and C++. Because of this heritage, C# has a high degree of
> fidelity with C and C++. Developers familiar with these languages can
> quickly become productive in C#."
> .......
> 
> "The modern design of C# eliminates the most common C++ programming errors.
> For example:
> 
> Garbage collection relieves the programmer of the burden of manual memory
> management.
> Variables in C# are automatically initialized by the environment.
> Variables are type-safe. "
> 
> Hmmm, sound familar?
> Here's the link
> http://www.msdn.microsoft.com/vstudio/nextgen/technology/csharpintro.asp
> 
> I look forward to you comments.
> 
> Don
> 
> --
> Believe nothing no matter where you read it or who has said it, not even if
> I have said it, unless it agrees with your own reason and your own common
> sense.
> --Buddha


From erik_wilsher at my-deja.com  Sat Jun 24 02:18:21 2000
From: erik_wilsher at my-deja.com (erik_wilsher at my-deja.com)
Date: Sat, 24 Jun 2000 06:18:21 GMT
Subject: List without duplicates?
References: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <8j1jr8$a2k$1@nnrp1.deja.com>

In article <VlS35.15102$Xx5.781310 at bgtnsc06-news.ops.worldnet.att.net>,
  "William Dandreta" <wjdandreta at worldnet.att.net> wrote:
> Is there a built in Python function that eliminates duplicates in a
list?

If you want to keep a list of unique values, try the kjBuckets module
from Aaron Watters  (http://www.pythonpros.com/arw/kjbuckets/).

EW


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


From BgPorter at NOacmSPAM.org  Wed Jun 21 13:12:26 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Wed, 21 Jun 2000 17:12:26 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <wk66r3cgek.fsf@turangalila.harmonixmusic.com> <CT545.13337$C44.783317@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <_H645.15280$lU5.120617@news1.rdc1.nj.home.com>

"William Dandreta" <wjdandreta at worldnet.att.net> wrote in message
news:CT545.13337$C44.783317 at bgtnsc05-> I have already had one experience
with a Python program taking 90 minutes to
> process an 18 MB file. By making what I consider minor changes in the
> program, it now takes 6 minutes.
>
> I've come to the conclusion that Python is not an inherently slow language
> but it is very easy to add overhead that makes Python programs run
extremely
> slowly. It is much more difficult to do that in the other languages that I
> use. So for the present, I am being overly cautious about adding overhead.
> As I become more familiar with Python, I'm sure that will dissippate.
...but that's true in _any_ language. The first chapter of Michael Abrash'
book "The Zen of Code Optimization" he shows a simple checksumming program
that gets a speed improvement of 50x by some very simple redesign. He shows
that attempting to 'optimize' a poor design by recoding it in assembly makes
a 'fast slow' program -- it's as fast as the design will allow, but still
slow.




From scherbi at bam.com  Tue Jun 27 09:24:06 2000
From: scherbi at bam.com (Bill Scherer)
Date: Tue, 27 Jun 2000 13:24:06 +0000
Subject: How to control Pmw.ScrolledText scrollbar position?
References: <8ire9f$rs4$1@nnrp1.deja.com>
Message-ID: <3958AAF6.AA75D44C@bam.com>

I had a need for this too, and, fortunately for both of us, I just figured
it out.

# create a Pmw.ScrolledText
self.textWindow = Pmw.ScrolledText(self.master)

#get a reference to the Tk.Text component
self.textWindowText =
self.textWindow._MegaArchetype__componentInfo.get('text')[0]

#to move to the end of self.textWindow call this:
self.textWindowText.see(END)


Enjoy!

-Bill

Maytag Repairman wrote:

> I have this program that I've written which uses Pmw.ScrolledText.
> There are buttons on my GUI which take data from entry fields and
> insert, replace, and append data in a data structure.  The ScrolledText
> is cleared and the updated data structure is displayed in a user-
> friendly format.  Unfortunately, the scrollbar always ends up at the
> top after this operation but the area of most interest to the user is
> the point where the operation on the data structure took place.  I
> can't figure out how to control the position of the scrollbar to move
> it to the area of interest.
>
> Yes, I do realize that I am doing this rather inefficiently by re-
> creating the full text each time but it works well enough for now.
>
> --
> Rob Cupples
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From bjorn at roguewave.com  Wed Jun 28 16:57:41 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 28 Jun 2000 14:57:41 -0600
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>
Message-ID: <395A66C5.E9AC4F3E@roguewave.com>

Note that all the problems you describe completely go away if you do:

  import AAA
  import BBB.AAA

IMHO you deserve any/all complications you get when you do "from foo
import *", and I strongly believe that we shouldn't do _anything_ to
encourage this use...

-- bjorn

Huaiyu Zhu wrote:
> 
> Given the following directory structure,
> 
> python/
>    AAA.py
>    BBB/
>       __init__.py
>       AAA.py
> 
> if sys.path contains 'python', then
> 
> from AAA import *
> from BBB.AAA import *
> 
> would do the right thing at most locations, except when pwd is python/ and
> when '.' is before 'python' in sys.path, in which case they both import from
> python/AAA.py.
> 
> Is this a fair description of current situation?  If so this makes package
> name spaces essentially flat. That is, in order to avoid name clashes one
> has to avoid identical module names even if they are buried deep inside
> packages.  This is even worse on Windows - its file names are caseless - as
> one has to avoid filenames with different cases as well.
> 
> This situation occurs quite often in practice.  For example, AAA could be a
> module for doing something.  BBB could be a wrapper package for doing things
> somewhat differently.  So within BBB/AAA.py one calls 'import AAA' to get
> the real job done, and oops, it imports itself!
> 
> Although it is not advisable to put '.' in path, especially at front, a lot
> of people actually do. And the functionality of a package should not depend
> on this detail.  Otherwise what happens if one imports several packages each
> expecting a different ordering of sys.path?
> 
> In Unix shells, however, this can always be overcome by using absolute
> paths, like /the/absolute/path/command, whatever the PATH is.
> 
> So now to my proposed solution.  Is it possible to have something like this?
> 
> sys.rootpath  = ['/usr/lib/python1.5', '/usr/lib/python1.5/site-packages']
> from rootpath.AAA import *
> from rootpath.BBB.AAA.import *
> 
> This guarantees that module names within packages are well separated.  One
> can still use
> from AAA import *
> to import from any module named AAA that happens to be in the path. IMHO,
> this would not break any existing code (unless it happens to contain the new
> keyword 'rootpath').  It would certainly facilitate much easier use of
> packages. The only real problem I can see of is that developers might (for
> what reason?) rush to occupy real estates under rootpath.  But since the
> standard distribution is centrally controlled and add-ons are user
> controlled, this shouldn't be a big problem.
> 
> --
> Huaiyu Zhu                       hzhu at users.sourceforge.net
> Matrix for Python Project        http://MatPy.sourceforge.net
> --
> http://www.python.org/mailman/listinfo/python-list



From stephen at cerebralmaelstrom.com  Sat Jun 10 01:56:54 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 10 Jun 2000 05:56:54 GMT
Subject: mxTextTools
Message-ID: <GMk05.52546$WS3.497538@typhoon.we.rr.com>

Okay, i'm learning mxTextTools on top of the dozen other things i'm doing at
once, at the moment, and have two questions:

With the commands that deal with 'sets', i'm assuming they're talking about
sets returned via the 'set()' function and its similar friends? But what
exactly IS a character set? Why use one instead of a string?

Secondly, 'sWordStart' -- I don't get the description of what it does. It
says it uses a Search Object to perform the required action, which is
fater.. but zero matching characters are allowed.. So, it only passes if
there is NOT a match?

Thanks

--S




From pinard at iro.umontreal.ca  Mon Jun  5 11:01:11 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 05 Jun 2000 11:01:11 -0400
Subject: Questions regarding design decisions in listobject.c
In-Reply-To: Courageous's message of "Sun, 04 Jun 2000 21:26:07 GMT"
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl>
	<393ACA13.FD2416F1@san.rr.com>
Message-ID: <oqvgzo9npk.fsf@titan.progiciels-bpi.ca>

Courageous <jkraska1 at san.rr.com> writes:

> Well, yeah, but [0:3] as "element zero through three, inclusive"

Read 0 up to 3, but not including it.  That notation is very, very handsome
and useful.  It is one of the jewels of Python.  You did not see it yet?
I fear you still write too much C, and too little Python :-) :-).

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





From thantos at chancel.org  Mon Jun  5 18:06:45 2000
From: thantos at chancel.org (Alexander Williams)
Date: Mon, 05 Jun 2000 22:06:45 GMT
Subject: Memory leak: assigning function object to instance variable
References: <8hh0pf$i7$1@phys-ma.sol.co.uk>
Message-ID: <slrn8jo93l.ojt.thantos@brimstone.mecha>

On Mon, 5 Jun 2000 20:55:59 +0100, John Parkey
<jparkey at nospamrapidbs.com> wrote:

>I have registered this bug, with code fragment below, but wondered if

This isn't, technically, a bug.  You create a cyclic reference to self
inside the instances, so the reference counting mechanism for
releasing memory can't let it go.  Standard, expected behaviour for
refcounting garbage collection.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From moshez at math.huji.ac.il  Wed Jun  7 10:24:18 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 7 Jun 2000 17:24:18 +0300 (IDT)
Subject: Calling external programs: ioredirection
In-Reply-To: <393e4e6f.72239515@nntp.interaccess.com>
Message-ID: <Pine.GSO.4.10.10006071723140.2293-100000@sundial>

On Wed, 7 Jun 2000, Thaddeus L. Olczyk wrote:

> I want to do this in order to do things similar to sh. For example:
> 
> wc -l *.h *.cpp|awk 'begin int i=0; {i+=$1 } end print i;}'
> 
> ( I know the script is probably wrong, but you get the idea. )
> 
> Any idea of how?

Yes, read about "os.popen" and "os.system". But the real answer is that
you *don't* want to do it -- Python has fine modules to do all those
things, and it's much easier and efficient to use them.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From parkw at better.net  Mon Jun 19 15:19:53 2000
From: parkw at better.net (William Park)
Date: Mon, 19 Jun 2000 15:19:53 -0400
Subject: setting environment variables
In-Reply-To: <slrn8ksrme.7ku.wh@frodo.uni-erlangen.de>; from wh@msgto.com on Mon, Jun 19, 2000 at 09:04:47PM +0200
References: <slrn8ksrme.7ku.wh@frodo.uni-erlangen.de>
Message-ID: <20000619151953.A2857@better.net>

On Mon, Jun 19, 2000 at 09:04:47PM +0200, Walter Hofmann wrote:
> 
> Hi,
> 
> I need to set an environment variable so that a python module which I'd
> like to use finds the right shared library.
> 
> I'm using Linux, so doing
> 
> LD_LIBRARY_PATH=/opt/kde2/qt/lib python test.py
> 
> works. But I would prefer to do it in python so that one can start my
> program by entering
> 
> ./test.py
> 
> However, setting LD_LIBRARY_PATH with os.environ within the python
> script (but before importing the library) is too late. The dynamic
> linker will not find the library this way.
> 
> I then tried to use
> 
> #!/usr/bin/env LD_LIBRARY_PATH=/opt/kde2/qt/lib python test.py
> 
> which, according to env's manpage should work. It doesn't, but hangs
> instead in a tight loop consoming 100% CPU. Neither works the more
> complex
> 
> #!/bin/bash -c 'LD_LIBRARY_PATH=/opt/kde2/qt/lib python ./test.py'
> 
> which gives
> 
> wh at frodo:~/programming/qtpython > ./test.py
> bash: - : unrecognized option
> Usage:  bash [GNU long option] [option] ...
> [...]
> 
> Finally, I found out that the following rather tricky (dual language)
> code works, however it is ugly to say the best:
> 
> #!/bin/sh
> LD_LIBRARY_PATH="/usr/lib/qt-2.1.1/lib" "exec" "python" "$0" "$*"
> del LD_LIBRARY_PATH
> # python code starts here
> 
> 
> My question:
> Is anyone on this list aware of a cleaner way to to this?
> 
> 
> Walter

I don't think you can change shared library path inside Python.  Since
you're running Linux, add '/opt/kde2/qt/lib' to /etc/ld.so.conf'.

--William



From parkw at better.net  Tue Jun 20 13:56:27 2000
From: parkw at better.net (William Park)
Date: Tue, 20 Jun 2000 13:56:27 -0400
Subject: Python and mSQL
In-Reply-To: <8io97i$d6u$1@newsg1.svr.pol.co.uk>; from jaford@watford53.freeserve.co.uk on Tue, Jun 20, 2000 at 06:21:56PM +0100
References: <8io97i$d6u$1@newsg1.svr.pol.co.uk>
Message-ID: <20000620135627.A990@better.net>

On Tue, Jun 20, 2000 at 06:21:56PM +0100, Jim Ford wrote:
> Hi
> 
> I use mSQL version 1.0.16 on Linux and want to write a script to access the
> database. I'm currently fiddling around with Perl and am starting to get
> somewhere, but what little I know about Python attracts me and I'm wondering
> whether it would be a better choice.
> 
> What is the current state of a Python module to access mSQL, please?
> 
> Regards: Jim Ford

Click to
    - http://www.python.org/topics/database/
    - http://www.python.org/download/Contributed.html and seach for
      'mSQL'.  (I found 2 items.)

--William



From thomas at xs4all.net  Fri Jun 16 10:52:22 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Jun 2000 16:52:22 +0200
Subject: Python segfault
In-Reply-To: <394A32C2.551B0290@stud.ntnu.no>; from petersc@stud.ntnu.no on Fri, Jun 16, 2000 at 03:59:30PM +0200
References: <20000616141903.A23985@foxboro.com> <20000616145836.O26436@xs4all.nl> <394A32C2.551B0290@stud.ntnu.no>
Message-ID: <20000616165222.P26436@xs4all.nl>

On Fri, Jun 16, 2000 at 03:59:30PM +0200, Peter Schneider-Kamp wrote:

> > On Fri, Jun 16, 2000 at 02:19:03PM +0200, Hans-Joachim Widmaier wrote:

> > > Why not simply make 'n_nchildren' an (32 bit) int? And, while at it,
> > > 'n_lineno' also? It would cost some memory, though.

> > Why bother ? It would do nothing to solve the real problem, which is that
> > there is a finite number of direct children to a node. The current number is
> > practically unreachable -- the example case of ('+'.join(['2+2'] * 8193)) is
> > hardly reminiscent of real code.

> Well, what about the n_lineno? Isn't that going to overflow
> in more realistic situations? 

Hmm, actually, yes. In fact, I was too optimistic in my original mail; the
script generated by

python -c 'print "x = 1\n"*33000'

crashes as well. That is, there is a limit of actual newline-seperated
statements, but I'm not sure if that's imposed by the lineno or by the
n_nchildren member. I don't really know anything about the parser, though,
only what I've quickly browsed, so I can't say if there are any real-world
examples of those limits being reached. (If it's a limit of the number of
lines of code in a script, or of toplevel-statements, or of
toplevel-statements until a block-statement, or whatever.)

This would probably be a good time for someone who knows what he's talking
about to enter ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jilanik at tin.it  Sun Jun  4 11:13:12 2000
From: jilanik at tin.it (Jilani Khaldi)
Date: Sun, 04 Jun 2000 17:13:12 +0200
Subject: python and web applications
Message-ID: <393A7208.23AF3E23@tin.it>

Hi All,
I am looking for some examples in python to init do some web programming
with this nice nice new (for me) language.  Could you please tell me
where to find some code?

Thank you.
--
<jk>
___________________________________________________
http://space.tin.it/scuola/jkhaldi
FreeBSD ~ The Power to Serve





From tom-main at REMOVEME.home.com  Tue Jun 27 11:50:51 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 27 Jun 2000 15:50:51 GMT
Subject: C++ style const's & types in Python
Message-ID: <v3465.71261$7o1.1704126@news2.rdc1.on.home.com>

I'm a C++ developer learning Python.

There are, of course, a few features of C++ that I miss (and many I don't)
and that I think are important for developing larger applications.

One is const types.  Particularly given that values are passed to methods by
reference.  But am I missing something?  Is there any way in Python to
specify that a value (other than a tuple) must remain constant?

Also, type declarations, while a nuisance in short programs, can be valuable
in larger programs.  It would be nice if Python names could optionally have
their types fixed.  As a substitute, I think I will start using Hungarian
notation to prefix my Python names in cases where I don't intend to
dynamically change their type.  Is this stupid?  Is there a better
alternative?

I realize that these issues are primarly a result of my C++ background, and
that I will be creating Python programs with a C++ style.

Thanks for your suggestions.

Tom.






From steel12659 at mail.com  Tue Jun 20 18:22:42 2000
From: steel12659 at mail.com (Steel Building Promo)
Date: Tue, 20 Jun 2000 18:22:42
Subject: Steel Building Clearance
Message-ID: <20000620222214.33EA71CEE7@dinsdale.python.org>

Looking for an industrial strength steel building?

Look no further!

We have 8 new clearance buildings, one of which may fit your needs!


Simply send your name and a written request for 
more information to: 

mailto:steel987654 at mailcity.com?subject=BUILDING_DETAILS
(for removal instructions, please see below) 

And you will receive information via email within 24 to 48 hours!
You will also receive exclusive information about our canceled and clearance buildings, so
be sure to check your email box! 

IMPORTANT: For verification purposes, you MUST 
provide your name and a request for 
more information or no further information will be sent
to you. Your written request could be something as 
simple as writing "Please send more info." 


mailto:steel987654 at mailcity.com?subject=BUILDING_DETAILS
(for removal instructions, please see below) 
_______________________________________________________
To be removed simply hit the REPLY button and put the word 
REMOVE in the subject or click on this link:
mailto:steel12659 at mail.com?subject=Remove



From tiddlerdeja at my-deja.com  Wed Jun 28 15:54:14 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Wed, 28 Jun 2000 19:54:14 GMT
Subject: file/re search and sub problem
Message-ID: <8jdl4t$r58$1@nnrp1.deja.com>

Any help appreciated...

I have a re:
wilf_re        = re.compile('[^A-Z]ebitda[^A-Z]', re.IGNORECASE)

And I do a search and replace
if wilf_re.search(file):

     # perform the substitution
     new_file = wilf_re.sub(" EBITDA", file)

This isn't really what I want. If I have
 ebitda</td>
It replaces it with
 EBITDA/td>


Can someone tell me how I go about getting:
 EBITDA</td>


Any help geatly appreciated.


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


From PClaerhout at CREO.BE  Thu Jun 22 06:25:07 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 12:25:07 +0200
Subject: CurrentUser for remote machine under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130120E@msgeuro1.creo.be>

Hello,

how can you get the current logged in user for a remote nt system using Python?

Pieter



From amitr at w-o-i.com  Mon Jun 19 01:20:09 2000
From: amitr at w-o-i.com (amitr at w-o-i.com)
Date: Mon, 19 Jun 2000 05:20:09 GMT
Subject: Python vs Perl
Message-ID: <8ikai5$v3d$1@nnrp1.deja.com>

hi,

I have been using Perl 5 for web development.
Now, I want to know if Python (with Zope) is better than Perl?

what r the strong points and pitfalls.

please help me.

thanks


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


From thomas at xs4all.net  Thu Jun 15 10:22:17 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 15 Jun 2000 16:22:17 +0200
Subject: PATCH: Augmented assignment
In-Reply-To: <skhovn87is473@corp.supernews.com>; from speed@?.com on Thu, Jun 15, 2000 at 02:11:03PM +0000
References: <20000609163937.D20700@xs4all.nl> <8hs4d8$33b$1@animus.fel.iae.nl> <20000610120811.H20700@xs4all.nl> <8ianhh$dro$3@newshost.accu.uu.nl> <skhovn87is473@corp.supernews.com>
Message-ID: <20000615162217.J26436@xs4all.nl>

On Thu, Jun 15, 2000 at 02:11:03PM +0000, Jake Speed wrote:

> Apparently the 'b' got flipped :^)

Duh. Sometimes I think I'm dislextyc :P Ohwell, the next patch will make it
'_ab__' :-) Thanx for the prod.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From kmself at ix.netcom.com  Wed Jun 28 01:51:41 2000
From: kmself at ix.netcom.com (kmself at ix.netcom.com)
Date: Tue, 27 Jun 2000 22:51:41 -0700
Subject: books????
References: <sla4ago0jev12@corp.supernews.com> <961961617.1744372225@news.rma.edu> <8j85gu$e86@autodesk.autodesk.com>
Message-ID: <dp3cj8.21o.ln@kmself.nntp.ix.netcom.com>

I'll second both suggestions.  The O'Reilly book is useful for guidance,
the New Riders book is a good reference.  For a *very* abbreviated
refernce, the ORA pocket reference is also handy.

Akira Kiyomiya <akira.kiyomiya at autodesk.com> wrote:
> I am using "Learning Python" from O'Reilly and "Python Essential Reference"
> from New Riders.  They are both pretty good books for beginners.

> Akira

> "Robert" <rhicks at rma.edu> wrote in message
> news:961961617.1744372225 at news.rma.edu...
>> On Sat, 24 Jun 2000, SugarKing wrote:
>> >Are there any good beginner books for Python? I'm itching to get my hands
> on
>> >any. Thanks
>>
>> There is:
>>
>> Learning Python
>> Learning Python in 24 hours
>> Python and Tkinter programming
>>
>> There should be links to these books in the library portion of the python
>> website. I have not used the LP in 24 hours but have heard good things
> about it.
>>
>> Robert



-- 
Karsten M. Self <kmself at ix.netcom.com>         http://www.netcom.com/~kmself
  Evangelist, Opensales, Inc.                       http://www.opensales.org
   What part of "Gestalt" don't you understand?      Debian GNU/Linux rocks!
     http://gestalt-system.sourceforge.net/      K5: http://www.kuro5hin.org
GPG fingerprint: F932 8B25 5FDD 2528 D595  DC61 3847 889F 55F2 B9B0


From bill at libc.org  Sat Jun 17 16:59:58 2000
From: bill at libc.org (Bill Anderson)
Date: Sat, 17 Jun 2000 14:59:58 -0600
Subject: Python 1.5.2 on HP-UX v11
References: <3948FE5E.5A433424@sec.noaa.gov>
Message-ID: <394BE6CE.A6F0D5F@libc.org>

j vickroy wrote:
> 
> Is anyone using the *thread* module in Python 1.5.2 on HP-UX v11?

yup.


From rjroy at takingcontrol.com  Sat Jun  3 11:00:50 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Sat, 03 Jun 2000 15:00:50 GMT
Subject: Why is my Tkintering so bad?
References: <XL%Z4.2161$gr4.67106@news1.rdc1.il.home.com>
Message-ID: <39391328.820786453@news1.on.sympatico.ca>

On Sat, 03 Jun 2000 04:24:23 GMT, "dsavitsk" <dsavitsk at e-coli.net>
wrote:

>example 1 does not work, example 2 does. anybody know why?
>more importantly, i need to read text from a ScrolledText (Pmw) and i can't
>do it.  The code i am attempting to use (or at least the relevant part) is
>in Example 3.  i am used to things like...
>x = Form.Text1.text (vb) so this is really foreign. essentially there is a
>text box and a button. at startup the text box reads from a text file (this
>works fine), then when the user pushes the button the contents of the text
>box are written to the output window.  It is this step of connecting code to
>events to visuals that i am not getting. any suggestions at all will be a
>big help and greatly appreciated.
>
>thanks,
>doug
>
## cut some stuff....

Your main problem here is variable scoping. st is defined int the
scope of MakeInterface and is not global. Thus it is not visible to
SaveEdits and SaveEdits will raise an exception.

If you want to do things this way then declare the variable st at the
module level and then use a GLOBAL declaration in the MakeInterface
function. Essentially the rules for global variables can be distilled
down to they can be read by any function in the module but must be
declared as GLOBAL in any function that wishes to assign to them
otherwise a new local variable is instantiated.

So by adding these two lines, your code should work.

That said, it is not a very pythonic way to do things.  Most python
programmers will create a class that encapsulates the behaviour. See
below for an example that demonstrates this.


># Example 3 ---------------------------------------
>from Tkinter import *
>import Pmw
>

st = None

>def SaveEdits():
>    x = st.get()
>    print x
>
>def MakeInterface():
	# this will tell the function to use the globally defined "st"
	# otherwise st will be local to this function
	GLOBAL st

>    flt = Frame(root)
>    st = Pmw.ScrolledText(flt, labelpos=N, label_text='Stuff',
>text_wrap='none')
>    st.importfile('TestData.txt')
>    st.pack(fill=BOTH, expand=1, padx=5, pady=5)
>
>    flt.pack(side=TOP, expand=YES, fill=X)
>
>    flb = Frame(fl)
>    cmdSave = Button(flb, width=20, text='Save Edits', borderwidth=3,
>relief=GROOVE, command=SaveEdits).pack(side=TOP, pady=5)
>    flb.pack(side=TOP, expand=YES, fill=X)
>
>root = Tk()
>root.title('window title')
>MakeInterface()
>root.mainloop()
>
>



from Tkinter import *
import Pmw
import shutil

class EditWidget:
    def __init__(self, parent, editfile):
        self.editfile = editfile
        self.bakfile = editfile + '.bak'
        flt = Frame(parent)
        self.scrolledtext = st = Pmw.ScrolledText(flt, labelpos=N,
label_text='Stuff',text_wrap='none')
        st.importfile(editfile)
        st.pack(fill=BOTH, expand=1, padx=5, pady=5)
    
        flt.pack(side=TOP, expand=YES, fill=X)
    
        flb = Frame(flt)
        cmdSave = Button(flb, width=20, text='Save Edits',
borderwidth=3,relief=GROOVE, command=self.SaveEdits)
        cmdSave.pack(side=TOP, pady=5)
        flb.pack(side=TOP, expand=YES, fill=X)

    def SaveEdits(self):
        x = self.scrolledtext.get()
        print x
        shutil.copyfile(self.editfile, self.bakfile)
        open(self.editfile,'w').write(x)
        
if __name__ == "__main__":        
    root = Tk()
    root.title('window title')
    ew = EditWidget(root,'TestData.txt')
    root.mainloop()



From wware at world.std.com  Sun Jun 25 16:26:26 2000
From: wware at world.std.com (Will Ware)
Date: Sun, 25 Jun 2000 20:26:26 GMT
Subject: Countdown
References: <3952952D.88E3588D@asme.org> <slrn8lc5qg.rqc.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <Fwq8s2.A8z@world.std.com>

Remco Gerlich (scarblac-spamtrap at pino.selwerd.nl) wrote:
> for i in range(len(thing)):
>    print thing[-i]

If thing = [1, 2, 3, 4, 5], then this piece of code will print
1, 5, 4, 3, 2. The code below has probably the less surprising
effect of printing 5, 4, 3, 2, 1.

for i in range(len(thing)):
    print thing[-i-1]
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From nhead at houbits.com  Fri Jun 30 02:01:26 2000
From: nhead at houbits.com (Nigel Head)
Date: 30 Jun 2000 06:01:26 GMT
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com> <ku66qtv945.fsf@lasipalatsi.fi> <395B6D35.DA83A7ED@estinc.com>
Message-ID: <8F6343327abbrvfgtyloup@131.176.85.36>

eric at estinc.com (Eric Lee Green) wrote in <395B6D35.DA83A7ED at estinc.com>:

...

>Last time I looked at ZOPE, it was not clear how to integrate new Python
>methods with it. A shame, it looks like a great product. Maybe I'll get
>a chance later on to really get into it and figure all that stuff out,
>it looks like it would solve some of the performance problems I've
>encountered using PostGreSQL and Python with Apache. 

While I agree with your feelings about the Zope documentation there are two 
(maybe more) drop in Postgres intreface "products" -- I can't speak for 
their robustness under high loads as my applications are trivial ... maybe 
worth a try for you all the same. You'll find the Zope mailing lists full 
extremely helpful people and the learning curve isn't quite as steep as it 
used to be ...

N



From shapr at uab.edu  Fri Jun  2 14:22:43 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 02 Jun 2000 18:22:43 GMT
Subject: Sorting a list
References: <snk8g856mf.fsf@motorola.com>
Message-ID: <393801C7.960D6E86@uab.edu>

Aaron Ginn wrote:
> 
> I'm a python newbie, so please be kind! :)
> 
> I'm trying to sort a list.  Fairly simple, but I'm not getting
> at all what I expect.  When I attempt to sort the list, the
> return value is None, and I can't understand why at all.
<snip>
> Why is sort returning an empty list?

because in this case, sort is expected to have the side effect of
sorting the list, rather than returning a sorted list.

>>> a = [3,2,1]
>>> a.sort()
>>> a
[1, 2, 3]
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt


From jmg at ecs.soton.ac.uk  Wed Jun  7 13:21:58 2000
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: Wed, 07 Jun 2000 18:21:58 +0100
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development 
 Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain>
Message-ID: <393E84B6.D80CD9D2@ecs.soton.ac.uk>

Manuel Gutierrez Algaba wrote:

> No, if every ones speak Dutch here, then why not :
> Dutch.comp.python.org

If  *everyone* speaks Dutch here, then there is no need. If anything, then
we should create a minority interest group: comp.lang.python.english.

> And now my own rant in Spanish:
>
> Python era un lenguaje casi perfecto, al menos hasta que empezaron
> a meterle tonterias sintacticas. Ahora estoy empezando a ver cual
> es el problema de python: los holandeses.

No son los holandeses, sino los demas que hablan holanes . . . y los que no
lo hablan y contestan en castellano :-)

> Al parecer estan tan
> excitados con la idea de python que no pueden evitar soltar la
> parida en Holandes. La verdad es que los holandeses son muy imaginativos:
> - Las cafeterias de drogas

Poco imaginativo. Liberal, quiza.

> - Tomates trangenicos.

Todos sin sabor.  Me fastidia muchissimo que aqui en Inglaterra es casi
impossible comprar un tomate con sabor, gracias a los holandeses.

> Bueno, ya en serio, si realmente no se pueden aguantar, que
> creen un Dutch.comp.python.org.

Lo que nos falta es comp.python.babel

Tal vez comp.python.python

De acuerdo, let's all revert to English !

Jacek





From kens at sightreader.com  Thu Jun 15 23:27:14 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:27:14 -0700
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com>
Message-ID: <39499E91.3F2A65A6@sightreader.com>



Courageous wrote:

> Re: python vs c++
>
> > However my C++ friends are badgering the crap out of me...
>
> Most of your C++ friends would be picking their tongues
> up off the floor if they saw the LOC of python programs
> in comparison to their C++ counterparts.

Expecially after you discount all declarations and braces from the C++
code.

> Python may be too slow for some tasks, but in the area
> of programmer productivity, very little beats it, and
> arguably the environments that do (e.g., lisp) have too
> many other negatives associated with them.
>
> C/

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/faed238b/attachment.vcf>

From petersc at stud.ntnu.no  Sat Jun  3 18:07:53 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 00:07:53 +0200
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <3937D345.CC0935D5@uab.edu> <39385DC2.BC0B4853@stud.ntnu.no> <slrn8jj4ot.om.thor@localhost.localdomain> <20000603234710.Q469@xs4all.nl>
Message-ID: <393981B9.10C8A96F@stud.ntnu.no>

Thomas Wouters wrote:
> 
> Eso es divertido, porque mi pescado preferido habla espa?ol :)

Ich hab meinen momentan im Fischglas, aber leider kann der Fisch kein
Holl?ndisch. Also hei?t es fr?hlich weiterraten.

> Dado el ese Guido y la otra gente que trabaja en Python (deja para darles un
> cierto cr?dito tambi?n! :-) han evitado de estropear Python hasta ahora,
> deja el corte ellos una cierta holgura y espera y ve cuidadosamente. Dudo
> que la arruinar?n eso f?cilmente.

Oversettningen fra fisken ligner meg naar jeg er fyllt. For eksempel:
"a certain comfort leaves to the cut they and hopes and sees carefully"

> Hey, Martijn, misschien dat er ergens een aangepaste babelfish draait die
> *wel* nederlands<->engels kan :)

Dat will ik dan ook.

no-hablo-espa?ol-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From phd at phd.russ.ru  Wed Jun  7 09:39:37 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 7 Jun 2000 13:39:37 +0000 (GMT)
Subject: Why not a "Deluxe" Python-distribution
In-Reply-To: <393f42e4.77383471@news.online.no>
Message-ID: <Pine.LNX.4.21.0006071336230.23668-100000@fep132.fep.ru>

On Wed, 7 Jun 2000, Thomas Weholt wrote:
> Wouldn`t it be nice to have a deluxe edition of Python, with lots of
> the adopted modules not in the standard distro? Like PIL, PST, code
> from the sigs, database drivers etc. 

   You are absolutely right! To some extent, it is here:
http://www.pythonware.com/downloads.htm#py152

   But it's certainly incomplete. Well, in the world of free / opensource
software it is simple: just do it yourself!
   Compile whatever you want, put it on your site, relax... and wait for
complains :)

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From mwh21 at cam.ac.uk  Tue Jun 20 10:04:19 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jun 2000 15:04:19 +0100
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu>
Message-ID: <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk>

bparsia at email.unc.edu (Bijan Parsia) writes:

> William Dandreta <wjdandreta at worldnet.att.net> wrote:
[snip]
> > In addition, there are 3 function calls in each replace. Logic
> > says eliminating one improve speed by 1/3.
> 
> Er.. no logic I know of :) Remember that "obvious" reasoning typically
> leads you astray when thinking about optimization. 

Hear hear.

> The key, as I have written, is the *lookup* of the method/function.

Are you really sure about this?  I was under the impression that it
was the cruft at the top of eval_code2 in ceval.c that was the major
overhead of a function call.  It certainly is in some circumstances,
though I guess it may be possible that in others method lookup
dominates.  Do you have numbers?

> These are *not* resolved at compile time, but at run time. They are
> *not* cached (unless you do so manually), they are performed from
> scratch each time.
> 
> > But on second thought, the one I eliminated was a call to a python.py
> > library module, the other 2 function calls are to functions in strop, a
> > compiled C 'module'. I'll have to test it to be sure.
> 
> Try saving the function to a local var and using it. I'd be interested
> to see what happens.

Does Python 1.2 have the local variable optimisation?  It would still
probably help though.

Cheers,
M.

-- 
    -Dr. Olin Shivers,
     Ph.D., Cranberry-Melon School of Cucumber Science
                                           -- seen in comp.lang.scheme


From cheeseballs at my-deja.com  Fri Jun 23 19:05:41 2000
From: cheeseballs at my-deja.com (cheeseballs at my-deja.com)
Date: Fri, 23 Jun 2000 23:05:41 GMT
Subject: Win32COM and Excel Charts
References: <8itpcd$ism$1@nnrp1.deja.com>
Message-ID: <8j0qfs$plm$1@nnrp1.deja.com>

In article <8itpcd$ism$1 at nnrp1.deja.com>,
  cheeseballs at my-deja.com wrote:
> Hi, has anyone successfully created an embedded chart within an Excel
> Worksheet using win32com.client?
>
> I'm able to create a chart in it's own Sheet with the following:
> >>> xlapp=win32com.client.Dispatch("excel.application")
> >>> xlapp.Visible=1
> >>> xlchart=xlapp.Charts.Add()
>
> then I try to add a chart in the active worksheet...
> >>> xlsheet=xlapp.Worksheets(1)
> >>> xlchartobj = xlsheet.ChartObjects.Add(10,10,100,100)
>
> but it doesn't look like Add is a method of the ChartObjects
> collection...
> Any ideas?
> Thanks, Drew
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

I'll answer my own question, I was missing the little parenthesis':
>>> xlchartobj = xlsheet.ChartObjects().Add(10,10,100,100)
                                     ^^


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


From petersc at stud.ntnu.no  Wed Jun 14 07:21:00 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 14 Jun 2000 13:21:00 +0200
Subject: Python and Video for Windows?
Message-ID: <39476A9C.DFE84D0@stud.ntnu.no>

I am currently working on a robot control project and I have
to use a Video for Windows compatible video camera.

Is there any possibility to get this working with Python?
Or is there some C/C++/Java based solution which I could
easily write an extension for?

All I need is really the raw image data. No fancy control
stuff or gamma correction.

don't-wanna-do-it-in-C++---please-save-me-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From hartmut at oberon.noris.de  Wed Jun 28 15:26:42 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Wed, 28 Jun 2000 20:26:42 +0100
Subject: "No module named _locale" (1.5.2 w/ RH 6.2)
References: <16600709@oberon.noris.de> <20000627160857.A32688@novara.avenue>
Message-ID: <16600711@oberon.noris.de>

David Porter (jcm at bigskytel.com) schrieb:

 >Do you have _localemodule.so ? I don't have any _locale.* on my system

No. Not a single file _locale*

+++hartmut


From nathanl at heagy.com  Mon Jun 12 01:45:38 2000
From: nathanl at heagy.com (Nathan Heagy)
Date: Mon, 12 Jun 2000 05:45:38 GMT
Subject: Putting Objects on the Web
Message-ID: <39447983.27115973@heagy.com>

I have created some classes in Python that I would like to put on the
web. I know how to create a cgi and use them in this way, but I need the
instances of the classes to be persistant so that attributes can be
changed via the web and remembered. Thus I have looked at pickle and
shelve, and they look very good. However I can't help thinking there is
another solution. Ideally I would be able to create a new instance for
each user, created via authentication methods, and each time that user
returned their instances would be remembered. Access to the methods of
the classes could be abstacted (by me) easilly. I have looked at Zope,
but was disappointed because it seemed to only use python methods.
ZPublisher looks very good, but I need some sort of authentication and
permission setting tied in so that users dont access methods they are
not supposed to. Also, I cannot see a way to create multiple instances
of an object with ZPublisher.

I hope this is clear enough (and not too verbose) to be understood.
Perhaps it is too late...

All comments and ideas welcome. I can easilly be more specific if anyone
would like more information.
-- 
Nathan Heagy
------------
Blow Your Beef Away
http://cowcomics.com


From cjc26 at nospam.cornell.edu  Sun Jun 11 20:52:03 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Mon, 12 Jun 2000 00:52:03 GMT
Subject: quick newbie question on variables. please answer
References: <39401B5F.122C6DB4@mediaone.net>
Message-ID: <slrn8k8d46.bdq.cjc26@localhost.mindriot.net>

* Toy <gee308 at mediaone.net> menulis:
| Im a programming newbie, so most of these concepts are basic to me.  I'm
| trying to write really basic  CGI programs just to get started.  How
| would I pass a text variable that a user entered to find a man page for
| a tool?

If you're writing a CGI script, then it's a little more complicated to
get input than it would be from a command-line script.  First, you need
to have a web page with a form on it, like the following:

<FORM ACTION="myscript.cgi" METHOD=GET>
<INPUT TYPE=TEXT NAME="cmd">
<INPUT TYPE=SUBMIT>
<INPUT TYPE=RESET>
</FORM>

The first INPUT element creates a text entry box called "cmd", while
the other two INPUT elements create the submit and reset buttons.

Then, in myscript.cgi, you need to use the cgi module to access the
value of "cmd":

import cgi

form = cgi.FieldStorage()
command = form["cmd"].value

See <http://www.python.org/doc/current/lib/module-cgi.html> for more
help with the cgi module.


| This is the code I wrote (startx is just a command I wrote in,
| but how would I write it so any tool could be inputed) the 3rd line od
| code is written wrong for sure:
| 
| #!/usr/bin/python
| import commands
| b = commands.getstatusoutput('man %x!') % startx

If you use the cgi code above, then this line would become:

b = commands.getstatusoutput('man %s' % command)

Note that you need to use %s instead of %x, since it's a string that
you're inserting, and that the "% command" part has to be >inside< the
parentheses.


| print = b
| 
|   I know that this is a real basic question, but I am 100% new to
| programming and I am self teaching.  Also, are ther large security risks
| if this program is on a web server(if so, how could I fix it?)?

Well you have to make sure that the user doesn't try to execute something
like:

man `rm -rf /; echo startx`

(Of course after running that, there wouldn't BE any man pages to read
;)

To catch this, you could use the string module to check and make sure
that the value input by the user is a single word, without any "funny"
characters like ` in it.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From petersc at stud.ntnu.no  Sat Jun 10 17:55:38 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 10 Jun 2000 23:55:38 +0200
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <8hqhhi$396$1@mail.pl.unisys.com> <8huc4t$gu7$1@slb2.atl.mindspring.net>
Message-ID: <3942B95A.939B3A1A@stud.ntnu.no>

Kent wrote:
> 
> Clive Tooth <clive at pisquaredoversix.force9.co.uk> wrote in message
> news:8hqhhi$396$1 at mail.pl.unisys.com...
> > Robert Israel wrote in message <8houlu$j9s$1 at nntp.itservices.ubc.ca>...
> >
> > >[...]
> > >There are two different algorithms here, for two different problems.
> > >
> > >The BBP algorithm can find a hexadecimal (or more generally, base
> > >2^k) digit of pi, faster than any known algorithm that would find
> > >all the digits up to this one.  This is the exciting one.
> >
> > Although BBP do mention that "This algorithm is, by a factor of
> > log(log(log(n))), asymptotically slower than the fastest known algorithms
> > for generating the n-th digit by generating all of the first n digits of
> > log(2) or pi."
> 
> I'm not getting this. It's "faster than any known algorithm that would find
> all the digits up to this one", but it's "asymptotically slower than the
> fastest known algorithms
>  for generating the n-th digit by generating all of the first n digits of
>  log(2) or pi". How can it be both faster and asymptotically slower? Does

I think the point is, that it is faster if you just desire ONE digit,
but if you want to calculate all the digits up to a certain limit you
are a bit slower than with a standard algorithm.

just-guessing-from-my-reading-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From hzhu at mars.localdomain  Thu Jun 15 20:19:46 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Fri, 16 Jun 2000 00:19:46 GMT
Subject: Why does time.gmtime() use Monday=0?
References: <8i9g2v$ecj$1@vife.cs.mu.OZ.AU>
Message-ID: <slrn8kisrd.pf.hzhu@mars.localdomain>

On 15 Jun 2000 12:46:55 +1000, Tim Bell <bhat at cs.mu.OZ.AU> wrote:
>
>Can anyone tell me why time.gmtime() (and time.localtime()) use the
>convention that Monday=0, rather than Sunday=0?  The Sunday=0 convention
>is used by C's gmtime(3), as well as both C and Python's strftime().
>Oh, and Python's strptime() uses Monday=0 also, but C's uses Sunday=0.
>
>Here's the relevant code from tmtotuple() in python-1.5.2/Modules/timemodule.c,
>line 262:
>
>	(p->tm_wday + 6) % 7, /* Want Monday == 0 */
>
>Why does Python go out of its way to be inconsistant, both with itself
>and with C?  (Or have I missed something here?)

Is this a misfeature of python, inspired by the desire of making every
sequence starting at 0, presumably for consistency's sake?

But consistency is illusive here: we count dates, months and years starting
at 1 while hours, minutes and seconds starting at 0.  If any, counting
weekdays the same as dates would be more consistent.  

Speaking of other languages (real ones): In Chinese the names for Monday to
Saturday are "Weekday one" to "Weekday six".  And the names for the months
(1 to 12) has been around for thousands of years.  Do we also want to count
January as 0?  :-)


Huaiyu


From kens at sightreader.com  Fri Jun 16 02:24:43 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 23:24:43 -0700
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <3949C82B.700B85F5@sightreader.com>

I did an ultra-generalized life program in python a couple years ago.  I haven't
touched it in about that long, but I can try to dig it up for you.  It allowed
/any/ set of rules imaginable, and I had implemented a couple kinds of
tri-colored hexagonal life, 3-D life (using OpenGL), and of course the
original.  I wasn't interested in speed, I just used it to learn python and
prototype some life variant ideas.

If you want to do a high speed implementation of life in python, try using a
blitter (though that would not be a beginners programming task).

Eight by eight matrix is:

>>> M = [[0]*8]*8

Usage is:

>>> M[7][7]=1
>>> M[7][7]
1

root wrote:

> newbie question.
>
> has anyone wrote the 8 non-attacking queen problem and conway's game of life
> in python? i try but face a few newbie problems.
>
> 1. conway's game of life:
>    how to have a matrix in python?
>    the following is not good;
>
>            r0=[0,0,0]
>            r1=[0,0,0]
>            ........
>
>            m=[r0,r1,....]
>
>    the matrix is fixed, which is not good. my solution use dictionary (i
>    know this is not the right way). the key is a tuple of the matrix
>    indices. the value is '0'. then base on game of life rule, update the
>    dictionary and print for next generation.
>
>            m={}
>            m[(0,0)] = 0, m[(0,1)] = 1,....
>
> 2. 8 non-attacking queen.
>
>    use backtracking. the function call itself back. how to do recursive call
>    in python? roughly, the C program I study look like this;
>
>    queen_count = -1
>
>    main()
>    {
>      addqueen();
>    }
>
>    addqueen()
>    {
>    queen_count++
>    if queen_count != 8
>         addqueen()
>    else
>         print result
>    }
>
>    how to pass a global variable (queen_count) to a function?
>
> many thanks in advance.
>
> zamri
>

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/4ebeac2b/attachment.vcf>

From danunder at columbus.rr.com  Sun Jun 11 09:55:08 2000
From: danunder at columbus.rr.com (Dan Underwood)
Date: Sun, 11 Jun 2000 13:55:08 GMT
Subject: Pythonwin: editor nits
Message-ID: <39439B04.76F15F39@columbus.rr.com>

I'm using build 129 on Win98SE. I find that the editor does not
(usually) remember my tab and indent settings. Nor does it remember that
I set "Use Tabs" on and "Show Whitespace" on. Actually, the options
dialog shows these settings as remembered but the editor seems to ignore
them for existing files. I do not want to deal with questions about why
I change tab and indent ;-) but I would like to know if I am just not
setting something right or are these just some bugs to be eventually
fixed. If a later build has already fixed them, "never mind".
Thanks.
Dan.


From jcm at bigskytel.com  Tue Jun 13 22:21:14 2000
From: jcm at bigskytel.com (David Porter)
Date: Tue, 13 Jun 2000 20:21:14 -0600
Subject: Does Python help with the no-Unix handicap?
In-Reply-To: <8i5p73$6eo$1@nnrp1.deja.com>; from david_ullrich@my-deja.com on Tue, Jun 13, 2000 at 04:58:21PM +0000
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i34pv$2ruu$1@nntp6.u.washington.edu> <8i5p73$6eo$1@nnrp1.deja.com>
Message-ID: <20000613202114.B7817@bigskytel.com>

* david_ullrich at my-deja.com <david_ullrich at my-deja.com>:

>     One of the very few things that bugs me about the Python
> docs... let's say one of the _incredibly_ few things that
> bugs me about the Python docs is where the explanation for
> how something works refers to a UNIX manpage.

Not that this aleviates your complaint, but several places on the net have
searchable man pages...

http://www2.linuxjournal.com/help/man.html


  david.

  



From db3l at fitlinxx.com  Tue Jun 13 15:06:31 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Jun 2000 15:06:31 -0400
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de>
Message-ID: <ur9a1mmdk.fsf@ctwd0143.fitlinxx.com>

Thomas Thiele <thiele at muc.das-werk.de> writes:

> >
> > Yup.  And that's precisely why Python looks better and better as your
> > programs get larger.  Python allows you to concentrate on your problem
> > rather than the language.  Python literally allows you to develop the
> > code for your algorithm in realtime, rather than designing your
> > algorithm and *then* writing the code to match your algorithm.
> >
> 
> The only difference I see between developing algorythms in C++ and
> Python is that python needs no time for compiling. But the advantage
> of C++ is that I get the most error messages during compilation (->
> earlier!)

This does raise an interesting question.  Of the total set of errors
(which "most" you receive during compilation), how many of them are
related to syntactic or language overhead versus algorithmic issues.

It seems to me that the former does not necessarily represent a
benefit to C++ over Python, because in my experience C++ has more
syntactic "sugar" that can increase errors in that respect, so the
fact that you get more of them during compilation isn't something that
translates over to Python anyway.

Some amount of such errors (say errors in type declarations or other
issues related to the strong, static typing of C++) can certainly
point to algorithm issues during compilation.  But again, many of
those issues arise from the requirement of strong typing an algorithm
in the first place, so need to be considered as errors whose existence
are largely a by-product of the development environment in use.

But true, there's the area of name errors during runtime, which my
first inclination was to consider a real problem with Python.  Even
with some of the 'lint'-like tools around, it didn't seem to be
considered a major problem by the community.  But the more code I
write the less concerned I become as I find that surprisingly it just
isn't showing up as a major problem, and it's easy enough to program
defensively against anyway.

But the latter (algorithm issues) are, I believe, just as likely to
show up in runtime in C++ as they are in Python.  And at that point,
Python makes it much easier to iterate, debug and maintain those
algorithms and designs.

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


From kain at dynup.net  Wed Jun  7 21:15:47 2000
From: kain at dynup.net (Scott Barron)
Date: Thu, 8 Jun 2000 01:15:47 GMT
Subject: I need a lesson.
References: <8hmn80$4vu$1@ionews.ionet.net>
Message-ID: <Fvt79q.8ry@boss.cs.ohiou.edu>

Hi Fred,

Maybe try something similar to this:

while 1:
	## Put your code to calculate speed here

	if raw_input ("Do another? ") == "n":
		break

That is a very basic infinite loop that will break out if you answer "n"
to the question.  What you should do next is find a good book that covers
basic programming concepts.  A good place to start nay be the book
"Programming Python".  If you can't afford to buy it, you may be able to
read it online at http://www.netlibrary.com .  Hope this hepled some.

-Scott

In article <8hmn80$4vu$1 at ionews.ionet.net>, "F. G. Brown"
<fgbrown at ionet.net> wrote:
> Greetings, I just recently started learning to program using Python, and
> to be honest when it comes to programming I'm almost dumb as a rock. 
> I've done some Qbasic, but I consider it a dead language as far as the
> future is concerned and wish to concentrate my efforts with Python.
> Here's the rub.  The following is a module called howfast.py that I
> can't get to repeat.  In Qbasic I would have used a goto at the end and
> it would have worked fine.  Of course there is no goto in Python and
> that where I can use anyone's help.  See if you can understand what I'm
> trying to accomplish below, it's a little scrambled towards the end, and
> tell me the error of my ways.  I have the book "Learning Python" but I'm
> still stuck.  If there are any questions I'll be glad to reply. Thanks,
> Fred
> ?????????????????????????????????????????????
> 
> howfast.py
> 
> a = input("What is the distance you traveled in miles?  ") b =
> input("How many seconds did it require?  ") c = a*3600.00/b print "You
> were going %d mph!" %c if c > 75:
>  print "Hey you better slow down!"
> d = raw_input ("Would you like to do another? (y)es or (n)o") if d ==
> 'y':
>  print 'Here we go!'
> else:
>  print 'See ya!'
> 
> 



From Cedric.Lefeuvre at tcomhp43.epfl.ch  Sat Jun 24 11:43:09 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Sat, 24 Jun 2000 17:43:09 +0200
Subject: an help for every function
References: <395078C0.5105D26E@tcomhp43.epfl.ch> <BT545.13336$C44.784772@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <3954D70D.8980E5FE@tcomhp43.epfl.ch>

William Dandreta wrote:

> Hi cedric,
>
> There is a Windows help file called Pythlp.chm that I use. You can download
> from the python webside. It has an index and you can seach for keywords.
>

No Windows please.....

--
C?dric Lefeuvre
SUPAERO 2A-Stagiaire EPFL
0041 (0)21 693 26 16
Fax: 0041 (0)21 693 26 83
 _ _ _/_  _
( (-(// /(


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000624/430ed00e/attachment.html>

From andre at beta.telenordia.se  Tue Jun 13 10:51:25 2000
From: andre at beta.telenordia.se (André Dahlqvist)
Date: Tue, 13 Jun 2000 15:51:25 +0100
Subject: importing module through variable
Message-ID: <8i5e9p$nm0$1@zingo.tninet.se>

Hi

I have a variable which is assigned the name of a module. Is it possible
to use this variable in an import statement? Someone told me I could use
"exec("import %s" %var) and while this works I am not sure how I can
access the module by it's name after that. Using something like 
var.modfunc() won't do it since 'var' is just a string containing the 
modules name.

// Andr?


From bent at clark.net  Fri Jun 30 16:19:33 2000
From: bent at clark.net (Ben Taylor)
Date: Fri, 30 Jun 2000 20:19:33 GMT
Subject: Python with cdaudio on Solaris 7/X86
Message-ID: <ph775.89756$5k2.182225@dfw-read.news.verio.net>

I can compile Python 1.52 on Solaris 7/X86 correctly with several
modules enabled. However, I'm having a problem enabling the
cdaudio module.  When I compile cdmodule.c, I get a lot of
errors that look like Python.h isn't getting read, although
I can see from the preprocessor output that is is included. 
I'm using libcdaudio-0.99.4.

Ideas?

Ben Taylor
bent at clark.net

Here are the error messages:

gcc   -I/opt/local/include -O2 -D_REENTRANT -I./../Include -I.. -DHAVE_CONFIG_H 
-c ./cdmodule.c
./cdmodule.c:43: parse error before `CDPLAYER'
./cdmodule.c:43: warning: no semicolon at end of struct or union
./cdmodule.c:44: warning: data definition has no type or storage class
./cdmodule.c:50: parse error before `cdplayerobject'
./cdmodule.c:50: warning: data definition has no type or storage class
./cdmodule.c:52: parse error before `{'
./cdmodule.c:56: parse error before `->'
./cdmodule.c:56: warning: data definition has no type or storage class
./cdmodule.c:58: parse error before `&'
./cdmodule.c:64: parse error before `cdplayerobject'
./cdmodule.c:64: warning: data definition has no type or storage class
./cdmodule.c:66: parse error before `{'
./cdmodule.c:70: parse error before `->'
./cdmodule.c:70: warning: data definition has no type or storage class
./cdmodule.c:72: parse error before `&'
./cdmodule.c:78: parse error before `cdplayerobject'
./cdmodule.c:78: warning: data definition has no type or storage class
./cdmodule.c:80: parse error before `{'
./cdmodule.c:89: parse error before `cdplayerobject'
./cdmodule.c:89: warning: data definition has no type or storage class
./cdmodule.c:91: parse error before `{'
./cdmodule.c:101: parse error before `&'
./cdmodule.c:107: parse error before `cdplayerobject'
./cdmodule.c:107: warning: data definition has no type or storage class
./cdmodule.c:109: parse error before `{'
./cdmodule.c:124: parse error before `&'
./cdmodule.c:130: parse error before `cdplayerobject'
./cdmodule.c:130: warning: data definition has no type or storage class
./cdmodule.c:132: parse error before `{'
./cdmodule.c:153: parse error before `cdplayerobject'
./cdmodule.c:153: warning: data definition has no type or storage class
./cdmodule.c:155: parse error before `{'
./cdmodule.c:157: warning: data definition has no type or storage class
./cdmodule.c:158: parse error before `status'
./cdmodule.c:158: warning: data definition has no type or storage class



From eviltofu at rocketmail.com  Wed Jun 28 18:29:37 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Wed, 28 Jun 2000 22:29:37 GMT
Subject: Module Name Problem
Message-ID: <eviltofu-C678D0.18293328062000@news.earthlink.net>

If I have a module called

maxvu.os

and I use

import os

Somehow, when I try to use the os calls or maxvu.os calls it gets 
confused. Is there a workaround besides renaming my module?


From erno at iki.fi  Tue Jun 27 13:14:31 2000
From: erno at iki.fi (Erno Kuusela)
Date: 27 Jun 2000 20:14:31 +0300
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk> <394f2d55.0@katana.legend.co.uk> <8inrg6$mm2$1@slb7.atl.mindspring.net> <kuem5juy3d.fsf@lasipalatsi.fi> <8jacu6$csg$1@slb1.atl.mindspring.net>
Message-ID: <kuaeg7ujuw.fsf@lasipalatsi.fi>

>>>>> "Aahz" == Aahz Maruch <aahz at netcom.com> writes:

    Aahz> The problem is that you literally know nothing about
    Aahz> /foo/bar because your permission problem is actually with
    Aahz> /foo.  Defining what to call that situation from the
    Aahz> program's POV is somewhat difficult; I'm comfortable with
    Aahz> saying, "no such directory".

the way i see it, it's just plainly false. also if you call
os.path.sidir('/foo/nonexistent/path2"), it should be a hard error
error in my opinion. (ignoring the fact that it's hard
to tell from the return value / error code from stat().)

  -- erno



From hzhu at knowledgetrack.com  Wed Jun 28 16:25:23 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 28 Jun 2000 20:25:23 GMT
Subject: (Serious?) package namespace problem (and a proposal)
Message-ID: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>

Given the following directory structure,

python/
   AAA.py
   BBB/
      __init__.py
      AAA.py
      
if sys.path contains 'python', then 

from AAA import *
from BBB.AAA import * 

would do the right thing at most locations, except when pwd is python/ and
when '.' is before 'python' in sys.path, in which case they both import from
python/AAA.py.

Is this a fair description of current situation?  If so this makes package
name spaces essentially flat. That is, in order to avoid name clashes one
has to avoid identical module names even if they are buried deep inside
packages.  This is even worse on Windows - its file names are caseless - as
one has to avoid filenames with different cases as well.

This situation occurs quite often in practice.  For example, AAA could be a
module for doing something.  BBB could be a wrapper package for doing things
somewhat differently.  So within BBB/AAA.py one calls 'import AAA' to get
the real job done, and oops, it imports itself!

Although it is not advisable to put '.' in path, especially at front, a lot
of people actually do. And the functionality of a package should not depend
on this detail.  Otherwise what happens if one imports several packages each
expecting a different ordering of sys.path?

In Unix shells, however, this can always be overcome by using absolute
paths, like /the/absolute/path/command, whatever the PATH is.

So now to my proposed solution.  Is it possible to have something like this?

sys.rootpath  = ['/usr/lib/python1.5', '/usr/lib/python1.5/site-packages']
from rootpath.AAA import *
from rootpath.BBB.AAA.import *

This guarantees that module names within packages are well separated.  One
can still use 
from AAA import *
to import from any module named AAA that happens to be in the path. IMHO,
this would not break any existing code (unless it happens to contain the new
keyword 'rootpath').  It would certainly facilitate much easier use of
packages. The only real problem I can see of is that developers might (for
what reason?) rush to occupy real estates under rootpath.  But since the
standard distribution is centrally controlled and add-ons are user
controlled, this shouldn't be a big problem.



-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From davec999 at my-deja.com  Mon Jun 26 10:52:39 2000
From: davec999 at my-deja.com (David Currie)
Date: Mon, 26 Jun 2000 14:52:39 GMT
Subject: Encryption modules
References: <8j7lnq$9jq$1@nnrp1.deja.com> <3dem5kwmmq.fsf@kronos.cnri.reston.va.us>
Message-ID: <8j7qne$de4$1@nnrp1.deja.com>

In article <3dem5kwmmq.fsf at kronos.cnri.reston.va.us>,
  Andrew Kuchling <akuchlin at mems-exchange.org> wrote:
> David Currie <davec999 at my-deja.com> writes:
>
> > Hi.  Does anyone know where to get the cryptographic modules
written by
> > Andrew Kuchling that are referenced in the Python library
> > documentation?  The link given for them is on Starship and seems to
be
>
> Try ftp://starship.python.net/pub/crew/amk/crypto .  The page was
> moved, and the redirect to the new location isn't working for some
> reason (Apache config, probably).
>
> --amk
>

Thanks, looks like exactly what I need...

Dave


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


From stephen at cerebralmaelstrom.com  Wed Jun 14 03:08:51 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 14 Jun 2000 07:08:51 GMT
Subject: tuple question
References: <3947255E.7CF77923@softhome.net>
Message-ID: <7cG15.59287$WS3.569927@typhoon.we.rr.com>

    This is a [insert descriptive word here] answer to your [insert
descriptive word here] question. I don't actually know the answer, but I can
guess at a couple possibilities based on the observation of the differences
between Python's lists and tuple's.

    A Tuple is of a fixed length, the length it was when it was created. A
list is of a dynamic length -- its actual size, I believe, from a memory-ish
standpoint, that lists are actually larger then they appear to be in the
Python environment. On the C level, my understanding from reading various
posts is that the list allocates additinoal memory for elements which may be
added in later, so that it doesn't have to allocate One Piece Of Memory
every time an element is added.
    A Tuple is immutable. Once created, with its given values, it can not be
changed. a = (1,2,3) can not have a[1] = 5 done. A list is mutable. You can
change the individual elements at any time.

    My unknowing guess? The overhead to create the truely 'dynamic' lists is
completely wasted in enough circumstances that it warrented making an array
which was fixed, in both size, and content.

    No clue if i'm right, but when I first came to Python, the whole concept
of an immutable array confused me enough that this is what my mind decided
Must Be The Right Answer (So, thus, it made me feel better about using them
:))

--S
etoffi <etoffi at softhome.net> wrote in message
news:3947255E.7CF77923 at softhome.net...
> *** post for free via your newsreader at post.newsfeeds.com ***
>
> admittedly this is a very [insert descriptive word here] question to ask,
but
> what is a tuple and why is it used (as opposed to an array, etc).  i
suppose the
> answer could be found in some comp sci book somewhere, but i dont have
access to
> those books.
>
> ps: i know:
>
> (x, y) = point.get()
>
> i just want to know what was the reason for creating/using tuples, more
from a
> language implementor standpoint than a witing-a-program-in-Python
standpoint...
>
> --
> Hopefully somebody will understand what I'm trying to say
>
>   **** Post for FREE via your newsreader at post.newsfeeds.com ****
>
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> *** Newsfeeds.com - The #1 Usenet Newsgroup Service on The Planet! ***
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>   http://www.newsfeeds.com     |         http://www.newsfeeds.com
>                                  |
> * Anonymous posting server!      | * Totally Uncensored!
> * SUPER Servers!                 | * Over 80,000 Newsgroups!
> * BINARIES ONLY Servers!         | * 16 seperate Newsgroup Servers!
> * SPAM FILTERED Server!          | * Instant access!
> * ADULT ONLY Server!             | * Multiple OC 3's and OC 12's!
> * MP3 ONLY Server!               | * 99% Article Completion!
> * MULTIMEDIA ONLY Server!        | * Months of Retention!
> * 7 UNCENSORED Newsgroup Servers | * Lightning FAST downloads!
>                                  |
>   http://www.newsfeeds.com       |         http://www.newsfeeds.com
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>         **** Point your newsreader to post.newsfeeds.com ****
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=




From michaelahoward at my-deja.com  Tue Jun 13 10:29:48 2000
From: michaelahoward at my-deja.com (Michael A. Howard)
Date: Tue, 13 Jun 2000 14:29:48 GMT
Subject: A simple question
Message-ID: <8i5gg5$ip$1@nnrp1.deja.com>

I am developing an application using Python/Tkinter that needs to look
similar to a VB application (a language I am far more familiar with) as
it will extend some functionality & we want the user interface to be as
close as possible to the existing system.

What I need to do is have a tabbed switchboard type menu screen with
buttons for each program that the user will interact with. This has
already been developed and went far faster than doing the same thing
with VB.
>From the switchboard the user will select the program they want to run
and it should open a new window for data entry/edit.

None of the code examples I look at show me how to do this.

Am I missing something really simple or is this task beyond the
capabilities of Tkinter or Python?



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


From tom-main at REMOVEME.home.com  Tue Jun 27 17:53:53 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 27 Jun 2000 21:53:53 GMT
Subject: more python evangelising...
References: <B57E791A.264E%eableson@mac.com> <Uj865.1905$iN5.408850@ptah.visi.com>
Message-ID: <Rn965.73297$7o1.1735998@news2.rdc1.on.home.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:Uj865.1905$iN5.408850 at ptah.visi.com...
...
> Since MS is a publicly held company, the people who run it
> have a fiduciary responsibility to the stockholders.  They are
> obligated to make them as much money as they can (legally, that
> is).  If they didn't, they could be held liable in court.

This is an important point.  This is the heart of capitalism.




From tim_one at email.msn.com  Tue Jun 27 23:21:11 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 27 Jun 2000 23:21:11 -0400
Subject: Guesstimate for 1.6 beta timeline?
In-Reply-To: <Cid65.2526$h8.109632@news-west.usenetserver.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOENBGGAA.tim_one@email.msn.com>

> The Python home page says that Python 1.6 (final) will be released
> around June 1, which is currently about 25 days ago, and it still
> looks like Python is in alpha.  Does anyone have some guesstimate
> as to when beta will be out?

Guido is back from his honeymoon, and everyone is working hard on the beta
even as I type.  Short-term prediction is fraught with dangers in this
field, but if nothing major goes wrong the first beta should be released
before this time next week.

and-if-not-you-get-your-money-back<wink>-ly y'rs  - tim





From achrist at easystreet.com  Fri Jun 16 03:19:39 2000
From: achrist at easystreet.com (Al Christians)
Date: Fri, 16 Jun 2000 00:19:39 -0700
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <39487299.4F30790C@commsecure.com.au>
Message-ID: <3949D50B.F417A20A@easystreet.com>

Ray Loyzaga wrote:
> 
> My company is currently sponsoring prizes for a course at the
> University of Sydney based upon extreme programming all using
> Python as the programming language. All the assignments look
> to be of high quality which is great ... except if you have
> to choose a winner out of them.
> 

Are you going to tell us about the programs?

TIA

Al


From ullrich at math.okstate.edu  Sun Jun 18 17:29:01 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 18 Jun 2000 21:29:01 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <01bfd93c$66c03180$2ace8ad1@daves-dell> <oO935.12054$Xx5.557215@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <01bfd96b$f7ddcea0$85a38ad1@daves-dell>

William Dandreta <wjdandreta at worldnet.att.net> wrote in article
<oO935.12054$Xx5.557215 at bgtnsc06-news.ops.worldnet.att.net>...
> Hi David,
> 
> The message is titled Python GREP(... you can take a look if you like. It
> was suggested that reg expr might be faster so give it a try.

	Sure enough, someone said "Also you might have better luck 
with the regular expression stuff (rather than the find command) and 
precompiling the string" and it seems clear from the context that
"better luck" means "faster". I doubt it's so. I could be wrong, it's
happened before, not that I can ever think of any examples.
(Note that comparing string.find() to re.search() is not the same
as comparing string.find() to the system's grep function...)

	Ok, try this. I spent a few minutes just now looking up how
a trivial regex works, not gonna look up the docs on profiler.py 
today. Instead I made a huge example that I can "time" by counting
"one, two, three...". You may want to start with less ado at
first and then crank the numbers back up if it goes by too fast:

import string
import re

r=re.compile('dog')

s='ado'*10000

l=[]
for j in range(1000): l.append(s[:])
l.append(s+'g')

def testre():
	localr = r
	for str in l: 
		res = localr.search(str)
		if res: 
			print res.start()
			break


def testfind():
	find=string.find
	for str in l:
		res = find(str,'dog')
		if res > -1:
			print res
			break

I don't _think_ I'm cheating here. On the machine I
tested this on calling testre() seems to take about
six seconds while testfind() seems to be less than 
three.

DU

> Bill
> 
> David C. Ullrich wrote in message
<01bfd93c$66c03180$2ace8ad1 at daves-dell>...
> >
> >
> >William Dandreta <wjdandreta at worldnet.att.net> wrote in article
> ><sy535.11434$Xx5.530580 at bgtnsc06-news.ops.worldnet.att.net>...
> >> I recently read a message that suggested that using regular
expressions
> >> might be faster than the find function.
> >
> > I would have guessed that if anything find would be faster, although
> >I certainly could be wrong. Did someone say that using a regular
expression
> >would be _faster_, or did they actually say it was more _powerful_ or
some
> >such?
> >



From hjwidmai at foxboro.com  Wed Jun 14 00:45:28 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Wed, 14 Jun 2000 06:45:28 +0200
Subject: Tkinter: Entry widget with history/completion
In-Reply-To: <39463E0F.97B65778@gssec.bt.co.uk>; from alan.gauld@gssec.bt.co.uk on Tue, Jun 13, 2000 at 03:07:27PM +0100
References: <20000613071201.A7987@foxboro.com> <39463E0F.97B65778@gssec.bt.co.uk>
Message-ID: <20000614064528.A11197@foxboro.com>

Alan Gauld:
> have you considered using a single line Text widget 
> instead of the Entry? It does have methods for accessing 
> the curent index position.

Not until now. But I will certainly give it a thought.

Thanks for the tip,
-- 
Hans-Joachim Widmaier



From joerg.wegener at sap.com  Fri Jun 23 10:55:11 2000
From: joerg.wegener at sap.com (Joerg Wegener)
Date: Fri, 23 Jun 2000 16:55:11 +0200
Subject: SAP R/3 and Python ?
References: <3953212C.84E3127F@lemburg.com> <8ivqtr$oo8$1@nntp9.atl.mindspring.net> <8ivsoe$7gu$1@news1.wdf.sap-ag.de> <8ivtiv$jo8$1@slb7.atl.mindspring.net>
Message-ID: <8ivtmc$8f2$1@news1.wdf.sap-ag.de>

Hi again,

"Aahz Maruch" <aahz at netcom.com> wrote in message
news:8ivtiv$jo8$1 at slb7.atl.mindspring.net...
>
> <grin>  Yet another application of Aahz's Rule:
>
> The best way to get information on Usenet is not to ask a question,
> but to post the wrong information.  --Aahz

Nice quote, I like it...

>
> Do note, please, that I was careful to say "virtual schema"; I know that
> SAP does not have a direct correspondence with the underlying DB.


Right, but to get access to this virtual schema, you need access to the ABAP
Dictionary, which again (sigh) is via RFC... (here we go again :-)

Bye
Joerg




From paul at prescod.net  Wed Jun 14 18:38:30 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 15 Jun 2000 00:38:30 +0200
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net> <memo.20000614223158.31901C@digitig.compulink.co.uk>
Message-ID: <39480966.C69FFDB7@prescod.net>

Tim Rowe wrote:
> 
> The construct
>         a = a + 1
> can confuse newbies too, provoking a response along the lines of "not for
> finite 'a' it doesn't!"
> 
> I reckon that's sufficient ground for a distinction between assignment and
> equality. '=' possibly isn't the best choice for assignment, though I much
> prefer it to "ADD 1 TO a GIVING a".

How about:

a <- a+1

pronounced "a gets a plus 1"

Nevertheless, I would NOT recommend that Python be so syntactically
innovative. a=a+1 only looks funny to mathematians who have never used
any other mainstream language -- not exactly our target audience!

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"Silence," wrote Melville, "is the only Voice of God." The assertion,
like its subject, cuts both ways, negating and affirming, implying both
absence and presence, offering us a choice; it's a line that the Society
of American Atheists could put on its letterhead and the Society of
Friends could silently endorse while waiting to be moved by the spirit
to speak. - Listening for Silence by Mark Slouka, Apr. 1999, Harper's





From david_ullrich at my-deja.com  Sat Jun  3 11:48:37 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sat, 03 Jun 2000 15:48:37 GMT
Subject: Bad Math Module Notation re: atan2
References: <010620001459577031%pecora@anvil.nrl.navy.mil> <pcovgzstc9b.fsf@math.ntnu.no>
Message-ID: <8hb9ci$kq4$1@nnrp1.deja.com>

In article <pcovgzstc9b.fsf at math.ntnu.no>,
  Harald Hanche-Olsen <hanche at math.ntnu.no> wrote:
[...]
> | Am I mising something?
>
> Yes.  The documentation (meaning the library reference) is wrong, but
> the implementation is right:
>
> >>> math.atan2(0,1)
> 0.0
> >>> math.atan2(1,0)
> 1.57079632679
>
> In fact, even the function's doc string is right (Py 1.5.2):
>
> >>> print math.atan2.__doc__
> atan2(y, x)
>
> Return atan(y/x).

    As opposed to the 1.5.2 library reference, which says

atan2 (y, x)
    Return atan(y / x).

Looks kinda the same to me... (maybe you got a different
version of the docs?)

> you-should-listen-to-what-they-do-but-not-what-they-say-ly y'rs,
> --
> * Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
> - "There arises from a bad and unapt formation of words
>    a wonderful obstruction to the mind."  - Francis Bacon
>


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


From jschmitt at vmlabs.com  Sat Jun 17 02:01:53 2000
From: jschmitt at vmlabs.com (John Schmitt)
Date: Fri, 16 Jun 2000 23:01:53 -0700
Subject: writing (Gnu)MAKE in Python
Message-ID: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>

Hi.

I'd love to take the diversion of writing Make using Python.  The things
that are keeping me from doing this are, in order:

- I'm not completely convinced that doing this is a good thing
- my employer has not asked me to, and I doubt they would see the value in
it that I do
- Make is big and complicated and reproducing all the functionality that we
need (complete with quirks and bugs) would be time-consuming
- if I couldn't demonstrate its value immediately the work would not be
appreciated, especially since we have many other projects worthy of my time

The benefits as I see them (in no order):

- provide a Make that really, really works
- Make would work the same everywhere!  wouldn't that be great?!
- not only would Make work the same way everywhere, it would take care of
foreslash and backslash differences too!
- Make bugs would be easier to fix
- new features would be much easier to add
- if I got it into our tool set, just maybe, Python will be accepted as a
(preferred?:-) development tool
- if I did it, it would be my opportunity to learn Python and to learn Make
fairly well
- did I mention portability?

In a smaller way, I would like to have Bash, sed, and grep as Python
programs rather than compiled C.

Anyone volunteering? :-)

John




From dana at mmi.oz.net  Thu Jun 15 11:04:01 2000
From: dana at mmi.oz.net (Dana Booth)
Date: 15 Jun 2000 15:04:01 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <85em60nozb.fsf@one.net.au> <8i8o8b$83i$0@216.39.141.36> <crigkss2m7ick2imqdctubg5d4lqjpvtd1@4ax.com>
Message-ID: <8iar91$gt6$0@216.39.141.36>

Dennis Lee Bieber <wlfraed at ix.netcom.com> wrote:

>> TRS-DOS or NEW-DOS? :)

DLB: 	The LS TRSDOS 6, or the Tandy original?

Good old Tandy/Radio Shack original that came with my TRS-80 III, complete
with 48k ram. :)

Ever try out CP/M on those old Osborne's or Kaypro's?

-- 
-----
Dana Booth <dana at mmi.oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From rumjuggler at cryptarchy.org  Sun Jun 18 17:36:29 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 18 Jun 2000 21:36:29 GMT
Subject: command-line parser
Message-ID: <kdgqkscjmqa1aqs78l6grsmkonfmjcpsb6@4ax.com>

"""
arglist.py

A utility for retrieving arguments passed via the command line.
"""

class Error(Exception):
    pass

    
class Arguments:
    """
Instantiate with tuples of acceptable options.  Tuples should have the form

    (short_form, long_form, accepts_value, default_value)

Either short_form or long_form can be None or a false value, if there is no
short
or long equivalent of the option, but not both (obviously).  If
accepts_value is true
and default_value is None, the instance will assume that a value is
necessary, and
raise Error if the option but no value is present.  Short_form and
long_form should
_not_ be passed with leading dashes.

__getattr__ is used to find out if an option is passed.  If the instance is
instantiated as

>>> arg = arglist.Arguments(
        ('h', 'help', None, None),
        ('p', 'port', 1, 5555),
        ('l', 'lalala', 1, None),
        ('a', 'anotheropt', 1, 21)
        )

and the command line options are --port -l --anotheropt=haha

>>> arg.process(['--port', '-l:2', '--anotheropt=haha'])

then the following should result:

>>> arg.h
>>> print arg.h
None
>>> arg.help
>>> print arg.help
None
>>> arg.port
5555
>>> arg.p
5555
>>> arg.l
'2'
>>> arg.lalala
'2'
>>> arg.a
'haha'
>>> arg.anotheropt
'haha'

It is possible to verify that arguments are a certain type using
__setattr__.
To verify that the argument to -a or --anotheropt is an int, you would do:

arg.a = int

This will raise a SystemExit exception if arg.a cannot be converted to an
int using the function int().
"""

    class Option:
        def __init__(self, opt):
            if '=' in opt:
                opt, self.arg = opt.split('=',2)
            elif ':' in opt:
                opt, self.arg = opt.split(':',2)
            else:
                self.arg = None
            if opt[:2] == '--':
                self.option = opt[2:]
            else:
                self.option = opt[1:]
                
    def __init__(self, *args):
        self.__dict__['dict'] = {}
        for short, long, hasvalue, default in args:
            if short and long:
                self.dict[short] = self.dict[long] = [hasvalue, default]
            elif short:
                self.dict[short] = [hasvalue, default]
            elif long:
                self.dict[long] = [hasvalue, default]
            else:
                raise Error, "Option must have either a short or a long
key"
        self.__dict__['nonopts'] = []
        
    def process(self, options):
        dict = self.dict
        for i in range(len(options)):
            if options[i][0] <> '-':
                self.nonopts.append(options[i])
                del options[i]
        for opt in map(Arguments.Option, options):
            option = opt.option
            if not self.dict.has_key(option):
                raise Error, "Unrecognized option %s" % option
            hasvalue, default = dict[option]
            if not hasvalue:
                dict[option][1] = default or 1
            else:
                if opt.arg is None:
                    if default is None:
                        raise Error, "Value must be provided for option %s"
% option
                    else:
                        dict[option][1] = default
                else:
                    dict[option][1] = opt.arg
    def get_nonopts(self):
        return self.nonopts
    def __getattr__(self, key):
        dict = self.dict
        if not dict.has_key(key):
            raise Error, 'No option %s' % key
        return dict[key][1]
    def __setattr__(self, key, _type):
        dict = self.dict
        if not dict.has_key(key):
            raise Error, 'No option %s' % key
        value = dict[key][1]
        try:
            _type(value)
        except ValueError:
            raise SystemExit
        except TypeError:
            raise SystemExit



From kpmurphy at my-deja.com  Thu Jun 29 16:55:04 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Thu, 29 Jun 2000 20:55:04 GMT
Subject: events are slow!
Message-ID: <8jgd2v$tif$1@nnrp1.deja.com>

does anyone know if it is possible to speed up the events in Tkinter?
What I mean is it currently samples maybe 10x per second (like
<B1-Motion>) and I want to increase that to 50x per second or something
big.  Thanks,

-->keith


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


From moron at Glue.umd.edu  Fri Jun  9 14:39:19 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 9 Jun 2000 14:39:19 -0400
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net>
Message-ID: <8hrdkn$3uf@stochastic.eng.umd.edu>

David said:
>On Fri, 9 Jun 2000 16:39:38 +0200, Thomas Wouters <thomas at xs4all.net>
>wrote:
>
>>As promised, here is my patch to add augmented assignment (+=, -= and
>>family) to Python. It's still Work In Progress, and probably requires
>>a rewrite of some magnitude before Guido will accept it.

>Why would the Python community want this?
[...]
>And it's not like it adds new functionality

But it does.  Accessing an attribute of an object or an item in a
container type can invoke arbitrary code, with arbitrarily freakish
side effects.  Evaluating x[17] might suck up hours of CPU time and
gibibytes of RAM, redefine builtin functions, launch the missiles, or
annoy your mother.  Having all of these things happen once, as in
x[17] += 1, is much preferable to, and potentially significantly
different from, having them happen twice, as in x[17] = x[17] + 1.
It's not just syntactic sugar that

>saves the lazy programmer a whopping one or two keystrokes.

Warren Focke


-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From mhammond at skippinet.com.au  Fri Jun  9 03:09:54 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Fri, 09 Jun 2000 07:09:54 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <vxW%4.1319$iJ6.12966@news-server.bigpond.net.au> <uog5b4m3e.fsf@ctwd0143.fitlinxx.com>
Message-ID: <6L005.122$cQ.2107@news-server.bigpond.net.au>

"David Bolen" <db3l at fitlinxx.com> wrote in message
news:uog5b4m3e.fsf at ctwd0143.fitlinxx.com...

> In particular, if the code uses broad exception clauses (for example,
> plain "except") in order to attempt recoveries and continue executing,
> then it's going to catch the KeyboardInterrupt exception this method
> tries to generate just as any other exception, and continue execution.
> If that execution is resulting in an infinite loop, well...

Actually, there are 2 things that can make it appear like it doesnt always
work:

* If an infinite loop is continually printing, we may have broken the
loop, but in the process of printing the resulting megabytes generated.
Being patient can solve this.

* The Keyboard interrupt happens, but Pythonwin sees it!  Even though your
code is in the tight loop, Pythonwin is still getting a little involved -
notably again for "print" statements.  So sometimes the break happens _as_
Pythonwin is printing, and Pythonwin just stops printing - but your code
keeps going!

Mark.





From wh at msgto.com  Mon Jun 19 15:04:47 2000
From: wh at msgto.com (Walter Hofmann)
Date: Mon, 19 Jun 2000 21:04:47 +0200
Subject: setting environment variables
Message-ID: <slrn8ksrme.7ku.wh@frodo.uni-erlangen.de>

Hi,

I need to set an environment variable so that a python module which I'd
like to use finds the right shared library.

I'm using Linux, so doing

LD_LIBRARY_PATH=/opt/kde2/qt/lib python test.py

works. But I would prefer to do it in python so that one can start my
program by entering

./test.py

However, setting LD_LIBRARY_PATH with os.environ within the python
script (but before importing the library) is too late. The dynamic
linker will not find the library this way.

I then tried to use

#!/usr/bin/env LD_LIBRARY_PATH=/opt/kde2/qt/lib python test.py

which, according to env's manpage should work. It doesn't, but hangs
instead in a tight loop consoming 100% CPU. Neither works the more
complex

#!/bin/bash -c 'LD_LIBRARY_PATH=/opt/kde2/qt/lib python ./test.py'

which gives

wh at frodo:~/programming/qtpython > ./test.py
bash: - : unrecognized option
Usage:  bash [GNU long option] [option] ...
[...]

Finally, I found out that the following rather tricky (dual language)
code works, however it is ugly to say the best:

#!/bin/sh
LD_LIBRARY_PATH="/usr/lib/qt-2.1.1/lib" "exec" "python" "$0" "$*"
del LD_LIBRARY_PATH
# python code starts here


My question:
Is anyone on this list aware of a cleaner way to to this?


Walter


From petersc at stud.ntnu.no  Sat Jun 10 17:15:26 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 10 Jun 2000 23:15:26 +0200
Subject: tuple creation in C extensions
Message-ID: <3942AFEE.5155481@stud.ntnu.no>

I have to create a tuple (of two integer values) in a
C extension. At the moment I use the following (rather
ugly in my eyes) construction:

t = PyTuple_New(2);
PyTuple_SET_ITEM(t,0,PyInt_FromLong(a));
PyTuple_SET_ITEM(t,1,PyInt_FromLong(b));
PyErr_SetObject(PyExc_IndexError,t);

where
PyObject* t;
long a,b;

Is there "A Better Way(TM)"?

thnx,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From ma9aba at bath.ac.uk  Tue Jun 13 11:56:59 2000
From: ma9aba at bath.ac.uk (Arkaitz)
Date: Tue, 13 Jun 2000 16:56:59 +0100
Subject: Two newbie questions
Message-ID: <20000613165659.A7828@basajaun>

Hi all,

I'm learning Python, as it seems everybody around here :-), and I found two 
problems that I can't find the solution to:
1. As the strings are inmutable, if I want to modify the order of the characters
in the string(e.g. sort them), what I do is:

st = "asdfasfasdf"
l = []
for c in st:
    l.append(c)

Then I process the list, and the problem is that I have no idea of how I can get
the string back, any help?
2. The second question is about random numbers, how can I get a random int between the minimum and maximum integers of the machine?
The next code doesn't worK
import sys, random

i = random.randint(-sys.maxint-1, sys.maxint)

Thanks in advance,

Arkaitz.



From petersc at stud.ntnu.no  Sat Jun 10 06:02:04 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 10 Jun 2000 12:02:04 +0200
Subject: ? Py 1.6 final
References: <Pine.OSF.4.10.10006091136370.28972-100000@prfdec.natur.cuni.cz>
Message-ID: <3942121C.CDC581B2@stud.ntnu.no>

>From Guido's announcement of moving Python development to BeOpen:

"There's one piece of bad news: Python 1.6 won't be released by June
 1st.  There's simply too much left to be done.  We promise that we'll
 get it out of the door as soon as possible.  By the way, Python 1.6
 will be the last release from CNRI; after that, we'll issue Python
 releases from BeOpen.com.Michal Bozon wrote:"

add to that: from all I know he might be on honeymoon right now ;-)

be patient or grab the CVS version,
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From collins at seal.aero.org  Sat Jun  3 13:36:41 2000
From: collins at seal.aero.org (Jeff Collins)
Date: Sat, 3 Jun 2000 10:36:41 -0700
Subject: Palm Python?
In-Reply-To: <8h8jmc$t9i$1@gaia.cdg.acriter.nl>
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl>
Message-ID: <14649.14534.184930.984180@malibu.aero.org>

I'm currently porting a 1.5.2+ (soon to be 1.6) to the Palm.
Hopefully, I'll be making an announcement soon.

Cees de Groot writes:
 > Sorry if this is a FAQ - I haven't been following this group for a while,
 > at least not as closely as I'd like. But rumors are that a Palm version
 > of Python is pending, and as I'm looking for a language to do little Palm
 > experiments with, I am just curious to know whether this is true.
 > 
 > The reason I'm asking is that I want to see how Palms can integrate with
 > Jini networks. The KVM is neat, but far from anything you would call a 
 > good environment for Palm apps (the showstopper being that you cannot switch
 > to and from KVM applications - they always re-initialize), and I rather 
 > invest my time in something that I may use for production later on. I
 > don't like (Pocket)C, so it's either Java or Python for me...

The Python VM is contained within a single GLib shared library.  Palm
apps are built using a freeze-like mechanism; a small wrapper
application loads and initializes the interpreter and the main python
module (stored as an application resource) is executed.

I'm unfamiliar with Jini and KVM - is a KVM application seperate from
a palm application?  Perhaps you could provide a few more details
off-line.



Jeff


 > 
 > -- 
 > Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
 > GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
 > Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/
 > -- 
 > http://www.python.org/mailman/listinfo/python-list
 > 



From eric at estinc.com  Fri Jun  9 11:26:39 2000
From: eric at estinc.com (Eric Lee Green)
Date: Fri, 09 Jun 2000 15:26:39 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <39410CF4.CEC33F2C@estinc.com>

Steve Mullarkey wrote:
> I have read in several places productivity claims of 5 to 10 times over
> 'C' and "C++". I would like to ask for some feedback from "C++"
> programmers who have moved to Python as to whether these estimates are
> realistic.
> 
> In "C++" I have :-

[Lots of tools, libraries, and pre-defined classes to ease the process.]

> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated. 

Sooner or later, no matter how comprehensive your tools and libraries, you
must write code. Python code is much more compact than C++ code. For example,
a netstream class took 3 pages of "C" code (would have taken the same amount
of C++ code, since no STL container class was applicable here).  It took half
a page of Python code. The Python code was written in 1 hour. The "C" code was
written in 4 hours. The two classes were functionally equivalent (one lives on
a client where Python is not available, one lives on a server where Python is
available).  

Our QuickStart product is written in C++. A product we are currently working
on is written (mostly) in Python (with some "C" modules and programs). So we
do have some experience with both. We would not have even started our current
project if C++ had been a requirement, because there is no way we could
deliver it with the manpower we have. (Even with Python it's a stretch -- Hey,
if anybody out there is looking for a good Python job, leave me email,
especially if you've done any sort of GUI work in the past, you're hired :-). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From cjc26 at nospam.cornell.edu  Thu Jun  8 22:53:36 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 09 Jun 2000 02:53:36 GMT
Subject: Marking hyperlinks in a Text widget
References: <8h04jo$lgh$1@zingo.tninet.se> <oq7lcc41zr.fsf@titan.progiciels-bpi.ca> <8h0kn5$6pv$1@cubacola.tninet.se> <slrn8j94er.ebv.cjc26@localhost.mindriot.net> <8h4d2e$ap4$1@cubacola.tninet.se> <slrn8jblag.jua.cjc26@localhost.mindriot.net> <8h5hvi$djt$1@cubacola.tninet.se> <slrn8ji7dm.5b1.cjc26@localhost.mindriot.net> <8hdphb$l03$1@zingo.tninet.se>
Message-ID: <slrn8k0n5c.74e.cjc26@localhost.mindriot.net>

* Andr? Dahlqvist <andre at beta.telenordia.se> menulis:
| 
| [snip] So instead of doing this, which would be pretty ugly, I 
| would prefer to use regular expressions for this. How would I use
| regular expressions to only find the links themselves, and not the
| stuff just in front of it?

If I'm understanding you correctly, you can use a group to match
the entire link, instead of just the prefix.  For example:

urlfinder = re.compile( "("                     # begin group
                        "(http://|ftp://)"      # match either prefix,
                        "[^\s)]+"               # then any number of
                                                # non-whitespace chars,
                                                # excluding ')'
                        ")"                     # end group

url.search("blah (http://www.cornell.edu) blah").group(0)

 => 'http://www.cornell.edu'


| Anyway, here's the code that I am not proud of which finds the URLs:
| 
| for word in string.split(text):
|     if string.split(word, "://")[0] in ("http", "ftp"):
|         pos = helpwin.textwidget.search(word, 1.0, END)
|         # Add a tag from the start to the end of the URL
|         textwidget.tag_add('url', pos, + pos + " + " + `len(word)`+ " chars")

Have you tried this on something with more than one URL in it?  It looks
like it will only tag the first URL, because when you search for "http"
or "ftp", you start over from position 1.0 each time.

I think it'd be easier to find all the URLs in the raw text first, then
insert into the text widget, tagging as you go along.


| And finally the code that has to find what URL we clicked on, and start
| the browser. I have not yet changed this to check for already running 
| browsers, but I will do that later. If there is an easier way of finding 
| these URLs I would love to hear it.
| 
| def browser(event):
|     import bisect
|     common_browsers = ['mozilla', 'netscape', 'lynx', 'w3m']
| 
|     # Find the ranges that have been tagged with the 'url' tag
|     ranges = event.widget.tag_ranges('url')
|     cursor_position = event.widget.index("@%d %d", event.x, event.y)
|     # Find out in range the cursor was clicked
|     slot = binsect.binsect(ranges, cursor_position)
|     # Get the url
|     hyperlink = event.widget.get(ranges[n-1], ranges[n])

This looks like it'd be the easiest way to do it, short of what I
suggested before (storing the URL with the tag itself).


|     for browser in common_browsers:
|         if not os.system(browser + hyperlink):
|             # Found browser, leave
|             break

You could probably search for browsers at the start of your application,
then cache the results, instead of looping through trying all possible
browsers each time.


-- 
cliff crawford    -><-    http://www.people.cornell.edu/pages/cjc26/
                          Synaesthesia now!            icq 68165166


From frankn=nuws at cs.vu.nl  Tue Jun  6 16:00:07 2000
From: frankn=nuws at cs.vu.nl (Frank Niessink)
Date: Tue, 6 Jun 2000 20:00:07 GMT
Subject: dictionary with complex keys
References: <393D54B6.EAF70234@magellan.umontreal.ca>
Message-ID: <8hjl87$n3l@cs.vu.nl>

Stefan Seefeld <seefelds at magellan.umontreal.ca> wrote:
> I'm trying to create a dictionary with the keys being lists.
> How can I do that ? The docs say that only simple types are
> possible as keys. Is there a way to achieve what I want ? May
> be some special operator I need to define...

Without knowing more about your lists, this is difficult to
answer. A simple solution would be to use tuple(mylist) as
key, e.g.:
>>> mylist = [1,2,3]
>>> dict = {}
>>> dict[tuple(mylist)] = 1
>>> dict
{(1, 2, 3): 1}

Another solution could be to subclass UserList and add your
own __hash__() function.

Cheers, Frank

-- 
"Ford," he said, "you're turning into a penguin. Stop it."
	-- Douglas Adams, 'The Hitch Hiker's Guide to the Galaxy'


From bill at libc.org  Wed Jun 28 19:44:21 2000
From: bill at libc.org (Bill Anderson)
Date: Wed, 28 Jun 2000 17:44:21 -0600
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net> <960673850.532727042@news.pacbell.net> <394768BA.1BCAD2B7@infercor.no> <0df101bfd5f8$6c2c1fe0$1906a8c0@fc.fenx.com>
Message-ID: <395A8DD5.CD352BCB@libc.org>

Emile van Sebille wrote:
> 
> ----- Original Message -----
> From: Paul Boddie <paulb at infercor.no>
> 
> <snip>
> 
> > What are the problems with case insensitivity? Being aware
> that some names which
> > might look very different actually refer to the same
> thing, because it is the
> > "pattern" of case in these names which is the only thing
> distinguishing them.
> >
> 
> <snip>
> 
> > Paul
> 
> This past weekend, I inherited a web site project laid
> out for NT using VBSript.  I re-hosted to linux with
> Chili!Soft.  My major problem so far has been case
> issues related to winxx/*nix file system.  In particular,
> quoted references to files were inconsistent in the code.
> AUTH.asp vs auth.asp vs AUTH.ASP. ;(  Windows liked this
> just fine.  Linux of course didn't.  These problems
> will continue to exist regardless of which way python
> moves as long as the *nixen distinguish file names
> based on case.


Or as long as non *nixen preserve yet ignore case. ;-)


From thomas at xs4all.nl  Thu Jun 29 09:00:23 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 29 Jun 2000 13:00:23 GMT
Subject: PyType_Type?
References: <Ebo45.13372$Uw3.865124@bgtnsc04-news.ops.worldnet.att.net> <slrn8l9p1v.elg.thomas@centurion.xs4all.nl> <qcp65.13064$kK.349754@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <slrn8lmi2m.qb9.thomas@centurion.xs4all.nl>

On Wed, 28 Jun 2000 15:53:58 GMT, William Dandreta
<wjdandreta at worldnet.att.net> wrote:

>>The type of a type object is not defined in a module, but in an Object
>file.
>>'Objects/typeobject.c' to be exact.
>>
>>Do you get any other error messages ? What version of Python ?

>Sorry for the module confusion, my C compiler (BC4.5) refers to .c files as
>modules .

>I am trying to compile Python 1.5.2 for DOS without a DOS extender. To do
>that I need to use overlays. I started out by including all the .c files in
>the source and made several changes and deletions. Currently I am left with
>the following modules, which all compile without errors and four linking
>errors:

Hm. I don't have the vaguest idea what 'overlays' are, nor how to compile
for DOS, with or without DOS extenders ;-) But I can tell you where to find
the symbols.

>Linking pyth152.exe:
>Linker Error: Undefined symbol _initstruct in module SOURCE\PC\CONFIG.C

This is the 'initstruct' function from the struct module. You don't seem to
compile the struct module (it isn't in the list of compiled C files you
included) so you might want to comment this out. Wether this'll result in a
working Python is something else, I don't know what uses the struct module.

>Linker Error: Undefined symbol _initsignal in module SOURCE\PC\CONFIG.C

The signal module. Same story as above, I guess, though I assume signals
are entirely different under native DOS. Signals might live in an entirely
different file, or they might be nonexistant. I do not know wether you need
this module or not.

>Linker Error: Undefined symbol _initnt in module SOURCE\PC\CONFIG.C

The 'nt' module. I seriously doubt you need this.

>Linker Error: Undefined symbol _crypt in module SOURCE\MODULES\CRYPTM~1.C

The 'crypt' module. You don't need this, it's there for systems that have a
crypt() system call. There's a native Python 'crypt.py' somewhere out there,
so you can just comment this module out.

>Any suggestions you might make would be appreciated.

Hope it helps ;)

Thomas.


From moshez at math.huji.ac.il  Sun Jun  4 00:53:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 4 Jun 2000 07:53:05 +0300 (IDT)
Subject: User Input?
In-Reply-To: <cnj_4.3811$r7.919327@news1.rdc2.tx.home.com>
Message-ID: <Pine.GSO.4.10.10006040752490.23740-100000@sundial>

On Sun, 4 Jun 2000, Michael Morrison wrote:

> My friend was talking about a programming asignment and I was telling him
> about python, so we decided to make programs in our different languages (his
> being QBasic :P) ...
> 
> Anyway, it dealt with user input, and from only having read Learning Python,
> I didn't know how to do it (and still don't).
> Is there an easy way to do user input in python?  I found a non-standard
> module, but it looked too complicated (not something I'd expect from the
> ease-of-use I've had so far).
> 
> Example: python blah.py
> Name? (type in name)
> Address? (type in address)
> etc....
> 
> Any help would be appreciated, thanks!

Read about the raw_input() function
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From scarblac-spamtrap at pino.selwerd.nl  Wed Jun 21 08:28:04 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 21 Jun 2000 12:28:04 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <8iq78m$u5m$1@nnrp1.deja.com> <8iqav3$om$1@nnrp1.deja.com>
Message-ID: <slrn8l1d1m.n23.scarblac-spamtrap@flits104-37.flits.rug.nl>

Fred Gansevles wrote in comp.lang.python:
> It just occured to me: If you'd like the enums to be 'global', you can
> use the following (also-quick-but-even-dirtyer hack):
> 
> def enum(*kw):
>     globals().update(kw)
> 
> enum(Tag = 0, Date = 1, Ammount = 2)

Or even the extremely horrible hack

Tag = 0; Date = 1; Amount = 2

;-)

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From phill at myriad.com  Wed Jun 28 01:11:55 2000
From: phill at myriad.com (Paul Hill)
Date: Tue, 27 Jun 2000 23:11:55 -0600
Subject: JPython and Java integration
References: <8j7h3m$5v18j$1@fu-berlin.de> <39579a03.17964799@news.clara.net>
Message-ID: <3959891B.23073A4@myriad.com>

"Laurence V." wrote:
> 
> On Mon, 26 Jun 2000 14:07:20 +0200, "Maurice le Rutte" <_delete_this_mlerutte at square.nl> wrote:
> 
> >I'm thinking of using JPython as our 'scripting'/plug-in language for our
> >product written in Java. With the risk of being flamed beyond recognition:
> >like MS's 'VBA', but then better.
> 
> If your product is a development tool for Java programmers, it makes more sense to use one of the Java interpreters as
> your scripting engine. Learning curve = 0. JPython learning curve = ?high? for people who don't know it.

The assumption you are making is that the target audience for the script writers are Java programers.
There are many applications where some quisi-end-user or less experienced programmer is just
adding a few extra business rules, or connecting a few objects together.  In that case
the scripting language of choice would be the one that looks the 'cleanest' to the
these users.

-Paul

-- 
Myriad Genetics: http://www.myriad.com/
Java FAQ: http://www.afu.com/javafaq.html (Section 9, Computer Dating)


From wjdandreta at worldnet.att.net  Sat Jun 17 21:12:32 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 18 Jun 2000 01:12:32 GMT
Subject: Module index information in machine readable form?
References: <394A930D.77CA9CB9@nih.gov>
Message-ID: <4mV25.6352$Uw3.421920@bgtnsc04-news.ops.worldnet.att.net>

Hi Jonathan,

If I understand your question, I did something similar for the C source code
files. In your case you can write a Python script to search for 'def' in
each of the .py files in the directories you are interested in.

Bill
Jonathan Epstein wrote in message <394A930D.77CA9CB9 at nih.gov>...
>
>
>Is there any way to obtain the module information (module names and the
>names of their methods) in a form which is suitable for machine
>processing?  I am interested in processing this data into a form which
>can be digested by ELSE:
>  http://www.lirmm.fr/doc/python/lib/modindex.html
>  http://members.xoom.com/pmilliken/
>
>Thanks,
>
>- Jonathan




From neelk at brick.cswv.com  Fri Jun  2 19:08:37 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 2 Jun 2000 23:08:37 GMT
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no> <3936EC02.60768A66@sightreader.com> <Pine.LNX.4.21.0006020157430.25547-100000@lade.src.no> <8h7ar6$nb8$1@slb6.atl.mindspring.net> <Pine.LNX.4.21.0006021828550.13804-100000@lade.src.no>
Message-ID: <slrn8jghjg.nd4.neelk@brick.cswv.com>

Joseph Santaniello <joseph at src.no> wrote:
> 
> I gave up on a locale way, but this is just what I needed. Thanks
> everybody!
 
The original post didn't show up on my machine, so I'm responding
to this followup. 

Let me caution you against using floating point numbers when dealing
with money amounts -- the rounding errors that can accumulate can
make reconciling your accounting records a PITA.

You are much better off storing money as an integer, denominated in
pennies (or whatever the equivalent minimal integral unit is for the
kroner). Or, if you really want to apparently use floating point, I'd
use a real fixed-point arithmetic class, an implementation of which
Tim Peters posted to the list in the past year. (Deja's archives are
down, as is the python.org search, or I'd give you a URL to
it. Hopefully Tim will followup with a repost....)



Neel




From trentm at activestate.com  Mon Jun  5 19:09:15 2000
From: trentm at activestate.com (Trent Mick)
Date: Mon, 5 Jun 2000 16:09:15 -0700
Subject: indirect references
In-Reply-To: <8hha1n$phl$1@nnrp1.deja.com>
References: <8hha1n$phl$1@nnrp1.deja.com>
Message-ID: <20000605160915.G11994@activestate.com>

On Mon, Jun 05, 2000 at 10:36:40PM +0000, shaleh at debian.org wrote:
> How do I do this in python:
> 
> foo = 'string'
> bar = foo
> 
> somehow assign to bar and change the value of foo.
> 
> 
I don't think I am sticking my foot in my mouth: Strings are immutable in
Python and passed/assigned by value (not by reference, which is what you are
asking for here). Arrays and dictionaries are passed by reference. Maybe you
could setup you data to use an array, pass a reference to it and change its
value, like this:

>>> foo = ['string']
>>> bar = foo
>>> bar[0] = 'hello'
>>> foo
['hello']
>>> bar
['hello']
						


Trent


-- 
Trent Mick
trentm at activestate.com



From urner at alumni.princeton.edu  Sun Jun  4 19:09:58 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 04 Jun 2000 16:09:58 -0700
Subject: Starship ever gonna fly?
Message-ID: <f9oljso6hl7r7blv64an77h8lfj6dhr6f9@4ax.com>

The loss of Starship was a real tragedy.  Glad to
read it's recovering.  But is it?  The same not-
so-reassuring text has remained unchanged for some
time.

Part of what made the Python community seem so vital
was all these add-ons (like PyOpenGL).  But with 
Starship down for the count, so many of the pointers
in Vaults of Parnassus are just dead, stone dead.

Whoever killed Starship did real and lasting damage
it seems.  Did anyone call the FBI?

Kirby




From ge at nowhere.none  Mon Jun 19 17:14:25 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 19 Jun 2000 21:14:25 GMT
Subject: SOCK_RAW with python socket library?
References: <slrn8ksreq.i2.tim.lavoie@goibniu.beyondtv.net>
Message-ID: <R2w35.457$iN5.49269@ptah.visi.com>

In article <slrn8ksreq.i2.tim.lavoie at goibniu.beyondtv.net>, Tim Lavoie wrote:

>I've been tinkering with some socket programming lately, basically trying
>out ideas I get from browsing W. Richard Stevens' "UNIX Network
>Programming". Anyway, I'd like to play in Python too, so I'm using the
>socket library functions. When I try to use SOCK_RAW though, I get:
>
>   socket.error: (93, 'Protocol not supported')
>
>I understand the "operation not permitted" if I try it as a non-root user,
>but why this? Is SOCK_RAW not supported yet? The docs mention that it isn't
>generally useful, but it is quite specifically useful for say, ping.

It's also quite useful if you're dealing with devices that
implement a non-standard protocol (I do have to deal with these
types of devices).  I currently have a small C program that
does the raw stuff, so I run it using
commands.getstatusoutput() everytime I want to send a command
to the device.  Raw socket support would be nice...


-- 
Grant Edwards                   grante             Yow!  World War III? No
                                  at               thanks!
                               visi.com            


From markx.daley at intel.com  Thu Jun  8 17:14:09 2000
From: markx.daley at intel.com (Daley, MarkX)
Date: Thu, 8 Jun 2000 14:14:09 -0700 
Subject: Obscure Exception message
Message-ID: <75F7304BB41CD411B06600A0C98414FC0A418B@ORSMSX54>

Hi, all!

I'm still trying to catch up on my Python reading (>50 messages!), but I
have a problem with a while loop I'm using.  Hope someone can help.

alphadict = {}
for i in range(len(result)):
	c = 1
	serialnum = result[i][6]
	try:
		while serialnum in alphadict[c]:
			c = c + 1
		alphadict[c] = alphadict[c] + [serialnum]
	except KeyError:
		alphadict[c] = serialnum
print len(alphadict)

The print len(alphadict) is there to let me know if it has done the job or
not.  Basically, I anticipate the length being >1.  Here is the exception
message.

Traceback (innermost last):
  File "C:\Program Files\Python\Pythonwin\pywin\framework\scriptutils.py",
line 237, in RunScript
    exec codeObject in __main__.__dict__
  File "C:\Program Files\Python\cit.py", line 63, in ?
    while `serialnum` in alphadict[c]:
TypeError: string member test needs char left operand


I have no idea what this is telling me, although the 'char' comment makes me
wish I had a flamethrower!

TIA

- Mark




From fsteele at mindspring.com  Tue Jun 20 16:45:30 2000
From: fsteele at mindspring.com (Frank Steele)
Date: Tue, 20 Jun 2000 16:45:30 -0400
Subject: Python in home automation
References: <fsteele-E5789E.13533715062000@news.inet-systems.net> <cBd25.38645$cH1.355549@typhoon.austin.rr.com> <m3r99swt4x.fsf@hare.demon.co.uk>
Message-ID: <fsteele-DDF9A7.16453020062000@news.inet-systems.net>

In article <m3r99swt4x.fsf at hare.demon.co.uk>, Les Smithson 
<lsmithso at hare.demon.co.uk> wrote:

> I have some Python scripts that control a TWSER & TW523. This includes
> low-level drivers, and a higher-level server that lets you control X10
> over a network. I also wrote a simple TkInter GUI for controlling my
> X10 devices.
> 
> A while ago I spoke to someone who hacked my code to work with a
> CM12U. We planned to rework this code to properly support both TW523
> and CM12U, but nothing ever came of it.
> 
> Is it time to resurrect this?
> 


I would certainly be interested in having a look and modifying these for 
my own (sinister) purposes, after which I would be happy to share 
changes back into the community.

I had one other response, so I suppose we could actually try to work 
together in an organized way to move the project forward....

Any other takers?

please-take-one-step-forward'ly y'rs,

Frank
fsteele at mindspring.com

<snip>

>     >> So, is anyone aware of Python modules or extensions to do X10
>     >> home automation? I'm interested hearing about them on _any_
>     >> platform, but I intend to implement on Linux and MacOS....
>     >> 
>     >> Thanks, Frank fsteele at mindspring.com


From pduffin at hursley.ibm.com  Wed Jun 21 08:19:10 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Wed, 21 Jun 2000 13:19:10 +0100
Subject: Can't we just stop bickering (CROSS POST BE CAREFUL WHEN REPLYING)
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org> <8igpab$7tf$1@slb2.atl.mindspring.net>
Message-ID: <3950B2BE.AF756B9F@hursley.ibm.com>

Apologies for the cross posting but I think that this is important.

For those not on comp.lang.tcl the quoted message was originally posted on
comp.lang.python under the title "Python 3000".

**** If you want to flame someone do it off line. ****

Aahz Maruch wrote:
> 
> In article <20000617153652.A2727 at dophnic.yi.org>,
> Derek Moeller  <moeller at dophnic.yi.org> wrote:
> >Now, I can't say I agree with posting that bit about Ruby as a follow-up to
> >mentioning Python-3k, as this was not what the original poster asked for.
> 
> Yes, and that's exactly the kind of thing some Ruby people have been
> doing for a while.  I'd feel exactly the same if someone posted a
> message requesting help with text processing and another person followed
> up with a message saying that Python would never be good at text
> processing and Perl was the only way to go.  That goes triple if there
> were a pattern of it.

It is not just the Ruby people doing it, just recently someone posted on
comp.lang.c++.moderated and comp.lang.tcl about how to provide a Tcl
front end to a C++ program. Note that is Tcl, not Tk. And someone posted
don't use Tcl, use Python because it has a Tk binding and "Tcl is crap".
Now, front end is often used in the context of a GUI front end and 
GUI + Tcl = Tk so that mistake I can live with, however I do object to
the "Tcl is crap" comment as it is just downright rude both to the
community and also to the original poster who we have to assume has taken 
at least a cursory look at Tcl and thinks that it may be suitable for doing 
what he wants.

Some observations

	Whether you like one language over another is for a large part 
	"subjective", not "objective" as while there are some differences
	in what they can do most languages are broadly equivalent so it
	then becomes a matter of "style" or "philosophy".

	Rather than accepting that we attempt to justify our "feelings"
	by giving rational explanations for why we prefer one language
	over another.

	Rational arguments based on "feelings" quickly turn into flame
	wars because when someone disagrees with us it hurts our
	"feelings".

	Trying to convince someone else who has used their language a lot
	and is comfortable with it that your language is better is for
	the most part a waste of time and also very disrespectful of
	them.

	No language is suitable for every job and the sooner we realise 
	that the better.

	Maybe there is a need for a MODERATED newsgroup such as 
	comp.lang.scripting onto which people can post references to
	the different scripting languages which are available.

	"Comparisons" between different scripting languages are often
	very poor because the author has already chosen the language 
	which they think is best and then attempt to justify that choice.

	These comparisons very quickly become out of date because the
	author has no incentive to keep up to date with the "other"
	language(s).

	All languages benefit from interaction with other languages and
	CONSTRUCTIVE discussions about what is PERCEIVED to be wrong with
	a language can often lead to improvements to that language.

	The communities associated with these languages should cooperate 
	rather than fight. I understand that in an Open Source world it
	is important to have "mind share" as that has an obvious impact
	on your development resource. However, there is sufficient
	overlap in functionality between the different communities for
	cooperation to yield better results. The following areas spring
	to mind but I am sure that there are others.
		Cross platform functionality (huge).
		Operating system access (huge).
		Graphical User Interface (large).
		Database access (huge).

	If there is one thing that Open Source teaches us it is that
	cooperation without reward works !!!!!!

As an example of the last few points, lets take Tk as an example simply
because it is about the only extension that I know which is used by Tcl,
Python, Perl, Ruby, ...

Note. The use of 'other' is not a derogatory comment.

All of those languages benefit from using Tk. However the integration of
Tk into each language is made more difficult by the fact that it requires
a Tcl interpreter to use it which increases the complexity and reduces
the performance for those 'other' languages. The solution to this problem
is to create a Tk library which does not depend on Tcl and then provide
language specific bindings to the library.

	Tcl would obviously have more integration problems than it does
	now but would benefit from any improvements made by the 'other'
	languages.

	The 'other' languages would benefit in terms of performance and
	easier integration and would also be able to invest more in
	improving it and fixing it.

So why is this not done, some possible quotes from those involved.

	"If any language can easily come along and use it then our language
	 won't have any competitive advantage over that one."

	"If our community does it then those other communities will use it
	 without giving anything back."

	"I'd rather write my own than use something developed by that
	 community."

If you look closely you will see that they are arguments against Open
Source itself but we all know Open Source works. Maybe Open Source
only works between people and not between communities of people, maybe
communities behave more like commercial companies than commercial
companies do.

I would love to see (and this is not my idea) all of our communities
working together to produce really simple building blocks which can be
used by all of us. e.g.
	Regular expression engine.
	Cross platform build system.
	Libraries of cross platform abstractions.
	Database access systems.
	GUI toolkits.
		:
		:
	The list goes on and on and on.

Maybe this is a forlorn hope, maybe nobody in the other communities wants
this, maybe I have been contaminated by Tcl glue ;-)......


From pbt at pipeline.com  Fri Jun 16 00:58:07 2000
From: pbt at pipeline.com (Philip Thorne)
Date: Fri, 16 Jun 2000 04:58:07 GMT
Subject: Ret status technique popen, popen2 (NT)
Message-ID: <3949a528.1161643@news.pipeline.com>

Can anyone give me a pointer on how to monitor the status of win32pipe
popen & popen2 calls? I want to get the return stat of the popen job
and to be able to detect if the popen2 job worked/started.

The os.* versions return the status's as expected on .close() but I
don't want to be tied to console apps so am using the win32pipe (129)
versions. These run ok but appear to return None in all situations (on
NT SP5 anyway).

The intent of my app is to pipe a process's output thru a second whose
task it is to analyse and react to the former's output _as it's
produced_. On completion the monitoring process outputs a message to
stdout that I need to catch.

I can get the status out of the popen job by means of 
   q = popen('job1 & echo job1 returned %ERRORLEVEL%')

which also serves to detect if the job failed to run at all but this
seems clunky. [As does reading from the popen line by line in a loop
to feed the popen2]. 

If the 2-way pipe fails to kick-off it takes a while before things go
wrong (if they do) e.g. I can still write to it:

>>> a,b=popen2('nosuch','t')
>>> print a
<open file 'nosuch', mode 'w' at 798e80>
>>> print b
<open file 'nosuch', mode 'r' at 798ec0>
>>> a.write('swallow\n')
>>> a.flush()
>>> a.close()

Although later on I (sometimes?) get an I/O Error on a write
complaining about an invalid parameter (itself confusing as I find I
can write an empty string '' without complaint immediately after
getting the error - which unfortunately I can't paste in as I typing
from memory).

It's simple to do using a co-process in a ksh (UWIN) which is sad as
I've been advocating python as a quick dev tool for NT & *nix (and in
this situation ksh won't do long-term).

Any ideas appreciated. ta
Phil B. 


From db3l at fitlinxx.com  Mon Jun  5 23:51:37 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Jun 2000 23:51:37 -0400
Subject: Network byte ordering question ...
References: <8hhjo6$8ev$1@watserv3.uwaterloo.ca> <uog5fsccv.fsf@ctwd0143.fitlinxx.com>
Message-ID: <uk8g3sbzq.fsf@ctwd0143.fitlinxx.com>

David Bolen <db3l at fitlinxx.com> writes:

> As you indicated you had looked at, you could probably use the struct
> module, and that may be the simplest approach.  One caution is that
> the module doesn't really guarantee much about the alignment of
> individual elements in a structure, so I would suggest slicing off the
> fields you need to decode and then unpacking them individually.

Following up on my own post, but after seeing the other response by
Mike Fletcher I went back and double checked the module and I was
mistaken - it does in fact offer standardized alignment, so the above
is incorrect if you explicitly set the byte order / alignment (such as
with the "!"  character).  So unpacking in full should be fine as long
as you specify the byte ordering.

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


From riaan at e.co.za  Fri Jun 23 12:42:04 2000
From: riaan at e.co.za (Riaan Booysen)
Date: Fri, 23 Jun 2000 18:42:04 +0200
Subject: Boa stalled? (GUI editor for wxPython)
References: <8iqbv4$4p$1@newsserver.rrzn.uni-hannover.de>
	    <3952A499.1339701@e.co.za> <8ivup5$dsq$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <3953935C.826AF3FB@e.co.za>


Bernhard Reiter wrote:
> I did not find any progress in the CVS.
> Is there a reason for not leeting people see the development drive?
> 
> So can I download a CVS version with all the improvements?

Well for awhile I had a few problems and I wasn't sure that the
changes were 'improvements'.

I'm hopefully updating the CVS this weekend, you're not the only
one asking ;)

-- 
Riaan                     >>> a='a=%s;a%%`a`';a%`a`
___________________________________________________
Boa Constructor - RAD GUI building IDE for wxPython
     http://boa-constructor.sourceforge.net


From MarkH at ActiveState.com  Sat Jun 24 08:10:10 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 24 Jun 2000 12:10:10 GMT
Subject: COM question
References: <39535377.6E6E74FA@stud.fh-regensburg.de>
Message-ID: <Cy155.3425$Sn2.17766@news-server.bigpond.net.au>

"Klaus Doerfler" <klaus.doerfler at stud.fh-regensburg.de> wrote in message
news:39535377.6E6E74FA at stud.fh-regensburg.de...

> I'm writing a simple COM server and I want to use early-binding, so how
> exactly does the server know about what the interface in the tlb-file is
> like and how do I have to use it in the server? Are there any naming
> conventions?

Python doesnt directly support creating a TLB for a Python COM server - so
your clients are forced to use late binding.

Mark.




From kens at sightreader.com  Wed Jun 14 07:56:18 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 14 Jun 2000 04:56:18 -0700
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <394772E2.973DE3AC@sightreader.com>

Now that's something I've not heard before.  Python is undisputably way easier
to learn than C++.  I can say this with no risk of starting a flame war.  I've
heard many (mostly invalid) claims about disadvantages of python, but I've never
heard anyone say it's difficult to learn.  In fact, I taught someone (with no
prior programming experience) python in one day (enough to write a small
program).  Honest.

C++ on the other hand is very difficult to learn.  There are common concepts in
C++ that trip up even the most seasoned programmer.  I have been attempting to
teach C++ to another friend of mine for a couple frustrating months.

Apparently your C++ book must be an extremely well written book indeed!  :-)
Either that or you were an expert C++ programmer in a previous life (an
interesting time paradox).

Nevertheless, I think I do know where you're coming from.  All programming
languages involve learning concepts that are foreign.  Humans have been
programming computers for less than a century and our brains simply aren't wired
for it.  It's amazing that programming is possible at all!

Here are some suggestions that might help you get started:

1. Use the interactive interpreter.  Type in examples from the book and see what
happens.  Try some variations until you understand how the example works.

2. If you are still baffled, ask questions here on this list.  You will get
several good answers within an hour or two.

3. Be patient.

jin choung wrote:

> howdy all,
>
> i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
> tried to pick up python before moving on to c++.  however, i was trying to
> work my way through the book LEARNING PYTHON and while it says that there is
> no need to have prior experience in languages like c or c++, i really found
> that that was not true.
>
> anyone else feel this way?
>
> i'm currently working my way through a 24 hour c++ book and i think this is
> going to work for me.  oh, well, will be back to contribute here in a couple
> of months... in the meantime, i'll just leech valuable nuggets of wisdom
> from you guys.
>
> thanks
>
> jin

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000614/cee7c235/attachment.vcf>

From fiona at sitegnome.com  Thu Jun 22 18:21:17 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 22 Jun 2000 22:21:17 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 22nd, 2000
Message-ID: <20000622222117.32464.qmail@synop.com>

Hi Guys,

Below are the entries that made it into http://python.faqts.com

I sent this mail yesterday but it doesn't seem to have made it to the 
newsgroup... so I send again.  Sorry if it turns up in duplicate.

regards,

Fiona


## New Entries #################################################


-------------------------------------------------------------
Is there a Win32 module that I can use to set directory privileges, that has the functionality of cacls.exe using python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3914
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond, Albert Hopkins

Check the win32security module plus the pywintypes module for direct 
access to security objects.

The "problem" is that Python makes no attempt to make it easier than it 
is in C/C++.  There is a fair bit of magic to weave, involving a number 
of steps to successfully create the ACLs, ACEs and SDs!  Getting the 
code wrong can cause a few problems ;-)

There are no "nice" wrappers around this stuff simply because I 
personally dont have the experience with them.  The few times I need to 
do security related things I search MSDN for C sample code, and 
translate it to the relevant win32security/pywintypes calls.

As an example of the verbosity required, below is some code from the 
book examples <plug>Chapter 16 - Windows NT Administration</plug> that 
simply creates a security descriptor ready to be applied to the 
necessary object.
(In fact, this is probably a reasonable percentage of what you need)

# A utility function that creates an NT security object for a user.
def CreateUserSecurityDescriptor(userName):
    sidUser = win32security.LookupAccountName(serverName, userName)[0]
    sd = win32security.SECURITY_DESCRIPTOR()

    # Create the "well known" SID for the administrators group
    subAuths = ntsecuritycon.SECURITY_BUILTIN_DOMAIN_RID, \
               ntsecuritycon.DOMAIN_ALIAS_RID_ADMINS
    sidAdmins = win32security.SID(ntsecuritycon.SECURITY_NT_AUTHORITY,
subAuths)

    # Now set the ACL, giving user and admin full access.
    acl = win32security.ACL(128)
    acl.AddAccessAllowedAce(win32file.FILE_ALL_ACCESS, sidUser)
    acl.AddAccessAllowedAce(win32file.FILE_ALL_ACCESS, sidAdmins)

    sd.SetSecurityDescriptorDacl(1, acl, 0)
    return sd


-------------------------------------------------------------
Is there a help or man function, like in unix-like OS, that we can use to describe any function of python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3910
-------------------------------------------------------------
Fiona Czuczman
Richard Chamberlain, William Dandreta

You can use docstrings which are there for most things.

import sys
print sys.__doc__

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

There is a Windows help file called Pythlp.chm that I use. You can 
download from the python webside. It has an index and you can seach for 
keywords.


-------------------------------------------------------------
How do I get Tkinter to resize frames etc. when their parents (e.g. a Toplevel instance) are resized?
http://www.faqts.com/knowledge-base/view.phtml/aid/3913
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain, John Grayson

It depends on the kind of manager you use, but I'll use pack as an 
example.

from Tkinter import *
root=Tk()
Frame(root,borderwidth=3,relief=GROOVE).pack(side=LEFT,fill=BOTH,expand=
YES)
root.mainloop()

If you look at www.manning.com/grayson you can download two chapters of
Python and Tkinter programming one of which is screen layout. (of course
you'll be much better buying a copy ;)


You'll also find over 20,000 lines of example code at the same site...


-------------------------------------------------------------
Is there some documentation for the python interface to postgresql?
http://www.faqts.com/knowledge-base/view.phtml/aid/3911
-------------------------------------------------------------
Fiona Czuczman
Rod Haper,Erno Kuusela

The PyGreSQL package ftp://ftp.druid.net/pub/distrib/PyGreSQL.tgz
(Latest released version) has a tutorial directory.

The README that comes with the module documents the postgresql-specific 
api. the most recent versions support DB-API also.


-------------------------------------------------------------
Why does 123.4+2=125.40000000000001?
http://www.faqts.com/knowledge-base/view.phtml/aid/3909
-------------------------------------------------------------
Fiona Czuczman
Ruud de Rooij

>From http://www.python.org/1.6/

  The repr() of floating point numbers now gives full (17-digit)
  precision. This means that a number like 8.1 (which cannot be
  represented exactly in binary) will show up as 8.0999999999999996 when
  displayed using repr(). This shouldn't break any code but it may
  confuse users, especially since the interactive shell uses repr() to
  display all floating point results. Use str() or explicit precision in
  string format (e.g. "%.12g" % x) to see the the more familiar, rounded
  form.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I get the current date and time?
How do I print out the current time?
http://www.faqts.com/knowledge-base/view.phtml/aid/3600
-------------------------------------------------------------
Mike Hostetler, Fiona Czuczman
http://www.python.org/doc/current/lib/module-time.html,Jeff

I didn't think that the official documentation was very clear on this 
(but it is there), so I'm adding it here.

If you do

>> import time
>> print time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(time.time()))

You'll see something like:

2000-06-08 15:05:09

To dissect this:

time.time() gets the time in seconds since the epoch

time.localtime(secs) creates a tuple

time.strftime(string,tuple) creates a string of the time, using the 
formatting in "string" (that is what the "%"'s are doing there).
To see the full table of formatting options, see 
http://www.python.org/doc/current/lib/module-time.html

Also, the system man page, ('man strftime')? documents a closer set of 
the formatting options actually supported by the libc, which is what the 
python interpreter actually calls into. this will come into play mostly 
on 'older' machines, but python runs on them too!






From greg at cosc.canterbury.ac.nz  Wed Jun 28 21:35:06 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 29 Jun 2000 13:35:06 +1200
Subject: float, '%.2f' help... please!
References: <39596682.108805343@news1.on.sympatico.ca>
Message-ID: <395AA7CA.45A3E581@cosc.canterbury.ac.nz>

chibaA at TinterlogD.Tcom wrote:
> 
> I need to display a price -
> rounded up to the proper cent.  I was using this line:
> 
> print '%.2f' % (price+0.005)

Just use '%.2f' % price. The formatting operator already 
does rounding of its own:

>>> for i in xrange(0,11):
...  x = i * 0.001
...  print x, "%.2f" % x
... 
0.0 0.00
0.001 0.00
0.002 0.00
0.003 0.00
0.004 0.00
0.005 0.01
0.006 0.01
0.007 0.01
0.008 0.01
0.009 0.01
0.01 0.01

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


From fiona at sitegnome.com  Wed Jun  7 13:32:27 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 7 Jun 2000 17:32:27 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 7th, 2000
Message-ID: <20000607173227.4962.qmail@synop.com>

Hi All,

Here are the latest entries into http://python.faqts.com

Cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Why do I get "case mismatch for module name ODBC" errors when using mxODBC?
http://www.faqts.com/knowledge-base/view.phtml/aid/3557
-------------------------------------------------------------
Steve Holden
mxODBC documentation

If you have installed the Windows extentions from the Win32
package developed by Mark Hammond, sometimes confusions arises
between the Win32 odbc.pyd file, usually found in <python>\Lib\Win32,
and the mxODBC distribution, usually found at <python>\Lib\ODBC.

The simplest fix is to rename the odbc.pyd file to Win32odbc.pyd,
although the installation notes for mxODBC do mention a couple of
other approaches.


-------------------------------------------------------------
Is there a way to conditionally pause a loop until some event (button press) takes place?
http://www.faqts.com/knowledge-base/view.phtml/aid/3566
-------------------------------------------------------------
Fiona Czuczman
Ken Seehof

Sounds like you need to use threads.  Read about the 'threading' module.

The idea is that you would run your loop in a separate thread from the 
GUI thread.  This has the additional benefit of keeping your GUI running 
while the loop is running.  When you pause, create a lock which gets 
released when the desired event occurs in the GUI thread..


-------------------------------------------------------------
How can x threads be started at once and stopped y seconds later, processing only the successfully returned threads?
http://www.faqts.com/knowledge-base/view.phtml/aid/3552
-------------------------------------------------------------
Fiona Czuczman
Gordon McMillan

Answer:

Threads have to stop themselves. You can't "kill" a thread without 
jeopardizing the integrity of the process.

For example:

Consider a situation where all the links from multiple URL's must be 
retrieved at the same time. By setting a time limit, slow sites don't 
slow down the whole process and appropriate error messages are returned 
stating which sites weren't completed in y seconds.

Solution:

You need to use a method of retrieving that recognizes time limits. In 
your case, you have 3 choices:

- a version of urllib where urlopen uses select with a timeout and 
aborts  if the timeout expires (Aahz may have done this?).
- a version of urllib that uses non-blocking sockets (then no threads 
needed).
- use separate processes instead of separate threads (because you *can* 
kill a process and properly release resources).

If none of those options are available to you - just ignore the thread 
if it's too slow.


-------------------------------------------------------------
Does Python support multi-threading?
What are the thread and threading modules for?
http://www.faqts.com/knowledge-base/view.phtml/aid/3563
-------------------------------------------------------------
Fiona Czuczman
Thomas Wouters

Yep.  Python does support multi-threading.

Multi-threading :-) the 'thread' module is the low-level thread 
interface, whereas the threading module is a higher-level, more 
object-oriented, Java-inspired (I do believe) interface. Threading is 
fairly basic, however, and not very widely available (partly because not 
all platforms support (proper) threading, and partly because it has to 
be explicitly enabled by whoever compiles Python, on the target machine.


-------------------------------------------------------------
Is there some method similar to seek in the urllib-module?
http://www.faqts.com/knowledge-base/view.phtml/aid/3551
-------------------------------------------------------------
Fiona Czuczman
Robert Roy

>From the documentation for urllib.urlopen(

This supports the following methods: read(), readline(), readlines(),
fileno(), close(), info() and geturl(). 

so no seek()...

nothing preventing you from doing 

x = urlopen('http://www.python.org')
# read into the file and throw away result
x.read(1024)
stuffIWant = x.read(256)


-------------------------------------------------------------
What is wxPython? How does it compare with Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/3565
-------------------------------------------------------------
Fiona Czuczman
Shae Erisson

wxPython is a GUI toolkit that's based on wxWindows:

It's at http://alldunn.com/wxPython/






From thomas at cintra.no  Tue Jun  6 03:32:22 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 06 Jun 2000 07:32:22 GMT
Subject: Seek in urllib.urlopen()?
Message-ID: <393ca88a.228109774@news.online.no>

Hi,

I was wondering if you could use some method similar to seek in the
urllib-module. If I want to read just the 256 bytes in the middle of a
file, not read the entire crap in front of those bytes, can I do that,
or do I have to read all the data in front too?

Thomas


From neilh at scintilla.org  Sat Jun  3 21:05:37 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 04 Jun 2000 01:05:37 GMT
Subject: mixed solution: unicase (unique allowed case) (was: Re:
References: <Pine.GSO.4.10.10006031335390.17165-100000@sundial> <memo.20000604004939.14417C@digitig.compulink.co.uk>
Message-ID: <BXh_4.4616$Hz.35258@news-server.bigpond.net.au>

> >>> class Foo:
> ...    pass
> ...
> >>> def bar(foo = None,
> ...    # /lots/ of other optional parameters
> ...    ):
> ...    # some implementation
> ...
> >>> baz = Foo()
> >>> bar(foo = baz)

   Wouldn't 'foo' be better used to describe its role rather than its type.

   bar(output = baz)

   Neil





From Wolfgang.Grafen at marconicomms.com  Wed Jun  7 04:20:23 2000
From: Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen)
Date: Wed, 07 Jun 2000 10:20:23 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
Message-ID: <393E05C7.45BA1F0@marconicomms.com>

Tim Rowe wrote:
>>In article <393CCEE6.E5C1C267 at marconicomms.com>,
>>Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen) wrote:
>
> > b) case-sensitivity with the awareness of same names. Just allow
> >
> > >>> Orange = 1
> > >>> Orange = 2
> >
> > but throw an exception when
> > >>> orange = 3
> > or
> > >>> ORANGE = 4 etc.
>
> I consider that bad behaviour. I have said elsewhere that there are
> programming idioms I consider to be good practice that use case to
> distinguish objects. I'd have no objection to an editor that pointed
out
> to me when I was doing it, but I would consider exceptions to be a
serious
> penalty for no real advantage. Are we going to make C/C++
> and Java case-insensitive, too? Or are these learners going to have to

> come to terms with the fact that "a" <> "A" at some point, and the
sooner
> the better?

That's right. I think the discussion about case-insensitivity arose
because
there is the possibility to program mistakes when misspelling the
variables
without getting any error. It is not a very good idea to restrict the
language
because of this. It is a program mistake which you can discover with
program
external and independant tools. Define some programming style guides,
better give idle / PythonWorks a customiseable style guide pane where
problems like this can be discovered by highlighting etc.

>
>
> > a more difficult task is to recognize spelling mistakes:
> > >>> ornge = 2
> > >>> oragne =3
> > >>> orange_colour = "orange"
>
> The trouble with that is, having been forbidden the use of canvas
after a
> from Tkinter import * (clashes with "Canvas"), now the confounded
thing is
> going to assume aCanvas is a typo, amend it to Canvas and throw it
out!

Make your tool as smart that it discovers both Canvas and canvas was
imported

by Tkinter and are not to be highlighted. But when you try to type
canvaS,
highlight it.

>
> We'd end up with /much/ too restricted a set of identifiers. As a Brit
I
> have enough trouble with color/colour, but I emphatically do /not/
want
> the compiler to tell me I can't use my native spelling of "colour".
>
> "Corrected in place" is a disaster in waiting, by the way. Did anyone
here
> ever use Comal? That "corrected" typo's in place, but as it didn't
know
> what the programmer intended it basically changed to random code. The
> result was that it replaced error messages with hard-to-find bugs.

Well, so don't use this option. For small and independant program units
written by beginners or children it might help.

>
> > When programming or reading the input of a non-experienced
programmer
> > who might think  012 is decimal 12 the input will be interpreted as
> > an octal number. I don't have any solution for this. Maybe there
should
> > be
> > a routine reading from stdin which doesn't interpret 012 as octal
> > number...
>
> A Python-aware editor can help with this. I set my C++ editor to
highlight
> octal in a different colour to decimal, which makes it easy to spot.

This is not a matter of an editor. I think a programmer knows what he
does
when he writes 012 but if the number is read from <stdin> as an input of

somebody who is not aware that there exists other number schemes than
decimal it definitely is a problem which you can not avoid with an
editor :)





From scarblac-spamtrap at pino.selwerd.nl  Sun Jun  4 05:39:32 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 4 Jun 2000 09:39:32 GMT
Subject: float does not parse '0.000000E+00'
References: <39397EFF.E1B76E6E@inc.it> <8hbvv0$40h$1@news.wrc.xerox.com> <393A21C9.E955293F@inc.it>
Message-ID: <slrn8jk8sk.hgn.scarblac-spamtrap@flits104-37.flits.rug.nl>

Paolo Redaelli wrote in comp.lang.python:
> Mark Jackson wrote:
> 
> > Have you checked to see if you might have rebound float in that namespace?
> What do you mean? I does not know the meaning of "rebound" in python
> slang, although I've used it for almost 2 years...
> 10x in advance for your attention

He means that one of your 'from mod import *' commands may have overwritten
the name 'float' in the current namespace. Rebound it, in other words. Or
maybe some other command did that?

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From fer at parnasus.kitenet.net  Mon Jun 19 06:09:49 2000
From: fer at parnasus.kitenet.net (Fermin Martinez)
Date: Mon, 19 Jun 2000 12:09:49 +0200
Subject: dpkg-python??
References: <khlii8.j62.ln@127.0.0.1> <394C27EE.1FEA81BD@uab.edu>
Message-ID: <dhrki8.ei.ln@127.0.0.1>

En Sun, 18 Jun 2000 01:35:34 GMT, Shae Erisson <shapr at uab.edu> nos contaba:

	>
	>If you're searching for documentation, I don't know :(
	>Probably documentation on the Debian package system and the source code
	>for dpkg-python would do it.

yes, I'm looking for documentation. I didn't find anything on the Debian
package.... so I'll try with source code.

thanks.

-- 
Chip, imposible comer uno, es horrible!
		-- Www.frases.com. 
www.clikeas.com/linux


From boud at rempt.xs4all.nl  Fri Jun  9 15:50:34 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 9 Jun 2000 19:50:34 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no>
Message-ID: <8hrhqa$afm$1@news1.xs4all.nl>

Thomas Weholt <thomas at cintra.no> wrote:
> Yo,

> PythonWin has some great stuff, like "partial" code-completion. I too
> have som problems with code gone haywire, sometimes Pythonwin gets
> stuck, but most of the time it`s a good alternative to IDLE.

Indeed - I'm quite jealous of Windows Python developers - I think
Pythonwin is great. A really nice editor (even if the 1.6a2 version
doesn't search & replace correctly - a simple bug I guess), good
class browser (but I don't know why I can't add a path to the Python
path with a simple toolbar button & common file dialog), and a very
nice feel to it generally - it just looks clean and unobtrusive.

I've spent a few weeks editing code meant to run under Linux at
my workplace, where I have to use Windows (didn't have any work
to do, so could work on my own projects), and I really liked it
a lot. Now if I could find time I'd like to clone it for Linux/PyQt,
with a gui editor thrown in - but I first want to get a 1.0 release
of Kura out!

Another curious thing - I've written reams of Python code without
the ability to run it (it uses both MySQL and PyKDE) and when I came
home it just worked - and I'm normally a write-a-few-lines-and-test-it
developer...

> I use Emacs under Linux and it works just fine. Just wish the
> code-completion-thingy in Pythonwin could somehow be implemented in
> Emacs. That`s probably never gonna happen, but an editor with these
> capabilities would help the python grow bigger, and Windows-converts
> like myself get over the habits gotten from programming in
> Delphi/JBuilder. Code-completion, Code-Insight, parameter-lookups ...
> It would be nice wouldn`t it??!

The lookups aren't that great - they only find what you've already used
in that file. I'd like loopup to look in every file that's included for
completion. But I'm not going to complain - it's good for my memory not
to have it ;-).


-- 

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


From aahz at netcom.com  Tue Jun 27 10:15:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 14:15:34 GMT
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk> <394f2d55.0@katana.legend.co.uk> <8inrg6$mm2$1@slb7.atl.mindspring.net> <kuem5juy3d.fsf@lasipalatsi.fi>
Message-ID: <8jacu6$csg$1@slb1.atl.mindspring.net>

In article <kuem5juy3d.fsf at lasipalatsi.fi>, Erno Kuusela  <erno at iki.fi> wrote:
>>>>>> "Aahz" == Aahz Maruch <aahz at netcom.com> writes:
>
>    Aahz> but that doesn't really matter; the point is that whatever
>    Aahz> internal function gets called on the *entire* path returns
>    Aahz> an error, therefore the path cannot be a directory.
>
>but it is indeed a directory - the os is just telling the you that 
>you are not permitted to have any information on that filesystem object
>(including whether it exists).

That's a matter of point of view: if you're looking at it from the POV
of the OS, you're correct; if you're looking at it from the POV of your
program, I don't think it's correct.  The problem is that you literally
know nothing about /foo/bar because your permission problem is actually
with /foo.  Defining what to call that situation from the program's POV
is somewhat difficult; I'm comfortable with saying, "no such directory".
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From jsboyle at my-deja.com  Thu Jun 15 13:58:01 2000
From: jsboyle at my-deja.com (jsboyle at my-deja.com)
Date: Thu, 15 Jun 2000 17:58:01 GMT
Subject: SWIG + NumPy - any examples ?
Message-ID: <8ib5eu$6ck$1@nnrp1.deja.com>

I have a number of C-routines that involve array manipulations that
I would like to make into Python extensions  that are Numpy
friendly.
The C-code is clean so SWIG seemed to be a easy path - except
for the Numpy arrays -
Beazley indicates that one can add typedefs such as NumPy
arrays to SWIG but does not explain it enough for a poor soul like
me to produce working code.
Path: news!uunet!ffx.uu.net!newsfeed.easynews.net!easynews.net!easynet-melon!easynet.net!diablo.theplanet.net!newsfeed.icl.net!netnews.com!newspeer.monmouth.com!nntp2.deja.com!nnrp1.deja.com!not-for-mail
Xref: news comp.lang.python:99799

If anyone has done this, I would appreciate an example so I can
get started.

If I am totally clueless, I would appreciate any advice.

Thanks
Jim


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


From aahz at netcom.com  Mon Jun 26 17:34:46 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jun 2000 21:34:46 GMT
Subject: learning-python question
References: <8j8ep3$tkn$1@nnrp1.deja.com>
Message-ID: <8j8i9m$k3f$1@nntp9.atl.mindspring.net>

In article <8j8ep3$tkn$1 at nnrp1.deja.com>,  <JeremyLynch at fsmail.net> wrote:
>
>count = 5
>
>def count_func()
>     print count
>     count = count + 1
>
>count_func()
>
>Why can count_func() not see count (it throws up a 'name' or
>'attribute' error)

See http://www.python.org/doc/FAQ.html#4.36
and http://www.python.org/doc/FAQ.html#4.57

Let us know if you need further help.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From jasonic at nomadicsltd.com  Wed Jun  7 13:17:09 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Wed, 7 Jun 2000 13:17:09 -0400
Subject: Why not a "Deluxe" Python-distribution
References: <Pine.LNX.4.21.0006071336230.23668-100000@fep132.fep.ru>
Message-ID: <sjt0l9h42t519@corp.supernews.com>

Yes, thanks for http://www.pythonware.com/downloads.htm#py152

A Deluxe pre-compiled Python is an excellent idea.
In a pragmatic sense,  I honestly think this is what step#1 of CP4E must
be..

Just give people one large super useful download without headaches which
will let them do lots of stuff right from the start. Many people who would
rather have a big initial download and then be able explore the abundance
and power without confusion. Of course full docs and examples are a good
idea too.
For those with decent bandwidth [DSL etc], the size is not an issue.
For those who do not have bandwidth they other options - for example I would
be quite happy to make DeluxePython CD's for people as long as they cover
the modest cost of materials, postage, and doingit.

-Jason

________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director

Oleg Broytmann <phd at phd.russ.ru> wrote in message
news:Pine.LNX.4.21.0006071336230.23668-100000 at fep132.fep.ru...
> On Wed, 7 Jun 2000, Thomas Weholt wrote:
> > Wouldn`t it be nice to have a deluxe edition of Python, with lots of
> > the adopted modules not in the standard distro? Like PIL, PST, code
> > from the sigs, database drivers etc.
>
>    You are absolutely right! To some extent, it is here:
> http://www.pythonware.com/downloads.htm#py152
>
>    But it's certainly incomplete. Well, in the world of free / opensource
> software it is simple: just do it yourself!
>    Compile whatever you want, put it on your site, relax... and wait for
> complains :)
>
> Oleg.            (All opinions are mine and not of my employer)
> ----
>     Oleg Broytmann      Foundation for Effective Policies
phd at phd.russ.ru
>            Programmers don't die, they just GOSUB without RETURN.
>
>




From kapil2000 at my-deja.com  Fri Jun 23 15:02:12 2000
From: kapil2000 at my-deja.com (kapil2000 at my-deja.com)
Date: Fri, 23 Jun 2000 19:02:12 GMT
Subject: Disabling Comboboxes
References: <8iqqb5$d4s$1@nnrp1.deja.com> <16ef54c0.7dd47049@usw-ex0105-035.remarq.com>
Message-ID: <8j0c7a$eke$1@nnrp1.deja.com>

Hey,

It worked great.  I basically had the same thing, except I messed up
some bindings...=P

Thanks!!

Kapil

In article <16ef54c0.7dd47049 at usw-ex0105-035.remarq.com>,
  Richard Chamberlain <richard_chamberlainNOriSPAM at ntlworld.com.invalid>
wrote:
> Hi Kapil,
>
> In the ToDo list for Pmw disabling is mentioned so hopefully
> we'll have a solution from the source before long. In the
> meantime...
>
> This is a bit of hack (in the derogatory sense of the word) but
> hey!
>
> So you basically want to import the myComboBox class and create
> an instance just as you would with Pmw. I've added a couple of
> methods - disable and enable suprisingly enough.
>
> Included below is some test code.
>
> It works in the same way as you mentioned, by unbinding the
> events, and rebinding them to enable it.
>
> Richard
>
> from Tkinter import *
> import Pmw
> root = Tk()
> root.title('Disabling ComboBox Hack')
> Pmw.initialise()
>
> class myComboBox(Pmw.ComboBox):
>     def disable(self):
>         # Rebind things to my empty handler
>         self.component('arrowbutton').bind('<1>',self.handler)
>         self.component('arrowbutton').bind('<3>',self.handler)
>         self.component('arrowbutton').bind('<Shift-
> 3>',self.handler)
>         self.component('entryfield_entry').configure
> (state='disabled',fg='grey')
>
>     def enable(self):
>         # bind the events back up to the original methods
>         self.component('arrowbutton').bind('<1>',self._postList)
>         self.component('arrowbutton').bind('<3>',self._next)
>         self.component('arrowbutton').bind('<Shift-3>',
> self._previous)
>         self.component('arrowbutton').configure(takefocus=1)
>         self.component('entryfield_entry').configure
> (state='normal',fg='black')
>
>     def handler(self,event):
>         # so it doesn't propagate the event
>         return('break')
>
> # Now test it
>
> comboentries = ("Pmw Should", "have a nicer way", "to
> disable", "a combo box")
> theEnabler = Button(root, text='Enabled',padx=20, pady=10)
> theEnabler.pack(expand=1, fill=BOTH, padx=8, pady=8)
>
> combobox = myComboBox(root, label_text='Disabling Combobox:',
> labelpos='wn',
>                         listbox_width=24, dropdown=1,
>                         scrolledlist_items=comboentries)
> combobox.pack(fill=BOTH, expand=1, padx=8, pady=8)
>
> combobox.selectitem(comboentries[0])
>
> def disable(event):
>     if theEnabler['text']=='Enabled':
>         combobox.disable()
>         theEnabler.configure(text='Disabled')
>     else:
>         combobox.enable()
>         theEnabler.configure(text='Enabled')
> theEnabler.bind('<Button-1>',disable)
> root.mainloop()
>
> Got questions?  Get answers over the phone at Keen.com.
> Up to 100 minutes free!
> http://www.keen.com
>
>


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


From aahz at netcom.com  Sun Jun 25 09:47:53 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jun 2000 13:47:53 GMT
Subject: Python with braces.
References: <395683ae.299281703@nntp.interaccess.com>
Message-ID: <8j52i9$joo$1@slb6.atl.mindspring.net>

In article <395683ae.299281703 at nntp.interaccess.com>,
Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
>
>Someone told me that there is an interpreter that can hanld Python
>using braces instead of indentation for defining blocks. Can anyone
>tell me where to find it?

I believe that information was given out on April First.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Why don't they make poppasmarts or didjeridon'ts?  --Aahz


From andre at UltraShell.Net  Thu Jun 29 06:17:48 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Thu, 29 Jun 2000 10:17:48 GMT
Subject: Problem with redirection, buffering and chaining scripts.
References: <395c3b52.183760656@nntp.interaccess.com>
Message-ID: <gnF65.8680$Q7.225830@news-east.usenetserver.com>

I never called a Python script from another Python script, but probably
it might be helpful to flush the buffer with file.close().
This is even necessary if you read and write to the same file within the
same script.
The buffers are flushed always when the script terminates, but .close()
is sometimes necessary explicitely.

-avs

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
> I'm having this problem chaining two scripts together 
> ( Windows NT ), demonstrated by the demo below.

> When script1.py calls script2.py nothing gets printed
> until the script finishes. Then the whole output of script2.py
> is flushed. Taking one of the command and executing it by hand,
> the output comes one line at a time ( as I want it ).

> Any idea of how to get script1.py to print output from
> script2.py as it comes?

> ----------------------script1.py--------------------------------
> import os,stat,time

> def scan_dir(path,func):
>     filenames=os.listdir(path)
>     for filename in filenames:
>         st=os.stat(path+'/'+filename)
>         if stat.S_ISDIR(st[0]):
>             scan_dir(path+'/'+filename,func)
>             func(path+'/'+filename)       

> def print_dir(x):
>     command="e:/python/python -u script2.py "+x
>     print command
>     for line in os.popen(command+' 2>&1','r').readlines():
>         print ">>",line,

> scan_dir("l:/",print_dir)
> ------------------script2.py--------------------------------------

> import os,sys,time


> fname=sys.argv[1]

> command="e:/bin/ls -l "
> command=command+fname+' 2>&1'
> print command
> for line in os.popen(command,'r').readlines():
>    print ">>",line,
>    time.sleep(1)

> ---------------------------------------------------------------------

-- 

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From gmcm at hypernet.com  Thu Jun 29 13:08:36 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 17:08:36 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194> <ffcmlsc6p4lqohdd6tequdu918l0temns4@4ax.com> <395B6027.B47094CA@letterror.com>
Message-ID: <8F6281530gmcmhypernetcom@199.171.54.194>

Just van Rossum wrote: 

><silly-proposal>
>If relative imports as we know them are shot in the head, an
>alternative spelling for relative imports could be invented.
>Eg. with a leading period:
>
>import .some.submodule
></silly-proposal>

The import SIG is comatose, but not dead.

I can see the usefulness of relative imports, and (very) infrequently, use 
them myself. But they have quite a few problems:
 - hiding modules / packages outside the current package
 - create the possibility that an app will have 2 instances of a module, 
thus screwing up module globals (and very hard to debug)
 - they slow down normal imports 

Their main feature is "ease of use", which can be pejoratively cast as 
"programmer laziness". 

not-that-I'd-ever-say-such-a-thing-ly y'rs

- Gordon


From greg at cosc.canterbury.ac.nz  Mon Jun 26 20:51:11 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 27 Jun 2000 12:51:11 +1200
Subject: boolean true and false values.
References: <slrn8l69kp.geg.hat@se-46.wpa.wtb.tue.nl> <memo.20000625230712.5257C@digitig.compulink.co.uk>
Message-ID: <3957FA7F.7C982483@cosc.canterbury.ac.nz>

Tim Rowe wrote:
> 
> > >what would you delimit it with? :-)
> >
> > {x,y,z} ?
> 
> That's going to get me confused...

Maybe that could be syntactic sugar for a dictionary which
maps the keys x, y and z to some default value, e.g. 1.
Then add some operators for doing set operations on
dictionaries of that form, and hey presto -- a set type
without adding a new type!

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


From aa8vb at yahoo.com  Tue Jun  6 12:32:24 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 06 Jun 2000 12:32:24 -0400
Subject: check for a running subprocess
In-Reply-To: <39369561.2633A7A9@gmx.net>; from dmoebius@gmx.net on Thu, Jun 01,
 2000 at 06:54:57PM +0200
References: <39369561.2633A7A9@gmx.net>
Message-ID: <20000606123224.A115482@vislab.epa.gov>

Dirk Moebius:
 |
 |I don't know whether that's a python or unix question, but:
 |
 |How do I check whether a subprocess of the current process is still
 |running? Let's say, the current process is called 'dial' and the
 |subprocess is called 'fetchmail'. The subprocess is not necessarily a
 |_direct_ successor of the current process.

If you know the pid, then see kill(2).  Call kill with a signal of 0.  In
C, the return value is set.  In Python, you'll probably get an OSError
exception if the process doesn't exist.

If you don't know the pid, I would find a way to get it and use the above.

Otherwise you have to resort to ugly popen( "ps -psargs" ) or /proc tricks
which are system-specific.

Though, if you have only one target OS with no chance of porting, then the
latter may be reasonable.

-- 
Randall Hopper
aa8vb at yahoo.com



From thomas at xs4all.net  Thu Jun 15 19:16:43 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Jun 2000 01:16:43 +0200
Subject: Problems with classes and __add__ and __mul__...
In-Reply-To: <8ibmn8$k5h$1@nnrp1.deja.com>; from abg@saturnsys.com on Thu, Jun 15, 2000 at 10:52:31PM +0000
References: <8ibmn8$k5h$1@nnrp1.deja.com>
Message-ID: <20000616011643.L26436@xs4all.nl>

On Thu, Jun 15, 2000 at 10:52:31PM +0000, abg at saturnsys.com wrote:

> I would like to do this...
> def __add__(self, other_instance):
>     return self.list + other_instance.list

> Which returns a list.  I would like to return an object of my class type, but
> it's not right obvious to me as to how to do that.

Well, the result of 'x + y' is exactly what you return in x's __add__
method, or if x doesn't have one, what y's __radd__ returns. So, if you want
'x + y' to return an instance, make __add__ return an instance:

class spam:
	def __init__(self, l=None):
		self.list = l or []

	def __add__(self, other_instance):
		return spam(self.list + other_instance.list)

If you're worried about 'spam' not being accessible to '__add__', because
you are defining the class outside of the global namespace, you can do this
as well:

	def __add__(self, other_instance):
		return self.__class__(self.list + other_instance.list)

A bit more clutter, but possibly more obvious.

> Or, would it be better to do this?  (spam is the class here...)

[ Other options here]

> Thanks for any help that you can give me.  It would be great if you emailed
> your responses to a_gurno at yahoo.com as I don't always get to check this group
> as often as I would like. Adam

The best thing to do would probably be to subclass the standard UserList :-)
This already emulates all list-like operations, like +/*, .append(),
.extend(), etcetera. You can simply subclass it to add your own
functionality. See the 'UserList' module.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From paul at prescod.net  Wed Jun  7 03:21:46 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 07 Jun 2000 02:21:46 -0500
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <Pine.SGI.4.05.10006061254220.22654-100000@lemur.magnet.com> <393DD854.56032E87@sightreader.com>
Message-ID: <393DF80A.BB53465C@prescod.net>

Ken Seehof wrote:
> 
> Java definitely has a niche: embeded software for household appliances.

I really don't see it having particular virtues there. Forth is more
efficient and more runtime flexible. Python is easier to code. Scheme is
smaller to implement. C has better real-time characteristics. Any of
them can speak TCP/IP.

Java is the language next door: nothing particularly wonderful, nothing
particularly terrible. Just there. On the other hand, if I am forced to
program in a non-Python language and have to choose between Perl, C, VB
and Java, Java is the least of all evils.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Alabama's constitution is 100 years old, 300 pages long and has more
than 600 amendments. Highlights include "Amendment 393: Amendment of
Amendment No.  351", "Validation of Laws Regulating Court Costs in
Randolph County", "Miscegenation laws", "Bingo Games in Russell County",
"Suppression of dueling".  -
http://www.legislature.state.al.us/ALISHome.html



From mtbros at tin.it  Wed Jun 28 18:37:12 2000
From: mtbros at tin.it (Mario Tola)
Date: Thu, 29 Jun 2000 00:37:12 +0200
Subject: Using zipfile.py
References: <3%d65.680$ZI2.26352@news1.rdc1.on.wave.home.com>
Message-ID: <8jdut0$l2s$1@nslave2.tin.it>

I know I'm late as you have already find the answere, but here you can find
also how to extraxt file from archives:

import zipfile
import glob, os

#Create a new zip file named new.zip and add to it
# all files in 'E:\' whose extension is '.bmp' without saving complete path
zf = zipfile.ZipFile (".\\new.zip", "w", 8)
for file in glob.glob ("e:\\*.bmp"):
     # First parameter is the file to add to the archive, second is the name
of
     # the file in archive (here is the first name without path).
     zf.write (file, os.path.split (file)[1])
zf.close ()

#Reopen the zipped file and extract the content in c:\
zf =zipfile.ZipFile (".\\new.zip", "r")
for file in zf.listdir ():
     newFile = open ("c:\\" + file, "wb")
     newFile.write (zf.read (file))
     newFile.close()
zf.close()

This work with python 1.5.2. I know in 1.6 ther's a zipfile.py module built
in,
but I don't know if his interface is the same as the old module.

-Mario

"ColBoy" <colinmeeks at home.com> ha scritto nel messaggio
news:3%d65.680$ZI2.26352 at news1.rdc1.on.wave.home.com...
> Could anybody give me a dummies guide to using the zipfile.py module.
Just
> a couple of examples of viewing a zips contents and extracting all of a
zips
> contents.  I know this is probably dead simple, but I'm still working my
way
> around Pythons syntax for stuff.  Many thanks in advance.
>
> Regards
>
> Colin
>
>




From aahz at netcom.com  Thu Jun 22 19:36:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jun 2000 23:36:00 GMT
Subject: python optimization
References: <8iu5dt$s6b$1@nnrp1.deja.com>
Message-ID: <8iu7t0$5jg$1@slb7.atl.mindspring.net>

In article <8iu5dt$s6b$1 at nnrp1.deja.com>,  <h__l at my-deja.com> wrote:
>
>This may have already been covered but I have not seen this particular
>question answered anywhere.  I am sure I missed this in the docs (or
>read the wrong ones) but what are the OPT options for optimization?
>
>I know of -O0  <turns it off?>
>and also -O2 <default it seems for the RPM I d/l for python 1.5>

Nope, -00 should be the same as -02; they both strip docstrings from
.pyc files for slightly faster loading (but not execution).  That's it
for optimizations.  (Unfortunately, I can't find docs for this right
now; I'm not even sure that -O2 is legal.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From schorsch at schorsch.com  Fri Jun  2 13:06:11 2000
From: schorsch at schorsch.com (Georg Mischler)
Date: Fri, 02 Jun 2000 17:06:11 GMT
Subject: Python for Autocad 0.3 - closer to usefulness
Message-ID: <8h8phn$uhf$1@nnrp1.deja.com>

Hi all,

contributing my part to the reanimation of the stairship,
I have uploaded the newest release 0.3 of py_acad.arx to:

  http://starship.python.net/crew/schorsch/py_acad/

Py_acad.arx makes Python available as an extension language
to Autocad in a way similar to Autolisp, and allows to control
many aspects of Autocad from within Python programs.

This release implements large portions of the ADS API, with the
most notable exceptions being symbol table/system dictionary
access and DCL dialog control. The rest of the functionality is
now in a state that actually makes it useful for many practical
purposes. As always, I'd be grateful for comments, suggestions,
patches and anything else you want to say about it.

For the lazy folks, I have created two binary packages for
R14/R15 respectively, so that you can play around with it
even if you don't feel like compiling an ARX extension yourself.


Have fun!

-schorsch

--
Georg Mischler  --  simulations developer  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


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


From root at rainerdeyke.com  Tue Jun 27 19:20:50 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 27 Jun 2000 23:20:50 GMT
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com>
Message-ID: <mFa65.595$To.223585@news1.rdc2.tx.home.com>

Greg Gallagher <ggallag at red-bean.com> wrote in message
news:8jb6jr$2uov$1 at news.enteract.com...
>
> I'm a little disheartened here... I know how to do this in Perl, but I'm
lost in Python and
> searching for 'complex data structures' is landing me nowhere.  Basically,
I want
> a dictionary whosevalues are lists of dictionaries.  Simple.
> I want to basically loop through some data and so something like this:
>
> while loop:
>    new_forward = {'ip' : ip_address,
>                   'hostname' : hostname,
>                   'mac_address' : mac_address}
>
>    forward[domain].append(new_forward)

This should work, assuming forward[domain] exists and is a list.

try:
  tmp = forward[domain] # Does forward[domain] exist?
except:
  forward[domain] = []
forward[domain].append(new_forward)

There is almost certainly a better way to check if forward[domain] exists,
but I'm too lazy to check.

>    forward[domain] = (forward[domain], new_forward)

This creates nested tuples - not what you want, I think.


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




From latha at wnmail.att.com  Fri Jun  2 16:02:32 2000
From: latha at wnmail.att.com (Latha Ravishankar)
Date: Fri, 02 Jun 2000 20:02:32 GMT
Subject: Help with SSL
Message-ID: <3938124A.9AC18D9F@wnmail.att.com>

Hi,
I am new to using SSL in python.. and am a little confused too...
I have the packages m2crypto and ossl..
I use the SSL module from here( import SSL)..
Questions:
    There seems to be another module/file - _m2cryptoc.py that is being
used, though I dont know where and a function over here is messing up..
I get a mismatch on public/private keys when I try to load a
certificate...Can somebody help me on that...

Also, is this the process of delaing with ossl:
1. Create a SSL instance.
2. Load the certificate.
3. Create a socket with the SSL instance.
4. Connect on the socket..

Any help will truely be apprecaited...
Thanks,
Latha



From nascheme at enme.ucalgary.ca  Fri Jun  9 12:36:44 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Fri, 09 Jun 2000 16:36:44 GMT
Subject: Finding cycles and garbage, was( RE: Python Productivity over  C++)
References: <CC59868A0A76D311B1F50090277B321C1F339F@VRTMAIL>
Message-ID: <slrn8k27av.1e1.nascheme@cranky.arctrix.com>

Mike Fletcher <mfletch at tpresence.com> wrote:
>Hmm, Neil's patch doesn't seem to be compatible with Stackless?

I had ported a patch a few months ago.  The current patch will
not apply without some work.

>I'm looking for something lower-level, like the VC++ feature
>where, at the end of the run, the debugger says "hey, as I was
>doing shutdown, I noticed that the following objects (including
>strings, floats, etceteras) were still in memory, that's
>probably not what you wanted".

Py_TRACE_REFS is what you want.  It should be available in
stackless too.  Look in Include/object.h.

    Neil


From michael at stroeder.com  Mon Jun 26 05:52:35 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 26 Jun 2000 11:52:35 +0200
Subject: chroot'ed demon process
Message-ID: <395727E3.D4C1388A@stroeder.com>

HI!

Is it possible to call a chroot() in a demon process running under
Python after reading all configuration data?

Ciao, Michael.


From shapr at uab.edu  Thu Jun  8 17:12:46 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 08 Jun 2000 21:12:46 GMT
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <393FB299.BED98501@muc.das-werk.de>
Message-ID: <39400C26.CE8B5B89@uab.edu>

Thomas Thiele wrote:

> And my wish is:
> 
> Something like a (optional) preprocesor that scans the whole file for unknown
> arguments and double defined functions (int the same class)

try kjpylint.py from the kwP package on http://www.chordate.com

It does that and more.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From erno at iki.fi  Sat Jun 24 14:31:43 2000
From: erno at iki.fi (Erno Kuusela)
Date: 24 Jun 2000 21:31:43 +0300
Subject: Controlling an external program with Python
References: <slrn8l87is.dpi.keating@mars.rcode.com.au>
Message-ID: <kuzoobue0g.fsf@lasipalatsi.fi>

>>>>> "Tony" == Tony Keating <keating at mech.uq.edu.au> writes:

    Tony> start the program, then send various strings to the program,
    Tony> read some output from the program then send some more
    Tony> strings to the program.

there are a couple of expect modules for python (so named because
there is an old tcl-based language called expect for doing just this).
parnassus lists 3 modules (<URL:http://www.vex.net/parnassus/
apyllo.py?find=expect>), i haven't
used any of them however...

  -- erno


From dsavitsk at e-coli.net  Sun Jun  4 04:50:05 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Sun, 04 Jun 2000 08:50:05 GMT
Subject: win32com and Excel
References: <3939D364.2EF22B13@zebrasoft.co.nz>
Message-ID: <1Lo_4.4084$gr4.138398@news1.rdc1.il.home.com>

"Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
news:3939D364.2EF22B13 at zebrasoft.co.nz...
> hi everyone,
>
> A couple o0f questions:-
> 1)
> can you access an excel (.xls) file via com without having excel loaded
> on that machine?

to the best of my knowledge, so long as you have the odbc 'stuff' installed
you can do it.  is there a reason for excel if you don't have it? it might
be easier to use access. the 'stuff' (for the jet engine anyway) is freely
downloadable from microsoft. please excuse me if i'm totally wrong here.  i
am imagining an excel file that might as well be a db table.  if this is not
what you are doing then i don't know anything about excel.  it would be very
un-microsoft like to share much of the excel object model with folks who
don't own it.
i know from distributing programs with ie embeded in them that if the user
does not have up to date ie that things fail.
in summery, in order to have the object model on your computer you must have
installed it at some point, that is, win95 doesn't magically have excel
2000's COM installed.  accessing via odbc will work, but com needs a backend
sorry this isn't really that helpful

> 2)
> once you've done:-
>  blah = win32com.client.Dispatch("Excel.Application")
> can anyone give details of
> a) howto open a particular s/s in the current directory
> and
> b) howto access particular cells in that particular s/s
> and
> c) howto close that s/s

none of this seems all that python specific. i would pick up some book on
com programming and translate from there. for instance...
    rs.AddNew
        rs![FieldName] = x
    rs.Update
in vb translates to
    rs.AddNew()
    rs.Fields('FieldName').Value = x
    rs.Update()
in python




From cgw at alum.mit.edu  Fri Jun  2 16:34:20 2000
From: cgw at alum.mit.edu (Charles G Waldman)
Date: Fri, 02 Jun 2000 14:34:20 -0600
Subject: Help, Image Filters with numpy
References: <393803A2.9AE23E9C@visionart.com>
Message-ID: <8h9284$rc2$1@info3.fnal.gov>

In article <393803A2.9AE23E9C at visionart.com>, Pete Shinners
<pete at visionart.com> wrote:
> i have some image data, greyscale, and i'd like to do some simple filter
> work on it. i'm trying to think of a "most optimized" way to get this
> done. i'm not sure about all the tools that numerical python gives me,
> so i thought i'd ask to see if parts of this are already taken care of.
> 
> say i define a quick soften filter...
> 
> soften = array([[1,2,1], [2,4,2], [1,2,1]])
> 
> how can i quickly/easily apply this to my image which is stored in
> another array? i've been hunting examples that would do this for me, but
> no luck
> 

The name for the operation you're looking
for is convolution.  Any book on digital image
processing will describe this - the book of 
Russ is an good general reference.

Numeric Python has a "convolve" method but it 
unfortunately only supports 1D convolution.  However
the filter coefficients you give describe a separable
kernel - i.e. the 2D convolution is a composition of
2 1D convolutions, in the row and column directions
(this is not true for all convolution kernels).

Also note that the Python Imaging Library ("PIL") has
builtin support for convolution filters.

Finally, if you are convolving with very large kernels,
it is more efficient to use Fourier transform techniques.



From stephen at cerebralmaelstrom.com  Thu Jun  1 08:09:44 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Thu, 01 Jun 2000 12:09:44 GMT
Subject: PyRun_String
References: <%FpZ4.31001$WS3.359957@typhoon.we.rr.com>
Message-ID: <cosZ4.31356$WS3.360373@typhoon.we.rr.com>

    Heh, nevermind. Found the answer in the demo directory where :)

Note to self: Do Search In Files *.c* for examples in before asking
questions :)

--S

Stephen Hansen <stephen at cerebralmaelstrom.com> wrote in message
news:%FpZ4.31001$WS3.359957 at typhoon.we.rr.com...
>     What do I enter into PyRun_String's "globals" and "locals" dictionary
> position if I want it to be in __main__?
>
>     E.g: I'm embedding Python into a program. I call in prior places
> PyModule_Import("languages"), and execute an arbitrary string or two
> (namely, atm, PyRun_SimpleString("sys.path.append('a-path')"))
>
>     I want to call "languages.Translate()" from __main__...
>
> Thanks,
>
> --S
>
>




From dalke at acm.org  Thu Jun  1 23:53:06 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 1 Jun 2000 21:53:06 -0600
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no> <3936EC02.60768A66@sightreader.com> <Pine.LNX.4.21.0006020157430.25547-100000@lade.src.no>
Message-ID: <8h7ar6$nb8$1@slb6.atl.mindspring.net>

Joseph Santaniello wrote:
>Ah, yes. But I'm still missing my commas [for 23,342.40]... A nice
> clean way to do that?

I think the following is only somewhat tricky, but perhaps the
cleanest.  (Okay, there's also supposed to be a way to do this with
locale, but I don't know how.)

def dollar(f):
  if f < 0:     # pull the sign out now so I know where to stop
    sign = "-"  # when going bacwards
    f = -f
  else:
    sign = ""

  s = "%.2f" % f
  # -6 is the thousand's place  --  987654.21
  # -3 goes back 1000 at a time
  # stop at 0 instead of -1 so "999.99" doesn't lead with a ",".
  for i in range(len(s)-6, 0, -3):  # -6 is the first thousands place
    s = s[:i] + "," + s[i:]         # Go back 1000 at a time
  return sign + s

>>> for i in (0, .9, 1, 10, 99.99, 100.00, 100.0, 999.99, 1000.,
  1000.00, 1000000000L, -0.01, -0.9, -1, -99.9, -100, -999.99, -1000.00,
  -123456789):
...  print i, dollar(i)
...
0 0.00
0.9 0.90
1 1.00
10 10.00
99.99 99.99
100.0 100.00
100.0 100.00
999.99 999.99
1000.0 1,000.00
1000.0 1,000.00
1000000000L 1,000,000,000.00
-0.01 -0.01
-0.9 -0.90
-1 -1.00
-99.9 -99.90
-100 -100.00
-999.99 -999.99
-1000.0 -1,000.00
-123456789 -123,456,789.00

                    Andrew
                    dalke at acm.org





From bvdpoel at uniserve.com  Sat Jun 17 16:53:42 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 17 Jun 2000 13:53:42 -0700
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com> <slrn8kn881.sdl.nobody@isis.visi.com> <394BA666.E5F3B19B@uniserve.com> <15O25.1379$i06.308670@ptah.visi.com> <394BC4CD.6432CBBF@uniserve.com> <iIQ25.1400$i06.313245@ptah.visi.com>
Message-ID: <394BE556.1D31DC6E@uniserve.com>

Grant Edwards wrote:
> 
> I am wondering if BitmapImage is somehow not adding the name of
> the created bitmap to the global list of "known" bitmaps, or if
> the name is somehow getting mangled so the name that Tk knows
> about is different than the one Python presents to the user.
> 
> Does anybody know how to see the list of "known" bitmaps?

To tell the truth, I've never used this option in tcl/tk, and I'm just
learning Tkinter...so, hopefully someone else here can help you.

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


From drew.mcdowell at msfc.nasa.gov  Wed Jun  7 14:43:38 2000
From: drew.mcdowell at msfc.nasa.gov (Andrew McDowell)
Date: 7 Jun 2000 18:43:38 GMT
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <393D3948.821E1DB6@prescod.net> <881B6C214E393C4E.E01D545A5E6A6673.A4C267911C483151@lp.airnews.net>
Message-ID: <8hm54q$anh$1@hammer.msfc.nasa.gov>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> And what software has the biggest buzz on the server side?
> Python-based Zope.  Not Beans, not even AOLServer or mod_perl,
> and certainly not JWS.
>

As much as I'd love to agree with you, I must point out that %99 of the
server side 'buzz' I hear involves Java Servlets and EJB.

-Drew





From mwh21 at cam.ac.uk  Mon Jun 19 06:29:21 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Jun 2000 11:29:21 +0100
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com> <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net> <20000619101816.X26436@xs4all.nl>
Message-ID: <m3u2eq55ha.fsf@atrus.jesus.cam.ac.uk>

Thomas Wouters <thomas at xs4all.net> writes:

> It won't matter a hoot in function calls, as all that happens string
> literals in a void context is that they get added to the functions'
> constants tuple (function.func_code.co_consts.)  And that happens at
> compiletime (which can be at runtime, of course, if you use exec()
> or import an uncompiled module, but if you do that often you
> shouldn't whine about performance ;-)

PyTrivia: isolated string constants are *not* added to co_consts,
they're just dropped.

Cheers,
M.

-- 
  Arrrrgh, the braindamage!  It's not unlike the massively
  non-brilliant decision to use the period in abbreviations 
  as well as a sentence terminator.  Had these people no 
  imagination at _all_?                 -- Erik Naggum, comp.lang.lisp


From stevena at permanent.cc  Fri Jun  9 12:57:09 2000
From: stevena at permanent.cc (Steven D. Arnold)
Date: Fri, 09 Jun 2000 12:57:09 -0400
Subject: = vs. == in Py3k
In-Reply-To: <m3n1kuvrcp.fsf@atrus.jesus.cam.ac.uk>
References: <4.3.1.2.20000609101313.01add2f8@phear.dementian.com>
Message-ID: <4.3.1.2.20000609124551.019adef8@phear.dementian.com>

At 03:52 PM 6/9/2000 +0100, Michael Hudson wrote:
>"Steven D. Arnold" <stevena at permanent.cc> writes:
>
>Rather better to give a SyntaxError *all* the time, no?  Then you
>learn not to do it pretty quickly.  I've never really understood calls
>for '-w'-type flags in Python; constructs of dubious character should
>be outlawed, not discouraged.

Actually, I rather like the C-style `=' in expressions.  (I am aware that 
no one, not even me, was proposing this.  I just happen to like it. ;-)  If 
such a feature were implemented, I'd definitely like to be able to be 
warned so that if I didn't really mean it, I could fix it.


>(Actually, I can see some merit in flagging things that are *usually*
>wrong, but have some virtue, eg. "do you *really* want to have that
>mutable default argument", but that could be a more static thing).

Well said.  Things that can NEVER be right should be errors.  Things that 
could be wrong but could be right should generate warnings, but only if you 
have set a warnings flag.


steve


--
Steven D. Arnold            Que quiero sera         stevena at permanent.cc
"Fear: The best way out is through."                     -- Helen Keller




From rob at hooft.net  Wed Jun 14 01:58:26 2000
From: rob at hooft.net (Rob Hooft)
Date: 14 Jun 2000 07:58:26 +0200
Subject: Can this be stopped from leaking memory?
References: <393FFFE7.1887276C@ebeon.com> <8hqjj1$ml4$1@nnrp1.deja.com>
Message-ID: <m27lbsolbx.fsf@hooft.net>

John Grayson writes:

 JG> However, this leaves the file descriptor dangling (further reading of
 JG> the documentation reveals:

 JG> Important: the caller is responsible for closing the file argument,
 JG> if it was not None, even when an exception is raised. ...

 JG> So, this is probably better:

 JG> import imp

 JG> fn = "mod.pyc"

 JG> for i in range(0, 20000):
 JG>     fd = open(fn, "rb")
 JG>     m = imp.load_module('Mod', fd, fn, (".pyc", "rb", imp.PY_COMPILED))
 JG>     fd.close()
 JG>     m.main()

Or, if you re-read the docs:

for i in range(0, 20000):
    fd = open(fn, "rb")
    try:
        m = imp.load_module('Mod', fd, fn, (".pyc", "rb", imp.PY_COMPILED))
    finally:
        fd.close()
    m.main()

Regards,

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


From nhodgson at bigpond.net.au  Wed Jun 14 21:26:24 2000
From: nhodgson at bigpond.net.au (Neil Hodgson)
Date: Thu, 15 Jun 2000 01:26:24 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
Message-ID: <8i9bbh$rv4$1@nnrp1.deja.com>

> I was looking at the Ruby editor/IDE for windows and was disappointed
with
> RubyWin, although it is definitely coming along. I noticed the
smalltalk
> like immediate mode (Type code, hit ctrl-J, results inserted into the
> buffer) but did not see any debugging features. Are such features
> (debugging, syntax highlight) in the works for this?

   At one stage the RubyWin page mentioned Pythonwin and Scintilla
although the page has changed since. If there is a /simple/ explanation
of Ruby's syntax and that syntax isn't too difficult, I'll write a
lexer for Ruby so it can be highlit in editors that use Scintilla (like
SciTE) or derivatives such as the wxStyledTextControl in wxWindows.

   I haven't explored Ruby much myself. An explanation of the syntax
like "its the same as C but with '#' for comments" would be easy to
implement. If it has complex Perl style quoting mechanisms then it will
be difficult.

   Neil


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


From sholden at bellatlantic.net  Tue Jun 27 23:55:16 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Wed, 28 Jun 2000 03:55:16 GMT
Subject: Guesstimate for 1.6 beta timeline?
References: <Cid65.2526$h8.109632@news-west.usenetserver.com>
Message-ID: <39597711.960D7169@bellatlantic.net>

The delay has been occasioned partly by a change in employment
on the part of a number of key players, and partly by a wish to
ensure adequate quality in the final release.

Would "real soon now" be any help? :-)

regards
 Steve

Matthew Cline wrote:
> 
> The Python home page says that Python 1.6 (final) will be released
> around June 1, which is currently about 25 days ago, and it still
> looks like Python is in alpha.  Does anyone have some guesstimate
> as to when beta will be out?
> 
> Thanks in advance.

-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From pmueller at decrc.abb.de  Mon Jun  5 07:33:26 2000
From: pmueller at decrc.abb.de (Peter Mueller)
Date: Mon, 05 Jun 2000 13:33:26 +0200
Subject: Problems with pickler in comination with pygnome/pygtk
Message-ID: <393B9006.A553F608@decrc.abb.de>

Hi,

I use the pickler to store the current object state within a small
cashbook application. With Redhat 6.1 everything works fine. After
upgrading to redhat 6.2 (python has the same version) I get the
following error message when the pygnome/pygtk modules are included in
my application:

>python cb.py

Traceback (innermost last):
  File "cb.py", line 414, in ?
    cb_obj = cashbook.load("test1", cb_obj)
  File "cashbook.py", line 27, in load
    cb = p.load()
  File "/usr/lib/python1.5/pickle.py", line 495, in load
    dispatch[key](self)
  File "/usr/lib/python1.5/pickle.py", line 551, in load_float
    self.append(string.atof(self.readline()[:-1]))
ValueError: invalid literal for atof(): 5000.0

The application code and the pickeled object are exacly the same for
Redhat6.1/6.2
And I can't leave out the pygnome stuff, because the app's gui is built
upon this libs.

Does anyone have an idea?

Peter



From breiter at usf.Uni-Osnabrueck.DE  Mon Jun 26 09:27:36 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 26 Jun 2000 13:27:36 GMT
Subject: Speed with Python
References: <QH645.94$QA1.13929@news.uswest.net>
Message-ID: <8j7lo8$adc$1@newsserver.rrzn.uni-hannover.de>

In article <QH645.94$QA1.13929 at news.uswest.net>,
	"Jeff Massung" <jmassung at magpiesystems.com> writes:
> Many threads with this kind of name have popped up, comparing the speed of
> Python to C, Java, etc. and trying to determine how fast some of the
> algorithms in Python are.
> 
> My question however, is how many of you who are Python gurus/junkies (no
> offence, I myself am a C and Forth junkie) :) use Python for speed intensive
> processing (not graphics, but databases and such)? Does efficiency of
> developement outweight the eventual speed of use?

Yes. Yes.
	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From kens at sightreader.com  Fri Jun 16 00:23:33 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 21:23:33 -0700
Subject: Starship - "I'm not dead."
Message-ID: <3949ABC5.57D6D0B@sightreader.com>

Well so far, I haven't been able to locate anyone to re-activate my FTP
account.  It's been a couple months since the crash and it's still not
completely up.  But I don't want to sound ungrateful; it's really been a
great free service.

What's the current status?  Is there progress being made?  Are there
people currently in charge of the project?  Is there any funding for it?

Should I switch to another host (I would have already done so had I
known it would take this long)?  If so, what are the best alternatives?
I could put it all on my own site, but I kind of liked the whos-who
aspect of the starship.

In any case, I'd like to thank all the people who made the starship
possible in the first place.
--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/4f90ae43/attachment.vcf>

From KatMouse at gmx.de  Thu Jun 29 17:02:55 2000
From: KatMouse at gmx.de (Eduard Hiti)
Date: Thu, 29 Jun 2000 23:02:55 +0200
Subject: Bug in PythonWin??
References: <8jfsv4$g53$1@nnrp1.deja.com>
Message-ID: <8jgdi4$9ht$15$1@news.t-online.com>

> Hi.  I've got this weird situation with PythonWin build 132.
>
> I'm working with a spanish language keyboard under NT 4 (SP4) and every
> time I hit one of the accent keys (the ones that expect a key to be
> pressed afterwards to add an accent to it) PythonWin crashes in the
> most horrible way (could not 'read' memory location bla, bla, bla.).
> The version of NT I'm running is in Spanish.
>
> Is this a bug or a misconfiguration (in PythonWin or NT itself)?  Does
> anyone else out there manage to run this on a similar configuration
> with no problems?
>

Same here for german keyboard. I contacted the author and he's probably up
to something.

    Eduard





From paul.robinson at quantisci.co.uk  Tue Jun  6 04:13:35 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Tue, 06 Jun 2000 09:13:35 +0100
Subject: SSL and POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com> <etdog5fcx7a.fsf@w20-575-109.mit.edu> <8hhttv$7r6$1@nnrp1.deja.com> <R_%_4.30$WI2.3203@newsfeed.avtel.net>
Message-ID: <393CB2AF.C920E7EC@quantisci.co.uk>


Phil Mayes wrote:
> 
> You're sending it like a GET.  IIRC, the URL is /cgi-bin/ccredit.cgi
> and the data is "0=1".  You'll also need a Content-Length header line
> with a value of 3:
>   h.putheader('Content-length', '%d' % len(data))
>   h.endheaders()
>   h.send(data + '\r\n')
> (Maybe the \r\n is redundant.)

You'll also need a:

h.putheader('Content-Type','application/x-www-form-urlencoded')

[and the \r\n is redundant.]

	Paul.

-----------------------------------
Business Collaborator Team
Enviros Software Solutions
http://www.businesscollaborator.com
-----------------------------------



From bjorn at roguewave.com  Tue Jun 20 19:57:20 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 20 Jun 2000 17:57:20 -0600
Subject: Removing items from a list in a loop
References: <3952fdf6.950540640@nntp.interaccess.com>
Message-ID: <395004E0.4F138B57@roguewave.com>

"Thaddeus L. Olczyk" wrote:
> 
> I would like to remove a bunch of items from a list *efficiently*.
> Essentially  I would like to do this:
> 
>    for x in list:
>         if Function(x):
>         list.remove(x)
> 
> This does not work. Presumably for this reason:
[snip]

the obvious way:

  for x in list[:]:  # create a copy
    if f(x):
      list.remove(x)

perhaps faster:

  for i in range(len(list)-1, 0, -1):
    x = list[i]
    if f(x):
      list.remove(x)

that traverses the list backwards, so should be more resistant to the
list changing...

-- bjorn



From DavidA at ActiveState.com  Thu Jun  8 13:30:04 2000
From: DavidA at ActiveState.com (David Ascher)
Date: Thu, 8 Jun 2000 10:30:04 -0700
Subject: Rich comparisons (was:  My wishlist for Python3K)
In-Reply-To: <m3wvk06s29.fsf@brant.geog.ubc.ca>
Message-ID: <PLEJJNOHDIGGLDPOGPJJOEDECHAA.DavidA@ActiveState.com>

I should add that if someone wants rich comparisons enough to dive into the
mess, I'll gladly help, give my view on what needs to be done, give my rough
patches, etc.  I don't want to hold back progress.

--david




From etoffi at softhome.net  Mon Jun 19 02:40:56 2000
From: etoffi at softhome.net (etoffi)
Date: Mon, 19 Jun 2000 02:40:56 -0400
Subject: <OT> New filemanager needs people with files
Message-ID: <394DC078.86C1FB02@softhome.net>

I am going to be hosting a project on sourceforge called Newfie. Tell you the
truth I'm not good at naming things. (If I got a dog, I would name it "dog". 
I'd rather not waste time thinking up names for filemanagers). On to more 
important things, it is currently in the planning stage right now.  If you ever
had a wish for a file manager, feel free to tell me.  Python scripting
will more than likely be included. All ideas will be evaluated. You wont be
sorry (unless you're stuck on Explorer).

http://sourceforge.net/projects/newfie/

--- Original Text  ---
Subject: New filemanager needs people with files
From: etoffi <etoffi at bigfoot.com>
Date: 2000-Jun-16 (1600) -0400

Ha ha ha.  Newfie (got a better name?) is a project which will be used to
surpass "* Commander" filemanagers.  It will have an extensible, pluggable
architecture, be crossplatform, among other great buzzword-type things.  Any
resources (suggestions, screenshots, sourcecode) are welcome.



From parkw at better.net  Fri Jun 16 12:50:43 2000
From: parkw at better.net (William Park)
Date: Fri, 16 Jun 2000 12:50:43 -0400
Subject: Where are the math functions?
In-Reply-To: <memo.20000615202830.52185A@digitig.compulink.co.uk>; from digitig@cix.co.uk on Thu, Jun 15, 2000 at 08:28:00PM +0100
References: <20000615134357.A1087@better.net> <memo.20000615202830.52185A@digitig.compulink.co.uk>
Message-ID: <20000616125043.A19134@better.net>

On Thu, Jun 15, 2000 at 08:28:00PM +0100, Tim Rowe wrote:
> In article <20000615134357.A1087 at better.net>, parkw at better.net (William 
> Park) wrote:
> 
> > In fact, GNU math.h includes many higher functions like
> >     j0, j1, j2, y0, y1, y2
> >     lngamma, gamma, ...
> >     erfc, erf, ...
> >     cbrt,
> > but Python 'math' module doesn't include them.  So, you have to create a
> > new module of your own.  
> 
> Is Python based on portable C? If so (I hope it is), then it's right not 
> to wrap non-standard libraries as standard. But as Gnu is open source, 
> it's a pretty handy place to get them! I needed them for an academic 
> project, and it's harder to cite gnu source, but the next ones I want for 

What do you mean "harder to cite"?  

--William


> real work I think I'll get that way!




From jkraska1 at san.rr.com  Wed Jun 28 21:13:15 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 29 Jun 2000 01:13:15 GMT
Subject: Pickle for C++?
References: <395A1F7E.4112B2A3@muc.das-werk.de> <vJp65.11228$227.232032@nnrp1.uunet.ca> <395A3049.2E783D21@san.rr.com> <Svq65.11254$227.232542@nnrp1.uunet.ca>
Message-ID: <395AA355.D6041119@san.rr.com>

> Suppose I implement Module X that can unpickle anything that Python can
> pickle.  I believe it is reasonable to assert that Module X must therefore
> be either (a) Python, (b) a fairly Large Subset or Reinvention of Python.

Typically what is done is that look-aside metadata is used
to create/interpret the data. Pickling isn't the right thing
to send to C++ if you really want to have the objects once
they're demarshalled to be C++ objects. The better thing to
do is to marshal them into some cross-platform cross-language
marshalling format.




C/


From jeremyhonl at ispchannel.com  Thu Jun 22 18:11:20 2000
From: jeremyhonl at ispchannel.com (Jeremy Honl)
Date: Thu, 22 Jun 2000 22:11:20 GMT
Subject: Python website
References: <3951fe9c_2@nnrp1.news.uk.psi.net>
Message-ID: <caw45.26$zP.37956@news1.onlynews.com>

I noticed it was down yesturday,  all seems fine now.

"Simon Brunning" <sbrunning at trisystems.co.uk> wrote in message
news:3951fe9c_2 at nnrp1.news.uk.psi.net...
> Has <http://www.python.org> disappeared, or is it just me?
>
> --
> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
> The opinions expressed are mine, and are not necessarily those of my
> employer. All comments provided "as is" with no warranties of any
> kind whatsoever.
>
>
>
>




From hgg9140 at seanet.com  Thu Jun  8 00:14:08 2000
From: hgg9140 at seanet.com (Harry George)
Date: 07 Jun 2000 21:14:08 -0700
Subject: graph layout module?
Message-ID: <m3n1kwizcf.fsf@wilma.george.org>

I have a shop scheduling task, thus roughly rectangular layouts
with task bars and pred/succ arcs between tasks.  Any convenient tools
for this?

Probably need a simple gui for debugging, but final result is a 36
inch printed plot.  I can imagine, e.g., generating XML for dia to
visualize, then print to postscript.

Needs to operate in both UNIX (AIX, SGI, Linux) and WinNT
environments.


-- 
Harry George
hgg9140 at seanet.com


From robin at jessikat.fsnet.co.uk  Sat Jun  3 19:11:05 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 4 Jun 2000 00:11:05 +0100
Subject: PERFORMANCE: list versus a "dlist"
References: <39398DC0.C04E9784@san.rr.com>
Message-ID: <X9B0IUAJCZO5Ew98@jessikat.freeserve.co.uk>

In article <39398DC0.C04E9784 at san.rr.com>, Courageous
<jkraska1 at san.rr.com> writes
>
>Python lists, which are in actuality tail-preallocated dynamic
>arrays, are compared to the performance of a similar dynamic
>array using a head and tail preallocation strategy. This
>construct is referred to as a "dlist" -- a double-ended list.
>
>The tail preallocation strategy for the dlist involved
>keeping actual length and allocated length, and doubling the
>allocated space whenever an addition would involve appending
>past the allocated region.
>
>The head preallocation strategy involved keeping a "virtual
>index" into the array which is initially set to a position
>into the allocated region of memory at a position corresponding
>to 1/3 of the way through. Certain critical conditions are
>monitored which will trigger a reset to 1/3 of the way through
>via either a reallocate/copy or a simple move, depending on
>circumstance.
>
>The performance of this construct is compared to python's
>list as a FIFO (insert at head, pop at tail). Results are
>reported as follows:
>
>#insert/delete       DLIST            LIST
>
>100                  .006             .006
>1000                 .006             .018
>10000                .065             .365
>
>As before, on containers containing < 100 items or so, the
>use of this additional datatype made no measurable difference
>in performance of the tests. However, at list sizes at or near
>1000 elements, the DLIST implementation broke forth as a clear
>winner, where scaleability of the dlist imeplementation seemed
>to indicate at or near O(1)+k cost for insertion at the head.
>
>This is an early implemenation of this datstructure, and none
>of the typical sequence manipulation code is available for it
>as of yet.
>
>
>
>
>
>C/
so how about also doing some random tests (or assuming I'm perverse) the
worst case for your dlist ie insert and remove in the middle.
-- 
Robin Becker


From mwh21 at cam.ac.uk  Fri Jun 30 20:16:25 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 01 Jul 2000 01:16:25 +0100
Subject: Problem with execv
References: <8jg6j3$o84$1@nnrp1.deja.com> <blbt0jzv8h.fsf@bitdiddle.concentric.net>
Message-ID: <m3n1k27lie.fsf@atrus.jesus.cam.ac.uk>

Jeremy Hylton <jeremy at beopen.com> writes:

> The correct calling convention is to supply a tuple or list as the
> second argument to execv.  Example:
> 
> >>> os.execv('/bin/ls', ('.',))
> [lists contents of current directory]

ITYM

>>> os.execv('/bin/ls',('ls','.'))

or some such.  There's also

>>> os.execlp('ls','ls','.')

and many other variants...

Cheers,
M.

-- 
7. It is easier to write an incorrect program than understand a
   correct one.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From tom-main at REMOVEME.home.com  Tue Jun 27 18:01:06 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 27 Jun 2000 22:01:06 GMT
Subject: more python evangelising...
References: <200006271406.AAA06218@mbuna.arbhome.com.au> <D9465.71358$7o1.1704294@news2.rdc1.on.home.com> <8japr5$1qb$1@nntp9.atl.mindspring.net>
Message-ID: <Cu965.73375$7o1.1736562@news2.rdc1.on.home.com>

I didn't say that nobody has ever done it (created a large system in Python)
or that it was impossible to do, but rather that it was not what Python was
designed for.

Tom.

"Aahz Maruch" <aahz at netcom.com> wrote in message
news:8japr5$1qb$1 at nntp9.atl.mindspring.net...> In article
<D9465.71358$7o1.1704294 at news2.rdc1.on.home.com>,
> Tom <tom-main at REMOVEME.home.com> wrote:
> >
> >But you're falling into the old trap of thinking that one language is
best
> >for everything.
>
> I think you'll find that very few Pythonistas think that way.
>
> >Python is excellent for what it was designed for, but this does not
include
> >large scale app development.
>
> Really?  Then what's your reaction to all the people building
> large-scale applications with Python?
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> The best way to get information on Usenet is not to ask a question,
> but to post the wrong information.  --Aahz




From moshez at math.huji.ac.il  Fri Jun  9 13:59:48 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 20:59:48 +0300 (IDT)
Subject: Problem with script and typing
In-Reply-To: <394123ad.89325968@nntp.interaccess.com>
Message-ID: <Pine.GSO.4.10.10006092056410.28247-100000@sundial>

On Fri, 9 Jun 2000, Thaddeus L. Olczyk wrote:

> 
> Here goes ( extraneous pruned ):
> 
> project =[]
> 
> class Project:
>        def IsNull(self):
> 	return self.name==""
> 	project.append(self)
        ^^^^^ This line is never reached ???????

> def FindProject(x):
>     for project in projects:
>         if project.name == x:
>             return x
>         null=Project("")
>         projects.remove(null)
          ^^^^^ projects are compared by identity, unless you supply an
          __cmp__ method

> def is_project(s):
>     if s[len(s)-3:] == 'prj':
>         project=FindProject(s)
> ## error here
>         if project.IsNull():
>             project=Project(s)
>         project.Scan()
> ----------
> error message AttributeError: 'None' object has no attribute 'IsNull'

You snipped too much: how are we supposed to know what you put in the
"projects" global variable?

> 1) I've banged my head against this type of typing problem before.
>     How do I get around it?

Put correctly typed objects in containers

> 2) Is there a better way of handling null objects? ( Most languages I 
>     know have objects which are explicitly null )

None is an object which is explicitly null. What more do you need?
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From v2ng at my-deja.com  Fri Jun 16 16:12:34 2000
From: v2ng at my-deja.com (v2ng at my-deja.com)
Date: Fri, 16 Jun 2000 20:12:34 GMT
Subject: win32com and Excel
References: <3939D364.2EF22B13@zebrasoft.co.nz> <B_u_4.111323$E85.2192336@news1.rdc1.md.home.com>
Message-ID: <8ie1n1$8d0$1@nnrp1.deja.com>

I'm having a problem instantiating my Python COM objects from VB6 and
Excel VBA.

My Python class is really small, a straight rip off of the PPOW
O'Reilly book.

class MoneyMarketCOMServer:
    _public_methods_ = ['testfunc']
    _reg_progid_ = "PythonMMPL.Utilities"
    _reg_clsid_ = "{186D0DF0-43BA-11D4-B806-00C04F6CFC5C}"

    def testfunc(self):
        return 'Hello world!'

if __name__ == '__main__':
    import win32com.server.register
    win32com.server.register.UseCommandLine(MoneyMarketCOMServer)


No big deal.

When I connect to it from VB or VBA using:

dim objCOM
Set objCOM = CreateObject("PythonMMPL.Utilities")
MsgBox (objCOM.testfunc())
set objCOM = Nothing

I get crashes _EXACTLY_ every other time I try to run my VB program.

I don't seem to have any problems using other COM objects from VBA.

Has anyone seen this problem before?

Vic



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


From dworkin at ccs.neu.edu  Fri Jun  2 16:22:20 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 02 Jun 2000 16:22:20 -0400
Subject: 1.6: TMTOWTDI (was Re: String.join revisited (URGENT for 1.6))
In-Reply-To: aahz@netcom.com's message of "31 May 2000 14:24:52 GMT"
References: <28052000.5@sanctum.jae.ddns.org> <8h15qr$rje$1@news.udel.edu>
	<mtzop76b2z.fsf@astron.berkeley.edu>
	<39346FAC.F2797EA5@cosc.canterbury.ac.nz>
	<8h37bk$taq$1@slb6.atl.mindspring.net>
Message-ID: <vnd3dmvbzpf.fsf@camelot-new.ccs.neu.edu>

aahz at netcom.com (Aahz Maruch) writes:

> I think this is one of the better arguments against ''.join().  The
> number of people who think this is "natural" is relatively small, *and*
> it goes against the Python philosophy of having only one obvious way to
> accomplish a given task.

If votes counted, I'd second this.

Other arguments aside, having both string.join() and ''.join() seems
unpythonic in its TMTOWTDIness, and making string.join() go away is
clearly not a good idea, so... 

Making this method of the string object be a 'don't use this unless
you mean it' sort of thing (''.__join(), for instance) would be much
more palatable.

-Justin

 



From hzhu at knowledgetrack.com  Wed Jun 21 19:21:00 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 21 Jun 2000 23:21:00 GMT
Subject: Python vs Perl
References: <8ikai5$v3d$1@nnrp1.deja.com> <394ebe95.255648282@news.online.no> <slrn8krpbf.gc.andyl@azaal.dircon.co.uk>
Message-ID: <slrn8l2jkd.q2.hzhu@rocket.knowledgetrack.com>

On 19 Jun 2000 09:18:39 GMT, Andy Leighton <andyl at azaal.dircon.co.uk> wrote:
>>
>>* Perl is harder to read than Python. This makes debugging and re-use
>>of other peoples code harder.
>
>Of course if you only know Perl, I would think Perl would be the easiest
>to read.  Although it would be fair to say that Python is more transparent
>than Perl.


The first time I read Python was when the RedHat installer produced an
exception.  I found it easier to read even though I only knew Perl at that
time.  Thank RedHat for introducing me to Python.  :-)

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From hat at se-46.wpa.wtb.tue.nl  Fri Jun  9 04:39:31 2000
From: hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp)
Date: 9 Jun 2000 08:39:31 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <slrn8k1erf.a5o.hat@se-46.wpa.wtb.tue.nl>

On Fri, 09 Jun 2000 05:12:56 GMT, Steve Mullarkey <stevemul at ozemail.com.au> wrote:
>I have been a 'C' programmer for the past 15 years and a "C++"
>programmer for the past 3 years. I'm not a genius but would modestly (?)
>class myself as reasonably competent.

I learned C about 11 years ago, and switched to C++ 4-5 years ago.

>I have read in several places productivity claims of 5 to 10 times over
>'C' and "C++". I would like to ask for some feedback from "C++"
>programmers who have moved to Python as to whether these estimates are
>realistic.

Depends on what you define as productivity. With a nice definition, no doubt
you can get these rates. On the other hand, I have way too many advertisements
promising me incredible productivity increasements, so I tend not to believe
any of these claims.

>In "C++" I have :-
>
>1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
>Profilers, Programmers Editor including Class Browser, etc..
>
>2. I use the STL (Standard Template Library) which gives powerful string
>and container classes.
>
>3. I use incremental compiling and linking. This reduces wait times to
>very small amounts.
>
>4. I can also use background compiling and linking.
>
>5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
>etc..
>
>I don't want to start a flame war but, given the above, I just can't see
>where the productivity increase is generated. I am genuinely interested
>to ask "C++" programmers :-

The big question is, are you happy with the above stuff ?

For example, why do you need a memory leak tool ? Are you really interested in
spending your time hunting down memory leaks, is that your goal in life ?
I think not.
I'd rather use a computer to perform tracking of memory usage, and let it
handle memory management for me.
A lot of the tools have an origin from the design ideas of C/C++, which is fast
execution on small machines. The price that you pay is that you have to do a
lot of management around it yourself.

Python starts at the other end. With an easy to learn language, take as much
management away from the programmer as possible. CPU time is extremely cheap
now, much cheaper than a few hours developer-time (you can buy a new PC if you
save 14 days of work !).

>1. What productivity increase do you achieve ?

I am not sure, I don't measure it. My basic problem is that everything I do has
to be multi platform.
Python is a bigger help than C++ in that respect, basically because it provides
an abstraction layer just like Java, while at the same time it is more open,
conceptually cleaner, and freely available.

However, development tools is not the entire story.
The main gain imho is maintainability after the program is `finished' (notice
this is still the main cause of costs for programs). Python is an easy
language, but it scales well. That means that others can read your programs
quite easily without sacrificing (word doesn't look right) the ability to
tackle real-world problems.
Also, since it is interpreted rather than compiled, the source is always
nearby. If necessary, you can even do some bug-hunting in the
production-environment rather than your own development environment (which is
important for me since most of what I do is prototyping).

>2. How long did you use Python before you achieved increased
>productivity ?

Hmm, that's tricky. I'd say -6 months.
I started building Python extensions in C++, before I could program Python. 6
months later, I wrote my first program in Python. Due to the excellent
documentation and books, learning was a matter of hours. After that I only
needed to look up some details on syntax.

Obviously, I picked Python because I had a problem that was suited to use Python
for. If I would have to do the same thing in C++, it would have cost me days
instead of hours. In that respect I had a huge productivity increase in minutes.
As said before, I don't believe these numbers, and I don't understand why
management loves these numbers. Imho, it is not unlike the productivity
increase you experience when you switch from hammer to screw-driver when trying
to insert screws.



Albert
---
Look ma, windows without Windows !!


From cut_me_out at hotmail.com  Sat Jun  3 18:10:26 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Jun 2000 18:10:26 -0400
Subject: Automatic reload()?
References: <8hbnt1$62i$1@news2.isis.unc.edu> <m3r9aexydb.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <etdd7lyo1pp.fsf@w20-575-109.mit.edu>

> I have this in my ~/.pythonrc:
> 
> def rr(keepem=sys.modules.keys()):
>     for i in sys.modules.keys():
> 	  if type(sys.modules[i]) is types.ModuleType and i not in keepem:
> 	      print "reloading",i
> 	      reload(sys.modules[i])
> 

That could cause some confusing problems if you're unlucky about the
order the modules get loaded in.  E.g.:

A.py:
######################################################################
import B, C
reload (B)
reload (C)

t = B.B ()

B.py:
######################################################################
import C

class B (C.C):

    def __init__ (self):
        C.C.__init__ (self)
        
C.py:
######################################################################
class C:

    def __init__ (self):
        pass
    
% python A.py 
Traceback (innermost last):
  File "A.py", line 5, in ?
    t = B.B ()
  File "B.py", line 6, in __init__
    C.C.__init__ (self)
TypeError: unbound method must be called with class instance 1st argument

The problem is, the C.C which B becomes a subclass of is the version
from before the reload, but the C.C in C.C.__init__ refers to the
version following the reload.

Alex.


From aahz at netcom.com  Thu Jun  8 22:35:27 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jun 2000 02:35:27 GMT
Subject: wishlist Python3K: subclass ints
References: <Pine.GSO.4.10.10006090447300.19646-100000@sundial>
Message-ID: <8hpl5f$ila$1@slb2.atl.mindspring.net>

In article <Pine.GSO.4.10.10006090447300.19646-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>
>No, you're missing my point. At the Python level, there should be a single
>number type. Asking whether a number is integral is fundemantelly
>different then asking whether it is even, and yet nobody advocates having
>odd and even numbers seperate classes deriving from the same base class.
>A number is a number is a number -- and should be a single type. This is
>where "optimize behind the scenes" comes in: we should have a "number"
>type, s.t., isinstance(1, number) as wel as isinstance(1.5, numer), is
>true. However, current implementation details say that it
>would be convinient implementation-wise if type(1)!=type(1.5) -- which is
>fine, just say that type(number) is implementation dependant, and cannot
>be relied upon.

Oh, I see.  Hmmmm....  That makes a lot of sense, yes.  I'm still having
a little trouble wrapping my head around how this works with
user-defined numeric types, but I'll let the people who are into
metaclass programming deal with it.  ;-)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From jhe at webde-ag.de  Thu Jun 29 12:38:53 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Thu, 29 Jun 2000 18:38:53 +0200
Subject: traceback string
References: <8jfnn5$1fpo$1@news2atm.raleigh.ibm.com>
Message-ID: <8jfu04$f36$1@pollux.ip-plus.net>

"Arint?" <shouldbe at message.com> schrieb im Newsbeitrag
news:8jfnn5$1fpo$1 at news2atm.raleigh.ibm.com...
> How do you print the traceback string from an exception
> try:
>    whatever()
> except:
            import traceback, string
            msg = string.join(traceback.format_exception(
                    sys.exc_info()[0], sys.exc_info()[1],
sys.exc_info()[2]), "")
            print msg





From Software at TheMysticEye.com  Mon Jun 26 01:04:02 2000
From: Software at TheMysticEye.com (Corvus)
Date: Mon, 26 Jun 2000 00:04:02 -0500
Subject: Python for Palm ?
References: <8j56ir$7fr$1@ctb-nnrp2.saix.net> <8j5n9o$grt$1@netnews.upenn.edu>
Message-ID: <8j6nne$4bv$1@slb6.atl.mindspring.net>

It's available for WindowsCE and it works pretty good...

--
Russ
www.TheMysticEye.com
Software at TheMysticEye.com


"Andrew Diller" <dillera at isc.upenn.edu> wrote in message
news:8j5n9o$grt$1 at netnews.upenn.edu...
>
> "D De Villiers" <ddevilliers at lando.co.za> wrote in message
> news:8j56ir$7fr$1 at ctb-nnrp2.saix.net...
> > Hello, Everyone
> >
> > I just started programming my Palm device and is looking for a version of
> > Python for the Palm. Does anyone know please know if such a product exist
> > and where I can find it ?
> >
> > Lennie De Villiers
>
> someday..... it will be here:
>
> http://www.tismer.com/research/stackless/
>
>




From 12klat at sightreader.com  Mon Jun  5 09:22:55 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Mon, 05 Jun 2000 06:22:55 -0700
Subject: stripping out unwanted characters in string
References: <393b1f58$0$95987@news.execpc.com>
Message-ID: <393BA9AF.BCC7F769@sightreader.com>


"Jerry F. Davis" wrote:

> Howdy:
>
> I have a string that has some unwanted characters in it.
> Octal 240 for instance.

>>> import string
>>> string.replace("s\240p\240a\240m","\240","")
'spam'

> I want to actually strip them out of my string.

Can't exactly do that.  Strings are immutable.  If you don't know what
I mean by immutable, find out (mutability is a key concept of python).

(Hint: if strings were not immutable, they could not be safely used as
keys in dictionaries.)

So you'll have to settle for:

>>> s = "s\240p\240a\240m"
>>> s = string.replace(s, "\240", "")
>>> print s
spam

> I have looked at string.replace, but can't seem to figure out how to do
> this.
>
> Thanks in advance.
>
> Jerry
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000605/10d16de1/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000605/10d16de1/attachment.vcf>

From dbroadwell at mindspring.com  Sat Jun 24 20:41:52 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sat, 24 Jun 2000 20:41:52 -0400
Subject: need insight...
References: <3954671F.51E53AE6@mindspring.com> <Zh_45.85985$WS3.752637@typhoon.we.rr.com> <slrn8la4f4.769.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <39555550.3E9013F8@mindspring.com>

>     Actually I don't think code even needs to make a copy or have the
> cwd.remove(item) call since it deletes cwd right after the for loop.
> That should save a little space and speed up the code a bit.
It dosen't.. you are right read the 'now for the flames' post that i
made, i did much of that, and am currently working on a recursive
version of the directory scanning, but am afraid that i ampuching my
limits, so give it time..

-- 
        ~We're all fallen angels who've forgotten how to fly~
        Remember how: http://www.cyber-action.com/oed/dbroadwell.html
--


From eric at estinc.com  Tue Jun 27 19:27:17 2000
From: eric at estinc.com (Eric Lee Green)
Date: Tue, 27 Jun 2000 23:27:17 GMT
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com>
Message-ID: <39593888.E7091570@estinc.com>

Greg Gallagher wrote:
> 
> I'm a little disheartened here... I know how to do this in Perl, but I'm lost in Python and
> searching for 'complex data structures' is landing me nowhere.  Basically,  I want
> a dictionary whosevalues are lists of dictionaries.  Simple.
> I want to basically loop through some data and so something like this:
> 
> while loop:
>    new_forward = {'ip' : ip_address,
>                   'hostname' : hostname,
>                   'mac_address' : mac_address}
> 
>    forward[domain].append(new_forward)

You may want to simply this by using classes. For example, here's a class
"IP_Address":

class IP_Address:
    def __init__(self,ip,hostname,mac_address):
         self.ip=ip
         self.hostname=hostname
         self.mac_address=mac_address

Then:
      new_forward=IP_Address(ip_address,hostname,mac_address)

Then:

      masq_table['forward'].append(new_forward)

then later:

   for ip in masq_table['forward']:
       ...do something here...


-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From mfletch at tpresence.com  Wed Jun 14 04:08:14 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 14 Jun 2000 04:08:14 -0400
Subject: TCP packet size?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33BB@VRTMAIL>

Standard approach in these cases is to use one of a (large number of
schemes, with large numbers of variations on each one) for encoding your
information.  Some simple examples:

    Data Length Encoding: length, data, length, data... --> an encoded
length value, followed by x bytes (or words, or what have you) of data.
Other end reads length, then reads until it has that length of information
(appending the results of reads until the total is equal to the desired
length, storing any remainder for later use if they are under length).  This
is most popular in binary formats and UDP protocols, where byte offsets seem
more natural and messages are discrete.  It is more efficient in most cases
than other systems.  It is somewhat fragile, in that if you lose the length
byte you can wind up foobared (since you can never again determine where
message boundaries are again in that stream).  In case of such failure, you
need to re-establish the stream/connection (expensive).  See any number of
formats from the C/C++ worlds, see also Python's pickle format (if I'm
remembering correctly).

    Markup: escaped data, escape character (or sequence), escaped data,
...--> read data until you get escape character (or set of characters, or
what have you), then process the message.  If the escape character is not
the last byte you received, then wait for the next bit of data and append,
repeat until you get a message, storing any remainder for the next
iteration.  This is more popular in human-readable data, and is what
XML-based protocols use.  See also Mime encoding (multipart).

    Control channel encoding: secondary channel gives indices into the
second as boundaries for messages.  Multi-media control protocols often use
this approach (they don't want to touch the multimedia stream, just let the
client know where to start/stop reading, etceteras).

HTH,
Mike
    

-----Original Message-----
From: chris [mailto:chris at rpgarchive.com]
Sent: Thursday, June 15, 2000 6:14 AM
To: Mike Fletcher; Python List
Subject: Re: TCP packet size?


I'll try a more specific explanation.  My application is pretty much a
multi-user application similar to Net Meeting or a chat room.  One major
piece of data that the users are sharing is a xml document (in memory not a
file).  One application is the host who accepts TCP connections from any
number of clients.   If a user changes the xml document, those changes are
sent to the host and then resent to each client.  Again, all connections are
TCP.   A problem arises when I send a large message (3000-4000 bytes).
Despite the fact that I give a large buffer size to socket.read() method, I
often receive that message dissembled.  Also, sometimes I'll find small
messages combined in the same read() call.  The messages just aren't
reassembled the way I thought TCP worked.  I'm hoping to find a way to
received one message at a time in its entirety, or determine the start and
end of my messages.    I'm not really using xml as a network protocol, its
just the data structure of the application.  I'd be happy to show you my
networking module if you like (you probably just cringed :) ) 
One way I tried to solve my problem was to break my messages into smaller
(1000 byte) messages that included a header.  However, even though I broke
the message up, it was not received in the same format.  Some of the message
were combined and broken at unpredictable positions. 
FYI, this might be silly to add, but I'm not creating a new TCP connection
for every message. I'm using the same connection repeatedly. 
Thanks for the help.  I'm kind of at my wits end. 
Mike Fletcher wrote: 



From dana at mmi.oz.net  Thu Jun 15 11:09:52 2000
From: dana at mmi.oz.net (Dana Booth)
Date: 15 Jun 2000 15:09:52 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net>
Message-ID: <8iark0$gt6$1@216.39.141.36>

Peter Immarco <android at micron.net> wrote:

PI: However my C++ friends are badgering the crap out of me and they were
PI: already ribbing me about VB. :)

Well, you deserve it for VB. :)

However, you can just let the Python speak for you. People at my shop are
amazed at how fast they receive program requests, and now MySQL client
add-ons since I began using Python only a few months ago.


-- 
-----
Dana Booth <dana at mmi.oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From thomas at cintra.no  Wed Jun 14 04:24:13 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 14 Jun 2000 08:24:13 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <394b3e7d.284293552@news.online.no>

Hi,

I`ve been programming for the last 7-8 years or so, in Pascal, Delphi,
Java, Perl, itsy-bitsy pieces of C/C++ and for the last 1,5 year
Python.

Python is by far the easiest language of the above, even with no
experience whatsoever. I tell all people who want to start programming
to at least start with Python, learn the ropes then move to something
else if Python doesn`t do what they need. Most of them I`ve talked
with still use only Python. :->

I`ve tried to implement a project in all of the programming languages
above. The development time was reduced by a ten fold when I moved to
Python. Some if this might be experience from the prior
implementations, but most of it is cuz Python is easy, fun and it
actually gets the job done. It also reduced the amount of code I had
to write, which reduced the time I had to look for bugs, so I could
focus on development of features and cool, nifty things.

Needless to say, I love Python, in a programming-nerdy-kind of way,
but I`ve tried alot of the alternatives, and they don`t come anywhere
near the easy of use and productiveness of Python.

Try Python once more. 

Thomas

On Wed, 14 Jun 2000 07:51:31 GMT, "jin choung" <jin.choung at gte.net>
wrote:

>howdy all,
>
>i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
>tried to pick up python before moving on to c++.  however, i was trying to
>work my way through the book LEARNING PYTHON and while it says that there is
>no need to have prior experience in languages like c or c++, i really found
>that that was not true.
>
>anyone else feel this way?
>
>i'm currently working my way through a 24 hour c++ book and i think this is
>going to work for me.  oh, well, will be back to contribute here in a couple
>of months... in the meantime, i'll just leech valuable nuggets of wisdom
>from you guys.
>
>thanks
>
>jin
>



From david_ullrich at my-deja.com  Fri Jun 30 11:57:05 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Fri, 30 Jun 2000 15:57:05 GMT
Subject: command-line history in win
References: <395c304e.1224185@news.student.lu.se> <8jhm03$qnm$1@nnrp1.deja.com>
Message-ID: <8jig08$dk7$1@nnrp1.deja.com>

In article <8jhm03$qnm$1 at nnrp1.deja.com>,
  David Currie <davec999 at my-deja.com> wrote:
> In article <395c304e.1224185 at news.student.lu.se>,
>   invalid at dev.null (Alexander K) wrote:
> > aloha!
> >
> > i did a backsearch for this on deja. found a question but no
answers.
> > and the question is: isnt there a historyfunction in the python
> > command-line interpreter for win?.
> >
> > you know, the bash-kinda thing.
> >
> >   thanks in adv. / alex k
> >
>
> It works for me (Python 1.5.2)...
>
> Dave

     Keen. Maybe I'm the only one who doesn't know:
_What_ works for you, exactly?

DU

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


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


From pecora at anvil.nrl.navy.mil  Thu Jun 15 07:20:02 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Thu, 15 Jun 2000 11:20:02 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu> <20000614184508.B26436@xs4all.nl>
Message-ID: <150620000720023861%pecora@anvil.nrl.navy.mil>

In article <20000614184508.B26436 at xs4all.nl>, Thomas Wouters
<thomas at xs4all.net> wrote:

> The docs already have that:
> 
>    PyObject* PyList_GetItem (PyObject *list, int index)
> 
>    Return value: Borrowed reference.
> 
>           Returns the object at position pos in the list pointed to by p.
>           If pos is out of bounds, returns NULL and sets an IndexError
>           exception.
> 
>    PyObject* PyList_New (int len)
> 
>    Return value: New reference.
> 
>           Returns a new list of length len on success, or NULL on
>           failure.
> 
>    PyObject* PyList_GetSlice (PyObject *list, int low, int high)
> 
>    Return value: New reference.
> 
>           Returns a list of the objects in list containing the objects
>           between low and high. Returns NULL and sets an exception if
>           unsuccessful. Analogous to list[low:high].
> 
> And:
> 
>    int PyList_SetItem (PyObject *list, int index, PyObject *item)
>           Sets the item at index index in list to item. Note: This
>           function `steals'' a reference to item.

Thomas,  Nice list.  But dumb/newbie question, when do I INCREF and
DECREF the above?  I would guess I INCREF the borrowed ref that's
returned and DECREF any of the above (new or borrowed) at the end of
the C routine, unless those pointers are returned to Python caller to
be used there.  Right?  Wrong?  ...help.

You know, just adding INC and DEC rules to the above in the docs would
go a long way.


From bwarsaw at python.org  Fri Jun  2 12:09:34 2000
From: bwarsaw at python.org (Barry A. Warsaw)
Date: Fri, 2 Jun 2000 12:09:34 -0400 (EDT)
Subject: Status of List
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com>
	<3937B59C.3AFE081E@bam.com>
	<3937B7C2.A012D3B2@bigfoot.com>
Message-ID: <14647.56382.657467.145793@anthem.python.org>

>>>>> "ST" == Steve Tregidgo <smst at bigfoot.com> writes:

    ST> Same here.  I checked the eGroups mirror of the list, and they
    ST> list just 15 messages for June so far:

    ST>   http://www.egroups.com/list/python-list/

    ST> Perhaps the news->mail gateway is broken?  Mail->news seems
    ST> okay -- Remarq is listing this thread...

    ST> Never mind, it leaves me more time to do some work ;^)

It's possible that CNRI's newsfeed is broken.  That's hard for me to
check directly now, and my cron script isn't functional anymore.  I'll
put in a request to have that checked.

-Barry



From jeremy at beopen.com  Tue Jun 13 18:10:02 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 13 Jun 2000 18:10:02 -0400
Subject: tuple creation in C extensions
References: <l03102801b569016fbd23@[193.78.237.174]>
Message-ID: <1011EAC254FBE96B.1C7D23CE0AAFAF2E.34C32DAFEA9F1880@lp.airnews.net>

Just van Rossum <just at letterror.com> writes:

> At 11:15 PM +0200 10-06-2000, Peter Schneider-Kamp wrote:
> >t = PyTuple_New(2);
> >PyTuple_SET_ITEM(t,0,PyInt_FromLong(a));
> >PyTuple_SET_ITEM(t,1,PyInt_FromLong(b));
> >PyErr_SetObject(PyExc_IndexError,t);
> >
> >Is there "A Better Way(TM)"?
> 
> t = Py_BuildValue("(ii)", a, b);
> 
> Just

This is a shorter-to-write, slower-to-run version.  If you use
Py_BuildValue, the code has to parse the format string in order to
build the object.  If you are coding in C because you're trying to
speed up a speed-critical part of your program, the code you posted is
the better way.

Jeremy


From petersc at stud.ntnu.no  Tue Jun  6 15:55:28 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Tue, 06 Jun 2000 21:55:28 +0200
Subject: dictionary with complex keys
References: <393D54B6.EAF70234@magellan.umontreal.ca>
Message-ID: <393D5730.26594CE5@stud.ntnu.no>

Stefan Seefeld wrote:
> 
> I'm trying to create a dictionary with the keys being lists.
> How can I do that ? The docs say that only simple types are
> possible as keys. Is there a way to achieve what I want ? May
> be some special operator I need to define...

Your problem is that keys have to be immutable to be used
in dictionaries. You can use tuple(somelist) as the key
though in most cases because it is immutable.

do-changing-keys-make-sense-?-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From richard_chamberlain at ntlworld.com  Tue Jun 20 02:10:31 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 20 Jun 2000 07:10:31 +0100
Subject: serial I/O
References: <394F01DD.E0CF3EB@stud.ntnu.no>
Message-ID: <cMD35.4116$fw6.82744@news6-win.server.ntlworld.com>

Hi Peter,

Have a look at ftp.python.org/pub/python/contrib/sio-151.zip

Richard

Peter Schneider-Kamp <petersc at stud.ntnu.no> wrote in message
news:394F01DD.E0CF3EB at stud.ntnu.no...
> Sorry if I am missing the obvious, but what module
> would I use to make serial I/O?
>
> In this case I need 2400 Baud 8N1 serial in- and
> output on a Win98 platform (sorry, the USB camera
> only works there) but a platform independent solution
> would be even better (or at least partly independent).
>
> thanks
> Peter
> --
> Peter Schneider-Kamp          ++47-7388-7331
> Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
> N-7050 Trondheim              http://schneider-kamp.de
>




From Sttscitrans at tesco.net  Tue Jun 13 14:57:43 2000
From: Sttscitrans at tesco.net (Iain Davidson)
Date: Tue, 13 Jun 2000 19:57:43 +0100
Subject: Wondering about Domingo's Rational Mean book"
References: <pn5djsceirmrdhdd2ncht9c26hv0q5gkdq@4ax.com> <ralyh3rlx4xa@forum.mathforum.com> <8hp1g8$47o$1@barcode.tesco.net> <8i475e$48n$1@nnrp1.deja.com>
Message-ID: <8i5vsn$llr$1@epos.tesco.net>

> "Iain Davidson" <Sttscitrans at tesco.net> replied:
>
> >I don't see why you choose [3/2, 4/3]
> >as the two initial fractions and why you choose
> >to multiply top and bottom by 2.
> >What you seem to be doing is choosing two fractions
> >a/b and c/d such that ac= 2l and bd = 1.

> The set of two initial fractions used in a rational process
> for approximating the square root of any number P  is:
> [a/b, P*b/a]
> It is clear that  [3/2, 4/3] and consequently all the other new sets
> satisfy the well known condition for generating reduced fractions  |ad-
> bc| = 1.

But then you have "seeded" the process to "get the right answer".
In the above case, a^2 - Pb^2 = 1, which is just Pell's equation.
If P was 61 or 109 , say, then it would be hard to find suitable initial
values.
For most values of a and b you would not get best approximations.
So you have to state a rule that gives the initial conditions explicitly
or finding a and b would be equivalent to solving Pell's equation.

It just seems to be a variant on the method used in schools to find square
roots. You take a 1xn rectangle, then take the average of the sides and
divide n by the average giving two new sides that are "more equal"
This has nothing to do with finding best approximations

If x^2 +ax - b =0
then x(x+a) = b
If you assume x = b and x'+a =1,
then x = b, x' = 1-a, and by taking mean
you can make x and x' converge to same value.
x^2 -n =0 gives special case above.

The same idea can obviously be extended to
higher degree polynomials but conditions for
convergence may be complicated.





From digitig at cix.co.uk  Mon Jun 26 17:10:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Mon, 26 Jun 2000 22:10 +0100 (BST)
Subject: boolean true and false values.
References: <8j7n6h01tnl@news2.newsguy.com>
Message-ID: <memo.20000626221039.43263D@digitig.compulink.co.uk>

In article <8j7n6h01tnl at news2.newsguy.com>, alex at magenta.com (Alex 
Martelli) wrote:

> Tim Rowe <digitig at cix.co.uk> wrote in message
> news:memo.20000625230712.5257C at digitig.compulink.co.uk...
>     [snip]
> > > Yes, I know, but why is it not in the language ?
> > > Quite a lot of algorithms rely on having sets. If you put them in a
> > > list, there
> > > is an order which is not always wanted.
> >
> > There is always an order, but you don't always know what it is. When 
> > you
> > write your set class you just document that if the user iterates over 
> > the
> > elements then the order in which they will be returned is undefined --
> > that's what happens in the C++ STL. The fact that the implementation 
> > has a
> 
> Good general concept, un-good specific example.  C++'s std::set 
> (originally
> inspired from STL) is based on an ordering relationship, and when you
> iterate
> from .begin() to .end() you get the elements in increasing order 
> according
> to
> just that relationship.  The current release of STL also has hash-based
> containers (which didn't make it into the C++ standard library), but the
> standard ones are sorted...

Oh well, at least I had the right idea!


From wjdandreta at worldnet.att.net  Sun Jun 11 19:54:50 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 11 Jun 2000 23:54:50 GMT
Subject: Missing source files?
Message-ID: <eFV05.5409$iy.446615@bgtnsc06-news.ops.worldnet.att.net>

I am trying to build Python 1.5.2 for DOS and my linker finds 3 undefined
functions:

PyMember_Get
PyMember_Set
PyBuffer_FromObject

The only thing I was able to find was:

DL_IMPORT(PyObject *) PyMember_Get Py_PROTO((char *, struct memberlist *,
char *));

which for my setting should be

PyObject * PyMember_Get()

which is useless. I checked the entire set of source files for 1.5.2 and
those functions don't seem to be there. Does anyone know where they are? And
if some sources files are missing from the latest package?

Bill





From cut_me_out at hotmail.com  Tue Jun  6 23:44:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 23:44:59 -0400
Subject: Newbie questions on pdb
References: <8hju62$s3m$1@ncar.ucar.edu>
Message-ID: <etdsnuq9mtg.fsf@w20-575-109.mit.edu>

Type 's<return>' a couple of times to get in to the debugging session
proper. 

Alex.


From db3l at fitlinxx.com  Wed Jun 14 00:06:30 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 00:06:30 -0400
Subject: How do you FTP ASCII if ftplib strips CR?
References: <8i5jhd+h2lu@eGroups.com> <uog55mk6r.fsf@ctwd0143.fitlinxx.com> <8i6lr7$s2o$1@nnrp1.deja.com>
Message-ID: <u66rc52k9.fsf@ctwd0143.fitlinxx.com>

echuck at mindspring.com writes:

> By losing access to the file object returned by open(), it gets gc'ed
> and therefore closed and flushed immediately. You only have to worry
> about standalone open() calls if you use JPython or think you will
> someday.

What about some newer version of CPython or some alternative to either
CPython or JPython?  Although of course you're right that without the
reference an implementation is free to GC (and CPython will), it's not
guaranteed - in fact it's specifically permitted not to happen in the
language reference.

But I'll fess up - my mind was partially thinking about scoping when I
was writing (expecting the object to be destroyed when the script
ended) :-)

Even with immediate GC, I don't believe there's anything documented
that implies the destructor for a file object must close() the object,
but it's of course a reasonable assumption.  I still think its safer
all around to be explicit when handling files, and take responsibility
for closing them when done.  Besides, what happens if the close fails?
A previous co-worker had an interesting data-loss bug related to not
checking return codes on close() :-)

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


From moshez at math.huji.ac.il  Tue Jun  6 10:37:00 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 6 Jun 2000 17:37:00 +0300 (IDT)
Subject: author of htmllib/formatter?
In-Reply-To: <m2og5ec2q9.fsf@netmeg.net>
Message-ID: <Pine.GSO.4.10.10006061736170.26389-100000@sundial>

On Tue, 6 Jun 2000, Les Schaffer wrote:

> Does anyone know who are the authors of the htmllib and formatter
> modules in the standard python distribution?
> 
> i am having some trouble with htmllib.HTMLParser and am trying to
> track down the problem in the code to no avail.

If you'd mention the problems, people here would be able to help you (and
quite likely, use the answers given to enhance the documentation)
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From borcis at geneva-link.ch  Wed Jun 21 12:13:47 2000
From: borcis at geneva-link.ch (Boris Borcic)
Date: Wed, 21 Jun 2000 18:13:47 +0200
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net> <Pine.LNX.4.10.10006211154500.875-100000@dl2sql9.princeton.edu>
Message-ID: <3950E9BB.D54C2DDB@geneva-link.ch>

Jan Kristian Haugland wrote:
> 
> On Wed, 21 Jun 2000, Jeff Massung wrote:
> 
> > Please, this interests me -- I've had differention equations, but that's as
> > far as I go. In theory, pi (and e) are infinite, but has that been proven
> > that there is not end to their decimals?
> >
> > Jeff
> 
> Someone else already answered your question, but what
> do you mean by "in theory, pi and e are infinite"?

Come on, it can obviously be defended as a shorthand for :

"In theory, an exact decimal representation of pi and e
 would be of infinite length".

Together with :

"I am not familiar with other meanings of 'infinite'
 applied to numerical values"

A legitimate state of affair, wouldn't you say ?

BB


From phd at phd.russ.ru  Fri Jun 16 09:07:37 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 16 Jun 2000 13:07:37 +0000 (GMT)
Subject: Suggestion: "Completing" HTTP support in Python.
In-Reply-To: <394A2434.2B23D7D@darmstadt.gmd.de>
Message-ID: <Pine.LNX.4.21.0006161306290.15416-100000@fep132.fep.ru>

On Fri, 16 Jun 2000, Robb Shecter wrote:
> I knew I'd be doing this work for myself, and didn't know if anyone else felt the
> way I do about the library.  It looks like there -are- people who'd want these
> features too, which is cool.  I'll post something when I've got something to
> show.

   Seems they wanted these features not much - no patches were proposed :)

> But thanks anyhow for the motivation, Oleg! :)

   Don't mention it :)))

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From tim_one at email.msn.com  Mon Jun 26 19:00:36 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 26 Jun 2000 19:00:36 -0400
Subject: need insight...
In-Reply-To: <3954671F.51E53AE6@mindspring.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEAJGGAA.tim_one@email.msn.com>

[posted and mailed]

[David Broadwell]
> ... statement the  'for item in directorylist:' seems to be skipping
> over entries in it's list for no apparent reason... (note: what is
> skipps changes if the list it is given chages ...

>         ...
>         for thing in subdir:
>             ...
>             if ...:
>                ...
>             else:
>                ...
>                subdir.remove(thing)
> ...

You're modifying subdir *while* you're iterating over it.  You can read the
full story on why you shouldn't do that in the Language Reference Manual
(see the section about "for" loops).  Try

    for thing in subdir[:]:

instead.  That is, iterate over a *copy* of subdir.  Then modifications to
the original won't interfere with the iteration (which is over a copy).

you-can't-spend-money-and-save-it-at-the-same-time-either<wink>-ly y'rs
    - tim





From robinporter72 at yahoo.ca  Mon Jun 12 10:25:35 2000
From: robinporter72 at yahoo.ca (Robin Porter)
Date: Mon, 12 Jun 2000 10:25:35 -0400 (EDT)
Subject: Sorts
Message-ID: <20000612142535.957.qmail@web1505.mail.yahoo.com>

Does anyone know what type of sort is used for the internal sort function in Python (quick, merge, insertion, etc.)?


Mr. Robin Porter
St. Lawrence College
Computer Engineering

Database Coordinator - Raiser's Edge

Alumni Development

 



---------------------------------
Do You Yahoo!?
Get your free @yahoo.ca address at Yahoo! Mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000612/998e8bfe/attachment.html>

From gansevle at cs.utwente.nl  Wed Jun 21 15:59:57 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Wed, 21 Jun 2000 19:59:57 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <8ir6ri$mj9$1@nnrp1.deja.com>

If you are still interested in using enum (or the nearest Python
equivalent I can think of) after the discussion about dictionaries and
tuple unpacking, I've home-brewn a module caled 'Enum' which gives you
something you might use.
Here it is:

-----------------------------------------------------------------------
# Enum.py - Enumerated types in python
# Author: Fred Gansevles <gansevle at cs.utwente.nl>

__version__ = 0, 0, 1

""" Enumerations in Python

    From 'The C++ Programming Language by Bjarne Stroustrup':

        An enumeration is a distinct integral type with named constants.

        For example,

            enum color { red, yellow, green=20, blue };

        makes 'color' an integral type describing various colors.

        The possible values of an object op type 'color' are 'red',
        'yellow', 'green', 'blue'; these values can be converted to the
        'int' values 0, 1, 20 and 21.

    Since names only can spring-into-existance by assigning it a value,
    either as a variable or a keyword, the above syntax is not legal in
    Python.

    The "closest match" would be something like:

        color = enum('red', 'yellow', green=20, 'blue')

    but, since keyword parameters *must always come last* this isn't
    legal either.

    The "next closest match" would be something like:

        color = enum('red', 'yellow', 'green', 'blue', green=20)

    which would be the equivalent of the C++ code above.

    If 'enum' would be an instance of a class, you could not only
    reference the individual values (i.e: "color.blue") but jou can
    also loop-over all the values of color (i.e.: "for col in color:")

    This is what I attempt to make here.

"""
import string

class enum:
    __init = 1
    def __init__(self, *args, **kw):
        value = 0
        self.__names = args
        self.__dict = {}
        for name in args:
            if kw.has_key(name):
                value = kw[name]
            self.__dict[name] = value
            value = value + 1
        self.__init = 0

    def __getitem__(self, item):
        name = self.__names[item]
        return getattr(self, name)

    def __getattr__(self, name):
        return self.__dict[name]

    def __setattr__(self, name, value):
        if self.__init:
            self.__dict__[name] = value
        else:
            raise AttributeError, "enum is ReadOnly"

    def __call__(self, name_or_value):
        if type(name_or_value) == type(0):
            for name in self.__names:
                if getattr(self, name) == name_or_value:
                    return name
            else:
                raise TypeError, "no enum for %d" % name_or_value
        else:
            return getattr(self, name_or_value)

    def __repr__(self):
        result = ['<enum']
        for name in self.__names:
            result.append("%s=%d" % (name, getattr(self, name)))
        return string.join(result)+'>'

def test():
    color = enum('red', 'yellow', 'green', 'blue', green=20)
    print color
    print color.yellow
    print color(21)
    for col in color:
        print color(col), '=', col

if __name__ == '__main__':
    test()

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From haugland at math.princeton.edu  Wed Jun 21 12:50:38 2000
From: haugland at math.princeton.edu (Jan Kristian Haugland)
Date: Wed, 21 Jun 2000 12:50:38 -0400
Subject: Nth digit of PI
In-Reply-To: <3950E9BB.D54C2DDB@geneva-link.ch> 
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net> <Pine.LNX.4.10.10006211154500.875-100000@dl2sql9.princeton.edu> <3950E9BB.D54C2DDB@geneva-link.ch>
Message-ID: <Pine.SUN.3.91.1000621124639.28859A-100000@math.princeton.edu>

On Wed, 21 Jun 2000, Boris Borcic wrote:

> Jan Kristian Haugland wrote:
> > 
> > On Wed, 21 Jun 2000, Jeff Massung wrote:
> > 
> > > Please, this interests me -- I've had differention equations, but that's as
> > > far as I go. In theory, pi (and e) are infinite, but has that been proven
> > > that there is not end to their decimals?
> > >
> > > Jeff
> > 
> > Someone else already answered your question, but what
> > do you mean by "in theory, pi and e are infinite"?
> 
> Come on, it can obviously be defended as a shorthand for :
> 
> "In theory, an exact decimal representation of pi and e
>  would be of infinite length".
> 
> Together with :
> 
> "I am not familiar with other meanings of 'infinite'
>  applied to numerical values"
> 
> A legitimate state of affair, wouldn't you say ?
> 
> BB

Yes, it's natural to expect that the statement "pi is
infinite" means "a decimal representation of pi would
be of infinite length", but the problem is that that
already answers the question he was posing. Perhaps
"in theory" here means something strange.



From katz at Glue.umd.edu  Mon Jun 12 16:13:06 2000
From: katz at Glue.umd.edu (Roy Katz)
Date: Mon, 12 Jun 2000 16:13:06 -0400
Subject: Python Productivity over C++
In-Reply-To: <3941B79D.AAE9ECAF@san.rr.com>
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <3941B10B.8E6B2B41@Lugoj.Com> <3941B79D.AAE9ECAF@san.rr.com>
Message-ID: <Pine.GSO.4.21.0006121559430.13331-100000@y.glue.umd.edu>

I wanted to test this out, so I built a project in Python.  It is a small
CPU emulator, implementing twelve or so basic instructions (push,
pop, mov, add, interrupt-execute, etc...).  The time invested in its
Python build amounted to two hours.  Mostly of the time was spent writing
new code. 

Then I moved it to C++.  In a nutshell, this required re-writing
everything using STL.  This took me about twice as long to implement (not
counting nitpicking over the syntax errors).

Then I move to C.  The process took about twice as long as the C++
attempt; not counting the time it took for me to write miscellaneous
libraries (stack, memory allocation), I noticed that I was spending far
more time hunting pointer bugs and syntax errors than writing new
code. oThis was the longest port; it took me about four times longer than 
it did in Python. 

At the same time, C was faster than (straight) Python by a factor of
ten.  Whatever, I now have three implementations of a virtual CPU to play 
around with :)


Roey Katz



On Sat, 10 Jun 2000, Courageous wrote:

> 
> > I find it easier and quicker to lay out high level design and in
> > experimenting with finding the best class/data schema. And I can then run
> > it. The syntactical constructs of C++ tend to distract from the design
> > thought processes.
> 
> I strongly agree. In python, I get to think entirely about the problem,
> without worrying about the language/programming environment one bit,
> really. Furthermore, and this is a MAJOR win, python programs are *small*,
> which makes them easier to maintain, easier to read, and faster to code
> in general. Python is also very syntactically lightweight, which has to
> me my favorite thing about it.
> 
> Right now, I strictly use it as a rapid prototyping environment. It's
> great for testing out concepts.
> 
> 
> C/
> 



From tim_one at email.msn.com  Tue Jun  6 17:30:21 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 6 Jun 2000 17:30:21 -0400
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun  5)
In-Reply-To: <060620001708472558%pecora@anvil.nrl.navy.mil>
Message-ID: <002701bfcffe$6b5c09e0$48a2143f@tim>

>     It's finally public; Guido, Fred Drake, Jeremy Hylton, and Barry
>     Warsaw are now working for BeOpen.
>         http://www.pythonlabs.com/letter.html
>
>     BeOpen ran an interview with Guido:
>         http://www.beopen.com/features/interviews/rossum_index.html

[Louis M. Pecora]
> Is the above good, bad or neither?

Speaking as an unbiased observer (I'm unemployed <wink>), it's *very* good,
and for the reasons given in the referenced URL.  What are you worried
about?

leaving-mortgages-and-taxes-out-of-it-ly y'rs  - tim





From rjroy at takingcontrol.com  Tue Jun  6 22:40:16 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 07 Jun 2000 02:40:16 GMT
Subject: Example of Python code (C code to Python code)
References: <Ych%4.758$0l2.6362@news2.tor.primus.ca>
Message-ID: <393db42b.82240546@news1.on.sympatico.ca>

On Tue, 6 Jun 2000 21:08:45 -0400, "Robin Porter" <mogul at primus.ca>
wrote:

Here is one way, fairly straightforward translation.
Does no error checking on input so if you enter an invalid value at
the prompt will raise exception.
##################

TaxRate = 0.23
MinPay = 300.0
RegHours = 40.0

HoursWorked = float(raw_input('Please enter the hours worked per week
==> '))
PayRate = float(raw_input('Please enter the hourly rate of pay ==> '))

if HoursWorked > RegHours:
    GrossPay = (RegHours * PayRate) + ((HoursWorked-RegHours) *
PayRate * 1.5)
else:
    GrossPay = HoursWorked * PayRate

if GrossPay > MinPay:
    NetPay = GrossPay - ((GrossPay-MinPay)*TaxRate)
else:
    NetPay = GrossPay
    
print 'Gross Pay', GrossPay
print 'Net Pay', NetPay    

##############
#  Just for fun
# Here is another way, playing around a bit to minimise line count

TaxRate, MinPay, RegHours  = 0.23, 300.0, 40.0

HoursWorked = float(raw_input('Please enter the hours worked per week
==> '))
PayRate = float(raw_input('Please enter the hourly rate of pay ==> '))

print 'Gross Pay', (min(HoursWorked,RegHours) * PayRate) +
(max(0,(HoursWorked-RegHours)) * PayRate * 1.5)  
print 'Net Pay', GrossPay - (max(0,(GrossPay-MinPay)) * TaxRate) 
 
##########
Bob


>Could someone please give me an example of some Python code.  I am just new
>to this language and would appreciate if you could re-write this simple
>piece of C code in Python.
>
>Thank you,
>
>Robin
>
>
>#include "stdafx.h"
>#include <stdio.h>
>#include <math.h>
>
>float
>HoursWorked,PayRate,Overtime,IncomeTaxes,GrossPay,NetPay,TaxPayable,Overtime
>Hours;
>
>const float TaxRate=.23;
>const MinPay=300;
>const RegHours=40;
>
>main()
>{
>
>TaxPayable=0;
>Overtime=0;
>OvertimeHours=0;
>
> printf("\nPlease enter the hours worked per week? ");
> scanf("%f",&HoursWorked);
>
> printf("\nPlease enter the rate of pay? ");
> scanf("%f",&PayRate);
>
> if (HoursWorked>RegHours)
> {
>  OvertimeHours =HoursWorked-RegHours;
>  Overtime = (OvertimeHours)*(PayRate*1.5);
> }
>
> GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime;
>
> if (GrossPay>MinPay)
> {
>  TaxPayable = (GrossPay-MinPay)*TaxRate;
> }
>
> NetPay = (GrossPay-TaxPayable);
>
> printf("\nThe gross pay is %f",GrossPay);
> printf("\nThe net pay is %f",NetPay);
>
>
> return 0;
>}
>
>
>



From Eckhard at pfluegel.fsnet.co.uk  Fri Jun  9 10:47:30 2000
From: Eckhard at pfluegel.fsnet.co.uk (Eckhard Pfluegel)
Date: Fri, 09 Jun 2000 15:47:30 +0100
Subject: print without newline
Message-ID: <39410382.C83492D2@pfluegel.fsnet.co.uk>

I want to print a string without a newline afterwards. The command

print "foo",

doesn't print a newline but prints a trailing space, which is not what I
want (I wnat *nothing* after the string). I tried the formated output

print "%s" % ("foo")

but this prints a newline as well.

I use Python 1.5.2 on Linux.

Does anybody have an idea? Thanks very much in advance.

Eckhard






From hzhu at mars.localdomain  Tue Jun 13 03:47:21 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Tue, 13 Jun 2000 07:47:21 GMT
Subject: Formating complex numbers
Message-ID: <slrn8kbpuf.116.hzhu@mars.localdomain>

Is there a way to format complex numbers like

x = (1 + 2j)/3.
print "%8.3g" % x

instead of using the following?

def format(x): return "%8s" % ("% .3g% +.3gj" % (x.real, x.imag))
print format(x)

More generally, can % be overloaded?  I'd like to see the syntax changed so
that %8.3G or %.4F (with upper case symbols) allow optional complex
arguments. The format %s does not allow one to prescribe the precisions.  At
the moment the upper case %G %E are used to given uppercase E in the output,
which is sort of pointless.

I can hear people say, "but the C lib ...".  Well, the C lib doesn't like
("%s" % 3) or ("%s" % math.exp) either but python still got it.


Huaiyu


From mrfx at etang.com  Sun Jun 11 16:07:59 2000
From: mrfx at etang.com (mrfx)
Date: Mon, 12 Jun 2000 04:07:59 +0800
Subject: IDLE  can be rewrited using wxPython?
Message-ID: <8i25l0$1u8g$2@news.cz.js.cn>

The current IDLE uses TCL/TK as GUI widget.
IDLE  can be rewrited using wxPython as GUI widget?






From andres at corrada.com  Mon Jun  5 20:19:16 2000
From: andres at corrada.com (andres at corrada.com)
Date: Mon, 5 Jun 2000 20:19:16 -0400
Subject: A Simple question about a slice of a string.
In-Reply-To: <8hgefr$sok$1@news2.acs.oakland.edu>; from Jay R. Wren on Fri, Jun 02, 2000 at 05:12:29PM -0400
References: <8hgefr$sok$1@news2.acs.oakland.edu>
Message-ID: <20000605201916.A1522@corrada.com>

On Fri, Jun 02, 2000 at 05:12:29PM -0400, Jay R. Wren wrote:
> I'm attempting to accomplish grabbing the last 6 characters of sys.argv[0]
> so that I can compare it to another string.
> If 'myprog' is called as /usr/local/bin/myprog.  It could be called from any
> path, as long as the 'myprog' portion is there, certain things will
> execute.(this is used as a symlink)
> 
> myprogname = sys.argv[0][ len(sys.argv[0])-10:len(sys.argv[0])
> 
> it works like this, but I think there is a better, more proper way of doing
> it.  can someone let me know.  Thanks.
> 
> -Jay
> 

I don't understand exactly what you want but if what you want to extract
"myprog" from whatever path you are given you could use the module os.path
to do so as follows:

os.path.basename( '/usr/local/bin/myprog' )

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From piet at cs.uu.nl  Fri Jun  2 05:36:53 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 02 Jun 2000 11:36:53 +0200
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why -- or why  not? (was Re: Damnation!))
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <393326D3.400AA248@cosc.canterbury.ac.nz>
Message-ID: <u1z2gl8zu.fsf@cs.uu.nl>

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

GE> Currently, Python provides no way for you to directly specify
GE> which one you want -- instead, it uses a heuristic based on the
GE> run-time types of the arguments. This clearly violates the principle
GE> of not trying to guess what the programmer meant. Therefore, it
GE> is un-Pythonic. QED.

I want to add another reason for a/b to mean floating division:

As it is now in Python, you can have a==b and c==d both being true, but
a/c==b/d being false. Which would be a bad surprise for most people.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at freeler.nl


From root at rainerdeyke.com  Fri Jun  2 17:29:12 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 02 Jun 2000 21:29:12 GMT
Subject: overriding import
Message-ID: <IGVZ4.2412$r7.584924@news1.rdc2.tx.home.com>

I am writing a C++ program with embedded python.  I want to redirect import
commands to read the file from a different source, possibly not a true file
at all but an entry in a zip file.  How would I do this?


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From 12klat at sightreader.com  Tue Jun  6 23:40:07 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 20:40:07 -0700
Subject: strange list behavior
References: <393D4811.9989DE96@magellan.umontreal.ca>
Message-ID: <393DC417.C4EA40C2@sightreader.com>

You should include enough relevant code and example usage to show the problem, in
particular the code where something is being assigned the value 'bar' and 'baz' would be
helpful.  Otherwise there is no way to guess what might be going wrong.

Perhaps this may give you some insight though:

>>> a = [1,2,3]
>>> b = a
>>> b[2] = 42
>>> b
[1, 2, 42]
>>> a
[1, 2, 42]

Remember that assignment is by reference, not by value.

Stefan Seefeld wrote:

> I'v run into a strange problem with lists. I'm not sure
> at all where the problem is so I'll provide some context:
>
> I'v a couple of classes holding string lists (which are scoped
> names):
>
> class Module:
>      def __init__(self, name): self.__name = name
> #...
>
> class Declarator:
>      def __init__(self, name): self.__name = name
> #...
>
> I create them from another object, which traverses (visits)
> an AST:
>
> class Visitor:
>     def visitDeclarator(self, node):
>         name = self.scope()
>         name.append(node.identifier())
>         self.__result_declarator = AST.Declarator(name)
>     def visitModule(self, node):
>         name = self.scope()
>         name.append(node.identifier())
>         module = AST.Module(name)
> #...
>
> what I *though* was happening is that I created temporary variables 'name'
> in the respective scopes of all the visitSomething methods and then hand over
> the objects to the newly created Node types (Module, Declarator, etc.)
> However, what I *observe* is that even though a Module is initialized with
> name : ['foo', 'bar']
> it may end up being
> name : ['foo', 'baz']
> i.e. it seems the list is not properly released but instead modified from the outside.
> printing out 'id(name)' reveals indeed that the various temporary 'name' objects
> in the Visitor class have all the same id, i.e. they are the same object. Given
> that the AST nodes are created with the list copied by reference, not by value,
> I should not wonder that the value changes. Is this a fault of GC ?
>
> Could anybody explain what's going on here and how I get the behavior I want ?
> Thanks a lot !
>
> Stefan
>
> _______________________________________________________
>
> Stefan Seefeld
> Departement de Physique
> Universite de Montreal
> email: seefelds at magellan.umontreal.ca
>
> _______________________________________________________
>
>       ...ich hab' noch einen Koffer in Berlin...

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/3297aedc/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/3297aedc/attachment.vcf>

From thomas at xs4all.net  Thu Jun 15 09:15:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 15 Jun 2000 15:15:19 +0200
Subject: tuple creation in C extensions
In-Reply-To: <150620000715397988%pecora@anvil.nrl.navy.mil>; from pecora@anvil.nrl.navy.mil on Thu, Jun 15, 2000 at 11:15:38AM +0000
References: <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu> <20000614184508.B26436@xs4all.nl> <200006141705.NAA14660@w20-575-109.mit.edu> <150620000715397988%pecora@anvil.nrl.navy.mil>
Message-ID: <20000615151518.F26436@xs4all.nl>

On Thu, Jun 15, 2000 at 11:15:38AM +0000, Louis M. Pecora wrote:
> In article <200006141705.NAA14660 at w20-575-109.mit.edu>, Alexander S
> Coventry <alex_c at MIT.EDU> wrote:

> > > The docs already have that:
> > > 
> > >    PyObject* PyList_GetItem (PyObject *list, int index)
> > > 
> > >    Return value: Borrowed reference.
> 
> Oy, Vey!  Borrowed reference (if you know that) means you "manually"
> increase the refcount, right?  New ref. means you decrease the
> refcount, unless that item is returned to the Python calling routine?

Yes. Borrowed reference means you can use it for a little while, for
instance to pass it to a function that doesn't steal a reference, but if you
hang on to it for a while (executing code that might trigger changes in the
object you borrowed it from) you should INCREF it.

New reference means you get a new object, and if you 'throw it away', you
should DECREF it. But if you pass it on, make sure whoever you pass it to
knows it has to DECREF it (the parent function, usually, or one of the
'stealing' functions.)

> I'm sure there's more than this. :-)

Well, yes :) A couple of functions 'steal' references. They take a PyObject
as an argument, and they do not return a PyObject, and they do not INCREF
themselves -- they steal your reference. You should INCREF the object
(*before* passing it to such a function!) if you want to keep using it
yourself.

That's about it, though. The functions that don't have a comment about
references do not return a reference, and do not steal the references to
passed-in objects. (They may INCREF and DECREF them, but never take your
reference away.) If you write a container-object, something that should hold
on to another object, you have to INCREF that other object. If you remove
the object from the container, you should DECREF it, unless you plan on
passing it on. If you retrieve the object from the container, but do not
remove it from the container, you should INCREF it before passing it on.

Unless you define your functions to steal or return borrowed references
ofcourse ;) In which case I streneously suggest you document them properly,
in full view of yourself, before you shoot yourself in the foot with one of
your own functions. 

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From scarblac-spamtrap at pino.selwerd.nl  Thu Jun 15 04:21:33 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 15 Jun 2000 08:21:33 GMT
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <slrn8kh4ch.8pq.scarblac-spamtrap@flits104-37.flits.rug.nl>

root at sirius.ftsm.ukm.my wrote in comp.lang.python:
> 1. conway's game of life:
>    how to have a matrix in python? 
>    the following is not good;
>    
>            r0=[0,0,0]
> 	   r1=[0,0,0]
> 	   ........
> 	   
> 	   m=[r0,r1,....]
> 
>    the matrix is fixed, which is not good.

So just use a loop to do the same thing.

m = []
for i in range(8):
   m.append([0]*8)
   
Makes a 8*8 list of lists.

> my solution use dictionary (i know this is not the right way). 

It works, it's not very critical here, I'd say that's good enough too.

> 2. 8 non-attacking queen.
> 
>    use backtracking. the function call itself back. how to do recursive call
>    in python?

Well, just call the function.

>    roughly, the C program I study look like this;
>    
>    queen_count = -1
>    
>    main()
>    {
>      addqueen();
>    }
>    
>    addqueen()
>    {
>    queen_count++
>    if queen_count != 8
>         addqueen()
>    else
>         print result
>    }
> 	
>    how to pass a global variable (queen_count) to a function?

By default, if you assign to a variable inside a function, Python assumes it
is a local. If you never change it but only use it read-only, Python assumes
it's a global. If you want to use the global, use the 'global <varname>'
statement.

Your program would become something like:

queen_count = -1

def add_queen():
   global queen_count
   
   queen_count = queen_count+1
   if queen_count != 8:
      add_queen()
   else:
      print result # It would die here of course, 'result' is unknown

if __name__ == '__main__':
   add_queen()


(The last two lines is the common way to do a 'main()' sort of thing; it
only runs if the script is run as the main program; if some other module
imports it, it doesn't run).

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
 10:19am  up 100 days, 22:32,  7 users,  load average: 0.24, 0.12, 0.21


From jeffrey at digicool.com  Thu Jun 22 11:51:14 2000
From: jeffrey at digicool.com (Jeffrey P Shell)
Date: Thu, 22 Jun 2000 11:51:14 -0400
Subject: Extending Python with Obj-C on MacOSX(server)?
References: <eviltofu-D90BAE.22541415062000@news.earthlink.net>
Message-ID: <8itc6u$18f$1@bob.news.rcn.net>

> Has anyone any examples of how I extend Python with Apple's Cocoa layer?

Since Objective C can live alongside / inside C and C++, a long way is to
use the Python C API to write extension modules that use Objective C code
inside of them to send messages to the Cocoa framework.

Objective Everything from TipTop (www.tiptop.com) is/was a product that more
directly integrated Python/Perl/TCL into the YellowBox (now Cocoa)
framework.  I doubt it will run on MacOS X DP3 or 4 (or even any of the
earlier developer previews).  I haven't been able to execute any .app's on
my MacOS X Server partion when in DP4.  But Objective Everything has not
been updated for a long time anyways (it has Python 1.4).

A more interesting solution that I've been just lightly experimenting with
is JPython.  As long as Apple's Cocoa classes are in your class path (which
can be set by:

setenv CLASSPATH .:`javaconfig DefaultClasspath`

), you get access to all of Cocoa as described in the Java references for
the AppKit and Foundation frameworks, ie:

from com.apple.yellow import application      #AppKit
from com.apple.yellow import foundation      #FoundationKit

I've just done a *little* bit of poking around.  I've gotten some strange
memory leak errors when trying to play around with
foundation.NSGregorianDate, and I don't know if that's because the MacOS X
Java is still in development or I just haven't gone through and configured
JPython ++ Friends properly.

When the new audio subsystem for MacOS X is closer to completion (hopefully
in the upcoming beta), I plan to start developing some small tools using
Cocoa.  I haven't yet decided on whether I'm going to learn Objective C or
Java or try to extend the new ProjectBuilder (as it gets more polished) to
use JPython (ie - recognize Python files, use 'jpythonc' to compile into
Java classes from the ProjectBuilder UI).  I mainly want to be able to use
Interface Builder to define my interfaces and be not-too-many steps removed
from the core Audio API's.





From tiddlerdeja at my-deja.com  Mon Jun 26 15:51:18 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Mon, 26 Jun 2000 19:51:18 GMT
Subject: JPython Threading module fully free-threaded???
Message-ID: <8j8c7d$rib$1@nnrp1.deja.com>

As JPython is implemented in Java, does anyone know if JPython's
Threading module is fully free-threaded?

My Python programming for Win32 says Python's Threading is not as it
maintains an internal lock that prevents more that one thread being
inside the interpreter at any one time. I just wondered if JPython was
different.

Any information appreciated.


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


From matt at nightrealms.com  Thu Jun 22 06:53:38 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Thu, 22 Jun 2000 02:53:38 -0800
Subject: Dynamically add methods to classes?
Message-ID: <Sql45.1512$9W1.105781@news-west.usenetserver.com>

Is is possible to dynamically add methods to classes in Python?

Thanks in advance.





From claird at starbase.neosoft.com  Tue Jun 20 09:28:40 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 20 Jun 2000 08:28:40 -0500
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news and  links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net> <394E96D2.5787F2A3@gmx.de> <kuu2eox2b4.fsf@lasipalatsi.fi>
Message-ID: <BD2582D99FFCD3B7.985616BDFFE4F835.AD3FA9354F384A51@lp.airnews.net>

In article <kuu2eox2b4.fsf at lasipalatsi.fi>, Erno Kuusela  <erno at iki.fi> wrote:
			.
			.
			.
>i wrote a despamming proxy for deja that makes the
>non-classic mode almost tolerable, but it would probably
>be illegal to open it for public use...
			.
			.
			.
I collect these, at 
<URL:http://starbase.neosoft.com/~claird/comp.infosystems.search/dejanews.html>.
I understand your hesitation.  Sometime I'll tell stories about
the range of responses corporations exhibit when they encounter
simple proxies we automation zealots are wont to write.  In the
meantime, yes, I certainly respect your inclination to keep yours
private.
-- 

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


From shapr at uab.edu  Wed Jun 21 15:08:40 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 21 Jun 2000 19:08:40 GMT
Subject: Conditional Compilation for bytecode?
References: <8iquvb$gtd$1@nnrp1.deja.com>
Message-ID: <3951139C.C336FF95@uab.edu>

Don wrote:
> 
> I have some code that needs to run on Win32 and IRIX.
> 
> I thought I read somewhere that python had a sort of conditional
> compilation that would generate different .pyc files depending on the
> directives in the .py file.
> 
> i.e.
> #ifdef WIN32
> CreateProcess(...)
> #else UNIX
> fork()
> #endif
> 
> But I can't seem to find this anywhere.  Do I need to do this?

Nope, you don't need this, UNLESS your code uses calls that are platform
specific, then you have to deal with it yourself :) This means
os.system('ls') will most likely work on a *nix box, but almost never on
a win32 box.
Very little of Python changes from OS to OS. There are platform specific
modules like win32com and the irix GL stuff which obviously won't work
on another platform.

> Does the byte code compiler optimize this so it doesn't check every
> time?

Yup, it checks. If you haven't modified the python source since the last
time the python source was compiled, then python uses the bytecode
compiled file. bytecode compiled files are .pyc files.

The conditional bytecode compilation that I'm aware of has only to do
with optimization. python -O removes assert statements and if __debug__:
blocks, python -OO removes assert, __debug__, and docstrings. As I
remember, python compiles source to .pyo files when -O or -OO is used.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From hungjunglu at hotmail.com  Wed Jun  7 19:03:10 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Wed, 07 Jun 2000 16:03:10 PDT
Subject: Tkinter: any drag-and-drop features, yet?
Message-ID: <20000607230310.27755.qmail@hotmail.com>

Hi,

I have not touched Tkinter for a while, now.

Is there any drag-and-drop feature now in Tkinter or any other Python public 
graphical interface? Sometimes I think the Unix community is so reluctant in 
incorporating drag-and-drop that the cross-platform GUI system will never be 
as competitive as Visual Basic.

regards,

Hung Jung

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From aahz at netcom.com  Mon Jun 19 19:32:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jun 2000 23:32:00 GMT
Subject: Function that knows the names of its actual parameters
References: <394a6395.7217524@news.btx.dtag.de> <8ijgdl$oh8$1@slb6.atl.mindspring.net> <394e45c3.513888@news.btx.dtag.de>
Message-ID: <8imahg$9l5$1@slb6.atl.mindspring.net>

In article <394e45c3.513888 at news.btx.dtag.de>,
Stefan Franke <spamfranke at bigfoot.de> wrote:
>On 18 Jun 2000 21:53:57 GMT, aahz at netcom.com (Aahz Maruch) wrote:
>>
>>Why not simply write a function that takes a dict as its parameter?
>>
>>Maybe if you gave a better description of an occasion you think you need
>>this feature we could give you a better answer.
>
>Sure - see snippet below. Being able to call SQL_update without doubling
>the arguments would be nice. 
>If we could pass more than one of these dicts to SQL_update(), it would even 
>be possible to specify the WHERE clause with current parameters as well.
>I'm enclosing the magic arguments here into $...$ for provocative
>reasons <wink>:
>
>>>> index=1
>>>> member_name='you'
>>>> is_current=1
>>>> key=23
>>>> SQL_update(table, $member_index, member_name$, $key, is_current$)
>UPDATE my_table SET index=1, member_name='you' WHERE key=23 AND is_current=1

I'm assuming "$member_index" above should have been "$index".  I guess
what I'd do is something like this:

sqlDict = {
  'table': 'my_table',
  'updateClause': {
    'index': 1,
    'member_name': 'you'
    },
  'whereClause': {
    'is_current': 1,
    'key': 23
    }
  }
SQL_update(sqlDict)

IME, any time there's a direct mapping between variables and SQL, you
wouldn't want to do the magic trick anyway; any time you want to really
parameterize your SQL, you'd already have most of the info in a list or
dict in the first place (or *should*, if you wrote the code correctly).

--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From boud at rempt.xs4all.nl  Sun Jun 11 09:44:15 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 11 Jun 2000 13:44:15 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no> <8hrhqa$afm$1@news1.xs4all.nl> <G2g05.688$cQ.9254@news-server.bigpond.net.au>
Message-ID: <8i053f$pc8$1@news1.xs4all.nl>

Mark Hammond <mhammond at skippinet.com.au> wrote:
> "Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
> news:8hrhqa$afm$1 at news1.xs4all.nl...

>> The lookups aren't that great - they only find what you've already used
>> in that file. I'd like loopup to look in every file that's included for
>> completion. But I'm not going to complain - it's good for my memory not
>> to have it ;-).

> This is amazingly hard to do correctly - and the hard bits are not related
> to Pythonwin!

I know ;-). I tend to write very simple code, though, where there's
almost always a sequence of

class myClass:
 def xyz(self, a, b, c):
 ...

class mySecondClass:

  def __init__(self):
	  self.x=myClass()
		self.x.xyz('a','b','c')

For those cases it should be doable, but when I spread out and add a
second base-class runtime to the base class of the classes I'm going to
instantiate, well, I can't imagine a way to reliably and quickly find
all the possible lookups ;-).

-- 

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


From rhicks at nospam.rma.edu  Fri Jun  9 13:22:43 2000
From: rhicks at nospam.rma.edu (Robert Hicks)
Date: Fri, 09 Jun 2000 17:22:43 GMT
Subject: win32all-132 available
References: <rh705.459$cQ.7688@news-server.bigpond.net.au>
Message-ID: <DJ905.8$RB1.2837@news>

"Mark Hammond" <mhammond at skippinet.com.au> wrote in message
news:rh705.459$cQ.7688 at news-server.bigpond.net.au...
: Ive just made win32all-132 available.  Everyone who thinks it isnt
: terrible is welcome to upgrade :-)

ROFL (and I mean in a good sense as I didn't like that post)

Bob




From bparsia at email.unc.edu  Thu Jun 15 19:57:59 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 15 Jun 2000 19:57:59 -0400
Subject: python is hard to learn without c
References: <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net> <memo.20000615202832.52185B@digitig.compulink.co.uk>
Message-ID: <1eca95m.7c2pcf1kp9klfN%bparsia@email.unc.edu>

Tim Rowe <digitig at cix.co.uk> wrote:

> In article <9yS15.2877$Gp6.138923 at dfiatx1-snr1.gtei.net>, 
> jin.choung at gte.net (Jin) wrote:
> 
> > but i think the reason the C++ book is working so well for me is that it
> > assumes nothing about prior knowledge in an OO language because C++ is
> > the
> > base upon which everything else is built.  
> 
> Oh no it's not! C++ drew on earlier OO languages, and there are plenty
> (Scheme, Eiffel, Smalltalk) that are quite unrelated.

Indeed! Thank goodness!

Another point, there are many different things one might expect out of a
book entitled *Learning Python* (or *Learning Foo*); here are some of
the broader classes:

        1) Learning Python while/as a means to learning *programming in
                general* given *no prior experience with programming*.
        2) Learning Python while/as a means to learning *OOP in
                general* given *no prior experience with OOP*, but 
                some experience with *programming*.
        3) Learning Python as yet another oop language under your belt.

Books which do a good job of meeting an expectation which falls under
one of these classes might well do a terrible job meeting other types.

The most reasonable assumption you can make about why a book makes
assumptions about prior knowledge is that it intended to meet one (or
more) of these classes of expectation. (Or others!)

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From wbrian2 at uswest.net  Wed Jun 14 20:11:08 2000
From: wbrian2 at uswest.net (Brian Wisti)
Date: Wed, 14 Jun 2000 17:11:08 -0700
Subject: Regular Expressions and Threads?
References: <3947D2F7.264BDC6E@uswest.net> <8i8q6s$72t$1@nntp9.atl.mindspring.net>
Message-ID: <39481F1C.419364F5@uswest.net>

Aahz Maruch wrote:
>
> I doubt your problem has to do with threads; I bet that if you took your
> code and ran it unthreaded you would have the same problem.  From your
> description, it sounds like you have a list variable that you're
> appending to and not resetting after you do the string.join().  If
> that's not it, I don't have any better ideas without code.
> --

You may be correct... however, the only way I was able to fix the
problem was with an ugly hack:

	re._cache.clear()

Ugly because it's undocumented, and that means I should keep my grubby
little paws away from it. Ugly or not, once I put this line in,
everything has gone peachy.

I did follow your advice, though, and it cleaned up a few minor bugs in
the program.  Thanks for the reminder!

Later,
Brian Wisti

----------------------------------------------------------------------
| "All you need is ignorance and confidence; then success is sure."  |
|                                                    - Mark Twain    |
----------------------------------------------------------------------
| Brian Wisti   |  wbrian2 at uswest.net  | http://www.COOLNAMEHERE.com |
----------------------------------------------------------------------



From ian_maurer at my-deja.com  Fri Jun  9 14:30:32 2000
From: ian_maurer at my-deja.com (Ian Maurer)
Date: Fri, 09 Jun 2000 18:30:32 GMT
Subject: Probably a dumb question
References: <394106F9.A58F87CE@micron.net>
Message-ID: <8hrd49$b51$1@nnrp1.deja.com>

In article <394106F9.A58F87CE at micron.net>,
  Peter Immarco <android at micron.net> wrote:
> Hi,  still waiting for my two books on Python from Amazon.  I know
> Python is a full programming language so it can be used on web servers
> for server side processing, but what about browser side processing?
Can
> I also use it to create an 'applet' that runs in a compatible browser?
> I'm guessing not likely because I don't think IE4+ or Netscape
> Communicator support Python but I'm hoping I'm wrong.
>
> Thanks,
> Peter.
>
>

If the platform you are developing for is windows then Internet
Explorer 4+ will work since it supports Active Scripting. You can plug
in python as your script language (instead of Javascript or VBScript)
and it should work. I never tried it, but it's in Mark Hammond and Andy
Robinson's Python Programming on Win32, so it must be true (page
437). :)

The major hurdle for deployment is that you will need to download and
install Python AND the win32 extensions onto the client machine.

Good Luck,
Ian Maurer

Get PDOS for an enhanced command line experience:
http://hammer.prohosting.com/~pyguy




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


From shapr at uab.edu  Fri Jun 16 06:52:04 2000
From: shapr at uab.edu (Shae Erisson)
Date: Fri, 16 Jun 2000 10:52:04 GMT
Subject: ?Extreme Programming in Python?
References: <Pine.BSF.4.10.10006152355270.4382-100000@chi.pair.com>
Message-ID: <394A0732.6396DC4@uab.edu>

Michal Wallace wrote:

> I pretty much write all of our code, so I've had little experience
> with pair programming.. But I would LOVE to see an editor wired up
> over the net so that two people could work on code at the same
> time. (If I knew anything about GUI, I'd probably try hacking IDLE
> myself)

VirtualPairProgramming Editors:

Emacs can make another window/frame on a separate X display, you end up
with multi-editing.
XEmacs 21 has 'make frame on display' on the file menu. If some other X
client allows you access, you can use hostname:display (like coyote:0
for me) to make another XEmacs window. with Emacs, both people can type
at the same time in entirely different places and/or buffers.

screen allows more than one screen process to attach to a single
process:
in two different xterms type:
screen vi #this starts screen and vi inside it
screen -x #this attaches to the running vi

There are probably more, I'd like to hear about it if you know of one
not listed.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
 squirrels furnished by Paul E Black, the world is safe


From wjdandreta at worldnet.att.net  Thu Jun 15 22:29:33 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 16 Jun 2000 02:29:33 GMT
Subject: List of functions called in Python source files?
References: <3Xc25.3507$Uw3.210307@bgtnsc04-news.ops.worldnet.att.net> <008f01bfd724$5d5e0320$1906a8c0@fc.fenx.com>
Message-ID: <hig25.4023$C44.230012@bgtnsc05-news.ops.worldnet.att.net>

Hi Emile,

I don't understand your answer. What does 'extracting all imports' mean?

May be if I explain what I am trying to do you can be more explicit.

I am attempting to compile Python 1.5.2 to run under DOS without a DOS
extender.

I took the WAT_DOS make file that came with the 1.5.2 distribution and
extracted the names of the 82 source files that it contained and entered
them into a project in Borland's BC4.5 C/C++ compiler.

After some problems were resoIved, I was able to get them to compile and
link but the exe was too large to load (>640K as expected).

Step 2 is to attempt to use Overlays to get 1.5.2 to run under DOS. To get
that to work, I need to know the dependency of the source files on each
other. So far I have got a list of the functions defined in each source file
along with the code size and data size each sources file uses (I extracted
this info from Borland's Tlink .map file using a Python program.)

To complete the Overlay strategy, I need a list of all the files used in
each source file. If I had that list I could check it against my functions
defined list.

I was thinking about writing a Python program to search through each source
file and pick out the functions called. I know it's possible to do that
because the C compiler has to do it but I don't have a clue how to start.

Any suggestions would be appreciated.

Bill
Emile van Sebille wrote in message
<008f01bfd724$5d5e0320$1906a8c0 at fc.fenx.com>...
>You're not going to find anything that will
>give you a definitive answer.  If you don't
>use dynamic importing, eval's or exec's in
>your source base, you may be able to get
>something useful by extracting all imports,
>and looking at the functions that may be
>used, or at least get a list of the modules
>accessed.
>
>Or-not-ly y'rs
>
>Emile van Sebille
>emile at fenx.com
>-------------------
>
>
>----- Original Message -----
>From: William Dandreta <wjdandreta at worldnet.att.net>
>Newsgroups: comp.lang.python
>To: <python-list at python.org>
>Sent: Thursday, June 15, 2000 3:39 PM
>Subject: List of functions called in Python source files?
>
>
>> I need to get a list of the external (ones not in the same
>source file)
>> functions called in each Python source file. Does anyone
>know an easy way to
>> do this?
>>
>> Bill
>>
>>
>> --
>> http://www.python.org/mailman/listinfo/python-list
>>
>
>




From anders.eriksson at morateknikutveckling.se  Tue Jun 13 11:19:11 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 13 Jun 2000 17:19:11 +0200
Subject: Where should I put my HTMLgen directory?
References: <7BD10B680501D411B9DF009027E06F32011ACA@EXCHANGE>
Message-ID: <12jcks4hl8u7shg7o9r0a2nkekscllgnmi@4ax.com>

On Tue, 13 Jun 2000 15:05:08 +0200, Max M?ller Rasmussen
<maxm at normik.dk> wrote:

>I am trying to install HTMLGen on my machine.
>
>My python is installed in:
>C:\Program Files\Python\Lib\HTMLgen
>

I will assume that you mean: C:\Program Files\Python

>I have unpacked HTMLGen into:
>C:\Program Files\Python\Lib\HTMLgen
>
>
>So obviously I am doing something wrong. I thought that Python traversed the
>Lib diretory recursively to find any modules. But no?
>

This is a bit confusing! I'm not sure myself on how it really work. 

>Should I put all the *.py* files in the root of Lib? That seems a bit messy
>to me.

I think you have 3 alternatives:

1)	Add the subdirectory to the PYTHONPATH (which on Windows is in
the registry)

2)	Create a package (create an empty file called __init__.py in
the HTMLgen directory)

3)	create a HTMLgen.pth file in the lib directory. This file
should only contain the directory name HTMLgen


//Anders
Maybe-some-Guru-could-create-a-document-explaining-python-import



From aahz at netcom.com  Sun Jun 25 19:27:39 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jun 2000 23:27:39 GMT
Subject: Countdown
References: <3952952D.88E3588D@asme.org> <slrn8lc5qg.rqc.scarblac-spamtrap@flits104-37.flits.rug.nl> <Fwq8s2.A8z@world.std.com>
Message-ID: <8j64hb$kma$1@slb6.atl.mindspring.net>

In article <Fwq8s2.A8z at world.std.com>, Will Ware <wware at world.std.com> wrote:
>Remco Gerlich (scarblac-spamtrap at pino.selwerd.nl) wrote:
>> for i in range(len(thing)):
>>    print thing[-i]
>
>If thing = [1, 2, 3, 4, 5], then this piece of code will print
>1, 5, 4, 3, 2. The code below has probably the less surprising
>effect of printing 5, 4, 3, 2, 1.
>
>for i in range(len(thing)):
>    print thing[-i-1]

I'd make that

    print thing[-(i+1)]

for clarity.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From fwang2 at yahoo.com  Mon Jun 19 14:38:51 2000
From: fwang2 at yahoo.com (oliver)
Date: Mon, 19 Jun 2000 14:38:51 -0400
Subject: string parameter replacement question
Message-ID: <Pine.LNX.4.10.10006191432420.3161-100000@worf.mcnc.org>

hi, 

I have a function defined to accept one string parameter, inside this
funtion, I need to call another function to take this string as part of
the parameter, I am confused how to make the replacement happen. See
following code:

def fun1(str):
  fun2("input=str")

Noted that fun2 is from a moudle and it is expecting such a string. But
the replacement won't happen with above structure, for example, fun1("me")
won't produce a fun2("input=me"). Can anyone help me out here?

Thanks a lot




From emile at fenx.com  Mon Jun  5 14:32:07 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 5 Jun 2000 11:32:07 -0700
Subject: search for a string?
References: <8h6211$ark$1@news-int.gatech.edu> <LbBZ4.31403$WS3.364156@typhoon.we.rr.com> <8h6na4$ql7$1@nntp9.atl.mindspring.net> <33971223.8ec7c617@usw-ex0103-019.remarq.com>
Message-ID: <0b7701bfcf1c$6192ae60$1906a8c0@fc.fenx.com>

> I just tried to look up where I read a comparison between
the
> two methods, and I couldn't find it, so I could be
imagining
> things. Can somebody maybe confirm this?
>
> TD

I just compared 5 different ways in python of combining
strings
and it looks like '%s...%s' % tuple is comparable to adding
for small numbers (5-10) of smallish strings (10-250 byte),
and gets significantly better with more (25+) and larger
(1k+)
strings, where it stays comparable to string.join and the
newer "".join.

Emile van Sebille
emile at fenx.com
-------------------





From jajs at NOSPAMretemail.es  Sat Jun 24 23:46:17 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Sun, 25 Jun 2000 05:46:17 +0200
Subject: python email system help
References: <3953944C.1934AD89@tcgfinancial.com>
Message-ID: <slrn8lb049.1ru.jajs@Pascual.es>

The Commonwealth Group dijo sobre python email system help,
en fecha: Fri, 23 Jun 2000 12:46:04 -0400

|Hi, Im an 18 year old network kid, who knows nothing about programming.
|I accidently made a suggestion at a meeting and now I must implement
|that suggestion, and I know nothing about progamming.  I need to write
|this program by Monday.  I started learning python about a month ago,
|but I've been too busy to make real progress. I need to write a program
|that:
|1st, dial internet,(if busy, or someone else is using, dial again later)
|
|2nd, get email for several accounts and put on local file server

Take a look at this (for Linux):
http://www.clickeas.com/linux/animail.html. 

Then download  the latest .tar.gz file.  The web page  is in spanish  but the
source ,  the docs,  and the output  of the program  is in  english (the
output is i18n'ed, with english by default language).

Also note that, albeit fairly complete and functional, it is still beta
quality software.

Any questions to: jajs at retemail dot es
-- 
http://www.clickeas.com/linux


From mwh21 at cam.ac.uk  Tue Jun  6 12:53:49 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Jun 2000 17:53:49 +0100
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <m3wvk2wy1u.fsf@atrus.jesus.cam.ac.uk>

"Richard P. Muller" <rpm at wag.caltech.edu> writes:

> Does Programming in Python Marginalize You?

I hope not.
 
> I love Python. Of the 10-15 programming languages I know, Python is
> the only language which I actually look forward to using. 

Then you should learn Common Lisp, Haskell, ocaml, scheme and
Smalltalk (? I don't know smalltalk, but it *looks* like fun).

> I wake up in the morning, ask myself, "Will I be using Python
> today?" and if the answer is "yes" I know that it's going to be a
> good day.
> 
> And yet, I worry that using Python marginalizes one. Python is hard to
> beat for writing applications that run on a single computer. But
> that's yesterday's computing model. The future of computing, heck, the
> present of computing, is in writing programs that run through a web
> browser. And the fact of the matter is that 99% of web browsers
> surfing the internet run Java and JavaScript, which gives these
> languages a huge advantage over Python, an advantage that Python's
> superior design may not be able to overcome. I have Ada-programming
> friends who mumble about Ada's superiority over C++; that may be the
> case, but does it really matter anymore?

That something of an assumption; there's a market for desktop
computers and applications yet (I hope!).

> I'm aware of CGI scripting with Python. It's great. But JavaScript
> pages load faster than CGI pages, and are easier to program, to
> boot. Plus, JavaScript can directly call Java Applet methods, enabling
> it to do functions that are simply impossible with CGI scripts.
> 
> I'm also aware of the JPython module, which promises to enable
> us to write applets in Python instead of Java. But who among us has
> actually *used* JPython? 

You might ask this question on the JPython-interest list...

> Or, for that matter, who has come across a random web page on the
> net that uses JPython applets?

How would you tell?

[schnipp]
> It would have been nice if Netscape had implemented Python's syntax
> for LiveConnect rather than Java, but it didn't happen that
> way. Think how wonderful it would have been to write PythonScript
> functions in web pages rather than JavaScript.

It's a bit hard to tell, but I don't think javascript (silly name...)
is actually that bad a language.  It's just that to actually do
anything with it involves interacting with Netscape Navigator and
Internet Explorer in their multiple incompatible versions.  Write it
in Python, it would still be a pain in the arse, if you ask me.

Tangentially, does anybody know of any implementations of JavaScript
(not necessarily connected to a browser) that let you type interactive
commands into a listener and genereally muck about with the language?

> What else should we be doing to make Python *the* programming language
> rather than *a* programming language?

Wait?  I think Python is getting there ... it's far more "in the
mainstream" now than it was a year ago, and I think the trend is
continuing.  Use Python, be successful with Python - these are the
best adverts surely?  I don't think there's a real need for a
Java-style hype monster.

optimistically-ly y'rs
Michael

-- 
  Of the four project development variables - scope, cost, time and
  quality - quality isn't really a free variable.  The only possible
  values are "excellent" and "insanely excellent", depending on
  whether lives are at stake.              -- Kent Beck, XP Explained


From jkraska1 at san.rr.com  Sun Jun  4 13:32:12 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 17:32:12 GMT
Subject: Um? *boggle* Silly question about listobject.c
References: <000e01bfcddf$d2e09d40$1d2d153f@tim> <3939E84B.8AE86E66@san.rr.com> <m3ln0lxvot.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <393A9341.BAFEBD5A@san.rr.com>

> > After running through a quick bit-shifting test, I discovered
> > to my chagrine that I was thinking for some dumb reason that
> > int was 16 bits.
> 
> It can be, eg. on m68k Macs with appropriate compiler settings.
> Isn't, on any commodity hardware newer than that, afaik.

Well, part of the problem is that routinely over the years,
I use long when I want 32 bits, short when I need to conserve
space (16 bits), and int when I don't care. I'd never really
thought for long what an int was. Funny, considering I've
been doing this business for a decade.

This is yet another one of those arguments in favor of
expressions like int16 and int32 working their way into the
language itself, IMO. Albeit I'm familiar with the reasons
why the committees have rejected such notions in the past.



C/


From moshez at math.huji.ac.il  Sat Jun  3 01:56:23 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 3 Jun 2000 08:56:23 +0300 (IDT)
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why --
 or why  not? (was Re: Damnation!))
In-Reply-To: <u1z2gl8zu.fsf@cs.uu.nl>
Message-ID: <Pine.GSO.4.10.10006030854260.14749-100000@sundial>

On 2 Jun 2000 piet at cs.uu.nl wrote:

> I want to add another reason for a/b to mean floating division:
> 
> As it is now in Python, you can have a==b and c==d both being true, but
> a/c==b/d being false. Which would be a bad surprise for most people.

This is hardly a reason for floating point division. If at all, it should
be a reason for rational division, but note that the real solution is to
simply raise an exception when a float is compared with something

>>> 1.0 == 1
TypeError: illegal types for builtin operation

keeping-python's-error-messages-cryptic-ly y'rs, Z.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From mal at lemburg.com  Fri Jun  9 11:38:56 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 09 Jun 2000 17:38:56 +0200
Subject: Why is Python so slow?
References: <r7%%4.1754$iy.137191@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <39410F90.B7834B66@lemburg.com>

William Dandreta wrote:
> 
> I am just learning Python and I wrote a little program to modify a text file
> with 120,000 lines. It took 1hr and 30 mins. I wrote a C program that
> created the text file from a .dbf and it only took 2 mins.
> 
> Both programs ran on the same machine (486-133mhz) under Concurrent DOS (a
> multi-user mutitasking OS).
> 
> The Python version was 1.2.

Python latest stable version is 1.5.2 -- perhaps you should
upgrade ? Python 1.2 was released sometime around 1993, IIRC.

If you want to run Python under DOS with extender, you should
consider using PythonDX:

	http://www.cuci.nl/~hnowak/python/python.htm

For fast text processing, check out mxTextTools which should be
included in PythonDX.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From richard_chamberlain at ntlworld.com  Sat Jun 24 09:37:39 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 24 Jun 2000 14:37:39 +0100
Subject: Event Binding and Variable Passing
References: <8j0g88$htm$1@nnrp1.deja.com>
Message-ID: <IQ255.447$F01.9287@news6-win.server.ntlworld.com>

Hi Kapil,

from Tkinter import *
root=Tk()
b = Button(root, text = 'hi')
b.pack()
x = 5
b.bind('<Button-1>', lambda event: fun(event,x))

def fun(event, x):
    print x

root.mainloop()

You should look at lambda to pass the variables you want.

Actually the code you have written you needn't pass x at all because fun
would have access to it anyway - is this part of a class? if not you don't
need the self you used in your example, if it is you'll need to fiddle with
my example a bit.

Richard
<kstar18 at hotmail.com> wrote in message news:8j0g88$htm$1 at nnrp1.deja.com...
> hi,
>
> I'm relatively new to python and am experimenting with some Tk binding.
> I was wondering how I would pass a variable to a function that is called
> when a Tk event occurs.
>
> For example,
> b = Button(root, text = 'hi')
> b.pack()
> x = 5
> b.bind('<Button-1>', self.fun(x))
>
> def fun(self, event, x):
>    print x
>
> This doesn't seem to work, and the compiler gives me complaints about
> 'self' and the variable being passed.
>
> Any help would be greatly appreciated.
>
> Thanks!
>
> Kapil
>
> This method doesn't seem to work.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From bparsia at email.unc.edu  Thu Jun 22 17:20:10 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 22 Jun 2000 17:20:10 -0400
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL> <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net> <1ech8lh.19co0dkcvrvuoN%bparsia@email.unc.edu> <m3pupc5tzw.fsf@atrus.jesus.cam.ac.uk> <1ecjb1r.27wv4iz2rfkpN%bparsia@email.unc.edu> <m34s6l4flf.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <1ecn1bz.17wl5nd1s63kc3N%bparsia@email.unc.edu>

Michael Hudson <mwh21 at cam.ac.uk> wrote:

> bparsia at email.unc.edu (Bijan Parsia) writes:
> 
> > Michael Hudson <mwh21 at cam.ac.uk> wrote:
> [snip]
> > > Does Python 1.2 have the local variable optimisation? 
> > 
> > What do you mean?
> > 
> > [snip]
> 
> Certainly at some point there was no LOAD_FAST opcode; every name
> lookup would first look into a local dict, then the global one, then
> the builtins.  Now references to locals are just indexes into (C)
> vectors; this is much quicker.

Ah! Thanks! Am I correct in surmising from this that the gap between a
local access success and a miss (but hitting somewhere up the chain) is
even greater (than before LOAD_FAST)? Hmm.

I think I shall refrain from further speculation about method caches ;)

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From embed at NOSPAM.geocities.com  Thu Jun 15 09:58:06 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Thu, 15 Jun 2000 09:58:06 -0400
Subject: Can I build a single file containing all the contents of LIB?
Message-ID: <Vg525.3907$227.93950@nnrp1.uunet.ca>

I have an embedded system which has no support for long filenames. It's also
a pain to upload the whole LIB subdirectory to this system.  The embedded
system has a FAT filesystem and only supports 8 character filenames with a 3
character extension. The first problem then is with the "exceptions.py" and
from there on.    Since my system recompiles for both native NT/Win32 and
embedded mode, what I would ideally like to do is have a python script
"bundle up" the standard Python library into a file, and then just upload
that bundle. Then I would like it if python would search the \LIB directory
FIRST for xxx.py and if not found, then search in the compressed (or not)
file for it.   I would like the file to be separate from the executable, not
linked in, so I can change the python library without recompiling the
executable.

Suggestions please?

Warren




From thomas at xs4all.net  Sat Jun 10 06:08:11 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Jun 2000 12:08:11 +0200
Subject: PATCH: Augmented assignment
In-Reply-To: <8hs4d8$33b$1@animus.fel.iae.nl>; from cfelling@iae.nl on Sat, Jun 10, 2000 at 03:07:52AM +0200
References: <20000609163937.D20700@xs4all.nl> <8hs4d8$33b$1@animus.fel.iae.nl>
Message-ID: <20000610120811.H20700@xs4all.nl>

On Sat, Jun 10, 2000 at 03:07:52AM +0200, Carel Fellinger wrote:

> What about __add_is__? This sounds like addish, a kind of add; it resembles
> its use, plus followed by is; would prevent confusion of where to place the
> r like in __r_inplace_add__ versus __inplace_radd__; and still is concise:)

Well, that last bit is a non-issue, at least as far as I'm concerned,
because an augmented assignment doesn't need a 'right-hand' hook: The
new hook is called to change the left-hand-side object in-place, and you
shouldn't expect the right-hand-side object to know how to do that to the
lhs object. *that* would be confusing ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From dag at orion.no  Fri Jun 23 04:05:56 2000
From: dag at orion.no (Dag Sunde)
Date: Fri, 23 Jun 2000 08:05:56 GMT
Subject: running pyton on NT command lines
References: <8ithrb$cj1$1@nnrp1.deja.com>
Message-ID: <ETE45.14711$R7.2479539@juliett.dax.net>

>From NT's helpfile on the subject:

"If the command is not an executable (.EXE) file,
you must precede the command with cmd /c;
for example cmd /c dir > c:\test.out."

You might also try to run it with the /INTERACTIVE
switch to allow it to interact with the desktop of the
current user. That might at least give you a clue on
what's going on...

HTH...

Dag Sunde

<ghost_man at my-deja.com> wrote in message
news:8ithrb$cj1$1 at nnrp1.deja.com...
> I have written several scripts that I want to run at some set time.  I
> have been trying to get them to fire with the AT command line
> scheduler, but keep getting errors.  I have tried several different
> variations to get it to run even putting the python executable in a
dos
> batch file and trying to run the dos batch file.  I also have the
> python registered in my Windows 2000 environment path.
>
> my command line looks something like this
>
> at 16:30 "myscript.py"
>
> it fails everytime.  I have also tried putting in a complete path to
> the python.exe.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From moshez at math.huji.ac.il  Fri Jun  9 07:30:33 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 14:30:33 +0300 (IDT)
Subject: [Snippet] Case Correction in Python
Message-ID: <Pine.GSO.4.10.10006091424110.23947-100000@sundial>

Hi!

Those of you who were in my lecture last Friday will remember I showed a
cute-little case correction function, which used global variables. I have
reworked that example into a class which can be adapted to different file 
mistakes you may want to allow.

Here it is for your enjoyment:

==================== cut here =================
# PathCorrect.py
# Released to the public domain by Moshe Zadka, 9.6.2000

import dircache, os, string

class PathCorrect:

        def __init__(self):
                self._cache = {}

        def correct(self, file):
                if self._cache.has_key(file):
                        return self._cache[file]
                ret = self._correct(file)
                self._cache[file] = ret
                return ret

        def _correct(self, file):
                if os.path.exists(file):
                        return file
                dir, file = os.path.split(file)
                dir = self.correct(dir)
                if not file:
                        return dir
                files = dircache.listdir(dir)
                if file in files:
                        return os.path.join(dir, file)
                file = files[map(self.hash, files).index(self.hash(file))]
                return os.path.join(dir, file)

        def hash(self, file):
                raise NotImplementedError("use subclassable")

class CaseCorrect(PathCorrect):

        def hash(self, file):
                return string.lower(file)

class UnderScoreCorrect(PathCorrect):

        def hash(self, file):
                return string.replace(file, '_', '')

def _test():
        def randomly_case(s):
                import random
                functions = [string.lower, string.upper]
                l = []
                for c in s:
                        l.append(random.choice(functions)(c))
                return string.join(l, '')
        c = CaseCorrect()
        for file in map(randomly_case,
                        map(os.path.abspath, dircache.listdir("."))):
                assert os.path.exists(c.correct(file))

if __name__=='__main__':
        _test()

====================== cut here =====================

Those of you paying attention might have noted I also presented a
reworked randomly_case function in the test function, for your silly mood
days.

# Use like:
from PathCorrect import CaseCorrect
c = CaseCorrect()
for file in sys.argv[1:]:
	try:
		print file+':', c.correct(file)
	except ValueError:
		print file, 'does not exist, casing aside'

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From hanche at math.ntnu.no  Thu Jun 29 16:18:44 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 29 Jun 2000 15:18:44 -0500
Subject: Transcendentals and Encryption (was Re: Nth digit of PI)
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <394a5ed7.83928743@10.1.1.28> <8idoj0$lea$1@nntp.itservices.ubc.ca> <394a680a.86283939@10.1.1.28> <8iqj0f$kk5$1@flood.xnet.com> <oqem5q6erw.fsf@titan.progiciels-bpi.ca>
Message-ID: <pcovgys9r6j.fsf@math.ntnu.no>

+ Fran?ois Pinard <pinard at iro.umontreal.ca>:

| There is an aleph1 number of real numbers

That is a statement of the continuum hypothesis, which is known to be
independent of ZFC (the Zermelo-Fraenkel set theory including the
axiom of choice, on which we tend to imagine all of modern mathematics
is based).

| and an aleph0 number of non transcendental numbers (since we can
| enumerate all algebraic equations).  Aleph1 > aleph0, an aleph1
| number of transcendental numbers remain.  Right?

For aleph1 substitute c, and all this is true (where c is the
cardinality of the continuum: c > aleph0 as Cantor showed with his
diagonal argument).

belatedly-'cause-I've-been-vacationing-in-Arizona-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From kipster_t at my-deja.com  Wed Jun  7 23:58:15 2000
From: kipster_t at my-deja.com (kipster_t at my-deja.com)
Date: Thu, 08 Jun 2000 03:58:15 GMT
Subject: Score another one for the Python Job Board!
Message-ID: <8hn5ko$9b0$1@nnrp2.deja.com>

Fellow Pythoneers:

  I just commenced on the next phase of my career,
due in large part to the Python Job Board.
While the position that I am in was not the
specific job that was posted on the board, the
fact that there was an entry on board alerted me
to the use of Python in a company that I did not
suspect was using it.  Responding to the posting
resulted in my resume being passed to a person who
 had a fairly similar position to fill.

 I'm moving from a large telecom where I had to
fly under the radar of an ill-informed standards
group for the last five years whilst I was
delivering web application solutions implemented
almost exclusively in Python. I often wondered if
there ever would be a more than a trivial size
market for Python practitioners and whether
betting all my professional chips on Python would
ever pay off.  Well, landing my new position was
almost certainly due to my lengthy experience
using Python and the bonus is that I won't have to
be flying under the radar any more.

And based on what I've seen so far, using Python
will be crucial in being able to rapidly develop
flexible, scalable, and _*maintainable*_ systems
that massage/transform/analyze/report
on/distribute data that now is measured in 10's of
gigabytes per day to what is destined to become
near terabytes per day.

So, for all you Pythoneers out there who aren't
in the position to start your own business, or
join a start-up, and are wondering when the surge
in  Python usage by mainstream business will
arrive, my little data point isn't much, but it
does add to the count.

Massive thanks to Guido for such a clean language
that's fun to use and thanks to all who post and
respond on comp.lang.python.  The educational
value of this newsgroup is far and away better
than any of the others I've been lurking in the
last 10 years!

ObMangledQuote: "The journey of a thousand miles
begins with a single step".
(speaker name omitted as I'm not sure Senator
McCarthy isn't just hiding in J. Edgar Hoover's
closet)

--
Kip Lehman
kipster<iz at diz po>americanisp.net

Sorry, no .sig.  It's almost certainly prohibited
by that intellectual property agreement I signed
when my dilithium crystals had been exhausted.


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


From 12klat at sightreader.com  Tue Jun  6 22:55:46 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 19:55:46 -0700
Subject: Example of Python code (C code to Python code)
References: <Ych%4.758$0l2.6362@news2.tor.primus.ca>
Message-ID: <393DB9B2.9C3FE703@sightreader.com>

That's easy.  Just delete all variable declarations and puctuation.

TaxRate=.23
MinPay=300
RegHours=40

def main():
    TaxPayable=0
    Overtime=0
    OvertimeHours=0

    HoursWorked = input("Please enter the hours worked per week? ")
    PayRate = input("\nPlease enter the rate of pay? ")

    if HoursWorked>RegHours:
        OvertimeHours =HoursWorked-RegHours
        Overtime = (OvertimeHours)*(PayRate*1.5)

    GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime

    if (GrossPay>MinPay):
        TaxPayable = (GrossPay-MinPay)*TaxRate

    NetPay = (GrossPay-TaxPayable)

    print "\nThe gross pay is %f" % GrossPay
    print "The net pay is %f" % NetPay

main()


The gross pay is 85.000000
The net pay is 85.000000

Please make check payable to:

Ken Seehof
4965 Cree Way
Boise, ID  83709

:-)

Robin Porter wrote:

> Could someone please give me an example of some Python code.  I am just new
> to this language and would appreciate if you could re-write this simple
> piece of C code in Python.
>
> Thank you,
>
> Robin
>
> #include "stdafx.h"
> #include <stdio.h>
> #include <math.h>
>
> float
> HoursWorked,PayRate,Overtime,IncomeTaxes,GrossPay,NetPay,TaxPayable,Overtime
> Hours;
>
> const float TaxRate=.23;
> const MinPay=300;
> const RegHours=40;
>
> main()
> {
>
> TaxPayable=0;
> Overtime=0;
> OvertimeHours=0;
>
>  printf("\nPlease enter the hours worked per week? ");
>  scanf("%f",&HoursWorked);
>
>  printf("\nPlease enter the rate of pay? ");
>  scanf("%f",&PayRate);
>
>  if (HoursWorked>RegHours)
>  {
>   OvertimeHours =HoursWorked-RegHours;
>   Overtime = (OvertimeHours)*(PayRate*1.5);
>  }
>
>  GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime;
>
>  if (GrossPay>MinPay)
>  {
>   TaxPayable = (GrossPay-MinPay)*TaxRate;
>  }
>
>  NetPay = (GrossPay-TaxPayable);
>
>  printf("\nThe gross pay is %f",GrossPay);
>  printf("\nThe net pay is %f",NetPay);
>
>  return 0;
> }

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/621f6d48/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/621f6d48/attachment.vcf>

From mbel44 at dial.pipex.net  Fri Jun 23 08:32:42 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Fri, 23 Jun 2000 13:32:42 +0100
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <JEk45.262$Sn2.1571@news-server.bigpond.net.au> <slrn8l3rpi.8ku.r2d2@mao.acc.umu.se> <m0o45.6859$5f7.130092@news-east.usenetserver.com> <slrn8l4cmn.e2m.r2d2@mao.acc.umu.se>
Message-ID: <g4m6ls0orjo2djeevc3i6ia2vg089mul8l@4ax.com>

(posted and emailed)

r2d2 at mao.acc.umu.se (Niklas Frykholm) wrote:

>>Try my version of python15.dll at http://www.concentric.net/~jjlucsy/

Does that version include the winsound module?

That module causes python15.dll to link to winmm.dll. Some people with
unusual sound hardware have reported that loading this can take some
time.



Toby Dickenson
tdickenson at geminidataloggers.com


From scastell at sas.upenn.edu  Fri Jun 30 12:07:22 2000
From: scastell at sas.upenn.edu (Steven M. Castellotti)
Date: Fri, 30 Jun 2000 11:07:22 -0500
Subject: Problem with sys.exit()
Message-ID: <395CC5BA.AE64BB75@sas.upenn.edu>

Greets all,

	I'm having some difficulty with a particular python script. I'm taking
gzipped log files and calling zcat (from the command line) with one
script (feeder.py), which pipes the output to the second script
(parser.py). The second script addes up few things, and I want it to
report the final total back to the feeder script. I tried passing it via
sys.exit() (ugly I know, but it seemed like a good idea at the time) but
when I use the numbers in the feeder script, the numbers are way off:

End of parser.py:

print total_registrations
sys.exit(total_registrations)

Relevant portion of feeder.py:

...

grand_total = grand_total + \
  os.system('zcat %s | python authlog_parser.py' % (filename) )
print "%i" % os.system('zcat %s | python authlog_parser.py' % (filename)
)
print "Grand Total: %s" % grand_total


But my output looks something like this:

3
768
Grand Total: 1024

	...where the "3" is the actual total getting passed into sys.exit(),
768 is what parser.py is recieving, and 1024 being the total.

	Any ideas about what might be wrong, or perhaps a better way to do
this? If there was a way to read a gzip'd text file directly into python
that'd probably take care of it, but why isn't this working?

-- 
Steve Castellotti
Systems Programmer
School of Arts and Sciences, University of Pennsylvania


From eric at estinc.com  Tue Jun 27 19:22:33 2000
From: eric at estinc.com (Eric Lee Green)
Date: Tue, 27 Jun 2000 23:22:33 GMT
Subject: more python evangelising...
References: <B57E791A.264E%eableson@mac.com> <Uj865.1905$iN5.408850@ptah.visi.com> <Rn965.73297$7o1.1735998@news2.rdc1.on.home.com>
Message-ID: <3959376D.DA8DC41F@estinc.com>

Tom wrote:
> "Grant Edwards" <ge at nowhere.none> wrote in message
> news:Uj865.1905$iN5.408850 at ptah.visi.com...
> ...
> > Since MS is a publicly held company, the people who run it
> > have a fiduciary responsibility to the stockholders.  They are
> > obligated to make them as much money as they can (legally, that
> > is).  If they didn't, they could be held liable in court.
> 
> This is an important point.  This is the heart of capitalism.

Gosh, I thought the heart of capitalism was making a better product at a
better price. Maybe I ought to re-think that.

Let's see:

Make a shoddy product and market the hell out of it while charging double the
price. That's the heart of capitalism, yeah.

Hey, it works for Microsoft. 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From cmedcoff at pyrasol.com  Thu Jun 29 13:11:28 2000
From: cmedcoff at pyrasol.com (Charles Medcoff)
Date: Thu, 29 Jun 2000 17:11:28 GMT
Subject: Python COM Server Type Library
Message-ID: <8jfvvc$iql$1@nnrp1.deja.com>

I think I already know the answer to this but I'll ask it anyway...

Is there anyway to generate a type library for a COM server object
written in Python?  I know this has implications as far as what type of
binding is used.  That being the case I suspect the answer is no.

The reason that I ask is that this capability may make it easier to
utilize Python COM objects from C++.  This is especially the case in
Visual C++ where one can use #import.

Any other sources of Python COM information would be helpful.

--
Charles Medcoff, MCP
Senior Systems Engineer
Pyramid Solutions
1850 Research Drive, Suite 300


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


From parkw at better.net  Sat Jun 10 20:34:42 2000
From: parkw at better.net (William Park)
Date: Sat, 10 Jun 2000 20:34:42 -0400
Subject: MySQL module
In-Reply-To: <8hult4$t1a$0@216.39.141.21>; from dana@oz.net on Sun, Jun 11, 2000 at 12:18:44AM +0000
References: <8hult4$t1a$0@216.39.141.21>
Message-ID: <20000610203442.A497@better.net>

On Sun, Jun 11, 2000 at 12:18:44AM +0000, dana_booth wrote:
> I've asked this before, but never did get a good response, so I hope that
> one of you reading this might be able to help...
> 
> The MySQLmodule1.4 does not compile properly on any platform. I've tried
> removing the '&'s on the three lines which puke during compile. On
> UNIX-like's Python won't make install when having done that, on Windows,
> it'll compile, but Python chokes when trying to import.
> 
> I've even tried using mSQL and PostgreSQL, on which the modules compile
> fine, but I need to use MySQL, the only backend whos Python module seems to
> be flawed in some way. :)
> 
> Anyway do any of you know of, or have made yourself, a module for MySQL?
> 
> -- 
> 
> -----
> Dana Booth <dana at oz.net>
> Tacoma, Wa., USA

There are 2 modules for MySQL:
    - MySQLmodule-1.4 (old -- must remove 3 '&')
    - MySQLdb-0.2.1 (new -- no problem)
Both compiled and installed on Slackware Linux.

--William



From aahz at netcom.com  Sat Jun 17 18:45:24 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jun 2000 22:45:24 GMT
Subject: string.join question (was Re: urllib question)
References: <Pine.LNX.4.10.10006171424430.19831-100000@mail.fizzylab.com>
Message-ID: <8igv24$5li$1@slb7.atl.mindspring.net>

In article <Pine.LNX.4.10.10006171424430.19831-100000 at mail.fizzylab.com>,
paul beard  <paul at fizzylab.com> wrote:
>
>The length isn't the issue. I'm not able to do the string.join for some
>reason. I need to feed many URLs from a common host. I have the URLs (from
>an access log) and want to just append the URL to the host. 

What happens when you do a string.join()?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From hohage at muenster.de  Tue Jun 27 03:22:31 2000
From: hohage at muenster.de (hohage at muenster.de)
Date: Tue, 27 Jun 2000 07:22:31 -0000
Subject: e-mail problem
Message-ID: <8j9knn+dmcu@eGroups.com>

Hello,
I get the python-list by digest.
My problem is that I'd like to have a program which sorts the e-mails
by thread or date etc.
Is there something like this or do I have to write it by myself?
            Thanks




From rupole at compaq.net  Thu Jun 15 18:47:31 2000
From: rupole at compaq.net (Roger Upole)
Date: Thu, 15 Jun 2000 22:47:31 GMT
Subject: Mapping a network drive
References: <2B1262E83448D211AE4B00A0C9D61B03013011E2@msgeuro1.creo.be>
Message-ID: <62d25.9656$18.905628@pouncer.easynews.com>

You should be able to use the COM interface for Windows Scripting host.
import win32com.client
wnt=win32com.client.Dispatch('Wscript.Network')
wnt.MapNetworkDrive('z:','//someserver/someshare')
To test it, just try to access the drive:
os.listdir('z:') or something similar.
       HTH
              Roger Upole

"Pieter Claerhout" <PClaerhout at CREO.BE> wrote in message
news:2B1262E83448D211AE4B00A0C9D61B03013011E2 at msgeuro1.creo.be...
> Hello to all,
>
> anyone an idea how I can map a drive using pythonwin, and see
> if it was successfull or not? I'm running python 1.5.2 on WinNT 4.
>
> Pieter
>




From lsmithso at hare.demon.co.uk  Tue Jun 20 18:26:41 2000
From: lsmithso at hare.demon.co.uk (Les Smithson)
Date: 20 Jun 2000 23:26:41 +0100
Subject: ANN: pymqi - A Python wrapper for IBM MQ
Message-ID: <m3n1kgvvj2.fsf@hare.demon.co.uk>

I would have posted this to comp.lang.python.announce, but I can't get
that newsgroup. This is a copy of what's on Parnassus.

Abstract: A Python wrapper for IBM's MQ middleware.

Details: Pymqi is a wrapper for IBM's Messaging & Queueing middleware,
MQ. It lets you connect to MQ Queues and put/get messages from a
Python script.

URL: http://www.hare.demon.co.uk/pymqi/pymqi.html

Download: http://www.hare.demon.co.uk/pymqi/pymqi.tgz

Requires: IBM MQ

Binaries: Linux

--
Les Smithson.




From eq3pvl at eq.uc.pt  Fri Jun 23 09:16:29 2000
From: eq3pvl at eq.uc.pt (Pedro Vale Lima)
Date: Fri, 23 Jun 2000 14:16:29 +0100
Subject: zope, python vs apache, perl
References: <8ish5s$kd4$1@nnrp1.deja.com>
Message-ID: <3953632D.C2ED78C1@eq.uc.pt>

Hello,

> I want to evaluate zope, python vs apache, perl.
>
> which one can be used where!
> strong points and pitfalls.

Well that's mostly a philosophical question. I'm not very
good in philosophy. What I can tell you is that you can use:

Apache as a web server, zope as an application server,
python and perl as development languages all together at
the same time.

So it's your lucky day you don't have to make a choice.

(I'm not sure if perl methods is already working with zope,
but DC is working in that direction).

regards,

pedro



From lsmithso at hare.demon.co.uk  Tue Jun 20 06:20:46 2000
From: lsmithso at hare.demon.co.uk (Les Smithson)
Date: 20 Jun 2000 11:20:46 +0100
Subject: Python in home automation
References: <fsteele-E5789E.13533715062000@news.inet-systems.net> <cBd25.38645$cH1.355549@typhoon.austin.rr.com>
Message-ID: <m3r99swt4x.fsf@hare.demon.co.uk>

I have some Python scripts that control a TWSER & TW523. This includes
low-level drivers, and a higher-level server that lets you control X10
over a network. I also wrote a simple TkInter GUI for controlling my
X10 devices.

A while ago I spoke to someone who hacked my code to work with a
CM12U. We planned to rework this code to properly support both TW523
and CM12U, but nothing ever came of it.

Is it time to resurrect this?

>>>>> "news-server" == news-server austin rr com <bgrigsds at khsdfklsf.com> writes:

    news-server> Sounds like an opportunity to me.  A good start would
    news-server> be to develop some low -level classes to control the
    news-server> more common devices.  That would simple enough... A
    news-server> MisterHouse type project would require substantial
    news-server> effort though.

    news-server> "Frank Steele" <fsteele at mindspring.com> wrote in
    news-server> message
    news-server> news:fsteele-E5789E.13533715062000 at news.inet-systems.net...
    >> I'm beginning to think about extending the home network to
    >> include some simple home automation/X10 functions. I've started
    >> looking around at products and software, and even went so far
    >> as to pick up the latest copy of <A
    >> HREF="http://www.tpj.com">The P*&l Journal</A>, which has a
    >> lengthy story on <A
    >> HREF="http://www.misterhouse.net">MisterHouse</A>.
    >> 
    >> MisterHouse is a cross-platform system, all in P*&l, that
    >> allows voice and web control of X10 components, and uses
    >> text-to-speech to communicate alerts to the home occupants.
    >> 
    >> Here's the problem/question -- if it's my system, I don't want
    >> it in P*&l. I would prefer to stick to the Python
    >> world. Unfortunately, in a quick sweep through the Vaults of
    >> Parnassus, The Daily-URL, Python.FAQts, and the O'Reilly Python
    >> dev-center, I don't see any modules, extensions, or projects on
    >> this.
    >> 
    >> So, is anyone aware of Python modules or extensions to do X10
    >> home automation? I'm interested hearing about them on _any_
    >> platform, but I intend to implement on Linux and MacOS....
    >> 
    >> Thanks, Frank fsteele at mindspring.com



-- 
===========================================================================
      Les Smithson, Open Network Solutions Ltd, London, England
       lsmithso at hare.demon.co.uk   http://www.hare.demon.co.uk
PGP: 2B83E8A5/21 89 33 D0 53 45 59 B9  A6 4E 0A A5 62 EF CB FB


From alan.gauld at gssec.bt.co.uk  Mon Jun 19 08:11:05 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 19 Jun 2000 13:11:05 +0100
Subject: 0041 ||| Help requested from newbie |||
References: <nkfrkss0a2ar68n165sok07pau6gegun71@4ax.com>
Message-ID: <394E0DD9.16B8CE23@gssec.bt.co.uk>

> I am following Alan Gauld's Python tutorial
> (http://members.xoom.com/alan_gauld/tutor/tutindex.htm) 

latest version (including the Spanish translation) is at:

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

The xoom one is not quite as regularly updated as the 
crosswinds one - and the latter is usually faster to access
too...

> ...and I am stuck
> in the midst of his "case study" grammar module. I get a consistent
> "loop over non-sequence" error (see below) that I can't get rid of.
> I've tried everything 

Well you could have tried mailing me via the link at 
the bottom of the tutorial page... :-)

The problem lies in the line:

groups = getCharGroups(infile)

Unfortunately a gremlin swallowed the line in getCharGroups 
that returns a value so the line above overwrites the groups 
list with garbage('None' in this case)... You can either:

add a line: 

return groups 

just before the 'except' of getCharGroups() or

change the groups = line to just 

getCharGroups(infile)

BTW, The next section of the tutorial creates a much 
better quality version of the grammar checker! :-)


> Would someone care to take the time to show me the light? I know it is
> some embarassingly simple mistake but I can't find the durn thing!

Sorry my mistake, I'd been meaning to update that page when i
first noticed the bug, but it had slipped my mind. I'll fix it
tonight, honest!

Alan g.


PS Your report stats function still has a reference 
to sys.argv[1] - you'll need to remove that...


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


From richard_chamberlain at ntlworld.com  Sat Jun 17 17:58:55 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 17 Jun 2000 17:58:55 -0400
Subject: Problem Installing/Using Windows Extensions
References: <2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1@lp.airnews.net>
Message-ID: <642A954DD517D411B20C00508BCF23B001296EA2@mail.sauder.com>

Hi David,

I had the same problem with a PC at work running W95, but not on any of the
others. I finally narrowed it down to be the only one without Internet
Explorer 5 installed. And from that I presume that DCOM needs to be
installed.

Hope that helps,

Richard


David Steffen <steffen at biomedcomp.com> wrote in message
news:2F466A4F08768158.E7F19F943AD91E75.0C469CA3C5F346B1 at lp.airnews.net...
> Howdy:
>
>   I'm trying to use Pythonwin with Python 1.52 under Windows 95 and am
> consistently running into a problem that I don't understand at all,
> probably because I know relatively little about Windows.
>
>   When I selected "COM Makepy Utility" from the Tools menu of
> Pythonwin, I got the following error message:
>
> """
> >>> Failed to execute command:
> from win32com.client import makepy;makepy.main()
> Traceback (innermost last):
>   File "C:\Program Files\Python\Pythonwin\pywin\framework\toolmenu.py"\
> , line 103, in HandleToolCommand
>     exec "%s\n" % pyCmd  File "<string>", line 1, in ?
>   File "C:\Program Files\Python\win32com\client\makepy.py", line 46, \
> in ?
>     import genpy, string, sys, os, types, pythoncom
>   File "C:\Program Files\Python\win32com\client\genpy.py", line 16, in ?
>     import pythoncom
>     Import Error: DLL load failed: A device attached to the system is \
> not functioning.
> """
>
>   I noted that what I had was version 1.25, and that version 1.32 was
> up on Mark Hammond's site on starship python, so I installed the newer
> version.  The installed completed, but with three error messages near
> the end:
>
> """
> Registering COM Server AXScript Engine
>
> Registration of the AXScript Engine COM server failed.
>
> Installation will continue, but this server will require manual
> registration before it will function.
>
> exceptions.ImportError: DLL load failed: A device attached to the
> system is not functioning.
> """
>
> ...with the other two messages being similar ones about the Python
> Interpreter COM server and the Dictionary COM server.
>
> I'd be delighted to RTFM, if I could find the relevant M (I've looked
> in the included Python docs and on the python.org and starship python
> websites).
>
> I'd be happy to register the servers manually, if that would solve the
> problem and if I had a clue as to what than meant.
>
> If there is a device not functioning on my laptop, I'd love to know
> that, but a disk scan indicates no problems and I haven't noticed any
> other problems with my laptop.  I'm thinking I may be missing a DLL,
> driver, or other software component; if anyone could suggest which one,
> I'd be pleased as punch.
>
> Thanks in advance!
>
> -David Steffen-




From no_spam at umich.edu  Thu Jun  8 17:19:23 2000
From: no_spam at umich.edu (Jeff)
Date: Thu, 8 Jun 2000 14:19:23 -0700
Subject: Pythonwin is terrible!
Message-ID: <8hp2ku$a25@news.or.intel.com>

Okay, 2 questions.  Seem like basic enough functionality to be in pythonwin,
but I dont see it.
1. How do you kill an infinite loop program without killing the pythonwin
environment?
2. How can you 'unload' a module so that you can overwrite the .pyd without
leaving the pythonwin environment?

If there are no answers to these questions, someone needs to work on
pythonwin's "development" capabilities.






From pinard at iro.umontreal.ca  Tue Jun  6 15:39:45 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 15:39:45 -0400
Subject: __getattr__ and __setattr__ troubles
In-Reply-To: =?iso-8859-1?q?Fran=E7ois?= Pinard's message of "06 Jun 2000 14:46:05 -0400"
References: <8gehq1$kq6$1@news1.xs4all.nl> <8ggo6m$2v6h$1@news.aros.net>
	<oq1z2ad4wi.fsf@titan.progiciels-bpi.ca>
Message-ID: <oqk8g2bnum.fsf@titan.progiciels-bpi.ca>

Fran?ois Pinard <pinard at iro.umontreal.ca> writes:

> * May I least get the default string representation, which was satisfying?

> * So, as soon as I use `__getattr__', I suddenly get a lot of other things to
>   define as well?  This would make `__getattr__' less attractive.

Let me reply to my own message for these last two points (points 3. and
4., points 1. and 2. still remain, but are "lightier").  The two problems
above are solved by merely inserting a:

    raise AttributeError

whenever the `__getattr__' function does not know how to handle the thing.
Simple and clean to me.  Sorry for my noise.

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





From eviltofu at rocketmail.com  Wed Jun 21 21:18:42 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Thu, 22 Jun 2000 01:18:42 GMT
Subject: Pickle Question
Message-ID: <eviltofu-9E08E0.21184121062000@news.earthlink.net>

If I unpickle from a non - pickled file or byte stream what error does 
Pickle throw? I've gotten both KeyError and EOFError, does it throw 
anything else?


From fuessler at solutec.de  Wed Jun 28 11:00:34 2000
From: fuessler at solutec.de (Holger M. Füßler)
Date: Wed, 28 Jun 2000 17:00:34 +0200
Subject: MD5 - Passwords with python
Message-ID: <6e7dj8.bvb.ln@iiab.soluma.de>

Hi!

Is there a library function to encode passwords crypt-style, but with md5 as
hashing-algorithm for python?
Do you know where I can find documentation about the format?

Thanks
Holger




From m.faassen at vet.uu.nl  Fri Jun  2 09:57:53 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Jun 2000 13:57:53 GMT
Subject: *blink* That's neat.
References: <GXJY4.28606$WS3.320758@typhoon.we.rr.com> <SZKY4.6213$wYl.227247104@newsb.telia.net> <8h07fv$bnc$3@newshost.accu.uu.nl> <UINY4.6227$wYl.223307264@newsb.telia.net> <8h0ne2$70i$2@newshost.accu.uu.nl> <3934242B.ECFDBAAE@endea.demon.nl> <8h1k0q$h04$2@newshost.accu.uu.nl> <wUfZ4.6380$wYl.229799424@newsb.telia.net> <8h4djp$pkc$1@newshost.accu.uu.nl> <jXnZ4.6415$wYl.227858944@newsb.telia.net> <8h6p4n$ecq$1@newshost.accu.uu.nl> <8h72ji$32e$1@slb0.atl.mindspring.net>
Message-ID: <8h8eh1$7c0$1@newshost.accu.uu.nl>

Arthur Siegel <ajs at ix.netcom.com> wrote:
>> Martijn Faassen wrote:
>>I was  imagining some controlled circumstances where there > aren't any
>> helpers around when the newbies get down to the actual business of
>> programming.

> Why  the insistence on the fiction that the Alice studies were
> conducted in the context of learning programming?

I didn't think we were implying this. We were just talking about actual
experiments that *could* be conducted to learn more about it.

That's not to say that the Alice studies don't have something to say about
language usability at all. The web page Fredrik pointed out earlier in this
thread is useful, for instance, and by the Alice folks.

Regards,

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


From hzhu at knowledgetrack.com  Wed Jun 14 20:43:56 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 15 Jun 2000 00:43:56 GMT
Subject: Wondering about Domingo's Rational Mean book"
References: <pn5djsceirmrdhdd2ncht9c26hv0q5gkdq@4ax.com> <ralyh3rlx4xa@forum.mathforum.com> <8hp1g8$47o$1@barcode.tesco.net> <8i475e$48n$1@nnrp1.deja.com> <8i5vsn$llr$1@epos.tesco.net> <8i7vo0$p8v$1@nnrp1.deja.com> <8i8cdd$ndh$1@barcode.tesco.net> <8i967u$oh3$1@nnrp1.deja.com>
Message-ID: <slrn8kg9rf.328.hzhu@rocket.knowledgetrack.com>

Sorry to burst your bubble. I just checked your web page and as far as I can
figure out, your RM is simply a weighted arithmatic mean weighted by the
denominator.  That's also the reason why it's not unique (because 2/3 and
4/6 will be weighted differently).

Whatever you are arguing with the other person, it does not look to me that
the topic is relevant to the Python programming language, which is what
comp.lang.python is for.  Thanks for taking the argument elsewhere.


On Wed, 14 Jun 2000 23:59:02 GMT, dgomezm at etheron.net <dgomezm at etheron.net>
wrote: 
>
>Again:
>If you have some doubts, something to say,  about the rational mean and
>the rational process, then all your arguments are straightaway directed
>to the CFs, Bernoulli?s, Newton and Halley?s method, because all of
>them are ruled by the rational mean (They are just rational processes)
>as has been trivially and obviously stated in my book and summarized in
>my home page:
>www.etheron.net/usuarios/dgomez/default.htm
>

-- 

Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From wjdandreta at worldnet.att.net  Fri Jun 23 03:43:16 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Fri, 23 Jun 2000 07:43:16 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net> <3gh3lsgg9dtkfgrsq1ml2i9jcfkknrb6ei@4ax.com> <Qfq45.15292$C44.879449@bgtnsc05-news.ops.worldnet.att.net> <tm05lskrbk1if0h39m3o7fee7j64ushhkg@4ax.com>
Message-ID: <oyE45.19062$Xx5.1052150@bgtnsc06-news.ops.worldnet.att.net>

Hi Anders,

>Why don't you set up a Linux server and a TCP/IP net with DOS clients.
>Then you could have the database on the server and you would have much
>more to choose from!

That would be a good idea but we are not using plain DOS. It is Concurrent
DOS, amultitasking/mulituser DOS system and each user has a terminal. It is
a legacy system that is no longer being upgraded. It does not support
networking.

>>
>It's probably nothing wrong with xDB, but I think it's rather a big
>task to port it to DOS and create a Python Wrapper. So maybe there are
>other solutions that are better?


I don't think it will be as difficult as you imagine. I have documentation
on the dbase file structure. I wrote a simple Python program today that
takes the data in a Dbase file an converts it to a comma delimited ASCII
file. (A copy is below, I haven't finished all the data types yet.)

The only part I haven't a clue about is the indexing. Our needs are limited.
If I can add to the database, update indexes and access data using indexes,
I would be satisfied. So, I would only need to port a small part of it.
-------------------------------------------
# converts .dbf file to comma delimited ascii.

from struct import unpack
from sys import exit

DBFfile = 'tier_5.dbf'
DBF = open(DBFfile,'rb')
hdra = DBF.read(32)
numRecs,hdrSize,recSize = unpack('lhh',hdra[4:12])
hdrbSize = hdrSize-32
hdrb = DBF.read(hdrbSize)
numFields = (hdrbSize-1)/32
if '\x0D' <> hdrb[-1]:
  print 'Cannot read DBF file, header is corrupted.'
  exit(1)
line = ''
varFields = {}
for i in range(11,32*numFields+11,32):
  if (hdrb[i] == 'C') or (hdrb[i] == 'N'):
    line = line + '%s, '
    varFields[i] = ord(hdrb[i+5])
  elif hdrb[i] == 'D':
    pass #date stuff here
  elif hdrb[i] == 'L':
    pass #logic stuff here
  else:
    line = line + 'NOT A TEXT FIELD ,'
line = line[:-2] + '\n'
txtFile = DBFfile[:-3]+'txt'
txt = open(txtFile,'w')
for i in range(numRecs):
  record = DBF.read(recSize)
  vf = ()
  j = 0
  k = 1  #first byte is a tag
  while j < 32*numFields:
    fieldSize = varFields[11+j]
    vf = vf + (record[k:k+fieldSize],)
    k = k + fieldSize
    j = j + 32
  txt.write(line % vf)
--------------------------------------------
Bill




From Jonathan_Epstein at nih.gov  Tue Jun 20 15:25:09 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Tue, 20 Jun 2000 15:25:09 -0400
Subject: Computer Algebra System
References: <8inpa4$8iv$1@vishnu.jussieu.fr>
Message-ID: <394FC515.972FAE0F@nih.gov>

Matlab has an interface to Maple (their "Symbolic Toolbox", I think).

It might be useful to study their API, and emulate it in a reasonable
(and hopefully legal) fashion in Python.

-Jonathan

Rodolphe Pollet wrote:
> 
> Is anyone aware of a Computer Algebra System (CAS) ---
> i.e a mathematical software that is able to derivate, integrate,
> simplify, find limits, ... analytically --- in python ?
> It should compete with Mathematica or Maple.
> 
> --
> Rodolphe POLLET
> Laboratoire de Chimie Theorique, Case 137,    |   pollet at lct.jussieu.fr
> Universite Pierre et Marie Curie, T 22-23,    |   tel:   01.44.27.42.11
> 4, Place Jussieu, 75252 PARIS CEDEX 05, FRANCE|   Fax :  44.27.41.17


From rhays at interaccess.com  Mon Jun 19 07:30:20 2000
From: rhays at interaccess.com (Bob Hays, Computer Geek)
Date: Mon, 19 Jun 2000 06:30:20 -0500
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com> <39520785.690843718@nntp.interaccess.com>
Message-ID: <3p0sksgnqhso5dsp1cmf5srffbeh42ov0g@4ax.com>

You've misunderstood some of my response, so I'll try to clear it up.

>>I've gotten in the habit of putting the comment DEBUG on such lines in
>>all programming languages; depending upon indention is very dangerous
>>for debug lines since someone else may reindent the line properly and
>>then it gets left in (since they didn't know it was a debug line).
>>
>I never check in code with debug lines in it, so I never have to worry
>about others changing it. That's the whole point, it makes it easier
>to remove.

Good for you - you've been perfect and never, due to time constraints,
checked something in that had a debug line in it.  Congrats!

>>>2) You can easily insert a tab introducing a bug.  Consider the
>>>following code ( similar to a bug I recently fixed ):
>>{Example skipped}
>>>Initially the line in error was indented correctly, but someplace
>>>along the line the extra tab was introduced. A lot of effort then had
>>>to go in to finding this bug. With begin...end or braces this is less
>>>likely to happen.
>>
>>Yep, this happens.  Its a tradeoff - speed vs. occasional extra work.
>What increases in speed do you get by having to type braces?

The code can be typed in much faster; its SLOWER to type in the
braces.

>>In addition, I had an example, in C++, where someone had commented out
>>a block of code, but then had an embedded inline comment inside that
>>block, and, to top it all off, the less-than-desired commented out
>>block would still compile - the guy worked on that bug for a day
>>before coming to me.  I loaded it into emacs and the color coding told
>>him immediately what had happened; he was very ashamed and angry, but
>>I told him sh*t happens:-).
>>
>But that's the point. His code was illegal, it wouldn't compile. Easy
>to catch with emacs, and the compiler will catch it too. The little
>typo I show is easy to make and perfectly legal code. Its hard to
>accidentally bump against a keyboard and make a similar nasty typo.
>EVen something that looks like it might work like }{ would have to be
>put in just the right spot.

No, you didn't read carefully.  His could *WOULD* compile, but it ran
somewhat wrong (obviously) and he wasted a sinful amount of time and
energy trying to find the problem.

>>>3) Finally. Python is a language where it is almost easy to write
>>>extensible programs. To illistrate what I mean consider this example
>>>of a python program I recently wrote:
>>
>>Here's were we will agree to disagree, I think.  I think its way
>>faster and easier to extend Python stuff than C++ (and Java) due to
>>the lack of compile/link cycle and due to inheritance inside of
>>Python.  Yep, you have to be careful how you indent, but if you have a
>>good editor it will cut/copy/paste as-is without reindenting.  I see
>>more problems due to bad cut/copy/paste reuse in all programming
>>languages than I do from mis-indenting so far.
>>
>I think you misunderstood what I said. I think that Python is not now
>easily extensible do to the  problems with whitespaces. If that were
>fixed then Python would be easily extensible. 
{lots removed}

But I do understand.  Good reuse would be to not copy but to reuse via
inheritence or reinvocation - copy/paste reuse is always fraught with
problems, and that's part of what OOP should help avoid (although I've
always thought structured programming (SP) could deal with this
problem also - its code size that limits SP).
--
Bob Hays                           "Discipline is never an end in
181 W. Madison                      itself, only a means to an end."
MS 4000                            -- King Crimson
Chicago, IL  60602                 "Whoever flees history will be
(312) 904-4668 (V)                  pursued by history."
(312) 904-5774 (F)                 -- Janusz Korczak
Bob.Hays at abnamro.com               "Faster! Faster! Until the thrill
rhays at interaccess.com               of speed overcomes the fear of
                                    death."
                                   -- Hunter S. Thompson
                                   "Security is very frequently the
                                    commencement of calamity." 
                                   -- Paterculus
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS d+(-)@ s+:+ a C++(+++)$ U$ P+$ L@ E+++$ W+++$ N++ !o-- K--? w$ !O 
M$ V@ PS+ PE !Y+ PGP t !5-- !X !R tv+ b++() DI+ !D- G-- e++ h--- r+++ 
y+++** 
------END GEEK CODE BLOCK------


From epesh at my-deja.com  Fri Jun  9 16:05:42 2000
From: epesh at my-deja.com (epesh at my-deja.com)
Date: Fri, 09 Jun 2000 20:05:42 GMT
Subject: Programming Poll
Message-ID: <8hrim7$fjv$1@nnrp1.deja.com>

I have a programming skill poll back online, at
http://cupid.suninternet.com/~joeo/poll.html, with the intent of trying
to verify some suspicions regarding various programming languages. It's
very informal; no tracking is done, so it's not scientific at all, but
the more data I have...

Please check it out.


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


From hungjunglu at hotmail.com  Tue Jun 13 18:55:54 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Tue, 13 Jun 2000 15:55:54 PDT
Subject: SV: Python Productivity over C++
Message-ID: <20000613225554.27650.qmail@hotmail.com>

--- In python-list at egroups.com, Neurocrat <neurocrat at o...> wrote:
>for (int i = 0; i < 10; ++i);
>         cout << "Hello" << endl;
>
>This, of course, prints "Hello" only once, not 10 times as we

When I started to use Python a few years ago, the thing that told me that 
Guido had some brain was exactly little tiny features like that.

Try to talk to some hardcore Perl or C++ programmers, and most likely they 
will balk back at you. "What? Python does not even support increment 
operator like i++? What kind of a retarded language is that?"

I have had hard time convincing to fellow C++ programmers that an assignment 
inside an "if" statement is a bad thing, to no avail. E.g:

    if ( a = getNumber() ) {...}

Sure, totally valid statement in C++, but why would anyone introduce such a 
stupid feature so you and your compiler get confused all the time with

    if ( a == getNumber() ) {...}

?

C++ programmer get bitten all the time with the == && || operators, and 
refuse to acknowledge that there is a problem.

Guido was smart enough to get rid of the confusing operators. You can't do

   if a = b:

and you have to use explicitly the "or" "and" keywords instead of the 
cryptic "&&" "||". And the superflous operators "+=" "++" are eliminated. 
(Why? If you are trying to save nanosecond execution time, or 3 bytes in the 
code by using Python, you are in the wrong business.)

And Perl people would issue you a stern warning about the indentation in 
Python, they'd say: "be very careful with the unusual indentation of Python, 
you miss one space and all hell breaks loose." Yeah right. I've been 
programming in Python for 3 years, now. It might have happened 2 or 3 times. 
Compare with Perl where every single program I wrote was a nightmare to 
debug, even if they are just a few lines. And guess what? No matter whether 
it is Perl or C++, I always do proper indentation and match up all the 
statements and parentheses neatly by hand, so do all the dozens programmers 
that I have worked with. So the parenthesis and the semicolons sit there for 
... I guess, to be pretty? :) For avoiding a bug that might happen once a 
year, Perl and C++ prefer that you introduce bugs that happen daily or even 
hourly. Great logics.

After I saw what Guido did with the == && || ++ operators, I knew some brain 
was put into Python. And since then, I have never looked back.

Hung Jung


________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From anders.eriksson at morateknikutveckling.se  Wed Jun 14 04:01:18 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 14 Jun 2000 10:01:18 +0200
Subject: When to import
Message-ID: <lieeks4gt73r3va4ad79j5v43lufufkqgu@4ax.com>

Hello!

Being a hard-to-die C programmer I would like to know if the Python
community has any recommendation about this.

When does the import take place?
Will there be any difference in the two cases below?
If I have a module that I very seldom use should I only import it when
needed (using an if statement) or is it better to allways import it?

// Anders



# case 1
import module1
import module2


rc = module1.function1()
if rc == 1:
	rc = module2.function1()

#case 2

import module1

rc = module1.function1()
if rc == 1:
	import module2
	rc = module2.function1()




From herzog at online.de  Wed Jun 28 15:57:41 2000
From: herzog at online.de (Bernhard Herzog)
Date: 28 Jun 2000 21:57:41 +0200
Subject: Q: memory ownership in python extensions
References: <3958e1ee.13405033@localhost>
Message-ID: <m3vgytd1e2.fsf@greebo.nodomain.de>

fstajano at uk.research.att.com (Frank Stajano) writes:

> In the C code, I malloc some memory, perform some processing that
> writes a string in it, then turn this C string into a Python string
> via Py_BuildValue. The question is: Who is responsible for freeing
> that memory? Example code fragment follows.
> 
>              ...
> 	ciphertext = (byte*) malloc(pLen+1); /* who frees this? */
> 	if (!ciphertext) {
> 		PyErr_NoMemory();
> 		return NULL;
> 	}
> 	encrypt_block(key, kLen, plaintext, pLen, ciphertext);
> 	return Py_BuildValue("s#", ciphertext, pLen);
> 
> It would be logical and symmetrical for the code who malloc'ed to be
> the one responsible for free'ing. 

And that's exactly what you have to do.

> For this to happen, Py_BuildValue would have to take a copy. But I
> strongly suspect that, for efficiency reasons, Py_BuildValue does not
> take a copy and just reuses the memory passed to it; in which case the
> free would be done by the Python system when the reference count of
> the Python object goes to 0.

Py_BuildValue simply passes the char* to PyString_FromStringAndSize
which copies the data pointed to. 

A string object is one contiguous chunk of memory which includes the
Python object head (for refcount, type pointer etc) and the actual
string content. There's no provision for a string object using memory
allocated outside the string object code.

> Which one of these is the actual behaviour of Py_BuildValue is
> something that isn't at all apparent from Chapter 1.9 of "Extending
> and embedding" version 1.5.2 (hint hint). 

Well, one could argue that the examples suggest that the strings are
copied. Some of them use string literals.

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


From aahz at netcom.com  Sun Jun 18 17:52:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jun 2000 21:52:57 GMT
Subject: Python 3000 -> No, but there's 'R***'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <8iib7d$73f$1@news.jump.net>
Message-ID: <8ijgbp$s5s$1@slb7.atl.mindspring.net>

In article <8iib7d$73f$1 at news.jump.net>,
Conrad Schneiker <schneiker at jump.net> wrote:
>
>  [...]

If anyone cares, my response to Conrad was posted only to
comp.lang.ruby.  I see no reason to continue the debate here, and I plan
for that to be my last post on this subject.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From dalke at acm.org  Wed Jun 28 06:16:25 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 28 Jun 2000 04:16:25 -0600
Subject: unexpected behaviour in class variables
References: <8j83ml$2s7$1@slb3.atl.mindspring.net> <8F5FCA994gmcmhypernetcom@199.171.54.194> <8j8vjr$v63$1@nntp9.atl.mindspring.net> <8F60A1C6Bgmcmhypernetcom@199.171.54.154>
Message-ID: <8jcj57$ec$1@nntp9.atl.mindspring.net>

Gordon McMillan:
>It ain't wrapped. Beat up Fred, he needs it <wink>.


Beat him up?  I'm a software guy and try to stay away from that
whole hardware concept.  How about I email him instead?

>Functionally, the difference is the need to provide an arg to the user-
>defined function that will take the value of "self". Or, perhaps, the
>inability to use one in a C extension. Maybe your original post spelled out
>some brilliantly deduced logical consequences of this apparently innocuous
>difference. If so, they slid right past me <wink>.


The consequence is I have to know the implementation of a callable object
in order to use it as a class variable.  It means I can't write a
drop-in replacement of one of my C libraries with an equivalent pure
Python module in the obvious way of implementing the C-based functions
using user-defined Python functions.  (I could do it by using a
singleton callable object.)

The only other place in Python that a use distinction arises between C
code and Python code is the type/class dichotomy, where you can't derive
from a type.  That's considered to be something which should be fixed,
and I feel this should as well.

Consider

class C:
  f = f

c = C()
print f(), c.f()

If this works, then f is not a user-defined function; it is either
implemented
in C or is a callable object.

On the other hand, if

print f(c), c.f()

works, then f is a user-defined function.  Ummm.. unless it takes *args,
but that's beside the point.

Now that I know it exists and works as documented, I'll live with it
and I'll change my code to work around it.

I still wanted to complain that it isn't obvious that that's the
appropriate behaviour, in the hopes that someone would tell me why it
make sense, or that someday (Py3K?) it will be fixed.

Also, I do realize it is a rare case where this will be a problem, so
I'll shut up now.

                    Andrew
                    dalke at acm.org





From moshez at math.huji.ac.il  Fri Jun  9 14:40:20 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 21:40:20 +0300 (IDT)
Subject: Problem with script and typing
In-Reply-To: <394735dc.93980406@nntp.interaccess.com>
Message-ID: <Pine.GSO.4.10.10006092137270.28574-100000@sundial>

On Fri, 9 Jun 2000, Thaddeus L. Olczyk wrote:

> >>         null=Project("")
> >>         projects.remove(null)
> >          ^^^^^ projects are compared by identity, unless you supply an
> >          __cmp__ method
> >
> I'm not sure exactly what you are trying to say here. The Project
> constructor is supposed to insert self into the list. So if you are
> trying to say the null may not be removed , I don't think so because
> they are the same instance.

I see. The Project constructor you posted did *not* put anything into any
global list.

>            def __cmp__method(self,rhs):
>                  return self.name == rhs.name
> in the Projects clas, you are right. I didn't know.

No, I meant it would be easier to put:

	def __cmp__(self, other):
		return cmp(self.name, other.name)

In the project constructor. But I didn't understand what you were trying
to do there, so never mind.

> Actually it is the return x which whould have read return project.

I didn't see any "return x". Posting short, but *complete*, code will get
you higher quality answers.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From robin at jessikat.co.uk  Thu Jun 22 06:45:25 2000
From: robin at jessikat.co.uk (Robin Becker)
Date: Thu, 22 Jun 2000 11:45:25 +0100
Subject: Dynamically add methods to classes?
References: <Sql45.1512$9W1.105781@news-west.usenetserver.com>
Message-ID: <K9+ZeBAF5eU5Ew$7@jessikat.demon.co.uk>

In article <Sql45.1512$9W1.105781 at news-west.usenetserver.com>, Matthew
Cline <matt at nightrealms.com> writes
>Is is possible to dynamically add methods to classes in Python?
>
>Thanks in advance.
>
>
>
use the new module
-- 
Robin Becker


From claird at starbase.neosoft.com  Wed Jun 28 16:40:57 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Jun 2000 15:40:57 -0500
Subject: PythonQt Binding Is is available?
References: <8jd4hk$nda$1@bw107zhb.bluewin.ch> <slrn8lk9lp.431.newsbunny@noether.freeserve.co.uk>
Message-ID: <A5C1390A3B25AD10.93FC9B73410028DE.9F19CEE1C9D01AAF@lp.airnews.net>

In article <slrn8lk9lp.431.newsbunny at noether.freeserve.co.uk>,
Martin P Holland <m.holland at noether.freeserve.co.uk> wrote:
>On Wed, 28 Jun 2000 17:05:35 +0200,
>Markus Meng <meng.engineering at bluewin.ch> wrote it all in the subject
>
>http://www.thekompany.com/projects/pykde/
			.
			.
			.
For more on the subject, see
<URL:http://starbase.neosoft.com/~claird/comp.windows.misc/Qt.html>.
-- 

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


From kajiyama at grad.sccs.chukyo-u.ac.jp  Wed Jun 28 13:50:25 2000
From: kajiyama at grad.sccs.chukyo-u.ac.jp (Tamito Kajiyama)
Date: 28 Jun 2000 17:50:25 GMT
Subject: Tkinter: problems running as a script.
In-Reply-To: Aaron Ginn's message of 26 Jun 2000 11:10:37 -0700
References: <sng0q05n4y.fsf@motorola.com>
Message-ID: <KAJIYAMA.00Jun29025025@pseudo.grad.sccs.chukyo-u.ac.jp>

Aaron Ginn <aaron.ginn at motorola.com> writes:
| 
| I'm unable to run a simple "Hello, world" python script, but I
| can run the script if I paste the lines one at a time into the
| python interpreter.
(snip)
| coronado <213> ~/bin/Tkinter.py
| Traceback (innermost last):
|   File "/home/ginn/bin/Tkinter.py", line 2, in ?
|     from Tkinter import *
|   File "/home/ginn/bin/Tkinter.py", line 4, in ?
|     root = Tk()
| NameError: Tk
(snip)
| What am I doing wrong here?

You need to change the filename of the script other than those
of Python modules including Tkinter.py.

Python adds to the module search path (i.e. sys.path) the
directory where the executable script lives.  In this case, 
~/bin is considered as a directory holding Python modules.
That's why your script imports itself instead of the standard
Tkinter module.

Regards,

--
KAJIYAMA, Tamito <kajiyama at grad.sccs.chukyo-u.ac.jp>


From anders.eriksson at morateknikutveckling.se  Tue Jun 13 11:19:11 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 13 Jun 2000 11:19:11 -0400
Subject: Where should I put my HTMLgen directory?
References: <7BD10B680501D411B9DF009027E06F32011ACA@EXCHANGE>
Message-ID: <642A954DD517D411B20C00508BCF23B0012968EC@mail.sauder.com>

On Tue, 13 Jun 2000 15:05:08 +0200, Max M?ller Rasmussen
<maxm at normik.dk> wrote:

>I am trying to install HTMLGen on my machine.
>
>My python is installed in:
>C:\Program Files\Python\Lib\HTMLgen
>

I will assume that you mean: C:\Program Files\Python

>I have unpacked HTMLGen into:
>C:\Program Files\Python\Lib\HTMLgen
>
>
>So obviously I am doing something wrong. I thought that Python traversed the
>Lib diretory recursively to find any modules. But no?
>

This is a bit confusing! I'm not sure myself on how it really work. 

>Should I put all the *.py* files in the root of Lib? That seems a bit messy
>to me.

I think you have 3 alternatives:

1)	Add the subdirectory to the PYTHONPATH (which on Windows is in
the registry)

2)	Create a package (create an empty file called __init__.py in
the HTMLgen directory)

3)	create a HTMLgen.pth file in the lib directory. This file
should only contain the directory name HTMLgen


//Anders
Maybe-some-Guru-could-create-a-document-explaining-python-import



From jon at dgs.monash.edu.au  Sun Jun  4 17:31:36 2000
From: jon at dgs.monash.edu.au (Jonathan Giddy)
Date: Mon, 5 Jun 2000 07:31:36 +1000 (EST)
Subject: Um? *boggle* Silly question about listobject.c
In-Reply-To: <000e01bfcddf$d2e09d40$1d2d153f@tim> from "Tim Peters" at Jun 04, 2000 12:46:17 AM
Message-ID: <200006042131.HAA19297@nexus.csse.monash.edu.au>

One nice thing about using longs is that it would fix this long-standing
bug on 64-bit machines.

Python 1.5.2+ (#1, Sep  2 1999, 14:43:56) [C] on osf1V4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.maxint
9223372036854775807
>>> [1] * sys.maxint
[]



From spamfranke at bigfoot.de  Sun Jun 18 17:47:18 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Sun, 18 Jun 2000 21:47:18 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com> <ug0qflns9.fsf@ctwd0143.fitlinxx.com> <slrn8kipri.pf.hzhu@mars.localdomain>
Message-ID: <394e422b.6558773@news.btx.dtag.de>

On Thu, 15 Jun 2000 23:28:39 GMT, hzhu at mars.localdomain (Huaiyu Zhu) wrote:

>>Ah, good point - try this variant - I switched to getattr() which
>>takes the inheritance into account (you could also write your own
>>recursive function that used Instance.__bases__ to look in the
>>superclasses).

Some thoughts:

Using callable(x) instead of "type(x) is types.Function/MethodType" would
make this still a bit more generic. 

In its ZPublisher.mapply module, Zope uses Python's introspection features
to figure out and match function signatures.

Stefan



From thomas at xs4all.net  Sun Jun 11 20:04:05 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 12 Jun 2000 02:04:05 +0200
Subject: Missing source files?
In-Reply-To: <eFV05.5409$iy.446615@bgtnsc06-news.ops.worldnet.att.net>; from wjdandreta@worldnet.att.net on Sun, Jun 11, 2000 at 11:54:50PM +0000
References: <eFV05.5409$iy.446615@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <20000612020404.O20700@xs4all.nl>

On Sun, Jun 11, 2000 at 11:54:50PM +0000, William Dandreta wrote:
> I am trying to build Python 1.5.2 for DOS and my linker finds 3 undefined
> functions:

> PyMember_Get
> PyMember_Set
> PyBuffer_FromObject

> The only thing I was able to find was:

> DL_IMPORT(PyObject *) PyMember_Get Py_PROTO((char *, struct memberlist *,
> char *));

> which is useless. I checked the entire set of source files for 1.5.2 and
> those functions don't seem to be there. Does anyone know where they are? And
> if some sources files are missing from the latest package?

PyMember_Get and PyMember_Set are defined in Python/structmember.c, and
PyBuffer_FromObject is defined in Objects/bufferobject.c.

The definitions don't seem to be wrapped in any #ifdefs, so they *should* be
included in your binaries.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From eviltofu at rocketmail.com  Wed Jun 14 22:31:00 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Thu, 15 Jun 2000 02:31:00 GMT
Subject: ?Extreme Programming in Python?
Message-ID: <eviltofu-A13457.22304614062000@news.earthlink.net>

Has anyone done any extreme programming in python? Care to share your 
experiences?

<http://www.extremeprogramming.org/index.html>


From trentm at activestate.com  Tue Jun 20 21:47:23 2000
From: trentm at activestate.com (Trent Mick)
Date: Tue, 20 Jun 2000 18:47:23 -0700
Subject: Enums and Python
In-Reply-To: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net>
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <20000620184723.A26354@activestate.com>

On Wed, Jun 21, 2000 at 12:56:16AM +0000, William Dandreta wrote:
> Is there a Python type like the C enum type?
> 
> It can be very useful in making code more readable.
> 

Nothing specific but you can fake iti (sort of):

(elem0, elem1, elem2, elem3) = range(4)


Trent

-- 
Trent Mick
trentm at activestate.com



From gmcm at hypernet.com  Thu Jun  8 22:15:54 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 09 Jun 2000 02:15:54 GMT
Subject: Still doing python scripting.
References: <AJT%4.29773$2X2.255734@newsread2.prod.itd.earthlink.net>
Message-ID: <8F4DE43A4gmcmhypernetcom@199.171.54.195>

Arint? wrote:

>I have a question about scripting.  I have py embedded in my app and
>what I wanted to know should I do Py_Initialize and Py_Finalize after I
>run every script.  

No.

>I ask because after I run one script if there was an
>error, that same error is reposted when I run another script.  I capture
>the stderr and stdout and store it.

Did you try clearing it? PyErr_Clear, I think.

- Gordon


From adams_s at lab.eng.usyd.edu.au  Fri Jun  9 10:05:48 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Sat, 10 Jun 2000 00:05:48 +1000
Subject: Newbie: Simple question about string slices
References: <8hpsl6$mve@newshost.fujitsu.com.au>
Message-ID: <8hqtas$st8$1@news1.wire.net.au>

Hi,



Ray Smith <Ray.Smith at fujitsu.com.au> wrote in message
news:8hpsl6$mve at newshost.fujitsu.com.au...

> s = "abcdefg"
>
> I assumed s[2:2] to be equal to 'c',
> but when I tried it it equals '' (empty).
> The way I read this is give me the slice from index 2 to index 2.

as I see it, it does do this - I explain myself at the end :-)

> I also "feel like" s[:2] + s[2:] should be equal to 'abccdef'

its just how the slicing has been defined to work, probably a result of the
first index being zero - last index beind len()-1
(It's late and I'm tired from studying for exams :-|

> My question is:
> If given a slice s[a:b]
> can I assume index "b" means upto but not including the index "b"?
>
> eg.  s="abcdef"
>
> s[:3] = "abc", meaning from index 0, upto but not including index 3
> s[3:] = "def", meaning from index 3 to the end of the string
>

yes

> I guess another way to ask my question is:
> Why is s[a:a] always '' (empty) ?

because you are going 'from' the _start_ index 'a' to the _start_ of the
index 'a' - if that makes sense.
i.e. if you think of it in physical terms, the total displacement is zero if
I go from where I am, to where I am, I haven't gone anywhere.

> Hope someone out there in Python land can make sense of my question!

made sense of it, though I'm not sure if I made sense :-)

> Thanks
> Ray Smith

no probs,

Steven

DISCLAIMER: I'm not that experienced at Python programming, and it's late
(or as my watch tells me, its now early)




From boud at rempt.xs4all.nl  Sat Jun 17 17:23:42 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 17 Jun 2000 21:23:42 GMT
Subject: Python class browser
References: <XKQ15.1$oL1.1552@paloalto-snr1.gtei.net> <3949BDA0.DF21A426@sightreader.com>
Message-ID: <8igq8v$5o$1@news1.xs4all.nl>

Ken Seehof <kens at sightreader.com> wrote:
> This is a multi-part message in MIME format.
> --------------971B39BFB91105C31B954C06
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit

> Brian wrote:

>> Anyone know of a class browser for Python?  Preferrably written using
>> wxPython?
>> Thanks!

> There's a good one that's part of PythonWin.  I haven't looked at the
> code though.  PythonWin and wxPython are very similar, so it shouldn't be
> too difficult to port.

> Let me know if you find a solution.  I'm gonna need it too in a couple
> months.

Python comes with pyclbr.py, which makes it really easy to write
a class browser. I ported the IDLE class browser to PyKDE in about
three hours; I then ported the PyKDE browser to Qt 2.x in half an
hour. Both can be had from my website.

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



From david_ullrich at my-deja.com  Wed Jun  7 13:39:07 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 07 Jun 2000 17:39:07 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <kk5rjski2pc4g4tor99tc2h5ursinprg80@4ax.com> <393D999B.E1150BCB@home.com>
Message-ID: <8hm1bp$ac4$1@nnrp1.deja.com>

In article <393D999B.E1150BCB at home.com>,
  Rajarshi Ray <rajarshi at home.com> wrote:
> Kirby Urner wrote:
> >
> > Rajarshi Ray <rajarshi at home.com> wrote:
> >
> > >Besides the gee-whiz aspect of such algorithms, do they have any
value?
> > >After all, what makes pi pi is ALL of its digits in a particular
> > >sequence.
> >
> > Supposedly there were some deep points about number
> > theory in this vicinity -- according to the web pages
> > anyway.  I'm no expert.
>
> Ah, that's what I wanted to hear. The importance of this discovery is
> then what makes pi amenable to such a calculation, and also if other
> transcendental numbers can be treated like this e.g. 'e'.

    You wonder if I've ever heard of "Google", but you're unable
to click on the links he provided, even after I suggested you do so.

> Don't you just hate presumptuous bastards like Ullrich who will jump
> your gun without even considering if you actually have a point??

   Jump what gun? I've done nothing but ask _questions_ about your
"point" - questions which you have simply refused to answer.

   When most people have points they're willing to explain when
people ask about the details.

>Perhaps
> you should be glad you're a non-expert; the more expert you get, it
> seems the more of a pompous ass you become too.
>
> > Anyway, the "gee whiz" factor isn't to be underestimated,
> > either.
> >
> > Kirby
>
> --
> Sir, I have found you an argument. I am not obliged to find you
> an understanding.
>
>                                                - Samuel Johnson
>


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


From zessin at my-deja.com  Tue Jun 13 05:48:03 2000
From: zessin at my-deja.com (Uwe Zessin)
Date: Tue, 13 Jun 2000 09:48:03 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <hs815.297$i06.89565@ptah.visi.com> <20000612212452.P20700@xs4all.nl> <TQd15.356$i06.100607@ptah.visi.com>
Message-ID: <8i500g$ku4$1@nnrp1.deja.com>

In article <TQd15.356$i06.100607 at ptah.visi.com>,
  ge at nowhere.none (Grant Edwards) wrote:
> In article <20000612212452.P20700 at xs4all.nl>, Thomas Wouters wrote:
>
> >> >Is it's base structure less related to any one specific operating
> >> >system?
> >
> >> Perl isn't so much related to an OS, as it is to some other
> >> utilities (awk,sed,grep) and languages (bourne shell).
> >
> >Yeah, right, and a chair isn't geared as much to a human as to a one-
> >to-two meter tall, bi-legged creature with two arms, a flat back and
> >bends in all the right places ;)
> >
> >I guess it depends on your definition of OS, and of UNIX ;)
>
> Well, believe it or not, I actually did most of my initial
> awk/sed/grep shellscript learning under VMS.  (For God's sake,
> don't tell anybody...)
>
> DEC used to have a product called DECShell than consisted of
> the entire set of version 7 shell tools (including the Bourne
> shell, sed, awk, lex, yacc, and a libc that did a fair job of
> emulating the Unix system calls).
>
> Process creation under VMS involved a lot more overhead that
> undre Unix, so shellscripts ran a bit slower than DCL command
> scripts (DCL seemed to somehow run scripts without starting a
> lot of new processes)

Well, DCL is permanently mapped in the second GByte of the virtual
address space, while programms (like editors) run in the 1st GByte
and get removed on image-exit. Control is then returned to DCL.
I don't know shell programming very much, but DCL has many built-in
functions so you don't need to activate an extra program [F$SEARCH()
vs 'find'].
(I don't want to start a discussion if one feature is 'better' or
 not, OK? Let's agree that both systems are different.)

There is a POSIX environment with a real fork(), but that came _much_
later than DEC/Shell.


I have found that Python and OpenVMS live together very good. Don't
know how much you have done with system services, etc..., but if
you are interested, have a look:

http://www.decus.de/~zessin/python/doc/refman/vms_sys_getquiw.html
http://www.decus.de/~zessin/python/doc/refman/vms_sys_sndjbcw.html
http://www.decus.de/~zessin/python/doc/demo/rms_number_of_keys.html

--
Uwe Zessin


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


From neilh at scintilla.org  Wed Jun 14 09:51:18 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 14 Jun 2000 13:51:18 GMT
Subject: Killing wxPython process in PythonWin
References: <8i7178$44c$1@nnrp1.deja.com> <8F536925Fgmcmhypernetcom@199.171.54.194>
Message-ID: <q5M15.4703$cQ.38625@news-server.bigpond.net.au>

> I hate to tell you this, but except for very, very simple scripts, I never
> run anything from within Pythonwin or IDLE. If I do run from inside the
> IDE, (say, to use the debugger), I generally expect I'll have to exit and
> restart between runs. I've chased too many "bugs" (that were really fixed,
> but something wasn't reloaded) to trust anything but a fresh start. And
I'm
> certainly not going to bother with the bizarre hacks people come up with
> (eg, to replace the "code" but not the data in "running" instances).

   Looks like what is needed is a 'run in new environment' option for
Pythonwin/IDLE. Or in a separate process?

   Neil





From rob at hooft.net  Wed Jun 14 01:49:06 2000
From: rob at hooft.net (Rob Hooft)
Date: 14 Jun 2000 07:49:06 +0200
Subject: Tkintr: how to seize focus in a transient?
References: <voR%4.1057$zm3.238149@ptah.visi.com> <7ks05.1254$zm3.340136@ptah.visi.com>
Message-ID: <m2d7lkolrh.fsf@hooft.net>

>>>>> "GE" == Grant Edwards <nobody at nowhere.nohow> writes:

 >> In a transient dialog window, how do I disable focus for the rest
 >> of the app?  I only want to prevent focus for the rest of the app,
 >> not the whole X11 system.

 GE> I found it. In case anybody else is wondering, you call the
 GE> grab_set() method of the transient window.

grab_set might fail during certain operations of the window manager.
In such a case you'll need to retry.

Hint: use Pmw, it handles all kinds of niceties like this for you.

Regards,

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


From moshez at math.huji.ac.il  Sat Jun 10 11:39:21 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 18:39:21 +0300 (IDT)
Subject: windll
In-Reply-To: <39425774.9A7400F0@videotron.ca>
Message-ID: <Pine.GSO.4.10.10006101837530.6252-100000@sundial>

On Sat, 10 Jun 2000, Jean-Yves Vinet wrote:

> On windows, "windll" helps me a lot. I wrapped
> a financial system and the Fame database. I want
> to do the same for Sun and Linux.
> 
> Is someone know how to do it on Linux/Unix ?

The "dl" module. Download the latest documentation from the Python site
for documnetation. This module is not always built: if your Python doesn't
have it, just re-install from source, and edit the "Setup" file to include
it.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From frankc at colconsulting.com  Sun Jun 18 15:16:59 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Sun, 18 Jun 2000 15:16:59 -0400
Subject: Circular import
Message-ID: <394D202B.1DF60864@colconsulting.com>

With the assumption that class A and class B reference each other, how
do you avoid the Python error with:

---------- Composite.py -----------

from Leaf import Leaf

class Composite:
	def __init__( self ):
		self.__theLeaf = Leaf(self)

-------- Leaf.py --------------

from Composite import Composite

class Leaf:
	def __init__( self, aComposite ):
		self.__theComposite = aComposit



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

		
-- 
Frank V. Castellucci


From LCortes at CoolLink.Net  Fri Jun 30 13:31:09 2000
From: LCortes at CoolLink.Net (Luis Cortes)
Date: Fri, 30 Jun 2000 11:31:09 -0600
Subject: [JOB HUNT] Searching for a python programming position
Message-ID: <395CD95D.4349B8CF@CoolLink.Net>

Hello,

    Picked up python about 6 months ago and I love it.  I have been
programming in C/C++ for 5 years with experience in Windows 9x/NT,
Linux, OS/2, and many flavors of UNIX, including different database like
MySQL, PostGres, DB/2, Sybase, etc.  I have also gone to the Zope
training course and have had some minor experience setting up a
website.   If someone out there has a short term position for me (until
October), please get in contact with me via email so that I can submit
my resume.

Thanks,
Luis.

(Lcortes at CoolLink.Net)



From donn at oz.net  Thu Jun 15 00:49:22 2000
From: donn at oz.net (Donn Cave)
Date: 15 Jun 2000 04:49:22 GMT
Subject: Hex to decimal
References: <j7Y15.2020$Uw3.137442@bgtnsc04-news.ops.worldnet.att.net>
Message-ID: <8i9n8i$9e5$0@216.39.151.169>

Quoth "William Dandreta" <wjdandreta at worldnet.att.net>:
| I have an ascii string like 0C44DH. Ids there a Python function to convert
| it to decimal?

Yes, everything except that trailing H anyway.  I'd use
string.atoi('c44d', 16), and I guess in the end that is
about the same thing as int('c44d', 16)

	Donn Cave, donn at oz.net


From bjorn at roguewave.com  Thu Jun 29 13:12:03 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 29 Jun 2000 11:12:03 -0600
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194> <ffcmlsc6p4lqohdd6tequdu918l0temns4@4a
Message-ID: <395B8363.F2D86F5B@roguewave.com>

Just van Rossum wrote:
> 
> Toby Dickenson wrote:
> >
> > gmcm at hypernet.com (Gordon McMillan) wrote:
> >
> > >I think the "relative import" feature should be shot in the head (2 posts
> > >today), but unfortunately most packages rely heavily on it.
> >
> > A relative import is just as useful as a relative directory: it allows
> > intra-module imports, without each module needing to know where it is
> > packaged.
> >
> > The real problem is that relative imports and absolute imports are
> > spelled the same way.
> 
> That's an interesting way of looking at it.
> 
> <silly-proposal>
> If relative imports as we know them are shot in the head, an
> alternative spelling for relative imports could be invented.
> Eg. with a leading period:
> 
> import .some.submodule
> </silly-proposal>

Combining attempts from a sibling thread (and not changing the current
meaning of import but introducing a way to spell global import):

  import global os

or perhaps,

  global import os

all done with existing keywords :-)  I prefer the latter since it would
make it possible to do:

  global import os, sys

look-ma-no-hands'ly y'rs
-- bjorn



From dhansen at cyberoptics.com  Fri Jun  2 18:36:07 2000
From: dhansen at cyberoptics.com (Dave Hansen)
Date: Fri, 02 Jun 2000 22:36:07 GMT
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com> <8h8jg3$d66$1@slb6.atl.mindspring.net>
Message-ID: <3938363a.87214117@client.ce.news.psi.net>

On 2 Jun 2000 15:22:43 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>In article <8h87u1$sga$1 at news2atm.raleigh.ibm.com>,
>Arint? <shouldbe at message.com> wrote:
>>
>>I am trying to write this escape seq to a device from python
>>"H\x1b\x23\x12art\x1b\x23\x01\n"
>>The \x1b\x23\x12 is supposed to be a command tell the device to do something
>>special with "art", but python is making it look like this \x1b\x23\x12a
>>"rt".  Am I missing something on how to work with escape sequences?  Is
>>there another way to write hex values in a string.  I rather us \x00 because
>>that is what are testers are familiar with.
>
>I'd like to see some of the actual code.  Firing up the Python
>interpreter produces correct resulte for me.

I didn't post the original question, but here's what I get
interactively:

   >>> t = "H\x1ba"
   >>> for c in t: print "%x" % ord(c)
   ...
   48
   ba
   >>> t = "H\x1b" "a"
   >>> for c in t: print "%x" % ord(c)
   ...
   48
   1b
   61
   >>>

Regards,

                              -=Dave
-- 
The opinions expressed in this post are my own and do not 
represent the views of CyberOptics Corporation.
Change is inevitable.  Except from a vending machine.


From thomas at xs4all.net  Fri Jun  9 16:01:52 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 9 Jun 2000 22:01:52 +0200
Subject: PATCH: Augmented assignment
In-Reply-To: <slrn8k2gq6.pi.hzhu@rocket.knowledgetrack.com>; from hzhu@rocket.knowledgetrack.com on Fri, Jun 09, 2000 at 07:16:28PM +0000
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net> <slrn8k2gq6.pi.hzhu@rocket.knowledgetrack.com>
Message-ID: <20000609220152.E20700@xs4all.nl>

On Fri, Jun 09, 2000 at 07:16:28PM +0000, Huaiyu Zhu wrote:

> >>As promised, here is my patch to add augmented assignment (+=, -= and
> >>family) to Python. It's still Work In Progress, and probably requires a
> >>rewrite of some magnitude before Guido will accept it.

> Greate!  I'll use them in the MatPy package as well.

Cool! Because I'm sure that helps the acceptance of the patch :)

> BTW, maybe __add_eq__ would be a better name than __add_ad__, as it
> describes the symbols instead of their meaning.

I thought about this, but decided 'eq' isn't the right word. If we go and
describe the symbols, it should be '__plus_eq__'. The __hooks__ all have a
name descriptive of their numerical function, so I would personally favor
'__inplace_add__'. It might be a bit long, but it's fairly hard to get more
descriptive. The only hurdle might be explaining what 'inplace' stands for,
but the concept has been obvious to me for too long to make a useful
statement about that ;-)

If '__inplace_add__' is too long, perhaps '__iadd__' is better: it
emphasizes that the hook, like __radd__, is a 'specialized' version of
__add__.

Also, using '__add_eq__' might suggest to people that it is also called for

x = x + 1

as it also contains both the 'add' operator and the 'eq' sight ;)
But Guido suggested __add_ad__ in his latest posting on this subject, and it
is ultimately his choice. If he comes up with a better name, or changes his
mind in favor of whatever suggestion whoever made, it's a simple case of
search-and-replace to fix it, anyway ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From websquadNOweSPAM at hotmail.com.invalid  Tue Jun 13 09:54:56 2000
From: websquadNOweSPAM at hotmail.com.invalid (hazrd)
Date: Tue, 13 Jun 2000 06:54:56 -0700
Subject: Zope on NT port problem
Message-ID: <046d0a4c.6e7264ed@usw-ex0106-046.remarq.com>

Hi all

How do I customise the port number that Zope installs to? I'm
running NT and there is already a service running on the default
port 8080.

On a side note, are there any 'products' like an auto-RSS/RDF
creation facility for Zope or squishdot? I'd like to be able to
easily syndicate stuff.

Thanks

.wes


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



From ge at nowhere.none  Fri Jun 16 12:16:37 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 16 Jun 2000 16:16:37 GMT
Subject: Tkinter Widget command function call
References: <3949E71A.AF53A1CF@muc.das-werk.de>
Message-ID: <Fps25.1221$i06.273052@ptah.visi.com>

In article <3949E71A.AF53A1CF at muc.das-werk.de>, Thomas Thiele wrote:

>I create some buttons dynamically in a for loop:
>
>for i in range(0, n):
>	MyButton = Button(frame, text="press me", command=self.CmButtonPressed) 
>	MyButton.pack(side=TOP)	
>
>And now I want to check witch button has called the CmButtonPressed -
>function.
>
>def CmButtonPressed(self, ???):
>	print "Button ", buttonnumber, "pressed"
>
>What is the best way to do this? Has anaboby an idea?

I'd use lambda to create the command functions on the fly, but
I'm an old Scheme geek so I think rather oddly about some
things.

Something like:

----------------------------------------------------------------------
from Tkinter import *

root = Tk()

def buttonPressed(n):
    print "button %d pressed" % n
    
for i in range(0,4):
    Button(root,
           text = "Button %d" % i,
           command = lambda j=i: buttonPressed(j)).pack()
           
root.mainloop()
----------------------------------------------------------------------



-- 
Grant Edwards                   grante             Yow!  Imagine--a WORLD
                                  at               without POODLES...
                               visi.com            


From elengvenis at uswest.net  Thu Jun 15 09:04:37 2000
From: elengvenis at uswest.net (Eric Lengvenis)
Date: Thu, 15 Jun 2000 08:04:37 -0500
Subject: PythonWin Debugging tools???
References: <Zc425.166$M85.9384@news.uswest.net>
Message-ID: <dw425.14$lY4.1297@news.uswest.net>

Never mind. I saw a different post that directed me to
 http://starship.python.net/crew/mhammond/win32/Downloads.html

There I downloaded the build 129 and patch. Now, the date is still April
13th, but all the functions are there.

Does anybody find it strange that the Python.org site is not the definitive
site for current builds?



Eric Lengvenis wrote in message ...
>I am very new to Python.
>I just picked up the book Python Programming on Win32.
>Early in the book, the authors describe the debugging features available.
>None of them are available in my version. I noticed that the date on my
>interactive window is April 13th 1999 and  the date on the figures in the
>book is June 14th 1999.
>
>Were the debugging features not included until this version? If so, how do
I
>get it. I went all over python.org and it definitely installed the April
>13th version. (I have installed a couple of times from different
downloads.)
>
>Please advise. Where do I get the debugging tools???
>
>Eric
>
>
>




From m.faassen at vet.uu.nl  Tue Jun  6 13:49:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 6 Jun 2000 17:49:37 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <393CCEE6.E5C1C267@marconicomms.com> <8hj17n$rj$1@nnrp1.deja.com>
Message-ID: <8hjdjh$nfm$1@newshost.accu.uu.nl>

benji_york at cal-na.com wrote:
[snip]
> I'm not saying this is enough of a reason to change the language, but I
> definitely see an advantage to not mixing naming conventions in my
> modules just because I want to use someone else?s module that doesn't
> name the same way I do.

> In a production environment this can be quite
> annoying (or it has been for me).

The advantage however when maintaining the code is that I can instantly
recognise word-glyphs (which include the case-spelling in my brain at
least).

> Example of what I would like to be able to do:

[snip]
> Along the same lines; identifiers could also ignore imbedded (not
> leading or trailing) underscores, so I could use dict.HasKey() instead
> of dict.has_key().

Besides my word-glyph problem, that would be terrible for search (and
replace). While searching is often case-insensitivite in editors, it's
definitely not often underscore insensitive, so this could get
very nasty. (even case-insensitivity already ruins the default
'grep', though of course you can use -i to ignore it).

> I think both of these suggestions have /some/ merit. After all,
> identifiers carry meaning, so why not ignore those parts of the
> identifier that carry little or no meaning to the reader.

For some readers however, they do carry meaning. The first letter of a
class being a capital, for instance. The conventions concerning underscores
are more subtle, but there must be several of those floating around as well.

[snip]
> Some have said that this would make it harder for people just learning
> Python, but in training the people where I work (long-time COBOL and
> Fortran guys) I've found that the differences between dict.has_key()
> and HTMLgen.TD() and pickle.dumps() causes some confusion.

But do you want them to introduce their long-time COBOL and Fortran
case spellings in their Python code? Or do you want them to learn
the Python idioms? While case-insensitivity indeed brings a disciplined
programmer more internal consistency in his own code even when using
external modules (question is whether Python's namespaces don't make
this point partially moot), case-insensivity will decrease cross-module
case-consistency significantly. If two people maintain the same
module, you'll also get case-consistency problems.

Trade-offs, trade-offs-ly yours,

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


From hgg9140 at seanet.com  Thu Jun 15 18:55:09 2000
From: hgg9140 at seanet.com (Harry George)
Date: 15 Jun 2000 15:55:09 -0700
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net>
Message-ID: <m3aegmh7w2.fsf@wilma.george.org>

1 room, 2 SGI's, 3 people looking over each other's shoulders.  3 days
to spec, code, unit test, and alpha system test.  Followed by
design/code review (2 more people).  Followed by small recodes per
review recommendations (mainly refactoring, eliminating c. 10% of
LOC).  

Subject:

Obtain preliminary design of hydraulics system from a lisp app,
delivered in XML (using that app's own dtd).  Convert to neutral
representation with its own XML dtd.  Use neutral rep to drive:

a) Convert to prolog-based schematics package (i.e., generate
predicates), and in turn convert prolog to XML in the DTD of an
analysis package.  (python stoped where prolog started, but we'll
probably eliminate the prolog phase and stay in python in near
future).

b) Convert to XML (another app-specific DTD), for a
fortran-based load analysis and tube sizing app.


Impressions:

a) Domain experts can read python and participate in parts of the code
review (the engineering rules, not the comp sci stuff)

b) Programmers whose expertise is in other languages can ramp up
quickly enough to play within a few days, but can't do the overall
arch of the python code in that time.

c) Refactoring was much easier (and therefore more likely) than in more
verbose languages.

d) XP is hype-of-the-month for practices sometimes known as egoless or
team-centered programming.  For years I've had a motto for project
mgmt: Schedule with a stopwatch, not a calendar.  XP just gives me an
excuse to move at a more comfortable pace.

e) Spiral develpment was critical -- we got a hello, world end-to-end
flow going and then added features.  

f) Hard part was of course working out the semantic conversions --
each app represented a different community with its own conceptual
view of the world.




Jerome Chan <eviltofu at rocketmail.com> writes:

> Has anyone done any extreme programming in python? Care to share your 
> experiences?
> 
> <http://www.extremeprogramming.org/index.html>

-- 
Harry George
hgg9140 at seanet.com


From jerome.quelin at insalien.org  Wed Jun 28 13:59:54 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Wed, 28 Jun 2000 19:59:54 +0200
Subject: [Tkinter] Centering a window in the screen ?
Message-ID: <962215194.1421627556@news.libertysurf.fr>

Hi all,

When using Tkinter, how to center a window in the screen? I tried the
following:

    def center_toplevel_in_screen(widget):
        top = widget.winfo_toplevel()
        sw = top.winfo_screenwidth()
        sh = top.winfo_screenheight()
        w = top.winfo_width()
        h = top.winfo_height()
        x = (sw - w)/2
        y = (sh - h)/2
        geom = '%dx%d+%d+%d' % (w, h, x, y)
        top.geometry(geom)

but this doesn't work: it centers the window but the width and height
returned by the functions are equal to 1.

So, here are some questions:
- Is there a "center-window" in the Tkinter module?
- Which functions am I to call to get the size of a window / a widget?
- What are all those winfo_* functions? Are we supposed to use them or
  should we use some more user-oriented functions (such as geometry(),
  configure(), etc.)?
- By the way, what are those tuples returned by widget.configure()?
- Also, is there a Tkinter FAQ? I went on www.python.org, but there
  isn't much information on Tkinter... Where can I find some (and if
  possible in a downloadable form)?

Thanks,
Jerome
--
jerome.quelin at insalien.org




From thomas at xs4all.net  Wed Jun 14 19:07:58 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 15 Jun 2000 01:07:58 +0200
Subject: Rude question about BeOpen.com
In-Reply-To: <Pine.HPX.4.10.10006141744570.3588-100000@merle.acns.nwu.edu>; from nickgeo@merle.acns.nwu.edu on Wed, Jun 14, 2000 at 05:47:15PM -0500
References: <Pine.HPX.4.10.10006141744570.3588-100000@merle.acns.nwu.edu>
Message-ID: <20000615010758.D26436@xs4all.nl>

On Wed, Jun 14, 2000 at 05:47:15PM -0500, Nicholas Geovanis wrote:
> How will the move of the Python crew to a commercial venture affect the
> Computer Programming For Everybody (CP4E) project?

Read the FAQ:

Q. 7. What is happening to CP4E (Computer Programming for Everybody)?

A. We love this project and are looking for industry sponsors to help us
   expand this multi-disciplinary work to include educators and usability
   experts. Contact sponsor at beopen.com if you are interested.

(http://www.pythonlabs.com/faq.html)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From mhammond at skippinet.com.au  Fri Jun  9 21:08:03 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 01:08:03 GMT
Subject: ActiveX scripting - does it work?
References: <NOd05.12902$e5.188416@news1.frmt1.sfba.home.com>
Message-ID: <Txg05.723$cQ.9083@news-server.bigpond.net.au>

This is a problem with WSH that comes with IE5.5 and/or Win2k - it was
brought to my attention about 2 weeks ago, but I since forgot :-(.  I
believe all other Active Scripting hosts are still OK.  Im looking into
this...

You can make it work by simply creating a .pys file, that just contains
the Python code - ie, just the 2 print lines in your example.  You will
then need to run "cscript.exe foo.pys" or "wscript.exe foo.pys", and it
will work correctly (this is the "old way" and still works).  Note that
you can't simply double-click on the .pys file and have it execute (but
you could arrange for that yourself, of course)

Mark.

"Joe Smith" <JoeSmith at bogusaddress.com> wrote in message
news:NOd05.12902$e5.188416 at news1.frmt1.sfba.home.com...
> I have the following file hello.wsf:
> <package>
> <job id="testp">
> <?job debug="true"?>
> <script language="python">
> import sys
> print 'hello world'
> print 'line 2'
> </script>
> </job>
> </package>
>
> It prints out " hello world" (notice beginning space).  It does not
print
> out "line 2".
> Then cscript crashes and brings up the "Application Error" dialog box
> because it memory at 0x00000024 was referenced.
> Then I see a VC++ run time error dialog box "abnormal program
termination".
> Then I see "Fatal Python error: PyEval_ReleaseThread: wrong thread
state" in
> the console window.
>
> I have no problem getting python to execute in ASP pages.  Is it
possible to
> use python in one of these wsf files?  If so, are there any examples?
>
>
>




From mjeuris at lib.ua.ac.be  Tue Jun 13 08:32:07 2000
From: mjeuris at lib.ua.ac.be (Marc Jeurissen)
Date: Tue, 13 Jun 2000 14:32:07 +0200
Subject: Pyexpat and iso-8859-1
Message-ID: <394629C7.4F7322D8@lib.ua.ac.be>

When I parse a XML-file containing the declaration '<?xml version="1.0"
encoding="iso-8859-1"?>' with Pyexpat, every iso-latin1 character is
being replaced by 2 new characters, the first of wich is nearly always
#195 (&Atilde;), the second one has a decimal value of 64 less than the
original character value.

Some examples:

#233 (&eacute;) becomes #195 + #169 (&copy;)
#231 (&ccedil;) becomes #195 + #167 (&sect;)
#239 (&iuml;) becomes #195 + #175 (&macr;)

Anyone knows what to do about this?

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


From kingwang98 at yahoo.com  Fri Jun 16 21:32:06 2000
From: kingwang98 at yahoo.com (Charles W.W.)
Date: Sat, 17 Jun 2000 03:32:06 +0200
Subject: Question?
Message-ID: <394AD516.C150F0AB@yahoo.com>

 


From seefelds at magellan.umontreal.ca  Tue Jun  6 14:50:16 2000
From: seefelds at magellan.umontreal.ca (Stefan Seefeld)
Date: Tue, 06 Jun 2000 18:50:16 GMT
Subject: strange list behavior
Message-ID: <393D4811.9989DE96@magellan.umontreal.ca>

I'v run into a strange problem with lists. I'm not sure
at all where the problem is so I'll provide some context:

I'v a couple of classes holding string lists (which are scoped
names):

class Module:
     def __init__(self, name): self.__name = name
#...

class Declarator:
     def __init__(self, name): self.__name = name
#...

I create them from another object, which traverses (visits)
an AST:

class Visitor:
    def visitDeclarator(self, node):
        name = self.scope()
        name.append(node.identifier())
        self.__result_declarator = AST.Declarator(name)
    def visitModule(self, node):
        name = self.scope()
        name.append(node.identifier())
        module = AST.Module(name)
#...

what I *though* was happening is that I created temporary variables 'name'
in the respective scopes of all the visitSomething methods and then hand over
the objects to the newly created Node types (Module, Declarator, etc.)
However, what I *observe* is that even though a Module is initialized with
name : ['foo', 'bar']
it may end up being
name : ['foo', 'baz']
i.e. it seems the list is not properly released but instead modified from the outside. 
printing out 'id(name)' reveals indeed that the various temporary 'name' objects 
in the Visitor class have all the same id, i.e. they are the same object. Given
that the AST nodes are created with the list copied by reference, not by value,
I should not wonder that the value changes. Is this a fault of GC ?

Could anybody explain what's going on here and how I get the behavior I want ?
Thanks a lot !

Stefan

_______________________________________________________              
              
Stefan Seefeld
Departement de Physique
Universite de Montreal
email: seefelds at magellan.umontreal.ca

_______________________________________________________

      ...ich hab' noch einen Koffer in Berlin...


From phd at phd.russ.ru  Thu Jun 15 11:54:54 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 15 Jun 2000 15:54:54 +0000 (GMT)
Subject: Suggestion: "Completing" HTTP support in Python.
In-Reply-To: <3948FB36.416393C5@darmstadt.gmd.de>
Message-ID: <Pine.LNX.4.21.0006151553280.8145-100000@fep132.fep.ru>

On Thu, 15 Jun 2000, Robb Shecter wrote:
> h = HTTP('my.host.name', '/loginscript', 8080)
> (code, msg, headers) = h.post()

import urllib
filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript")

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From mwh21 at cam.ac.uk  Wed Jun 14 15:14:56 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 14 Jun 2000 20:14:56 +0100
Subject: stdout
References: <NDBBJMANPNIHDHDJJBHMKEGMCMAA.Michaell.Taylor@reis.com>
Message-ID: <m366rc2hxr.fsf@atrus.jesus.cam.ac.uk>

"Michaell Taylor" <Michaell.Taylor at reis.com> writes:

> I am having a bit of trouble redirecting output from a file back to the
> terminal.  I have:
> 
> import sys
> f=open('c:\output.txt', 'w')
> sys.stdout=f
> blah..
> blah..
> blah..
> sys.stdout='<stdout>'
> print 'finished'
> 
> 
> which does not work.  I've searched the docs but only see references for
> directing to files (which is helpful) but none for redirecting to a screen.
> The last two lines aren't doing the trick.  Any ideas?

You need to save the original sys.stdout somewhere, eg.

oldstdout = sys.stdout
try:
    f=open('c:\output.txt', 'w')
    sys.stdout=f
    blah....
finally:
    sys.stdout = oldstdout

HTH,
Michael

-- 
  On the other hand,  the following areas are subject to boycott
  in reaction to the rampant impurity of design or execution, as
  determined after a period of study, in no particular order: 
    ...                              http://www.naggum.no/profile.html


From elecnix at step.polymtl.ca  Fri Jun 30 00:58:22 2000
From: elecnix at step.polymtl.ca (Nicolas Marchildon)
Date: Thu, 29 Jun 2000 23:58:22 -0500
Subject: Object-oriented programming applied to Web Sites
Message-ID: <5VU65.1500$EA1.166637@news.total.net>

Hi,
As a website programmer, I am concerned about coding with the object
philosophy. Languages such as Java, Perl, and PHP supports classes &
objects. Does anyone knows of a resource (web, book) that covers that
particular subject? The greatest difference between normal applications it
that the client usually requests many different pages, that interract
betwwen themselves.

I began coding with the following structure:

---start of file

switch($action) {
	case "":
		displayForm(); break;

	case "save":
		saveForm($formParams); break;
}

exit;

function displayForm() {
	$objDB = new QueryTool($host, $user, $pass);
	$objDB->query("SELECT * FROM tablename");
	while($row = $objDB->fetchRow()) {
		print $row['COL_NAME'];
	}
	...
}

...

---end of file

Also, I created a class that "this" page should use to access to the
database. Normally, we should use oracle_connect(), but with that class, I
simply create a DB object, calling its connect, query, fetchRow, and close
methods. This hides oracle, so we can change to whatever DB we want, in
just a few hours, because we only need to swap the oracle class code to the new class.

Actually, I create two classes:

class QueryTool_oracle {
	function connect(...
}

class QueryTool extends QueryTool_oracle {
	// Nothing to add, just a class renaming trick
}

An other class is for controlling access with a username+password. Each
file that needs authentification should include the file "verify_login".
That file uses an object to login, verify that the user is logged, and
logoff: class auth {
	function login($username, $password); function verify(); function
	logout();
}

That class uses a cookie and a DB connection to remember who the user is.
The cookie is a long random unique ID. The DB allows to to the translation
cookie-->user_id.

This is how I program my websites for now, but I am wondering if there is
not a nicer way. I want very modular and reusable code.

Any comments on my programming style? Any resources that discusses that
kind of software?

Thanks for your help,

Nicolas Marchildon
Web Developper
Montr?al, Qu?bec, Canada


From gangli at msn.com  Thu Jun 22 14:16:29 2000
From: gangli at msn.com (gangli at msn.com)
Date: Thu, 22 Jun 2000 18:16:29 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <JEk45.262$Sn2.1571@news-server.bigpond.net.au> <slrn8l3rpi.8ku.r2d2@mao.acc.umu.se> <m0o45.6859$5f7.130092@news-east.usenetserver.com> <slrn8l4cmn.e2m.r2d2@mao.acc.umu.se>
Message-ID: <8itl5b$f80$1@nnrp1.deja.com>

Is your python installed on network drive?
NT can be very slow to load file from network.
Also try put python15.dll in the same directory as python.exe's.
If the python15.dll is not in the same directory as the python.exe's,
NT has to search it by following the path setup by path evn.
If your path setup has the network path in front, NT will search network
directory first. If the python15.dll is in C:\winnt\system32 and your
path set it in the first place to search, the search still take time
given that system32 usually has over 1000 files in it.

In article <slrn8l4cmn.e2m.r2d2 at mao.acc.umu.se>,
  r2d2 at mao.acc.umu.se (Niklas Frykholm) wrote:
> >Try my version of python15.dll at http://www.concentric.net/~jjlucsy/
>
> Sorry, I still get the same slow startups.
>
> // Niklas
>


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


From porter at et.byu.edu  Thu Jun 22 11:59:26 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Thu, 22 Jun 2000 09:59:26 -0600 (MDT)
Subject: Frozen socket. (fwd)
Message-ID: <Pine.GHP.4.21.0006220957370.6540-100000@don.et.byu.edu>

OK, I have been able to duplicate the problem that I was having, but I
still don't know what to do about it.  First things first, I suppose.

I made two small programs, a server and a client, and I got the client to
"freeze" in the exact same way.  My test server accepts three messages,
but after that, it freezes.  For some reason that I don't understand, the
client keeps sending messages (about 1000 of them, just like before), even
though the server is not receiving any.  After about 1000 messages, it
freezes, just like the program I am debugging.

Why is the client able to keep sending messages if the server is not
receiving them?  

In short, what is really happening, and how do I safeguard against it?

Here is the source for the two test programs I am talking about.
sample client:
from socket import *
import time 
sock = socket(AF_INET, SOCK_STREAM)
sock.connect("127.0.0.1",2778)
for i in range(5000):
   sock.send("This is a string of approximately the correct
length.          "
   time.sleep(0.001)
   print i

sample server:
from socket import *
s = socket(AF_INET, SOCK_STREAM)
s.bind(""<2778)
s.listen(5)
conn, addr = s.accept()
counter = 0
while 1:
   counter = counter + 1
   data = conn.recv(1024)
   if couter > 3:
      while 1:
         pass
   if data:
      print data

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------

On 16 Jun 2000, David Bolen wrote:

> "C. Porter Bassett" <porter at et.byu.edu> writes:
> 
> > I have looked around, but I haven't been able to figure out why a program
> > would choke on a send() command.  I don't even know if it is because of
> > something I am doing on my end or something happening on the server
> > end.  If it is my fault, how do I fix it?  If it is because of something
> > happening on the server side, how do I compensate?
> 
> Under default conditions, send() may block if there is no room to
> locally buffer the information you want to send.  It's not that it has
> "choked" - just that it's waiting for room to become available.  You
> can set a socket to non-blocking mode, but then you have to handle a
> failure to send() all of your data and keep retrying until it is
> accepted.
> 
> In most cases, this occurs because you have gotten far ahead of the
> receiver - enough so to back up the TCP window size across the network
> and your internal socket buffers.  As long as the systems are still
> actually sending data, it will eventually catch up and you can keep
> sending.
> 
> However, the fact that you say it happens regardless of how quickly
> you send out the strings makes me wonder if the other side is really
> receiving them at all.  Are you absolutely positive that you are
> performing the initial communication properly so that the receiving
> side is really receiving your data?  It may be just that it takes you
> about that many strings to fill up your internal queues while the
> remote system does not receive any data.
> 
> I suppose another possibility is something in the data is causing a
> problem for the remote system such that after receiving that initial
> amount it becomes busy and fails to continue receiving.
> 
> Do you have any other applications that perform this function that you
> might be able to compare/contrast with - having some sort of control
> or pre-existing behavior to compare against might give you some hint
> as to what is going on.  But send() blocking is normally just a
> symptom of something larger happening, and less likely to be a problem
> in the actual send() call itself.
> 
> --
> -- David
> -- 
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 








From poo at stick.org  Tue Jun 13 22:33:41 2000
From: poo at stick.org (Poostick)
Date: Wed, 14 Jun 2000 02:33:41 GMT
Subject: Python + libnet? Link Level Networking?
Message-ID: <3946eb3b.2033142@news.va.mediaone.net>

I am trying to learn network programming and Python together and have
a question.  I am looking for the ability to do data-link layer
frame/packet construction, much like the library by route "libnet"
provides for C (and there is a Perl interface as well).  I could just
use libnet but unfortunately my programming skills don't progress much
farther than compiling C or modifying makefiles/headers, and Perl is
cryptic junk AFAIC.  I am able to use the excellent ip classes to make
raw IP packets and put them in a socket but sockets always come in at
least at the network layer, right? (I'm asking, don't know).  But I
need to write packets (okay I'm playing with Cisco Discovery Protocol)
which are a CDP packet inside of a SNAP frame on Ethernet.  Any ideas?





From wjdandreta at worldnet.att.net  Mon Jun 19 21:30:18 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 20 Jun 2000 01:30:18 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com> <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net> <394E0751.A6D6ACB8@gssec.bt.co.uk>
Message-ID: <KOz35.14013$Xx5.686654@bgtnsc06-news.ops.worldnet.att.net>

Hi Alan.

It's running under ConCurrent DOS, which is a mutli-user/multi-tasking OS
that does not support DOS extenders. The DOS versions of 1.5.2 on the Python
website all use DOS extenders and the 1.2 binaries are the only ones that
work (so far).

Bill

Alan Gauld wrote in message <394E0751.A6D6ACB8 at gssec.bt.co.uk>...
>William Dandreta wrote:
>> It is Python 1.2 because the later versions will not run
>> on my machine.
>
>Can you say why? I had no problem getting python 1.5.2 running
>on my old 50MHz 486 laptop under DOS.
>
>Just curious.
>
>Alan g.





From pj at sgi.com  Tue Jun 27 17:30:28 2000
From: pj at sgi.com (Paul Jackson)
Date: 27 Jun 2000 21:30:28 GMT
Subject: Python  Article Announcement
References: <002d01bfe043$f610f3c0$0101a8c0@one>
Message-ID: <8jb6dk$3g3tt$1@fido.engr.sgi.com>

|> http://www.linuxdev.net/features/articles/ldpytk1_1.shtml=20

Looks like a space char got encoded at the end of that URL.

The following URL works better for me:

    http://www.linuxdev.net/features/articles/ldpytk1_1.shtml
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From dsavitsk at e-coli.net  Tue Jun 20 00:54:35 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Tue, 20 Jun 2000 04:54:35 GMT
Subject: Current date and time
Message-ID: <fOC35.10326$dF.445003@news1.rdc1.il.home.com>

what is the easiest way to get the current date and or time?
something like
>>> print Now()
is what i am hoping to find.
thanks
doug




From herzog at online.de  Thu Jun 29 13:31:32 2000
From: herzog at online.de (Bernhard Herzog)
Date: 29 Jun 2000 19:31:32 +0200
Subject: stderr and stdout
References: <qzI65.16987$_b3.366473@newsread1.prod.itd.earthlink.net>
Message-ID: <m3itusjswb.fsf@greebo.nodomain.de>

"Arint?" <shouldbe at message.com> writes:

> Oh I print stdout and err like so:
>    char *pix = PyString_AsString(temp);
>    Py_XDECREF(temp);
>    appcallback()->status(pix);
> where temp is either errorIO or stdIO.

You better swap the last two lines. pix points to memory owned by tmp
and it could be freed by the DECREF. Why XDECREF, btw? If tmp were NULL
PyString_AsString would segfault.

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


From android at micron.net  Sat Jun 10 23:59:01 2000
From: android at micron.net (Peter Immarco)
Date: Sat, 10 Jun 2000 23:59:01 -0400
Subject: Python and ADO 2.5?
Message-ID: <39430E84.2679DD45@micron.net>

I am thinking of replacing my VBScript ASP's with Python.  VBScript is
excellent for integrating with ADO (Active Data Objects) databases.  Is
Python as facile?

Also I'd like to hear from others that are using Python on Windows 2000
Server edition servers with IIS 5.0 which is my platform.  If the answer
to the above question is yes then I'll be using Python quite a bit on
the server side.  My two books on Python are due to show up on Monday
some I'm rarin' to go.

Thanks,
Peter.



From petersc at stud.ntnu.no  Wed Jun 14 08:39:38 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 14 Jun 2000 14:39:38 +0200
Subject: Can't get NumPy to Install
References: <upqeksgfc39indo8mg3dsc4vfj8oghdr2d@4ax.com>
Message-ID: <39477D0A.C2510C3C@stud.ntnu.no>

Scott Hensey wrote:
> 
> I'm having trouble installing NumPy. Distutils-0.8.2 seemed to install
> without any errors, but when I tried to install Numerical-14 I get the
> following error message.  I'm running RH 6.1.

I have heard there is a new version of Numeric which has this
fixed.

But if you're fine with Numerical-14 you can just change install_path
to extra_path in the setup.py file. It's just a rename.

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From aahz at netcom.com  Thu Jun 15 08:54:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 12:54:57 GMT
Subject: initializing a class
References: <3948a566_5@goliath.newsfeeds.com>
Message-ID: <8iajn1$aib$1@slb6.atl.mindspring.net>

In article <3948a566_5 at goliath.newsfeeds.com>,
Janos Blazi <jblazi at netsurf.de> wrote:
>I need several constructors for a class. So would this pseudo code be in
>order:
>
>class x:
>  ...
>
>class y:
>  ...
>
>class z:
>  def __init__(self,a)
>      if isinstance(a,x):
>         z.t=x.t1+x.t2
>      elif isinstance(a,y):
>         return z(y.x)
>      else:
>         ...
>
>When the constructor for class z is called, its argument is checked. If the
>argument is class x, then the instance is constructed directly. Otherwise,
>when the argument is class y then the constructor is called again with a
>class y argument.

Well, I'm not going to address the issue of "return z(y.x)", because
that gets into metaclass issues I'm not comfortable with.  However, I'm
pretty sure you meant to type

  self.t = a.t1 + a.t2

unless you really meant to define class-level attributes instead of
instance-level attributes.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From jhauser at ifm.uni-kiel.de  Tue Jun 13 03:33:02 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 13 Jun 2000 09:33:02 +0200
Subject: NumPy arrays: how to remove extra axes
References: <slrn8kaov1.1ar.hzhu@rocket.knowledgetrack.com> <m2og56k256.fsf@spiff.home.com>
Message-ID: <87u2eydoi9.fsf@ifm.uni-kiel.de>

Tim Hochberg <tim.hochberg at ieee.org> writes:

> hzhu at rocket.knowledgetrack.com (Huaiyu Zhu) writes:
> 
> > For arrays in Numeric, is there a general way to remove axes of length 1?
> > For example I have an array of shape (1,3,1,4) I want to change it to (3,4).
> > Thanks
> > 
> > 
> > Huaiyu Zhu                               hzhu at knowledgetrack.com
> 
> Here's a function I just tossed together that should do what you
> want. It's only barely tested, so I'd beat on it a bit before trusting
> it.
> 
> -tim
> 
> --------------------------------------
> 
> from Numeric import *
> 
> def compactAxes(A):
>     """Return an array with axes of length 1 removed"""
>     index = []
>     for length in shape(A):
>         if length == 1:
>             index.append(0)
>         else:
>             index.append(slice(None))
>     return A[tuple(index)]
> 
> --------------------------------------
> 
The above function always makes a copy of the data. The next one only
changes the shape.

def DelAxis(m):
    """
    Removes all axis with length one
    """
    sh = m.shape
    new_shape=[]
    for axis_length in sh:
	if axis_length > 1:
	    new_shape.append(axis_length)
    return Numeric.reshape(m,new_shape)

HTH,
__Janko


-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From dgoodger at bigfoot.com  Thu Jun 29 22:47:07 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Thu, 29 Jun 2000 22:47:07 -0400
Subject: Constants and Enums
In-Reply-To: <Pine.GHP.4.21.0006291234020.9839-100000@don.et.byu.edu>
References: <Pine.GHP.4.21.0006291234020.9839-100000@don.et.byu.edu>
Message-ID: <B581826A.5993%dgoodger@bigfoot.com>

See the thread 'Tricks to do "enums"?' starting 2000-05-07, in which I
suggested:

    (VAL1,
     VAL2,
     VAL3) = range(3)

There were several other good suggestions too.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From jwbnews at scandaroon.com  Mon Jun 26 11:35:39 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 26 Jun 2000 08:35:39 -0700
Subject: float safety clarification
References: <Pine.BSF.4.10.10006260856360.16661-100000@chi.pair.com>
Message-ID: <jwbnews-F2AF64.08353926062000@news.olympus.net>

In article <Pine.BSF.4.10.10006260856360.16661-100000 at chi.pair.com>, 
Michal Wallace <sabren at manifestation.com> wrote:

> Hey all,
> 
> I'm building a shopping cart, so I want to be able to store and work
> with prices. I've always been told that you shouldn't use floats for
> money. From past discussions here, I even understand why floats can be
> inaccurate... But in actually trying it out, I can't seem to produce
> any real-world situations relevant to my cart where using floats would
> screw me up... I guess I'm asking for a sanity check here.
> 
> Are floats safe to use for money values if you're only going to be
> adding and subtracting them, or multiplying them?
> 
> I tried this:
> 
> >>> for x in range(100):
> ...    print "%.2f" % (float(x)/100),
> 
> And it looks right. I haven't gotten any funny results with just
> adding or subtracting.. 
> 
> Here are the operations I need to perform:
> 
>   - assign a price
>   - multiply price by (whole number) quantity
>   - add a bunch of prices together
>   - calculate a sales tax
>   - possibly some day give a percent discount 
> 
> I thought sales tax might be a problem, but I haven't been able to
> come up with a problem where a percent sales tax represented as a
> float causes any float problems, either.
> 
> Basically, I can't find a reason not to use floats. 
> Am I asking for trouble?
> 
> Cheers,
> 
> - Michal

You'll probably be OK as long as you don't compare (with 0.00, or with 
any other specific value).  Are you doing something like offering free 
shipping on orders of $100 or more?  Comparison against 100 or 100.00 
could produce a wrong answer if the sum of the items is slightly less 
than 100.  Same for free shipping "over $100"...you might have a total 
"equal" to 100 for which sum > 100.00 is true (less bad:  you aren't 
failing to give something you should have, landing in FTC hell).

I wouldn't do it...I see no reason to work with decimal money with a 
method which expresses only 4% of the possible 100 different cents 
values exactly (.00, .25, .50, .75).

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From olczyk at interaccess.com  Fri Jun  9 14:30:36 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Fri, 09 Jun 2000 18:30:36 GMT
Subject: Debuggers.
Message-ID: <39483789.94409484@nntp.interaccess.com>

Basic question what are the debuggers available for python.
( pdb is OK, but I wish I could restart a project. )
Everyone seems to say that pythonwin comes with one, but I don't see
it. OTOH I have v 1.5.1


From no_spam at umich.edu  Wed Jun  7 16:58:50 2000
From: no_spam at umich.edu (Jeff)
Date: Wed, 7 Jun 2000 13:58:50 -0700
Subject: Capturing Extension output (newbie)
Message-ID: <8hmd2b$17e@news.or.intel.com>

Hi, I am trying to use python to extend some C functions someone wrote long
ago.  In all the examples I have seen on the web and in tutorials, none of
them use anything like "printf" !!  How can I get python to capture stuff
that the C function sends to STDOUT???

Jeff




From 12klat at sightreader.com  Wed Jun  7 01:30:32 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 22:30:32 -0700
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu> <7wc%4.7928$r7.1560935@news1.rdc2.tx.home.com>
Message-ID: <393DDDF8.DE0401BB@sightreader.com>



Rainer Deyke wrote:

> Richard P. Muller <rpm at wag.caltech.edu> wrote in message
> news:393D2797.4E4EE7CC at wag.caltech.edu...
> > What else should we be doing to make Python *the* programming language
> > rather than *a* programming language?
>
> Write an optimizing compiler that increase performance by a factor of at
> least 100 and decreases memory usage by a factor of at least 100.  Python in
> its current incarnation is too inefficient to be useful as primary language
> for shrink-wrapped software.

Because of it's dynamic nature, Python simply can't be optimized all that much
(read about all that type inference stuff and you'll see what I mean).
Fortunately that just doesn't matter!

98% of code does not need to run fast.  You write the other 2% in C.
98% of all objects (on a per-object basis) do not need to be packed tight.  You
implement the other 2% in C.

Conclusion: The primary language does not need to be efficient.

Any questions?

> --
> Rainer Deyke (root at rainerdeyke.com)
> "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/ed2f011e/attachment.vcf>

From ge at nowhere.none  Mon Jun 26 17:05:28 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 26 Jun 2000 21:05:28 GMT
Subject: SMTP with Python 1.6
References: <VqK55.1715$iN5.345609@ptah.visi.com> <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com> <8j7n8d$4u64$1@www.univie.ac.at> <8j80qa$5d0c$1@nntp6.u.washington.edu>
Message-ID: <sAP55.1769$iN5.361158@ptah.visi.com>

In article <8j80qa$5d0c$1 at nntp6.u.washington.edu>, Donn Cave wrote:
>Quoth ge at nowhere.none (Grant Edwards):

>>> 'Localhost' will not work. You have to make this the address of
>>> a valid SMTP-server.
>>
>> Gosh, on all my computers, "localhost" is the address of a
>> valid SMTP-server.  Are your computers all broken?
>
>Some choice, broken vs. broken into.

The only SMTP server I run on a machine accessible to the
outside world is qmail.  I am quite confident that it is not a
point of vulnerability.

>SMTP service is only required for hosts that accept or redistribute mail.
>If you know "localhost" is going to work, that's fine, but in the general
>case it's liable not to, and sys-admin isn't even likely to be very
>sympathetic.

I guess I'm spoiled.  I exped all of my computers to be willing
to accept e-mail.  Most of them won't relay incoming SMTP
requests, but that's by choice.

-- 
Grant Edwards                   grante             Yow!  Now my EMOTIONAL
                                  at               RESOURCES are heavily
                               visi.com            committed to 23


From gmcm at hypernet.com  Tue Jun 13 16:40:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 13 Jun 2000 20:40:32 GMT
Subject: Paring down Python (physically)
References: <39465328.2352344@news.ark.com>
Message-ID: <8F52AC723gmcmhypernetcom@199.171.54.194>

[posted and mailed]

Alex Walker wrote:

>I've been tasked with a question of how to use Python on a WinNT
>system where isn't installed.  Of the options available (and I don't
>thing I've discovered them all yet,) installing with our app just
>those Python files absolutely required to run a script is the
>front-runner.  


http://www.mcmillan-inc.com/install1.html

- Gordon


From cut_me_out at hotmail.com  Fri Jun 16 20:35:10 2000
From: cut_me_out at hotmail.com (Alex)
Date: 16 Jun 2000 20:35:10 -0400
Subject: Idle question about process threads (nothing to do with editors.)
Message-ID: <etdaeglqh4x.fsf@w20-575-109.mit.edu>

Hi.  If I create an instance of a subclass of threading.Thread and start
it without binding it to a reference (like 'My_thread ().start ()')
there must be some references to it that keep it from getting garbage
collected.  Does anyone know what they are?

Alex.


From gmcm at hypernet.com  Thu Jun 15 08:16:03 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 15 Jun 2000 12:16:03 GMT
Subject: Modulator.py in Pythonwin?
References: <eviltofu-B92575.07393815062000@news.earthlink.net>
Message-ID: <8F545D546gmcmhypernetcom@199.171.54.154>

Jerome Chan wrote:

>I noticed that modulator.py is missing in the tools folder of Pythonwin. 
>Is there a way I can get it without downloading the entire source code 
>for Python?

First, Pythonwin is part of the Win32 extensions and doesn't have a tools 
folder. Second, yes, you'll need the source distribution of Python. While 
you now *can*, on Windows, build extensions without access to the source, 
you won't be able to debug, and you won't be able to grep.

like-shooting-skeet-blindfolded-ly y'rs

- Gordon



From andres at corrada.com  Wed Jun 28 11:06:19 2000
From: andres at corrada.com (andres at corrada.com)
Date: Wed, 28 Jun 2000 11:06:19 -0400
Subject: Controlling Acrobat Reader with COM
In-Reply-To: <s7khls0in7rrfbqnqdo4cha6chh9qqhstr@4ax.com>; from  on Tue, Jun 27, 2000 at 12:07:08PM -0400
References: <200006262159.AA3991339306@mail.mamey.com> <s7khls0in7rrfbqnqdo4cha6chh9qqhstr@4ax.com>
Message-ID: <20000628110619.C29827@corrada.com>

On Tue, Jun 27, 2000 at 12:07:08PM -0400,  wrote:
> On Mon, 26 Jun 2000 21:59:51 -0400, "Andres Corrada-Emmanuel"
> <andres at mail.mamey.com> wrote:
> 
> >Hi,
> >
> >I'm trying to develop an application that will print (as in a paper copy) a PDF file automatically on Win32. I've used Mark Hammond's genpy utility to create a wrapper around "Acrobat Control for ActiveX". I'm able to get a handle on Acrobat but cannot execute any of its registered methods:
> >
> >>>> from win32com.client import Dispatch
> >>>> acro = Dispatch('PDF.PdfCtrl.1')
> >>>> acro.LoadFile(fileName='C:\Temp\ReadMe.pdf')
> >Traceback (innermost last):
> >  File "<interactive input>", line 1, in ?
> >  File "C:\Program Files\Python\win32com\gen_py\CA8A9783-280D-11CF-A24D-444553540000x0x1x3.py", line 30, in LoadFile
> >    return self._oleobj_.InvokeTypes(0x2, LCID, 1, (11, 0), ((8, 0),),fileName)
> >com_error: (-2147418113, 'Unexpected failure', None, None)
> >
> >I'm very much a COM newbie, so this traceback means very little to me. Could someone enlighten me on what is going wrong in this case? I have no problem controlling Word with my setup so I'm stumped on this one.
> >
> >Andres Corrada
> Try typing these strings directly into the interpreter.
> 
> >>> print 'c:\test\readme.pdf'
> c:	est
> eadme.pdf
> >>> print "c:\test\readme.pdf"
> c:	est
> eadme.pdf
> >>> print 'c:\\test\\readme.pdf'
> c:\test\readme.pdf
> >>> print 'c:/test/readme.pdf'
> c:/test/readme.pdf
> -- 

Thanks for pointing out that I was not escaping the backslashes. The error,
however will not go away when I use

acro.LoadFile(fileName='C:\\Temp\\ReadMe.pdf')

I've noticed that (I'm using PythonWin) when I get to the first parenthesis
on LoadFile, the editor offers the hint:

fileName=<PyOLEMissing object at 108c1010>

Does that have anything to do with the subsequent error?

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From parkw at better.net  Tue Jun 20 01:03:56 2000
From: parkw at better.net (William Park)
Date: Tue, 20 Jun 2000 01:03:56 -0400
Subject: Current date and time
In-Reply-To: <fOC35.10326$dF.445003@news1.rdc1.il.home.com>; from dsavitsk@e-coli.net on Tue, Jun 20, 2000 at 04:54:35AM +0000
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com>
Message-ID: <20000620010356.A892@better.net>

On Tue, Jun 20, 2000 at 04:54:35AM +0000, dsavitsk wrote:
> what is the easiest way to get the current date and or time?
> something like
> >>> print Now()
> is what i am hoping to find.
> thanks
> doug

>>> import time
>>> time.ctime(time.time())
'Tue Jun 20 01:02:03 2000'

--William



From jimsp at ichips.intel.com  Wed Jun 21 13:14:04 2000
From: jimsp at ichips.intel.com (jimsp)
Date: Wed, 21 Jun 2000 10:14:04 -0700
Subject: Swig and micro$oft __int64 typemaps
Message-ID: <3950F7DC.50C3341@ichips.intel.com>

Has anyone created typemaps for SWIG for Micro$oft __int64 to bind them
to

Python long ints?

searched deja, python.org, Python FAQ, Parnassus, the swig mailing list
etc...

James Spears
jimsp at ichips.intel.com




From aahz at netcom.com  Mon Jun 12 14:24:07 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jun 2000 18:24:07 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <8i39s7$3u4$1@nntp9.atl.mindspring.net>

In article <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf at 4ax.com>,
 <mkx at excite.com> wrote:
>
>So my question is this: Is Python a more "Windows-friendly" language?
>Is it's base structure less related to any one specific operating
>system? Is learning Python an exercise in learning the language, and
>not a Unix tutorial? Can I be successful even if I do not use Unix ;)
>?

There are some parts of Python that are somewhat more Unix-friendly
(dealing with processes, primarily).  Overall, though, Python is about
as platform-neutral as anything else available, and I would definitely
say that Python is more Windows-friendly than Perl.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From echuck3 at my-deja.com  Thu Jun 15 09:18:30 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Thu, 15 Jun 2000 13:18:30 GMT
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com> <01bfd2ed$42b67880$29cf8ad1@daves-dell> <39463BA7.43A3CD65@gssec.bt.co.uk> <39474D1D.F959969C@prescod.net> <8F545A89Agmcmhypernetcom@199.171.54.154>
Message-ID: <8ial2l$opl$1@nnrp1.deja.com>

In article <8F545A89Agmcmhypernetcom at 199.171.54.154>,
  gmcm at hypernet.com (Gordon McMillan) wrote:
> Paul Prescod wrote:
>
> >Try double clicking a PDF document with a space in its path. Adobe
> >Acrobat 4 will not find it. It's annoying as hell because Internet
> >Explorer puts files from the Web in the "Temporary Internet Files"
> >directory.
>
> That's the installer's fault, not Acrobat 4's. Just (double)quote the
%1 in
> the open action for the file type.
>
> I get a big kick out of MS's habitual use of paths with embedded
spaces.
> Sort of a barometer of hubris.
>
> - Gordon

Perhaps they do it on purpose so that programmers will get their act
together and support pathnames with spaces. As someone already pointed
out, this has been supported for many years on UNIX and I guess about 5
on Windows.

I don't care for Microsoft, but I have to say it's the developers who
have screwed up this time.

What's wrong with the developers at Adobe that [a] they can't handle a
space and [b] their pre-release testing didn't find this?

-Chuck


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


From rumjuggler at cryptarchy.org  Sun Jun 18 19:15:18 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 18 Jun 2000 23:15:18 GMT
Subject: command-line parser
References: <kdgqkscjmqa1aqs78l6grsmkonfmjcpsb6@4ax.com> <20000618151125.A22384@activestate.com>
Message-ID: <v1mqksou1vnlilrr2l4d7f8bfhqk1r29qi@4ax.com>

On Sun, 18 Jun 2000 15:11:25 -0700, Trent Mick <trentm at activestate.com>
wrote:


> Some comments below.
[comments excised].

Some advice taken, and it's a lot more readable, or seems so to me.
An instance of the Option class can take a callback that gets called on its
value, so if a particular option's value needs to be an int, say, you could
instantiate it with

Option(<short>, <long>, 1, val_cb = int)

If the callback raises an exception, the option's help() function prints
its description

# arglist.py
class Error(Exception):
    pass

class Option:
    def __init__(self, short, long=None, has_value=None, default_value =\
                 None, val_cb = None, description = None):
        self.short = short
        self.long = long
        self.has_val = has_value
        self.val_cb = val_cb
        self.value = default_value or None
        self.description = description
        
    def get_keys(self):
        return self.short, self.long
    
    def handle(self, val):
        if self.has_val:
            if val:
                self.value = val
            elif self.value is None:
                    raise Error, 'Option %s requires a value' % opt
        else:
            self.value = 1
        if self.val_cb:
            try:
                self.value = self.val_cb(self.value)
            except:
                self.help()
        del self.short, self.long, self.val_cb, self.has_val

    def help(self):#rather simpleminded, yes?
        if self.description:
            print
            print self.description
    
        
class Arguments:
    def __init__(self, *opts):
        self.__dict__['nonopts'] = []
        self.__dict__['dict'] = {}
        for opt in opts:
            short, long = opt.get_keys()
            if short and long:
                self.dict[short] = self.dict[long] = opt
            elif short:
                self.dict[short] = opt
            else:
                self.dict[long] = opt

    def process(self, options):
        for opt in options:
            if opt[0] <> '-':
                self.nonopts.append(opt)
                options.remove(opt)
        for opt,val in map(splitstrip, options):
            self.dict[opt].handle(val)

    def __getattr__(self, key):
        if not self.dict.has_key(key):
            raise Error, 'No option %s' % key
        return self.dict[key].value

    def get_nonopts(self):
        return self.nonopts
            

def splitstrip(opt):
    return optsplit(optstrip(opt))

def optstrip(opt):
    while opt[0] == '-':
        opt = opt[1:]
    return opt

def optsplit(opt):
    if ':' in opt:
        return opt.split(':',2)
    elif '=' in opt:
        return opt.split('=',2)
    return opt, None


From redaelli at inc.it  Mon Jun  5 10:52:52 2000
From: redaelli at inc.it (Paolo Redaelli)
Date: Mon, 05 Jun 2000 16:52:52 +0200
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
Message-ID: <393BBEC4.5BC343FB@inc.it>

Look at the this short program:

=======
#!/usr/bin/env python

s = "0.000000E+00"
print "Converting",s," into: ", float(s)
print "Importing gtk"

import GDK
from gtk import *
from gnome.ui import *
import GdkImlib
print "Converting",s," into: ", float(s)
=======
Here's its output:
=======
[paolo at aldebaran paolo]$ python float.py
Converting 0.000000E+00  into:  0.0
Importing gtk
Converting 0.000000E+00  into: 
Traceback (innermost last):
  File "float.py", line 11, in ?
    print "Converting",s," into: ", float(s)
ValueError: invalid literal for float(): 0.000000E+00
=======

The only lesson I can extract from it is that GTK breaks float!

-- 
 ______________________________
/        Paolo Redaelli        \
|        redaelli at inc.it       |
| http://gnupaghe.netpedia.net |
\------------------------------/



From coronel_jadonNOSPAM at teleline.es  Mon Jun 26 05:47:13 2000
From: coronel_jadonNOSPAM at teleline.es (JD)
Date: Mon, 26 Jun 2000 09:47:13 GMT
Subject: Lista de correo sobre Python
References: <Bi%35.2093$Hc7.11930@telenews.teleline.es> <8iu290$puu$1@nnrp1.deja.com>
Message-ID: <BEF55.24$8V.213@telenews.teleline.es>

Un foro sobre Python en:
python_dev-subscribe at egroups.com

Muchos usuarios dispuestos a resolver tus dudas, a ayudarte y deseando que
compartas tus conocimintos.


Un saludo.





From petersc at stud.ntnu.no  Thu Jun 15 01:56:50 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Thu, 15 Jun 2000 07:56:50 +0200
Subject: free python hosting?
References: <8geo5g$3rh$1@bob.news.rcn.net>
Message-ID: <39487022.5C2D0F11@stud.ntnu.no>

Z 3 Penguin wrote:
> 
> i saw a while back in c.l.p and in the Knowledge Base a thread about python
> hosting.  anyone know about a free hosting provider that offers python?

got to http://python.faqts.com
you'll find the result there.

now-if-I-could-get-free-zope-hosting-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From akuchlin at mems-exchange.org  Fri Jun  2 11:01:43 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: Fri, 2 Jun 2000 11:01:43 -0400 (EDT)
Subject: Upcoming Starship changes
Message-ID: <200006021501.LAA11161@amarok.cnri.reston.va.us>

Some more changes will be happening on Starship this weekend; it'll be
moving to a new machine hosted by BeOpen.  The new machine is larger
and faster, and BeOpen's staff will worry about keeping the machine
running.

The plan is to make the transition over this coming weekend, June 3-4,
in several steps.  Starship will stay up and working while the new
machine is being configured.

1) Everyone's files will be copied from the current starship.

2) The login message will then be changed, to warn users that
subsequent changes to their files on the old Starship won't be carried
over.

3) The new machine will be configured with FTP and HTTP and e-mail.

4) The DNS records will changed so starship.python.net and
www.python.net point to the new machine.  DNS changes usually take a
day or two to propagate, so by the middle of next week everyone should
be accessing the new machine, assuming we don't run into any snags
along the way.

The old starship will still be accessible through its IP address:
209.50.192.113.

What this means to you:

* If you're not modifying files on Starship over the weekend, we hope
  you won't have to do anything.  Once the DNS changes have propagated,
  'ssh starship.python.net' will get you to the new machine, and
  everything will be much the same.

* If you're going to be modifying things over the weekend, pay close
  attention when you log in, so you know when we've copied the files.
  If you modify your files after we've copied them, it'll be your
  responsibility to copy the changes from the old machine to the new
  one.

* If you only access starship in order to download files, you don't
  care about all this.  The old starship will stay up, and at some
  point the IP address pointed to by starship.python.net will change,
  but that shouldn't be noticeable to you.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
And then it crumbled in his hand. It was just dust, sand, a glittering
multicolored sand that fell away into the chilly wind at the end of the world.
  -- Dream uncreates the Land, in SANDMAN #36: "Over the Sea to Sky"





From jepler.lnk at lnk.ispi.net  Wed Jun  7 19:21:36 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Wed, 07 Jun 2000 23:21:36 GMT
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
References: <393BBEC4.5BC343FB@inc.it> <slrn8jptf0.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl> <393D1155.D067A821@decrc.abb.de> <m33dmqygdj.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <slrn8jt5ck.779.jepler.lnk@potty.housenet>

On 06 Jun 2000 16:32:40 +0100, Michael Hudson
 <mwh21 at cam.ac.uk> wrote:
>My money's on so bizzarre interaction with libc, myself.

I'm guessing that 'import gnome.ui' sets the locale, including the
numeric format:
	$ env LC_NUMERIC=en_US python -c 'import gnome.ui; print float("1.1")'
	1.1
	$ env LC_NUMERIC=de_DE python -c 'import gnome.ui; print float("1.1")'
	Traceback (innermost last):
	  File "<string>", line 1, in ?
	Value Error: invalid literal for float(): 1.1

Jeff


From kpmurphy at my-deja.com  Thu Jun 29 15:53:24 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Thu, 29 Jun 2000 19:53:24 GMT
Subject: bbox?
References: <8jg776$ooa$1@nnrp1.deja.com>
Message-ID: <8jg9fc$qls$1@nnrp1.deja.com>

In article <8jg776$ooa$1 at nnrp1.deja.com>,
  Keith Murphy <kpmurphy at my-deja.com> wrote:
> does anyone know what bbox is?  I see it in the context of the
> PhotoImage put(data, bbox).  Thanks,
>
> Keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

The bounding box is returned as a 4-tuple defining the left, upper,
right, and lower pixel coordinate.

found it...

here's what i was doing..
self.photo.put(tuple(["#000000"],), tuple([5,5,6,6]))
it draws a single black pixel at (5,5)

-->keith


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


From Alessandro.Bottoni at think3.com  Wed Jun 21 08:34:17 2000
From: Alessandro.Bottoni at think3.com (Alessandro Bottoni)
Date: Wed, 21 Jun 2000 14:34:17 +0200
Subject: Any Python project similar to Jserv and Cocoon?
Message-ID: <6D8A17398E28D3119F860090274DD7DBF13C74@pces.cadlab.it>

Hi Snake-tamers,
lately, I have been playing with XML, Apache, Jserv and Cocoon (see
http://www.apache.org , http://java.apache.org and http:/xml.apache.org ).

All of these tools are written in Java (and this makes almost impossible to
make a "distribution" CD-ROM for them, because of the strict licence policy
of Sun...).

As you know, Java and Python are somehow similar so I wonder if anybody is
thinking to use Python for building a set of XML server tools, more or less
equivalent to Jserv and Cocoon, for the Apache web server.

Does anybody know of any project like that?

Thanks in advance for your attention.
----------------------------------------------------------------
Alessandro Bottoni (Alessandro.Bottoni at Think3.com)
Web Programmer @ Think3 inc. (www.think3.com)
I do not speak for think3 and they return the favour




From cut_me_out at hotmail.com  Mon Jun 19 18:25:58 2000
From: cut_me_out at hotmail.com (Alex)
Date: 19 Jun 2000 18:25:58 -0400
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news and   links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net> <394E96D2.5787F2A3@gmx.de>
Message-ID: <etd8zw15mvd.fsf@w20-575-109.mit.edu>

Down the bottom of the page, you will see a link saying 'View original
usenet format.'  If you click on that, you can get the proper formatting
for python scripts.

http://www.deja.com/getdoc.xp?AN=635109618&fmt=text

But I wish the Dr. Dobb's weekly would give these links instead of the
original Deja links.

Alex.


From wh at msgto.com  Mon Jun 26 19:01:28 2000
From: wh at msgto.com (Walter Hofmann)
Date: Tue, 27 Jun 2000 01:01:28 +0200
Subject: learning-python question
References: <8j8ep3$tkn$1@nnrp1.deja.com>
Message-ID: <slrn8lfo68.8ck.wh@frodo.uni-erlangen.de>

On Mon, 26 Jun 2000 20:34:43 GMT, JeremyLynch at fsmail.net <JeremyLynch at fsmail.net> wrote:
>
>Why can count_func() not see count (it throws up a 'name'
>or 'attribute' error)

see http://www.python.org/doc/FAQ.html question 4.57

Walter



-- 
begin  LOVE-LETTER-FOR-YOU.txt.vbs
Hi! I'm a signature virus. Please help me spread!
end


From aahz at netcom.com  Wed Jun 14 11:28:32 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 15:28:32 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de>
Message-ID: <8i88b0$hlc$1@nntp9.atl.mindspring.net>

In article <3946760F.C5E46FD3 at muc.das-werk.de>,
Thomas Thiele  <thiele at muc.das-werk.de> wrote:
>Thomas Thiele deleted the attribution for Aahz:
>>
>>   I do have to wonder whether you've yet learned to think in
>> Python; it's possible that you're trying to write C++ code in Python,
>> the way some Perl people find writing Python difficult.
>
>May be. But what are typical mistakes?  How should a REAL python
>program looks like.  I design classes like in C++, thats true. But I
>also use typical python features like lists, dicts, map etc. I'm not
>trying to write everything by myself.

Let me give you an example to illustrate what Martijn wrote:

Suppose you needed a file object to connect to some chunk of memory.  In
other languages, you would need to inherit from the Python file object
in order to ensure that random functions that expect file objects would
still work.  In Python, that's completely unnecessary.  It's probably a
lot simpler to create your own class, complete with open(), close(),
read(), and write() methods -- and because you've maintained *interface*
consistency, you can use your new class *ANYWHERE* a file object would
go.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From aahz at netcom.com  Thu Jun 29 09:24:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jun 2000 13:24:00 GMT
Subject: newbie questions about imports, threads, and optimizations
References: <8jd7f1$kic$1@murrow.sp.trw.com>
Message-ID: <8jfilg$9sr$1@slb6.atl.mindspring.net>

In article <8jd7f1$kic$1 at murrow.sp.trw.com>,
Bryce Sellers <bryce.sellers at trw.com> wrote:
>
>Second, are there any tutorials or even reference manuals out there on using
>Python with threads?  I haven't so much as even seen any sample code,
>although I admit I haven't looked very hard.

I've posted a few things (unfortunately, DejaNews is doing poorly as an
archival service these days), but there's nothing up on a web site yet.
After I do my presentation at OSCON, I'm planning on whipping up
something for the web.  If you've used threads in another language, it's
pretty easy to use Python threads; even if you've never used threads,
it's still easy, but you need to know a few tricks.

In any event, take a look at the threading module (*NOT* thread) and see
if that's enough to get you started.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From spex66 at my-deja.com  Tue Jun 20 08:15:56 2000
From: spex66 at my-deja.com (spex66 at my-deja.com)
Date: Tue, 20 Jun 2000 12:15:56 GMT
Subject: ReRe: Kjbuckets-Module, Python 1.6 and NT
References: <394E4B02.5F4A0AB4@szs.ira.uka.de>
Message-ID: <8inn9m$3q5$1@nnrp1.deja.com>

Hi Robert,

Robert Kern <kern at caltech.edu> schrieb in im Newsbeitrag:
394E751F.DE9E5FF4 at caltech.edu...
> That is, kjbuckets has several extension types.  There will be lines
> like the following (taken from the FAQ, not kjbuckets):
>
>   statichere PyTypeObject bstreamtype = {
>       PyObject_HEAD_INIT(&PyType_Type)
>       0,
>       "bstream",
>       sizeof(bstreamobject),
>   ...
>
>
> Replace &PyType_Type with NULL.  Then add something like the following
> line to the initialization module:
>
>       bstreamtype.ob_type = &PyType_Type;
>
> Repeat for all extension types.

work for this problem. I added your mentioned code here (maybe helpful
for
Thomas Freytag, or for me, if I'm totally wrong :))

/* internal allocation function for table wrappers */
static object * newWrapper(long expectedsize, enum BucketFlag flag)
{
  /* allocate one wrapper */
  TableWrapper *wp;
  Dprint(("WnewWrapper\n"));
  wp = NEW(TableWrapper, 1);
  if (wp == NULL) {
    return err_nomem(); /* allocation failure */
  }
  switch (flag) {
  case SETFLAG:
    kjSettype.ob_type = &PyType_Type;
    wp->ob_type = &kjSettype; break;
  case DICTFLAG:
    kjDicttype.ob_type = &PyType_Type;
    wp->ob_type = &kjDicttype; break;
  case GRAPHFLAG:
    kjGraphtype.ob_type = &PyType_Type;
    wp->ob_type = &kjGraphtype; break;
  default:
    err_setstr(SystemError, "invalid internal table flag");
    return NULL;
  }


compiled the code without an error or a warning, into kjbuckets.dll and
dropped it in the Python/DLLs directory

then I get:
>>> from kjbuckets import *
Traceback (innermost last):
  File "<pyshell#1>", line 1, in ?
    from kjbuckets import *
ImportError: dynamic module does not define init function
(initkjbuckets)

any ideas??? the initkjbuckets is declared:
void
initkjbuckets()
{
  initmodule("kjbuckets", kjbuckets_methods);
}

/* end of kjbuckets module by Aaron Watters */



thanks

Peter Arwanitis
(=PA=)


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


From olczyk at interaccess.com  Tue Jun 20 19:03:29 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Tue, 20 Jun 2000 23:03:29 GMT
Subject: Output from a python script to a calling program.
Message-ID: <3951f692.948648187@nntp.interaccess.com>

When I call a python script from another program, I notice several
things.

1) It sort of bunches the output. Specifically it sits and waits for a
large block of output, then prints, stops printing, sits and waits
then prints the next batch.
2) Sometimes it seems that parts of the output are printed in
different order them I would expect.

Any idea of how to fix this?


From mhammond at skippinet.com.au  Wed Jun 14 15:33:05 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:33:05 GMT
Subject: Python and Video for Windows?
References: <39476A9C.DFE84D0@stud.ntnu.no>
Message-ID: <8i8mlg$c84$1@nnrp1.deja.com>

In article <39476A9C.DFE84D0 at stud.ntnu.no>,
  peter at schneider-kamp.de wrote:
> I am currently working on a robot control project and I have
> to use a Video for Windows compatible video camera.
>
Pythonwin has a demo that uses the MS ActiveMovie (or whatever it is
called) control.  This can play arbitary media clips from within
Pythonwin.  By default, it tries to play the first .avi file it finds
in your Windows directory.

Im not sure if that is what you were asking, tho!

Mark.


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


From fer at parnasus.kitenet.net  Sun Jun 18 10:15:16 2000
From: fer at parnasus.kitenet.net (Fermin Martinez)
Date: Sun, 18 Jun 2000 16:15:16 +0200
Subject: dpkg-python??
Message-ID: <khlii8.j62.ln@127.0.0.1>

well, does anybody knows where could I find documentation about this module???
It doesn't seems to exist anything...

thanks.


-- 
Chips de computadora, disponibles en dos sabores: batata y silicio.
		-- Www.frases.com. 


From phd at phd.russ.ru  Thu Jun  8 12:12:34 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 8 Jun 2000 16:12:34 +0000 (GMT)
Subject: PLEASE HELP ON SOLARIS!!!
In-Reply-To: <393FBD20.E4664BD4@usa.alcatel.com>
Message-ID: <Pine.LNX.4.21.0006081611570.796-100000@fep132.fep.ru>

On Thu, 8 Jun 2000, Sara Fowler wrote:
> Is ANYBODY out there trying to run Python on Solaris?  If so, how did
> you get it to build?  

   I run it on 2.5.1. No problems - just ./configure and make...

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From adams_s at lab.eng.usyd.edu.au  Tue Jun 20 00:39:49 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Tue, 20 Jun 2000 14:39:49 +1000
Subject: Help: raw_input() Error
Message-ID: <8ims8l$som$1@news1.wire.net.au>

Hi, (this is using PythonCE - a windowsCE version of Python 1.5.2)

when I try and use raw_input() it gives a ValueError: I/O operation on a
closed file.

Question: how do I work around this?
I've never encountered this problem on any other Python interpreter

thanks, Steven




From akuchlin at mems-exchange.org  Thu Jun  8 09:37:32 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Jun 2000 09:37:32 -0400
Subject: ANNOUNCE: early alpha Quixote Web development tool
References: <200006080135.VAA04781@mira.erols.com> <393F860A.EAC32B61@eq.uc.pt>
Message-ID: <3d1z28z42r.fsf@amarok.cnri.reston.va.us>

Pedro Vale Lima <eq3pvl at eq.uc.pt> writes:
> If i get it right this will be a dtmless, zodbless, aquisitionless zope.
> Just a Zpublisher-like ORB with a ASP-like template language.
> Why not (re)use ZORB?

Too complicated.  One of the overriding goals for Quixote was to
choose simplicity at every decision point (well, *almost* every
decision point).  It does use the HTTPRequest and HTTPResponse objects
from Zope, though.

--amk


From eric at estinc.com  Thu Jun 29 11:36:53 2000
From: eric at estinc.com (Eric Lee Green)
Date: Thu, 29 Jun 2000 15:36:53 GMT
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com> <ku66qtv945.fsf@lasipalatsi.fi>
Message-ID: <395B6D35.DA83A7ED@estinc.com>

Erno Kuusela wrote:
> >>>>> "Eric" == Eric Lee Green <eric at estinc.com> writes:
>     Eric> Insert speed is horrible -- inserting data takes
>     Eric> forever and a half, though PostGreSQL 7.0 now has a bulk
>     Eric> import facility that may make it faster (I haven't had a
>     Eric> chance to benchmark it yet).
> 
> hmm. there are some things that can make inserts slow:
> 
> 1) there is a switch telling postgresql whether it should flush to
> disk after each transaction. this calls the flush() system call. its
> implementation is very inefficient on linux.  if you are running on
> linux and can live with the reduced robustness, turn that feature off.

Unfortunately, when you do this, you lose the ability to guarantee
consistency. Without some redesign of the PostGreSQL storage manager (which is
underway, BTW), it is very easy to get an inconsistent table this way. For
some applications this does not matter (I don't care TOO much if my cookies
table gets corrupted!). For other applications (like payroll :-) it definitely
DOES matter.  
 
> 2) if you are using sql insert statements without surrounding
> begin/commit (ie in autocommit mode), a transaction for each insert
> makes things slow.

Correct. I have made inserting data as much as 10 times faster by batching
inserts within a transaction. A word of warning, though: there *IS* a limit to
transaction size. That bit me once when I wanted to add 165,000 records in one
fell swoop. Sorry!
 
> 3) using sql insert statements is not itself the fastest thing int the
> world. the pygresql connection objects have a inserttable() method for
> doing something similar (i think) to what COPY does with the "psql"
> interactive interactive terminal.

This is new for PostGreSQL 7.0, I believe. I have not yet benchmarked it. 

> try these, i bet you will have much better luck.

Yes, I am aware of all the tricks for 6.x and below. You can make insert
performance rise from totally unusably to merely horribly slow by using them.
You cannot, however, in any way approach the performance of most commercial
databases or of MySQL, all of which can be as much as 10 times faster than
PostGreSQL at inserts (depending on indexes, etc.) even WITH all the tricks of
the trade. BTW, adding a single index on one 'TEXT' field to the table that I
added 165,000 records to raised the insert time from under 5 minutes to 20
minutes.... 
 
>     Eric> And, finally, the documentation is cryptic, incomplete, and
>     Eric> scattered all over the place... I had to reference two
>     Eric> different manuals in order to figure out how to create (and
>     Eric> destroy) a PL/SQL stored procedure, for example.
> 
> the book on the postgresql web site covers all of thiss stuff
> pretty nicely.  <URL:http://www.postgresql.org/docs/awbook.html>.

Glad to see the book is coming along. Last time I looked, it had only two
chapters :-). 

> as CGI anyway), but the mod_python and zope docs are pretty nice
> (httpdapy docs are not totally nonexistant either).

Last time I looked at ZOPE, it was not clear how to integrate new Python
methods with it. A shame, it looks like a great product. Maybe I'll get a
chance later on to really get into it and figure all that stuff out, it looks
like it would solve some of the performance problems I've encountered using
PostGreSQL and Python with Apache. 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From sblakey at freei.net  Wed Jun 14 18:51:02 2000
From: sblakey at freei.net (Sean Blakey)
Date: Wed, 14 Jun 2000 15:51:02 -0700
Subject: How to prevent illegal definition of a variable in objects?
In-Reply-To: <8i8ujt$hop$1@news2.isis.unc.edu>; from sdhyok@email.unc.edu on Wed, Jun 14, 2000 at 05:48:45PM -0400
References: <8i8ujt$hop$1@news2.isis.unc.edu>
Message-ID: <20000614155102.E16232@freei.com>

On Wed, Jun 14, 2000 at 05:48:45PM -0400, Daehyok Shin wrote:
> How can I prevent illegal definition of a variable to an object?
> For instance,
> 
> class A:
>     __init__(self):
>             self.x = 1
> 
>     __setattr__(self, name, value):
>             if name == "x":
>                     self.x = value
>             else:
>                 raise AttributeException
> 
>     __getattr__(self, name):
>             if name == "x":
>                 return self.x
>             else:
>                 raise AttributeException
> 
> >>> a = A()
> >>> a.x = 10
> >>> a.y = 20 # I want to raise an AttributeException.
> 
> Peter
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

The code you posted seems to do what you want, provided you
a)add the 'def' keyword to your function definitions
    and
b) define AttributeException (perhaps you meant AttributeError?)

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
	"Uncle Cosmo ... why do they call this a word processor?"
	"It's simple, Skyler ... you've seen what food processors do to food,
right?"
		-- MacNelley, "Shoe"



From bill at libc.org  Wed Jun 28 19:33:07 2000
From: bill at libc.org (Bill Anderson)
Date: Wed, 28 Jun 2000 17:33:07 -0600
Subject: Which linux distros have Python by default?
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial> <%TS15.655$i06.206156@ptah.visi.com> <20000614175918.A5625@dophnic.yi.org> <RVW15.694$i06.216628@ptah.visi.com> <394BE553.99FBC464@libc.org> <gHT25.1412$i06.318267@ptah.visi.com>
Message-ID: <395A8B33.B3F30123@libc.org>

Grant Edwards wrote:
> 
> In article <394BE553.99FBC464 at libc.org>, Bill Anderson wrote:
> 
> >> >Yeah, but the distribution maintainers will be forced to include it as
> >> >a default to be considered seriously.
> >>
> >> Do most of the "consumer" distros even include kernel sources
> >> by default?  I'm not positive, but I don't think RH includes
> >> kernel sources in the standard "workstation" configuration.
> >
> >In the case of Redhat (and I believe Mandrake), it is basically
> >irrelevant, since the installer and many admin tools require python,
> >making it a mandatory install. They do not install the devel files, but
> >python is available.
> 
> Right.  In order to end up with a RH installation without
> Python and Tkinter (and snack) you've got to actually know it's
> there and then intentionally uninstall it after your system is
> up and running.

Yup, and use the --force option to RPM (several other packges requrie
it) ;)

> 
> I just bought a 2.2G removable-media SCSI drive (Castlewood's
> ORB) -- which if it works as advertised is going to be really
> handy.  So, once I've got that going, maybe I'll try out some
> of the other big "Consumer" names (Mandrake, Caldera, Suse,
> Corel) just to see what you get if you pick the default answers
> to all of the intall questions.

My Mandrake users are telling me it is a default installed item.

There is a Linux-orb mailing list, though I don't have the URL ATM. I
can get it after 05-July if you can't find it. AIUI, the ORB basically
works better in Linux ;) Of course, it also depends on which connector
interface it uses. SCSI being the better one :)
 
> I think I can assume that Debian and Slackware users will know
> whether or not they have Python/Tkinter and how to install it
> if they want to point-and-click insted of "vi /etc/rpshsi.conf".

I would agree with that assumption.


From mwh21 at cam.ac.uk  Tue Jun 13 11:26:05 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 2000 16:26:05 +0100
Subject: importing module through variable
References: <8i5e9p$nm0$1@zingo.tninet.se> <20000613160606.W20700@xs4all.nl> <20000613162155.A12328@beta.telenordia.se>
Message-ID: <m3ya4938mq.fsf@atrus.jesus.cam.ac.uk>

=?iso-8859-1?Q?Andr=E9_Dahlqvist?= <andre at beta.telenordia.se> writes:

> > You want the __import__ function:
> > 
> > module = __import__(var)
> 
> I misphrased the question. Inside my module I have a dictionary that I
> need to access. And the name of this dictionary is stored in a
> variable. If I do:
> 
> import the_module
> 
> How can I then access the dictionary inside this module, if the name of
> the dictionary is stored in a variable?

getattr(the_module,var)

Python is gloriously regular in some ways...


> Or is there some equivalence to "from the_module import
> the_dictionary" when you're using __import__?

Probably.  But you want the above, if I understand you properly...

Cheers,
M.

-- 
41. Some programming languages manage to absorb change, but 
    withstand progress.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From aahz at netcom.com  Sat Jun 10 02:15:40 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 10 Jun 2000 06:15:40 GMT
Subject: readlines() question
References: <Pine.GSO.4.10.10006100839410.1511-100000@sundial>
Message-ID: <8hsmec$jp7$1@nntp9.atl.mindspring.net>

In article <Pine.GSO.4.10.10006100839410.1511-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>On 9 Jun 2000, Aahz Maruch wrote:
>>
>> And if you're *really* paranoid ;-)
>> 
>> try:
>> 	my_file = open("blah")
>> 	...my_file.read, my_file.readlines, etc.
>> finally:
>> 	my_file.close()
>
>Not really. 

Never mind.  I got try/except and try/finally mixed up.  (I don't
usually use try/finally.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From buzzard at urubu.freeserve.co.uk  Sun Jun 25 12:09:36 2000
From: buzzard at urubu.freeserve.co.uk (Duncan Smith)
Date: Sun, 25 Jun 2000 17:09:36 +0100
Subject: Reloading modules
Message-ID: <8j5atk$ijs$1@newsg4.svr.pol.co.uk>

>>> from Graph import *
.
.
.
>>> reload(Graph)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
TypeError: reload() argument must be module

What do I need to do to successfully reload a module?  I'm currently wasting
a lot of time re-starting PythonWin each time I want to test a changed
module and I can't find the answer.  Thanks

Duncan




From wjdandreta at worldnet.att.net  Sun Jun 18 15:42:04 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 18 Jun 2000 19:42:04 GMT
Subject: Why is Python so slow?
Message-ID: <gC935.12021$Xx5.555547@bgtnsc06-news.ops.worldnet.att.net>

I posted a message a while ago about a Python script I wrote that took 90
mins to run when a similar job in C++ took only 2 mins.

With help from some experts in this news group, most notably, Peter
Schneider-Kamp, I was able to reduce the time to 37 mins.

Today I was looking at the script and I noticed that the only string
function I was using that was not in strop was replace. I took a look at it
and found:

replace(x,y) = joinfields(splitfields(x,y),'')

When I replaced, replace with the joinfields(splitfields()) in strop, much
to my surprise, the time for the job dropped to 6 minutes.

I am posting this because I thought this info might be useful to others.

Bill




From wlfraed at ix.netcom.com  Wed Jun 28 00:23:53 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 27 Jun 2000 21:23:53 -0700
Subject: float literal too large
References: <ulsntz2kra.fsf@runner.princeton.idaccr.org> <8jat5b$2t4@stochastic.eng.umd.edu>
Message-ID: <88uilsoh9r8atep913uklft51kacb3nov1@4ax.com>

On 27 Jun 2000 14:52:27 -0400, moron at Glue.umd.edu (Warren B. Focke)
declaimed the following in comp.lang.python:

> hardware (pretty much anything other than a Cray (or VAX, but I don't
> think they make those anymore (unless alphas have a VAX math mode)))

	Alphas running VMS and FORTRAN can be compiled in a mode to
produce VAX "D-Float" (though execution is slower) in place of "native"
"G-Float" (let us not get into "H-Float").

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From timaratz at lisco.com  Mon Jun 26 09:42:54 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Mon, 26 Jun 2000 08:42:54 -0500
Subject: smtplib
In-Reply-To: <39568BC4.96EBE8AE@acm.org>
Message-ID: <001501bfdf74$6dff1fe0$14b745c6@timaratz>

Thank you for SMTP.

Yes, I figured out that smtplib requires the subject and headers to be
imbedded in the message. I'm new to Python and I love the language. But I
think that the design of this module reflects poorly upon Python. Almost
every CGI programmer who looks at Python will want to use this module. It
would only take a few lines of code to make it much more usable.

I'm saying this not to flame Python, but to encourage those responsible for
it's distribution to look more critically at what they are putting out. I
hope that this module is atypical of the quality of this product. I'd like
to see Python become more popular. There are many factors involved in that,
but the quality of the modules is certainly one of them.


> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Gregory Tucker
> Sent: Sunday, June 25, 2000 5:46 PM
> To: Peter Timaratz
> Cc: Python List
> Subject: Re: smtplib
>
>
> It looks like the subject is included in the message body. See the attched
> code from Starship.
>
> The following code will call it (I think).
>
> import SMTP
>
> mailhost="tkd.att.ne.jp"
> mailfrom="tuckerg at acm.org"
> mailto="tuckerg at acm.org"
> subj="Test Message"
> msg="""
> Blah
> Blah
> Blah
> """
>
> s=SMTP.SMTP(mailhost)
> s.send_message(mailfrom, mailto, subj, msg)
> s.close()
>
> Regards,
> Greg
>
> P.S. I am getting very little mail off the Python list. Is it something
> wrong with me?
>
> Peter Timaratz wrote:
> >
> > I'm using smtplib with a CGI script. I'm puzzled as to why this module
> > doesn't support sending a subject for the e-mail.
> >
> > And the e-mail I receive doesn't show the to address in my
> e-mail client.
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
>
> --
>  .---. .---. .--- .---.   |  Gregory Tucker, Tokyo, Japan
>  |---. | |-< >--- |---.   |
>  `---' `-'`-'`--- `---'   |  "Our Father, who art in Redmond,
>  My opinions are my own.  |  William by thy name..."




From bozon at natur.cuni.cz  Fri Jun  9 05:38:08 2000
From: bozon at natur.cuni.cz (Michal Bozon)
Date: Fri, 9 Jun 2000 11:38:08 +0200
Subject: ? Py 1.6 final 
Message-ID: <Pine.OSF.4.10.10006091136370.28972-100000@prfdec.natur.cuni.cz>

Where the hell is it?
It's June 8.

Michal Bozon
bozon at natur.cuni.cz



From MarkH at ActiveState.com  Mon Jun 19 20:58:40 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 20 Jun 2000 00:58:40 GMT
Subject: SOCK_RAW with python socket library?
References: <slrn8ksreq.i2.tim.lavoie@goibniu.beyondtv.net> <20000619195809.B13763@quark.emich.edu>
Message-ID: <4lz35.3142$qz4.24485@news-server.bigpond.net.au>

"Jeff" <jam at quark.emich.edu> wrote in message
news:20000619195809.B13763 at quark.emich.edu...
> On Mon, Jun 19, 2000 at 07:08:03PM +0000, Tim Lavoie wrote:

> if all you really want is to implement 'ping' with python, this page:
> http://the-tech.mit.edu/~jeremy/python/

Cool.  Id never seen that before.  The ping program appears to work
correctly as-is with Python 1.6 on Windows!  The traceroute program
appears to have a few simple bugs not related to Windows at all, so
presumably this would also work...

Mark.





From pecora at anvil.nrl.navy.mil  Tue Jun  6 17:08:46 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Tue, 06 Jun 2000 21:08:46 GMT
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun  5)
References: <BA7ED618087F752B.EE3239CC079D7980.5169D95D2EE3FD08@lp.airnews.net>
Message-ID: <060620001708472558%pecora@anvil.nrl.navy.mil>

In article
<BA7ED618087F752B.EE3239CC079D7980.5169D95D2EE3FD08 at lp.airnews.net>,
Andrew M. Kuchling <akuchlin at mems-exchange.org> wrote:

    It's finally public; Guido, Fred Drake, Jeremy Hylton, and Barry
    Warsaw are now working for BeOpen.
        http://www.pythonlabs.com/letter.html

    BeOpen ran an interview with Guido:
        http://www.beopen.com/features/interviews/rossum_index.html



Is the above good, bad or neither?


From jkraska1 at san.rr.com  Sun Jun 11 17:56:17 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 11 Jun 2000 21:56:17 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil>
Message-ID: <39440B9E.88920EF9@san.rr.com>

> Yes, usually when you _create_ a Python object in C you have to handle
> the references or call routines that do that for you. Trouble is, it is
> something that is not always clear on how to do it.  I would say you
> need to INCREF your tuple after creation, unless you know the routine
> that creates it already does that.

I was under the impression that the creation of the tuple in C
created with a ob_refcount of 1 already. While I don't have the
time at the moment to verify this, the original poster or someone
else can confirm or deny my wild imagination. :)-

> If you return the tuple, do
> nothing.  If you do not return it, you should DECREF it.

My wild imagination agrees with you on the DECREF part. :)-




C/


From nospam at bitbucket.com  Mon Jun  5 17:58:53 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Mon, 5 Jun 2000 14:58:53 -0700
Subject: MemoryError crashes within try clause
Message-ID: <%oV_4.26$WI2.2691@newsfeed.avtel.net>

The following code sample is designed to consume memory, and crashes
under Win98.  If the try clause is removed or moved to a calling
function, a MemoryError is generated as expected??!

This is not an academic question; I arrived here through tracking hangs
in a beta release under low-memory conditions.  Any clues or insights
gratefully received.

Thanks, Phil

Configuration: Python 1.5.2, Pentium, 64MB, Win98 with virtual memory
turned off.

---------- code sample follows --------
"""
%1 caused an invalid page fault in
module %2 at %3!.04x!:%4!.08x!.
Registers:
EAX=%1!.08x! CS=%2!.04x! EIP=%3!.08x! EFLGS=%4!.08x!
"""

def GetBigList():
    size = 1 << 20
    while size:
        try:
            return [None] * size
        except:
            size = size / 2

def UseMostMemory():
    s = 'x' * 32000
    list = GetBigList()
    print 'UseMostMemory list size =',len(list)
    try:
        for x in xrange(len(list)):
            list[x] = s+str(x)
            print x
    except:
        print 'failed at',x
    return list

def Eat():
    try:
        list = GetBigList()
        print 'Eat list size =',len(list)
        eaten = UseMostMemory()
        x = 0
        while x < len(list):
            list[x] = str(x)
            if x %100 == 0:
                print x
            x = x + 1
    except:
        pass

try:
    Eat()
except:
    print 'fail'
--
Phil Mayes    pmayes AT olivebr DOT com
Olive Branch Software - home of Arranger
http://www.olivebr.com/





From jam at quark.emich.edu  Fri Jun  2 19:56:07 2000
From: jam at quark.emich.edu (Jeff)
Date: Fri, 2 Jun 2000 19:56:07 -0400
Subject: scripts not completing...
In-Reply-To: <39371b5e.376157555@news1.on.sympatico.ca>; from chibaA@TinterlogD.Tcom on Fri, Jun 02, 2000 at 02:28:05AM +0000
References: <39371b5e.376157555@news1.on.sympatico.ca>
Message-ID: <20000602195607.B15032@quark.emich.edu>

On Fri, Jun 02, 2000 at 02:28:05AM +0000, chibaA at TinterlogD.Tcom wrote:
> Other than having a runaway script... is there any other routine
> reason why a script would not terminate properly?  (I'm running python
> scripts as CGI through apache).  I noticed that script processing gets
> really slow after a while, and noticed that there were still processes
> still running with that script.  
> 
> Is there anyway to make the system automatically kill those processes
> (apache, python scripts) which have been idle for a long time (and
> probably hanging, or running away)?
> 
> TIA
> 
> kc

while there probably *is* a way to kill a hung process, that approach is, in
my opinion, prone to error.. if I was in this position, I would examine the
code itself to try to fix the bug rather than hacking around it (which, when
deadlines are involved, is sometimes the only way to do things). 

do you have access to the source code with this problem? what you describe
sounds like some kind of non-terminating loop. 

since CGI is involved, what type of error do you get on the browser when
this happens? does it say "No Data", or "Internal Server Error", or does it
just get slow?

what else is running on the machine that might be the cause of the problem?

what platform are you having this problem on? the specifics of
troubleshooting for a win32 version of apache (for example) would be
different from a UNIX-like platform.

hope that helps..

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 



From anil.singh at patni.com  Wed Jun 21 09:53:39 2000
From: anil.singh at patni.com (Anil Singh)
Date: Wed, 21 Jun 2000 19:23:39 +0530
Subject: Cold Fusion and IBM Websphere using Python as CGI ?
Message-ID: <00ad01bfdb88$1a5e8320$8235a8c0@pc-p2457.patni.com>

Hi guys,

We have developed a Web application using Python as CGI on IIS on Windows NT 4.0.
The Application layer of our website uses C exe for various services like 
database connectivity etc. We are trying to move to other webservers specifically to
Cold Fusion Server 4.0 or IBM Web Server 1.0 in near future. But we don't want to 
change the CGI script language from Python. We would like to know whether Python Scripts 
are compatible to Cold Fusion Server 4.0 & IBM IBM Web Server. If it require changes 
what will be the nature of changes ?

We would appreciate if the you can help us in identify the Python related Setup 
changes also for the above mentioned Servers. 

Thanking You in advance in anticipation of early reply. 
Regards,

Anil Singh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000621/41084ca0/attachment.html>

From mjhand at concentric.net  Thu Jun  8 16:36:38 2000
From: mjhand at concentric.net (Manus Hand)
Date: Thu, 08 Jun 2000 14:36:38 -0600
Subject: String.join revisited (URGENT for 1.6)
References: <hjoY4.6071$wYl.223614464@newsb.telia.net> <39328edc$0$18384@personalnews.de.uu.net> <_pyY4.10102$Za1.160118@newsc.telia.net> <3932e628$0$6681@personalnews.de.uu.net> <3932F323.C73E0404@san.rr.com> <3932F5DC.E00C1642@san.rr.com> <ZoDY4.6196$wYl.229414400@newsb.telia.net> <39333CC3.32805F5@cosc.canterbury.ac.nz> <saMY4.6222$wYl.223847936@newsb.telia.net> <8h087l$bnc$4@newshost.accu.uu.nl> <20000530182454.E29991@exeter.exeter.org>
Message-ID: <394003D6.366D0B0@concentric.net>

I really hate to jump in here, since so many people have already criticized
the length and content of this thread, but well, here I am.  And I have three
points to make.

First...
As one of those fools who grab the alpha version and recode everything
I've ever written to it, I have already gotten used to ''.join() [and I like it].
The view that someone posted here, that ','.join(list) can be viewed as
sending a "message" (each list member) to the string object ',' was all I
needed to get my head around the (admittedly initially strange) construction.

Second...
The fact that the string module will continue to exist (BTW, I really like
not having to regularly type "import string" when I start any code effort!)
is inherently a breach of the "proper" Pythonic view on TMTOWTDI.
It is my understanding that string.upper() will still exist, and I have not
seen this being argued as a reason not to support ''.upper() in 1.6.
''.join() seems to be the only case where TMTOWTDI is argued.
And if string.upper(y) is going to be equivalent to y.upper(), it only
makes sense to me from a consistency standpoint that string.join(x,y)
be equivalent to y.join(x).  [I gotta admit, I did like the ability to default
the separator and get a space, but this is a small trade-off for not having
to import string all the time.]

Third...
Before I learned that it was ''.join(list), I myself posted here wondering
why it was not [].join(string).  As I wrote above, I am now used to it
(and comfortable with it) being the way it is, but one thing about this
continuing debate has kind of always bothered me.  One argument
against it being [].join(string) is that this would mean implementing a method
for lists that only works on lists of a certain type (i.e., lists consisting entirely

of strings).  No one seems to question that, so I will (and I'm sure I'll
be corrected, which is fine).  But as the devil's advocate, tt just seems to me
that [].join() could be implemented such that each member of the list is str()'ed
or repr()'ed and then stuck together in the result string.  I imagine that one of
the fatal flaws in this, though, is how self-referential lists would be handled:
   >>> x = ['a']
   >>> x.append(x)
   >>> x.join()        # I would argue we get an exception?
If that is the only issue (which is being voiced as, "this would force
a method on lists to only work on CERTAIN lists") then I will understand
the major argument against [].join() at least a LITTLE more.

In short, I am (like anyone cares) in favor of standardizing on the ''.join() that
1.6a2 offers, and [purely for backwards compatibility, though I'm one of those
fools who recode everything, like I said] retaining the string module despite the
fact that it now means there will be two ways to do certain things.

In order to make THAT internally consistent in the language, we would want
(would we not? :-) to introduce a "list" module:
   >>> import list   # yeah, that would be a problem, given the builtin list()
   >>> x = [1]
   >>> list.sort(x)   # now an alternate way to write x.sort()
and a "dict" module:
   >>> import dict
   >>> x = { 1: 2 }
   >>> dict.keys(x)   # now an alternate way to write x.keys()
   [1]

I'm getting loopy.

Manus




From bparsia at email.unc.edu  Wed Jun 21 16:24:05 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Wed, 21 Jun 2000 16:24:05 -0400
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl> <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu> <394FFEF4.4B7A2765@prescod.net>
Message-ID: <1ecl20o.weekaz14o3gx7N%bparsia@email.unc.edu>

Paul Prescod <paul at prescod.net> wrote:

> Actually, I think that a method cache for Python has been implemented.

Really? How interesting.

> I
> think that the main reason that it isn't "standard" is that Guido wanted
> to preserve the simplicity of the Python *implementation*. In general
> this is important because we want many people to poke around in Python's
> internals and not just a tiny inner circle. 

I'd be interested to know whether, as has been indicated, the cache
makes things *especiallyh* complex. <sigh/>I'd love it if someday we
could get someone (er...C literate :)) to compare the code for the
Python VM and the Squeak VM, since one of the big design goals for the
Squeak on is (likewise) implementation simplicity.

> Of course, any particular optimization could be so attractive that it
> overwhelms this concern...but in general Guido does not take the
> approach where you try to cram in every cool optimization you can find.

Oh, definitely. (I have definitely strayed from empircal grounds :)) I
guess I've thought that given the example of Squeak (having a method a
cache, having a reputedly simple implementation, and seeming to benefit
greatly from the cache), and the very high cost of function/method
lookup in Python, that a method cache in Python would be a big win
without (necessarily) compromising implementation simplicity.

But, thus far, my speculation seems to be mistaken :). Thomas hasn't
*quite* convinced me on the complexity, yet, but probably because I
don't understand him! He (and others) have also suggested that the
speedup might not be that great. I wonder, however, if that's in part
due to avoiding coding patterns that would be greatly penalized by the
lookup times. (Of course, even if that were true, it doesn't help with
the task of speeding up *existing* Python code.)

If there is method cache code lying about, it might be fun to share it
out, even if it doesn't make it into the regular release.

(Indeed, having a base, simple, wildly ported version and a hacked to
help, crufty, speedy version of the VM would seem to have a fairly wide
audience :))

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From 12klat at sightreader.com  Tue Jun  6 08:47:41 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 05:47:41 -0700
Subject: How do I create extensions to Python in C?
References: <BlAW4.3473$DQ3.245148@news1.atlantic.net> <392BE0ED.12E4DA0E@cfdrc.com> <00ea01bfc594$02d2e2f0$4500a8c0@thomasnb> <393C95BD.77108EE@san.rr.com> <8hieq9$k1m$1@gw.localdomain>
Message-ID: <393CF2ED.D4AF6A31@sightreader.com>

Anything can be done with wrappers around wrappers around wrappers.

// C++...
class X
{
//...
   int foo();
}

// ... wrapped in C struct ...
typedef struct
{
 PyObject_HEAD
 X *pX;
} x_obj;

...

PyTypeObject x_type = {
 PyObject_HEAD_INIT(&PyType_Type)
 0,                              // ob_size
 "x_obj",                         // tp_name
 sizeof(x_obj),           // tp_basicsize
   ...
};


# ... wrapped in python class
class X:
    def __init__(self):
        self.x = new_x()
        ...

The details are left as an exercise for the student.

You should look at Extension Classes too.  It allows you to effectively make
your extensions behave like python classes that you can inherit from, etc.

Here are some references (search for these in Parnassus, etc.):
  PyExtWizard (if you are using Windows)
  Extension Classes (a more complete way of doing what I have described)
  SWIG ("automatically" wraps c into python extensions)
  Extending (look for this in the documentation)

Sameh chafik pro wrote:

> It is okay, but why coud we dot to translate a class write in C++. The
> problem that the standard way is to use the PyTypeObject and define the
> constructor as a function, but python don't recognize the translation as
> class but as type object.morality if i do the translation of a class "X" it
> is'nt posible in python to use it as a derivate class:
> class A(X):
>     def dosomething(self):
>     .....
>     ....
>
> is not possible.
>
> "Courageous" <jkraska1 at san.rr.com> a ?crit dans le message news:
> 393C95BD.77108EE at san.rr.com...
> >
> > > http://starship.python.net/crew/arcege/extwriting/pyext.html
> >
> > _Python Essential Reference_ also has a very nice section which
> > summarizes many of the available primitives. If you're going
> > to be doing this yourself, I recommend you take a look at
> > xxobject.c in the python source. It's an empty template for
> > creating a python extension module.
> >
> >
> >
> > C/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/10db6a03/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/10db6a03/attachment.vcf>

From moshez at math.huji.ac.il  Fri Jun  9 12:56:20 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 19:56:20 +0300 (IDT)
Subject: Probably a dumb question
In-Reply-To: <394106F9.A58F87CE@micron.net>
Message-ID: <Pine.GSO.4.10.10006091955420.27416-100000@sundial>

On Fri, 9 Jun 2000, Peter Immarco wrote:

> Hi,  still waiting for my two books on Python from Amazon.  I know
> Python is a full programming language so it can be used on web servers
> for server side processing, but what about browser side processing?  Can
> I also use it to create an 'applet' that runs in a compatible browser?
> I'm guessing not likely because I don't think IE4+ or Netscape
> Communicator support Python but I'm hoping I'm wrong.

Well, the Time Machine at work once again: they all support Java, and
Python can run inside JVMs: see http://www.jpython.org
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From trentm at activestate.com  Mon Jun 26 20:07:53 2000
From: trentm at activestate.com (Trent Mick)
Date: Mon, 26 Jun 2000 17:07:53 -0700
Subject: learning-python question
In-Reply-To: <8j8ep3$tkn$1@nnrp1.deja.com>
References: <8j8ep3$tkn$1@nnrp1.deja.com>
Message-ID: <20000626170753.C2492@activestate.com>

On Mon, Jun 26, 2000 at 08:34:43PM +0000, JeremyLynch at fsmail.net wrote:
> hi,
> I have just started learning python, although I am already confident in
> c/c++, and I finding some of python's features confusing. This may seem
> like a *very* stupid question so ignore it if you want to :)
> My problem simplifies to:
> 
> count = 5
> 
> def count_func()
>      print count
             ^^^^^
This reference to count is raising a NameError. This is because the global
namespace is not searched when in a Python function. You have to explicitly
declare any global variables used in a function with the 'global' keyword,

>      count = count + 1


so.... this is what you want:

c:\temp> python
>>> count = 5
>>>
>>> def count_func():
...     global count
...     print count
...     count = count + 1
...
>>>
>>> count_func()
5
>>> count_func()
6
>>>



As well, you have to have a colon after the 'def count_func()'.


Trent


-- 
Trent Mick
trentm at activestate.com



From spenrose at my-deja.com  Wed Jun 14 13:11:31 2000
From: spenrose at my-deja.com (spenrose at my-deja.com)
Date: Wed, 14 Jun 2000 17:11:31 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <394b3e7d.284293552@news.online.no> <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net>
Message-ID: <8i8ebi$5et$1@nnrp1.deja.com>

I was working through Learning Perl when I saw an ad for a programming
internship in Python. I went out and bought Learning Python, brought it
to the interview, and got the internship, which lead quickly to a
full-time job coding Python. Learning Python is a great book; I've read
it carefully twice and I still go back to certain sections periodically
to bone up on areas of the language I haven't had a chance to use yet.
That said, I am really glad that I had read Learning Perl first. That
book (specifically the first 50-100 pages) felt much more aimed at
first-time programmers, and helped me with the concepts I would
encounter in Python.

FWIW, I agree with those who think Python an ideal choice for a first
language--it worked for me! It's just the orientation of the two
O'Reilly books I'm referring to here.

In article <zGI15.8$Gp6.2456 at dfiatx1-snr1.gtei.net>,
  "jin choung" <jin.choung at gte.net> wrote:
> howdy,
>
> ummm, is there any book that you reccommend for a rank beginner?  that
> learning python book is REALLY not for beginners.
>
> thanks.
>
> jin
>
>


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


From akira.kiyomiya at autodesk.com  Tue Jun  6 16:55:42 2000
From: akira.kiyomiya at autodesk.com (Akira Kiyomiya)
Date: Tue, 6 Jun 2000 13:55:42 -0700
Subject: Why Do We Use This?
Message-ID: <8hjogr$d14@autodesk.autodesk.com>

Okay, what is the difference between these?

If these are same, whey do we bother using the "long" version such as
__len__(a)??

Sorry, the books I am reading do not explain very well about this issue.

a = [1,2,3,4,5,6]
len(a)                                # __len__(a)
x = a[2]                            # __getitem__(a,2)
a[1] = 7                            # __setitem__(a,1,7)
del a[2]                            # __delitem__(a,2)
x = a[1:5]                         #__getslice__(a,1,5)
a[1:3] = [10,11,12]         # __setslice__(a,1,3,[10,11,12])
del a[1:4]                        # __delslice__(a,1,4)

Thanks

Akira




From tim_one at email.msn.com  Sat Jun  3 00:44:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 3 Jun 2000 00:44:48 -0400
Subject: Copyright changes with move from CNRI to BeOpen?
In-Reply-To: <39375113.5AF5BC0B@gte.net>
Message-ID: <000601bfcd16$7308e2c0$b42d153f@tim>

[Harry Tanovich]
> Will the copyright move to a more GPL style thing now?

The details are still being worked out, but I'll tell you this:  if BeOpen
had showed any sign of wanting to make the Python license materially more
restrictive, I would not have accepted the job there (GPL is fine for many
things, but Python isn't one of them).  To the contrary, the BeOpen
PythonLabs team would like to *reduce* the legal folderol that accreted
under CNRI's tenure.  However, at this precise moment in time, I'm
unemployed, so don't take this as an Official Statement from anyone -- least
of all me <wink>.

if-i-wanted-to-speak-for-myself-i'd-hire-me-ly y'rs  - tim


PS:  Note that the JPython license has always been stiffer than CPython's,
     and that may not be changeable.





From wjdandreta at worldnet.att.net  Tue Jun 20 18:36:37 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 20 Jun 2000 22:36:37 GMT
Subject: List without duplicates?
Message-ID: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net>

Is there a built in Python function that eliminates duplicates in a list?

I have been sorting the list and compairing adjacent elements and if equal
deleting one:
-------------------------------
mylist.sort()
    j = len(mylist)-1
    while j > 0:
      if mylist[j] == mylist[j-1]:
        del mylist[j]
      j = j - 1
------------------------------------
I recently tried using a dictionary. The list is the keys and I put and
empty string for the entry
--------------------
mydict = {}
for e in mylist:
  if mydict.has_key(e): continue
  else: mydict[e] = ''
------------------------------------
Is ther a better way?

Bill




From MarkH at ActiveState.com  Mon Jun 19 20:01:24 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 20 Jun 2000 00:01:24 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <uitv9qb72.fsf@ctwd0143.fitlinxx.com> <slrn8kobd5.1vs.hzhu@mars.localdomain> <04b401bfd960$7a9f95f0$6401a8c0@home> <8iltfj$hm9u$1@fido.engr.sgi.com>
Message-ID: <ovy35.3129$qz4.24350@news-server.bigpond.net.au>

"Paul Jackson" <pj at sgi.com> wrote in message
news:8iltfj$hm9u$1 at fido.engr.sgi.com...
>
> |> 'Tabs will be replaced with %d spaces\n Are you sure
> |> Y/N:'%tabSize)
>
> But a tab isn't 8 spaces, but rather the number of
> spaces needed to get to the next multiple of 8 column,
> which might be less than 8 spaces (s/8/tabSize/g, if
> you must).

I agree, except you made the common mistake of using "8" where you meant
"the next tab-stop" ;-)

Mark.





From aahz at netcom.com  Tue Jun 13 12:02:52 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 13 Jun 2000 16:02:52 GMT
Subject: Paring down Python (physically)
References: <39465328.2352344@news.ark.com>
Message-ID: <8i5lvc$s88$1@nntp9.atl.mindspring.net>

In article <39465328.2352344 at news.ark.com>,
Alex Walker <alwalker at mars.ark.com> wrote:
>
>I've been tasked with a question of how to use Python on a WinNT
>system where isn't installed.  Of the options available (and I don't
>thing I've discovered them all yet,) installing with our app just
>those Python files absolutely required to run a script is the
>front-runner.  I've looked all over the Python web site, the docs, and
>my books but haven't come across any info the help me pare down Python
>to a size, other than just hacking off modules I'm not calling.   Any
>suggestions on a better way or where next to look for helpful info is
>appreciated ?   Thanks!

Look for mxCGIPython at
http://starship.python.net/~lemburg/mxCGIPython.html
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From israel at math.ubc.ca  Fri Jun 16 13:36:32 2000
From: israel at math.ubc.ca (Robert Israel)
Date: 16 Jun 2000 17:36:32 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <394a5109.80394991@10.1.1.28> <394A5795.F4DE98FB@san.rr.com> <394a5ed7.83928743@10.1.1.28>
Message-ID: <8idoj0$lea$1@nntp.itservices.ubc.ca>

In article <394a5ed7.83928743 at 10.1.1.28>,
Tim Dixon <tdixon.no at spam.fwi.com> wrote:
>On Fri, 16 Jun 2000 16:33:59 GMT, Courageous <jkraska1 at san.rr.com>
>wrote:


>>> Suppose, however, that you could get digit 'n' of pi directly (or e,
>>> or any other inifinte, nonrepeating sequence).  All I have to do is
>>> communicate which digit of pi the message key starts at, and I can
>>> generate the rest of the key.
>>
>>This is a collosally bad idea, IMO.

>Just for curiosity, why?

Because the real key is the number of the digit that you start at, and
that's pitifully small by cryptographic standards.  The time to generate
the n'th digit is not much less than the time to generate all the digits
up to the n'th, given that you have lots of memory available.  So the
cryptanalyst with a computer somewhat bigger than yours could try all
the keys you could be using.  

Robert Israel                                israel at math.ubc.ca
Department of Mathematics        http://www.math.ubc.ca/~israel 
University of British Columbia            
Vancouver, BC, Canada V6T 1Z2


From krussell4 at videon.home.com  Thu Jun 15 22:23:58 2000
From: krussell4 at videon.home.com (Kevin Russell)
Date: Fri, 16 Jun 2000 02:23:58 GMT
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com> <8iab6a$idm$1@hunter.orchestream.com> <8iarpa$gt6$2@216.39.141.36>
Message-ID: <39498FCF.AF8C0D13@videon.home.com>

Dana Booth wrote:

>
> Lisp is the default macro language of AutoCAD. The lead engineer at my shop
> swears by it. She is very productive, hence her title as lead engineer. :)

That wouldn't be "lead" as in the metal, would it? :-)





From wjdandreta at worldnet.att.net  Tue Jun 27 18:17:03 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Tue, 27 Jun 2000 22:17:03 GMT
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com>
Message-ID: <zJ965.4612$AM4.284145@bgtnsc04-news.ops.worldnet.att.net>

Hi Greg,

Have you tried

  forward[domain] = new_forward?

Bill
>
>   forward[domain].append(new_forward)
>

>Greg Gallagher
>greg at red-bean.com
>http://www.red-bean.com/
>




From wjdandreta at worldnet.att.net  Sun Jun 18 23:22:05 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Mon, 19 Jun 2000 03:22:05 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <Icd35.5663$_54.106037@news1.rdc1.md.home.com>
Message-ID: <xlg35.9509$C44.539145@bgtnsc05-news.ops.worldnet.att.net>

Hi Eric,

>Seems to me that you can prove your own assertion.  Simply create another
>copy of your code and remove all comments.  I would be quite suprized if
you
>found speed up -- but see for yourself.

I thought that is what I did but I see now it is not. I'll test it exactly
as you said.

>Any reason you are not using more recent versions of python ?

The latest versions of the binaries are all compiled to run with a DOS
extender which is not supported on my machine.

I am in the process of compiling 1.5.2 to fit in 640K but it requires
overlays, which I have never used before, and since I only work on it in my
spare time, it's taking a while.

Bill




From btang at pacific.jpl.nasa.gov  Thu Jun 15 15:25:18 2000
From: btang at pacific.jpl.nasa.gov (Benyang Tang)
Date: Thu, 15 Jun 2000 12:25:18 -0700
Subject: Numeric python
References: <Pine.LNX.4.10.10006141123440.26946-100000@intel3.wag.caltech.edu>
Message-ID: <39492D9E.F71039DD@pacific.jpl.nasa.gov>

This is my collection of links related to numPy. Some may not be there anymore.

Subject: python for sciences

numPy
http://sourceforge.net/project/?group_id=1369

simple instruction for installing numPy
http://www.oreillynet.com/python/2000/05/03/install_num_python.html

DISLIN, a plotting package for many languages, and python too. python version is free.
http://www.linmpi.mpg.de/dislin/

ScientificPython, bundling NumPy and other stuff together
http://starship.python.net/crew/hinsen/
http://starship.python.net/crew/hinsen/ScientificPythonManual/Scientific.html
http://starship.python.net/~hinsen/index.html

FPIG ? Fortran to Python Interface Generator
http://cens.ioc.ee/projects/f2py2e/

linking python and C and LAPack
http://www.fysik.dtu.dk/~schiotz/

NCAR's work on python
http://www.python.org/workshops/2000-01/proceedings/papers/van_andel/van_andel.htm


Fortran and python
http://www.foretec.com/python/workshops/1998-11/proceedings/demos/bishop/bishop.html

multiPack, for numPy
http://oliphant.netpedia.net/multipack.html

scientific python for windows
http://starship.python.net/crew/kernr/binaries/ScientificPython-2.0b1w.zip

many of numPy packages
http://oliphant.netpedia.net/
http://www.python.org/topics/scicomp/numbercrunching.html

NumPy home page
http://xfiles.llnl.gov/python.htm
http://sourceforge.net/project/?group_id=1369

Konrad Hinsen's distribution
http://starship.python.net/crew/hinsen/scientific.html
and his tutorial for NumPy
http://starship.python.net/crew/hinsen/ScientificPythonManual/ScientificPython.html

LLNL's computer help
http://www.llnl.gov/computing/unixenviron/index.shtml

physics and python
http://www.python.org/workshops/1997-10/proceedings/beazley.html

science plotting
GNUPlot: http://monsoon.harvard.edu/~mhagger/download/index.html
http://starship.python.net/crew/jhauser/plot-res.html, listing various python plotting modules

numerical recipe in python
http://www.python.org/topics/scicomp/recipes_in_python.html

another collection
http://oliphant.netpedia.net/



Michael Todd Feldmann wrote:
> 
> Could anyone point me in the right direction to get Numeric python?  I
> have a linux box and tried
> 
> http://www.python.org/download/download_linux.html
> 
> and went to
> 
> http://andrich.net/python/
> 
> which states that "This site is closed." since things may be outdated.  I
> am guessing there is somewhere else to find this and other RPM files.
> 
> Thanks a bunch!
> 
> Mike
> 
> --------------------------------------
> Michael Todd Feldmann
> Beckman Institute (139-74)
> California Institute of Technology
> Pasadena, California 91125 USA
> Phone:(626)395-2722, FAX:(626)585-0918
> email: feldmann at wag.caltech.edu
> --------------------------------------

-- 
<> Benyang Tang                                   office: 818/354-0327 <> 
<> 300-323, JPL                                      fax: 818/393-6720 <> 
<> 4800 Oak Grove Drive               mail: btang at pacific.jpl.nasa.gov <> 
<> Pasadena, CA 91109, USA                                             <>


From cartho at mordor.ch  Thu Jun  8 17:27:05 2000
From: cartho at mordor.ch (Cyrille Artho)
Date: Thu, 08 Jun 2000 23:27:05 +0200
Subject: TypeError: unsubscriptable object
References: <393ECFC3.879BC357@mordor.ch> <8hmp6i$ets$6@newshost.accu.uu.nl> <393F488B.37F13682@mordor.ch> <393feff5$1@news2.sireco.de>
Message-ID: <39400FA9.3DED5876@mordor.ch>

Thank you, this was the bug. Indeed, the error message was very
confusing. I hope that error messages will be better in Python 1.6.
Especially for a language that is used for teaching, accurate error
messages will be crucial.

I fixed all the other bugs, too, I think. However, the network produces
a strange output - maybe it is because the amount of neurons were very
limited for that example. It was a quick hack anyway, but it was a good
way to learn Python. If anybody is interested in the (working) script,
please let me know.
--
Regards,
Cyrille Artho - http://artho.com/ - Tel. +41 - [0]1 - 313 08 92
Give a man a fish, and you feed him for a day.
Teach a man to fish, and he'll invite himself over for dinner.
		-- Calvin Keegan


From thomas at xs4all.net  Tue Jun  6 09:50:07 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 6 Jun 2000 15:50:07 +0200
Subject: GTK breaks float! (related to "float does not parse 0.00000E+00")
In-Reply-To: <393BBEC4.5BC343FB@inc.it>; from redaelli@inc.it on Mon, Jun 05, 2000 at 04:52:52PM +0200
References: <393BBEC4.5BC343FB@inc.it>
Message-ID: <20000606155006.G469@xs4all.nl>

On Mon, Jun 05, 2000 at 04:52:52PM +0200, Paolo Redaelli wrote:
> Look at the this short program:

> =======
> #!/usr/bin/env python
> 
> s = "0.000000E+00"
> print "Converting",s," into: ", float(s)
> print "Importing gtk"
> 
> import GDK
> from gtk import *
> from gnome.ui import *

> import GdkImlib
> print "Converting",s," into: ", float(s)
> =======
> Here's its output:
> =======
> [paolo at aldebaran paolo]$ python float.py
> Converting 0.000000E+00  into:  0.0
> Importing gtk
> Converting 0.000000E+00  into: 
> Traceback (innermost last):
>   File "float.py", line 11, in ?
>     print "Converting",s," into: ", float(s)
> ValueError: invalid literal for float(): 0.000000E+00
> =======

> The only lesson I can extract from it is that GTK breaks float!

Wrong lesson. The lesson you should extract from this is 'Do Not Use From
<module> Import *'. No excuse, just do NOT use it. If really you want
gdk.SomeThing available as 'SomeThing' instead of 'gdk.SomeThing', use 'from
gdk import SomeThing'.

'from <module> import *' just breaks too much stuff :P

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From info at info.com  Fri Jun  2 08:58:58 2000
From: info at info.com (info at info.com)
Date: 2 Jun 2000 05:58:58 -0700
Subject: Find computer, software, accessories
Message-ID: <74505327.7F9502A0.79045bf2cc68c1ecd40d71d36b8b7baa@info.com>

Locate now: Computer equip
	    Software
	    accessories
	    Games

At: http://www.locate-now.com/adcomp1.htm

From tismer at tismer.com  Sat Jun  3 19:04:14 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sun, 04 Jun 2000 01:04:14 +0200
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com>
Message-ID: <39398EEE.C18EC987@tismer.com>


Courageous wrote:
> 
> > I'm against a linked list deque implementation for Python.
> > But I'm plus 1 on doing it as an extension of lists.
> 
> For what it's worth, I'm implementing a dynarray which
> leaves buffer room on both the head and tail as we speak.

Whow :-))

> Note that this slows down random access ever-so-slightly,
> as you have to have a "virtual index" to account for the
> actual and current positions of the "beginning" of the list.

Sure. There is no free lunch at all.
For that reason, I also wouldn't propose to change
lists for this feature, but offer this as an addition.
If you want it, you get it, at a slight cost for every
operation of course.
How about deque(list) as initializer?

> However, the cost is very small: an extra add on every
> access.

I think it is a bit more; well it depends on the
implementation. I thought of a list, with a size
controlled by the deque, padded with Py_None.

> My current algorithm is as follows:
> 
> 1. Store a "virtual index" into the list, allowing buffer
>    room at the head.

Yup.

> 2. On an insert at the head, simply decrement the virtual
>    index and insert.

Yup.

> 3. On an insert operation at the head, if the virtual index
>    would go negative on the insert, double the allocation
>    space of the array and place the actual used space of
>    the array 1/3 of the way into the array.

Great. I'd also love to get the 1/3 rule into standard lists.

> 4. On removal at the head, simply increment the virtual
>    index.

Yeah.

> 5. If the virtual index becomes larger than 2/3 of the
>    allocated space, move the actual used space of the
>    array 1/3 of the way into the array (this last is to
>    avoid an array which would grow infinitely on append
>    to tail and removal from head).

We have something in common...

> 6. The growth rate of the array will always be 100%
>    increase (as compared to python's 50/500 heuristic,
>    IIRC).

Blah. No more comments, just do it. It is the right
track; I love it.

> 7. You'll be able to specify an initial size for the list.
> 
> This technique should have the same complexity as python
> lists with the additional benefit of O(1) removal/insert
> at the head without sacrificing O(1) random access. I'll
> do a complete performance test as before, of course.
> 
> As per usual, the source will be entered into the public
> domain.

As per unusual, I will incorporate this into Stackless
Python immediately if it is written as clear as you described
it.

still-waiting-for-your-job-offer-ly y'rs - chris :-)

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



From garry at sage.att.com  Wed Jun 28 09:04:37 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 28 Jun 2000 13:04:37 GMT
Subject: Tutorials at IPC9
References: <39591877.535A96BC@sage.att.com> <8jb6o7$us1$1@slb7.atl.mindspring.net>
Message-ID: <3959F7E5.33863F5@sage.att.com>

Aahz Maruch wrote:

> To what extent should people not at IPC8 involve themselves in this
> discussion?

i didn't mean to exclude anyone.  i'd prefer participation
from anyone with an idea who cares enough to post.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From see at message.body  Sat Jun 17 21:05:42 2000
From: see at message.body (Sam Penrose)
Date: Sat, 17 Jun 2000 18:05:42 -0700
Subject: Optimizing Object Instantiation (was Re: How Fast Does Python Run?)
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net> <39499C91.3A1586BE@sightreader.com> <8if21t$4u5$1@slb1.atl.mindspring.net> <20000617120003.S26436@xs4all.nl>
Message-ID: <see-0F7889.18054217062000@news.dnai.com>

A recent article in this forum linked to an analysis of the
performance of several popular languages, including Python.
The article finds that instantiating an object in Python is
a particularly expensive operation relative both to other
Python operations and to instantiating an object in other
languages. My experience is that moving c. 1000 lines of 
CPython code from a relatively procedural to a relatively
object-oriented approach incurs a definite performance hit
on vintage 1998 Intel hardware running Linux (Python 1.5.2).

1) Does informed opinion agree that object instantiation is
noticeably performance-expensive in CPython? (Unfortunately,
I can't seem to find the article in question.)

2) If so, does Python 1.6 address this issue? Will later versions?

Note: I thoroughly agree with Thomas' comments on the virtues
of what he calls the Python Way.


In article <20000617120003.S26436 at xs4all.nl>, Thomas Wouters 
<thomas at xs4all.net> wrote:

<snip>
> You can probably speed up Python quite a bit if you concentrate
> on the common task and do not mind fawlty or flawed 
> results for the little used features.
> 
> Fortunately (in my opinion), that isn't the Python Way. 
> There is still room for improvement without losing accuracy
> and simplicity (of Python) but that *will* cost, in 
> simplicity-of-CPython: The current optimization (which
> translates 'load_name' opcodes into 'load_global' or 'load_fast'
> depending on wether they seem to be local or global) is a good
> example: The actual added code is not that large or complicated,
> but when you start adding new features, you have to keep those
> 20-odd lines of code in mind, or you'll screw it up. Add 10 more
> of those 'performance hacks' and adding to CPython can become a
> real pain.
> 
> Also, the load_fast-translation-cycle is skipped if the code block has a
> 'from module import *' or 'exec' statement in it, because the
> compiler/optimizer does not know what names it imports, and hence not 
> what names will be actually local when the code executes.
> 
> Optimized-Python-contradicts-'simplicitly'-both-ways-ly y'rs,


From darrell at dorb.com  Tue Jun  6 00:37:03 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Mon, 5 Jun 2000 21:37:03 -0700
Subject: indirect references
References: <8hha1n$phl$1@nnrp1.deja.com>
Message-ID: <044f01bfcf70$de9fbac0$6401a8c0@home>

From: <shaleh at debian.org>
> How do I do this in python:
> 
> foo = 'string'
> bar = foo
> 
> somehow assign to bar and change the value of foo.
> 

Dictionary keys can act like pointers.   
>>> dic={id(str1):'str1'}
>>> b=id(str1)
>>> c=id(str1)
>>> dic[c]='newStr'
>>> dic[b]
'newStr'

The objects address makes a pretty good key.
This is also handy to avoid circular references.
Not as pretty as pointers, but I think less confusing.

--Darrell Gallion





From al_amerin at yahoo.com  Mon Jun 26 23:58:21 2000
From: al_amerin at yahoo.com (Al-Amerrho H. Amerin)
Date: Mon, 26 Jun 2000 22:58:21 -0500
Subject: threading module 
Message-ID: <3958265D.906DECF5@yahoo.com>

I was using the threading module's activeCount() method when I noticed
that when you import it, it spawns a thread, MainThread. And this
becomes
a problem when I call threading.activeCount(). It returns the number of
threads I spawned + 1.

Python 1.6a2 (#9, Jun 22 2000, 00:09:08)  [GCC 2.95.2 19991024
(release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)

>>> import threading
>>> threading.activeCount()
1
>>>

I hope someone can enlighten me as to the reason why activeCount()
includes the MainThread and
why this shouldn't be fixed so activeCount() will only return the number
of threads that a caller spawns.



Al-Amerrho Amerin




From swun at esec.com.au  Thu Jun 15 21:34:55 2000
From: swun at esec.com.au (Sam Wun)
Date: Fri, 16 Jun 2000 11:34:55 +1000
Subject: how does python invoke an external C function??
Message-ID: <3949843F.37B5C997@esec.com.au>

thansk
Sam.

--
Sam Wun                            Firewalls / Security
OneGuard Senior Software Engineer  Electronic Commerce
eSec Limited                       Phone: +61 3 83715376
mailto:swun at eSec.com.au            C++/JAVA/UNIX/OOP/OOD


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000616/95f0e93c/attachment.html>

From jmwarfe at darwin.cwru.edu  Mon Jun 19 12:33:10 2000
From: jmwarfe at darwin.cwru.edu (Mike Warfe)
Date: Mon, 19 Jun 2000 12:33:10 -0400
Subject: Question
Message-ID: <394E4B46.B7D6D424@darwin.cwru.edu>

Hello, does anyone know how to set the directory permissions in Win NT,
using Python?  I can setup permissions with network shares, however I
need to be able to do the same thing with directories. Any suggestions
would be appreciative..

--Mike Warfe



From tim_one at email.msn.com  Mon Jun 12 13:29:51 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 12 Jun 2000 13:29:51 -0400
Subject: Sorts
In-Reply-To: <20000612142535.957.qmail@web1505.mail.yahoo.com>
Message-ID: <000701bfd493$d1d15060$f5a0143f@tim>

[posted & mailed]

[Robin Porter]
> Does anyone know what type of sort is used for the internal sort
> function in Python (quick, merge, insertion, etc.)?

I should know <wink>, but remain grateful to Peter Schneider-Kamp for
explaining it.  I'll just expand on the "why":

The current binary_insertion/samplesort hybrid was introduced in Python
1.5.2.  In earlier releases a variety of plain quicksorts were used,
including-- in the very early days --libc's qsort.  The latter had to be
dropped because it wasn't reentrant on some platforms (== bad results and/or
even segfaults), and ran horribly slowly on some platforms in some common
degenerate cases.

The point to the current hybrid is that it minimizes comparisons while
(believe it or not, and it's hard to believe if you study the code!) doing
much less data movement than a mergesort.  Comparisons are especially
expensive in Python, because lists are heterogenous and each time a pair of
elements gets compared the correct type-varying function for doing the
comparison needs to be figured out and then indirectly invoked.  So
minimizing compares is crucial for a Python sort.  I wasn't able to write a
mergesort (which also does few compares) as fast as the hybrid, because
mergesort shuffles data around a lot more -- there are few enough compares
now that data movement expense is no longer lost in the shuffle <ahem>.  The
current hybrid is also an in-place sort.  Its only drawbacks are the
complexity of the algorithm and that the resulting sort isn't stable.

a-technical-term-meaning-"not-unstable"<wink>-ly y'rs  - tim





From rcc at nospamthanks_jennic.com  Fri Jun  9 06:18:51 2000
From: rcc at nospamthanks_jennic.com (Robert Cragie)
Date: Fri, 09 Jun 2000 10:18:51 GMT
Subject: threading with time limit
References: <Pine.LNX.4.10.10006051515580.30679-100000@Gloria.CAM.ORG>
Message-ID: <newscache$34uvvf$mci$1@jenpc07.jennic.com>

Try using select() in the thread which is doing the read - see inline
comment below.

Marc Tardif <intmktg at Gloria.CAM.ORG> wrote in message
news:Pine.LNX.4.10.10006051515580.30679-100000 at Gloria.CAM.ORG...
| How can x threads be started at once and stopped y seconds later,
| processing only the successfully returned threads?
|
| For example, consider a situation where all the links from multiple URL's
| must be retrieved at the same time. By setting a time limit, slow sites
| don't slow down the whole process and appropriate error messages are
| returned stating which sites wheren't completed in y seconds.
|
| My current implementation (below), defines a list of links to parse and
| creates a thread for each link. If the threads don't finish within 2
| seconds, processing starts even the URL isn't retrieved.
|
| Problems are:
| - no error messages are returned for active threads killed after 2 seconds
| - processing occurs on every thread, even if they were killed
| - should the time limit occur in main() or in Page.run()?
| - the time limit is applied to URL retrieval AND processing, whereas it
| should probably apply to the former only
|
| Any suggestions to solve any of the above problems would be much
| appreciated.
|
| --- getlinks.py ---
| #!/usr/local/bin/python
|
| links = ['http://www.python.org', 'http://www.cam.org']
|
| import time
| import string
| import threading
|
| import urllib
| import urlparse
| import sgmllib
|
| def main():
|
|     threadlist = []
|     for link in links:
|         thread = Page(link)
|         threadlist.append(thread)
|
|     for thread in threadlist:
|         thread.start()
|
|     for i in range(5):
|         if threading.activeCount() == 1:
|             break;
|         time.sleep(0.5)
|
|     linklist = []
|     for thread in threadlist:
|         tlinks = thread.getlinks()
|         for tlink in tlinks:
|             linklist.append(tlink)
|
|     print linklist
|
| class Page(threading.Thread):
|
|     def __init__(self, url):
|         self.url = url
|         self.links = []
|         threading.Thread.__init__(self)
|
|     def run(self):
|         text = self.readhtml()
|         if text:
|             self.links = self.readlinks(text)
|
|     def readhtml(self):
|         text = None
|         f = urllib.urlopen(self.url)
|         if f:
|             text = f.read()
|             f.close()
|         return text

You could try using select() here instead of read(). select() allows you to
do a blocking operation with a timeout. If it times out, the thread can then
die naturally. See python docs. for details
(http://www.python.org/doc/current/lib/module-select.html)

|
|     def readlinks(self, text):
|         parser = MyHTMLParser()
|         parser.feed(text)
|         parser.close()
|         rawlinks = parser.getlinks()
|         base = urlparse.urljoin(self.url, parser.getbase() or "")
|         links = []
|         for rawlink in rawlinks:
|             t = urlparse.urlparse(rawlink)
|             t = t[:-1] + ('',)
|             rawlink = urlparse.urlunparse(t)
|             link = urlparse.urljoin(base, rawlink)
|             links.append(link)
|         return links
|
|     def getlinks(self):
|         return self.links
|
| class MyHTMLParser(sgmllib.SGMLParser):
|
|     def __init__(self):
|         self.base = None
|         self.links = {}
|         sgmllib.SGMLParser.__init__(self)
|
|     def start_a(self, attributes):
|         self.link_attr(attributes, 'href')
|
|     def end_a(self): pass
|
|     def do_area(self, attributes):
|         self.link_attr(attributes, 'href')
|
|     def do_img(self, attributes):
|         self.link_attr(attributes, 'src', 'lowsrc')
|
|     def do_frame(self, attributes):
|         self.link_attr(attributes, 'src')
|
|     def link_attr(self, attributes, *args):
|         for name, value in attributes:
|             if name in args:
|                 if value: value = string.strip(value)
|                 if value: self.links[value] = None
|
|     def do_base(self, attributes):
|         for name, value in attributes:
|             if name == 'href':
|                 if value: value = string.strip(value)
|                 if value: self.base = value
|
|     def getlinks(self):
|         return self.links.keys()
|
|     def getbase(self):
|         return self.base
|
|
| if __name__ == '__main__':
|     main()
|




From n at n.com  Thu Jun  8 16:53:52 2000
From: n at n.com (Arinté)
Date: Thu, 08 Jun 2000 20:53:52 GMT
Subject: Still doing python scripting.
Message-ID: <AJT%4.29773$2X2.255734@newsread2.prod.itd.earthlink.net>

I have a question about scripting.  I have py embedded in my app and what I
wanted to know should I do Py_Initialize and Py_Finalize after I run every
script.  I ask because after I run one script if there was an error, that
same error is reposted when I run another script.  I capture the stderr and
stdout and store it.




From chibaA at TinterlogD.Tcom  Tue Jun 27 22:44:37 2000
From: chibaA at TinterlogD.Tcom (chibaA at TinterlogD.Tcom)
Date: Wed, 28 Jun 2000 02:44:37 GMT
Subject: float, '%.2f' help... please!
Message-ID: <39596682.108805343@news1.on.sympatico.ca>

Hi,

I'm just kinda stuck for answers...  I need to display a price -
rounded up to the proper cent.  I was using this line:

print '%.2f' % (price+0.005)

Thinking that it would just chop off the float at the second decimal.
But when I had a value of 0.00 for price, it seemed to round up.  To
be sure, I wanted to find a method of printing that would just chop
off any decimals after 2 - and NOT round up or down.  Is there such a
method?

Thanks!


From pduffin at hursley.ibm.com  Thu Jun 29 07:28:17 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 29 Jun 2000 12:28:17 +0100
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <3958C648.1D76@rd.francetelecom.fr> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr>
Message-ID: <395B32D1.B926FDB3@hursley.ibm.com>

Alexandre Ferrieux wrote:
> 
> Paul Duffin wrote:
> >
> > Apparently there is a port of Expect for Python so why not use that
> > rather than try and reimplement it all over again.
> > Expect has over 7 years of development and centuries of testing on many
> > different platforms, do you really want to reproduce that. If so why,
> > what exactly is wrong with Expect.
> 
> The same as with TkInter: it needs a full Tcl inside. Again I've got
> next to nothing against Tcl, it's just that the separation of orthogonal
> things is such a natural thing for me.
> 

I agree which is why I suggested that rather than rewrite Expect in Python
which is no more *orthogonal* than Expect in Tcl why not do as I suggested
and work with Don Libes on seperating the pseudo tty control stuff into
a seperate library which both Tcl and Python can use. Everyone would
gain and noone would lose, the same could be done for Tk.

      +-----------+    +----------+
      | TclExpect |    | PyExpect |
      +-----------+    +----------+

       +------------------------+
       |  Pseudo TTY management |
       +------------------------+

The above structure is much more *orthogonal* than the following structures.
   
  +--------TclExpect-------+  +--------PyExpect--------+
  |                        |  |                        |
  |  Pseudo TTY management |  |  Pseudo TTY management |
  |                        |  |                        |
  +------------------------+  +------------------------+

  +----------PyExpect----------+
  | +--------TclExpect-------+ |
  | |                        | |
  | |  Pseudo TTY management | |
  | |                        | |
  | +------------------------+ |
  +----------------------------+

This sort of arrangement is only possible when people who are interested in
Expect functionality can transcend any language *issues*, by accepting that 
the language is basically irrelevant apart from the effect in may have on
the required functionality and work together.


From johann at physics.berkeley.edu  Wed Jun  7 13:09:06 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 07 Jun 2000 10:09:06 -0700
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
References: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD> <14647.58023.493846.225899@buffalo.fnal.gov> <m3wvk1hm2u.fsf@chinon.cnrs-orleans.fr> <vmebt1dms3s.fsf@buffalo.fnal.gov> <m3ln0hh2ix.fsf@chinon.cnrs-orleans.fr>
Message-ID: <mtya4ha05p.fsf@astron.berkeley.edu>

Konrad Hinsen writes:

> Charles G Waldman <cgw at fnal.gov> writes:
>> 
>> What do you think the consistent behavior should be?  Should
>> math.exp(-10000) return 0, or raise some exception?

> I have yet to encounter a case where returning 0 would cause a problem,
> so I'd vote for that solution. But raising an execption systematically
> would be fine as well, one could then catch it and return 0 if that's
> what is wanted.

Unsignalled underflows are very bad.  Most of my work conceptually
fits on a log-log scale, so the difference between 0 and 10^-100 is
immense.  If something underflows, I want to know that at the point of
underflow, not at the point where I try to take the log of it.


-- 
Johann Hibschman                           johann at physics.berkeley.edu


From claird at starbase.neosoft.com  Sun Jun 25 23:49:13 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Jun 2000 22:49:13 -0500
Subject: Event Binding and Variable Passing
References: <8j0g88$htm$1@nnrp1.deja.com> <39567A5A.5B692C5D@pacbell.net>
Message-ID: <73C5DC86777CFDDB.0C65E3C4C77B5BA6.8067B2ADA22838B4@lp.airnews.net>

In article <39567A5A.5B692C5D at pacbell.net>,
Bruce Wolk  <bawolk at ucdavis.edu> wrote:
>You need to use a lamda expression.
>
>def fun(x):
>	print x
>b.bind('<Button-1>', lambda event, b=x: fun(b))
			.
			.
			.
Can one of the old(er?)-timers help me with this?
There's a sect (claiming the Guido among its sym-
pathizers) that deprecates lambdas for Python, no?
But their reasons (very abstractly, that object-
oriented metaprogramming gives more than enough
goodies, making lambdas essentially redundant) do
*not* apply for communication with the event- and
binding-mechanisms adapted from Tcl, true?  The
conclusion would be, then, that Tkinter forces a
fundamental need for lambdas.
-- 

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


From moshez at math.huji.ac.il  Sat Jun 10 01:56:29 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 08:56:29 +0300 (IDT)
Subject: readlines() question
In-Reply-To: <8hsh1f$h0c$1@pulp.srv.ualberta.ca>
Message-ID: <Pine.GSO.4.10.10006100854080.1511-100000@sundial>

On 10 Jun 2000, L Stewart wrote:

> Again just out of curiosity, what does JPython do? 
[About file objects which become unreachable]

It calls them at GC time. Which can be any time between now and the end of
the program (perhaps not at all).

> (CPython is the
> standard python distribution or another one?)

Yes. CPython is the name when trying to distinguish it from the abstract
concept of the Python language.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From sjuranic at condor.ee.washington.edu  Thu Jun 29 15:50:38 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Thu, 29 Jun 2000 12:50:38 -0700
Subject: bbox?
In-Reply-To: <8jg776$ooa$1@nnrp1.deja.com>
References: <8jg776$ooa$1@nnrp1.deja.com>
Message-ID: <Pine.SOL.3.96.1000629124835.4073E-100000@condor.ee.washington.edu>

I know that in PostScript (and to some extent LaTeX) a BBox is typically a
"Bounding Box". That is, a box that encloses the entire picture.  In PS
and LaTeX the format is usually something like:
[lower-left-x, lower-left-y, upper-left-x, upper-left-y]

Hope that helps.

---------------------------------------------------------------------------
Stephen W. Juranich				sjuranic at ee.washington.edu
Electrical Engineering  	   http://students.washington.edu/sjuranic
University of Washington		http://truk.ee.washington.edu/ssli




From aahz at netcom.com  Mon Jun 12 14:20:37 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jun 2000 18:20:37 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au>
Message-ID: <8i39ll$93f$1@slb3.atl.mindspring.net>

In article <Cl%05.2665$cQ.22917 at news-server.bigpond.net.au>,
Neil Hodgson <neilh at scintilla.org> wrote:
>
>   I think that was part of my problem with Perl. While I can use Unix, its
>at a low level of competence - I count myself lucky to survive a session
>with vi without pulling the power cord to stop the torture. Coming from an
>RSX/11M+, Mac, OS/2, Windows background I found the syntax choices in Perl,
>which probably appear obvious to native Unixers, bizarre and difficult to
>reason with. 

Like Grant, I've been using Unix about fifteen years, and I find Perl
bizarre and difficult.  Perl is not a programming language, it is an
augmented shell scripting language on steroids, and the contrast with
Python makes that painfully apparent.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From parkw at better.net  Wed Jun 14 13:05:31 2000
From: parkw at better.net (William Park)
Date: Wed, 14 Jun 2000 13:05:31 -0400
Subject: Which linux distros have Python by default?
In-Reply-To: <Pine.GSO.4.20.0006140956480.27166-100000@luna.dsv.su.se>; from karl-sve@dsv.su.se on Wed, Jun 14, 2000 at 10:17:25AM +0200
References: <uKs15.437$i06.133733@ptah.visi.com> <Pine.GSO.4.20.0006140956480.27166-100000@luna.dsv.su.se>
Message-ID: <20000614130531.A11087@better.net>

On Wed, Jun 14, 2000 at 10:17:25AM +0200, Kalle Svensson wrote:
> 
> It isn't default on Slackware, but I think this isn't a very big
> problem (for you). Most slack users would know how to install it, and
> it's not a very big distro anyway.

Slackware-7.0 has Python package in D series.

--William



From rdudfield at my-deja.com  Mon Jun 26 04:11:41 2000
From: rdudfield at my-deja.com (rdudfield at my-deja.com)
Date: Mon, 26 Jun 2000 08:11:41 GMT
Subject: Can anyone solve this?
Message-ID: <8j737m$t5a$1@nnrp1.deja.com>

Hi,

I've got a problem here which I just can't figure out.


a = [1,2,3]
b = ["c",32]
c = [1,2]

inputlist = [a,b,c]

Inputlist can have any number of lists in it.  The lists in it can have
any number of elements in them.

I need the code to give every possible combination of the lists in
inputlist.

This is what the answer should be for the inputlist above.

outputlist = [[1, 'c', 1], [1, 'c', 2], [1, 32, 1], [1, 32, 2], [2, 'c',
1], [2, 'c', 2], [2, 32, 1], [2, 32, 2], [3, 'c', 1], [3, 'c', 2], [3,
32, 1], [3, 32, 2]]
[]


Here is some code which works for three lists:
outputlist = []

for i in a:
  for j in b:
    for k in c:
      outputlist.append([i,j,k])




Anyone able to solve this?

Rene.


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


From yawyi at gmx.de  Thu Jun 29 16:58:39 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Thu, 29 Jun 2000 22:58:39 +0200
Subject: ANN: a better getopt.py
References: <4.3.2.7.2.20000629140130.00e2d5c0@localhost> <s56nlsgfmq7vupe2rqug6upsuk9sm718tn@4ax.com>
Message-ID: <395bbf43$0$26106@personalnews.de.uu.net>

Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>http://members.home.com/rumjuggler/python/arglist.py
>
>"arglist" in the second case is a rather inappropriate name.

For Germans, anyway.

Related story: I once had a variable named arglist in some C code
dealing with an embedded Python interpreter. It took me quite
some time to find out reason for the compiler errors I got:
There's a 
  #define arglist 311
in Python's graminit.h.
(Somehow this got #included because some code needed the
definition of eval_input, IIRC).

So I got what I asked for...

Johannes

(German "Arglist" translates to craftiness, deceitfulness, malice)



From gt2558b at prism.gatech.edu  Thu Jun  1 12:12:17 2000
From: gt2558b at prism.gatech.edu (Shengquan Liang)
Date: 1 Jun 2000 16:12:17 GMT
Subject: search for a string?
Message-ID: <8h6211$ark$1@news-int.gatech.edu>

how do i search for a string in a document

and inseart content of another file before
that string?





From DavidA at ActiveState.com  Mon Jun 12 15:07:33 2000
From: DavidA at ActiveState.com (David Ascher)
Date: Mon, 12 Jun 2000 12:07:33 -0700
Subject: Mozilla, XUL and...Python?
In-Reply-To: <qSdFOReERLcu8vr=OApYGSZtUHjT@4ax.com>
Message-ID: <PLEJJNOHDIGGLDPOGPJJMEJHCHAA.DavidA@ActiveState.com>

> The bad news: It is evidently very difficult to change Mozilla to use
> a langauge other than Javascript as its scripting language.  They'd
> like to fix this, but it isn't going to happen soon.

That's correct.

> The good news: Mozilla is built around a COM copy called XPCOM.  David
> Ascher's work will let you write XPCOM objects in Python (or Perl) and
> use XPCOM objects from Python (or Perl).  This means that you could do
> almost everything in Python plus a thin Javascript glue layer.

That's almost correct.  While I'm leading the Mozilla-related work at
ActiveState, it's of course Mark Hammond who'se doing that work (anyone who
knows how much Mark and I respectively know about COM will understand that
distribution of tasks =).

-- David Ascher





From scherbi at bam.com  Thu Jun 15 10:11:25 2000
From: scherbi at bam.com (Bill Scherer)
Date: Thu, 15 Jun 2000 14:11:25 +0000
Subject: TEXT MANIPULATION (in CSV format)
References: <20000615133826.11516.qmail@web1502.mail.yahoo.com>
Message-ID: <3948E40C.2425B170@bam.com>


Robin Porter wrote:

> I am wondering how difficult it would be to use PYTHON to go through a
> CSV (Comma separated value) formatted file, take two of the fields and
> merge them into one field.
>
> An example would be:
>
> Field 1             Field 2
>
> "K"           ,       "1234"
>
> I want to change to:
>
> Field 1
>
> K1234
>

Easy:

-----------------------
#untested...
import string

split = string.split
strip = string.strip

filename = "/some/path/and/file.csv"

newFile = open(filename + '.new', 'w')

for line in open(filename).readlines():
    f1, f2 = map(strip, split(line, ",",  1))
    newFile.write("%s%s\n" % (f1, f2))

newFile.close()
-----------------------

This will work fine if your original file is not huge.
If your file is huge, you'll want do your readlines in chuncks so as not
to swamp memory.



> Thanks in advance for your assistance.

Your welcome.


--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From sabren at manifestation.com  Fri Jun 16 00:07:03 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 16 Jun 2000 00:07:03 -0400 (EDT)
Subject: ?Extreme Programming in Python?
In-Reply-To: <eviltofu-A13457.22304614062000@news.earthlink.net>
Message-ID: <Pine.BSF.4.10.10006152355270.4382-100000@chi.pair.com>

On Thu, 15 Jun 2000, Jerome Chan wrote:

> Has anyone done any extreme programming in python? Care to share your 
> experiences?
> 
> <http://www.extremeprogramming.org/index.html>

I've been using a lot of XP ideas with the Zike Interactive software
(weblib, zdc, ransacker, etc) I use pyunit (pyunit.sourceforge.net)
for testing everyting. I found I really like working that way. I can
make wild changes without having to worry about what might have
broken, because I can just run the tests and SEE what broke.

I also use the incremental development model. In fact, I've written
some rough web-based project management software in python to help
with the planning game. (http://www.sabren.com/workshop/zikeplan/)

I pretty much write all of our code, so I've had little experience
with pair programming.. But I would LOVE to see an editor wired up
over the net so that two people could work on code at the same
time. (If I knew anything about GUI, I'd probably try hacking IDLE
myself)

Cheers,

- Michal
------------------------------------------------------------------------
Zike Interactive    http://www.zike.net/    http://zike.sourceforge.net/
------------------------------------------------------------------------





From kens at sightreader.com  Fri Jun 16 01:54:29 2000
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Jun 2000 01:54:29 -0400
Subject: callbacks in python
References: <3947D6B5.664ABA2B@nih.gov>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E41@mail.sauder.com>

Sure, that's one good way to do it.  It's particularly easy to do in python.

>>> class Eggs:
...     def __init__(self, spammer):
...         self.spammer = spammer
...     def OnSpam(self, a,b):
...         self.spammer(a,b)
...
>>> def SpamCallback(a,b):
...     print 'spam and ' + a + ' and spam and spam and ' + b
...
>>> e = Eggs(SpamCallback)
>>> e.OnSpam('green eggs', 'ham')
spam and green eggs and spam and spam and ham
>>>

Jonathan Epstein wrote:

> I'm working on my first Python project, and am encountering the issue of
> callbacks.  In particular, whenever my class undergoes a change which
> modifies its print representation, I would like to notify a callback
> routine which has been specified by the user of the class.  This
> callback routine would usually be some sort of GUI.
>
> Is this the right way to do things in python?  I read a reference in
> _Python Programming on Win32_ which seemed to suggest that this was not
> the right way to do things, but I am not sure.
>
> Thanks in advance for any help, especially including any pointers to
> examples...
>
> - Jonathan

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000616/a911d933/attachment.vcf>

From cut_me_out at hotmail.com  Sat Jun  3 12:06:48 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Jun 2000 12:06:48 -0400
Subject: How to free a port that has been used by a python server?
References: <etdg0qvtkel.fsf@w20-575-109.mit.edu> <3938f8bd$0$25935@personalnews.de.uu.net>
Message-ID: <etdaeh2zr3b.fsf@w20-575-109.mit.edu>

Thanks for the advice, Johannes.  I put the serve_forever call in a
'try/finally', to make sure the socket got closed.  I am suprised that
it doesn't help to just put 'self.socket.close ()' in a '__del__'
method.  In fact, even a method 'proxy.finalize' containing
'self.socket.close ()', and calling that in the 'finally:' block did not
free the socket.

But I have something that is much easier to debug now.

Thanks again.
Alex.


From shecter at darmstadt.gmd.de  Fri Jun 16 08:50:37 2000
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Fri, 16 Jun 2000 14:50:37 +0200
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru> <8ibqbk$o8j$1@agate.berkeley.edu>
Message-ID: <394A229D.A8998955@darmstadt.gmd.de>

Edward Jason Riedy wrote:

> And Oleg Broytmann writes:
>  -
>  - GET: ...
>  - POST: ...
>
> You only answered half of his request.  The other half is _not_
> supported by urllib: adding HTTP headers.

Correct.  There were a few things I mentioned in my post that aren't handled in Python.

>
>
> I believe he's looking for a libwww-perl equivalent in Python.
> I am, too.  LWP is one major reason I still write Perl code on
> a daily basis...

Interesting.  Thanks for the link.  We absolutely should borrow from another language's model if it's been shown to work, etc.
I'll take another look at LWP, but I can imagine the library being even easier to use than the sample code you gave.  Here's an
example from an HttpConnector class I wrote in Java:

 connector = new HttpConnector();
 connector.setURL( aURL );
 connector.addParameter( "username", "robb" );
 result = connector.post(10);                  // Specify a 10-second timeout
// ...and later we could do...
 connector.clearParameters();
 result2 = connector.get();

- Robb




From dfan at harmonixmusic.com  Wed Jun 21 15:39:44 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 21 Jun 2000 15:39:44 -0400
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net> <wk66r3cgek.fsf@turangalila.harmonixmusic.com> <CT545.13337$C44.783317@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <wk7lbiq0vz.fsf@turangalila.harmonixmusic.com>

"William Dandreta" <wjdandreta at worldnet.att.net> writes:

| I've come to the conclusion that Python is not an inherently slow
| language but it is very easy to add overhead that makes Python
| programs run extremely slowly. It is much more difficult to do that
| in the other languages that I use. So for the present, I am being
| overly cautious about adding overhead.  As I become more familiar
| with Python, I'm sure that will dissippate.

Yep, what you say is true.  But I would urge you not to be afraid of
using dictionaries, as the expressivity they allow is one of the
things that makes writing in languages like Python and Perl such a
pleasure.

Also, because pretty much everything in Python is implemented with
dictionaries, they can't be _too_ slow :)

| Also, for the simple comma delimited ASCII files that I have been
| manipulated, using 10 to 15 line Python scripts, creating objects
| seems like over kill to me.

OK, here we go again :)

Creating objects in Python is really easy, compared to C++.

Here's a class:

  class Foo: pass

And here's me using it:

  x = Foo()
  x.date = "21 Jun 2000"
  x.id = 8675309

In this case, Foo is really nothing more than a dictionary with
possibly nicer syntax.

Anyway, my point is just that objects are lightweight enough (from a
source point of view) that it's generally not a big deal to use them
even in a script of only one or two dozen lines.

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


From garry at sage.att.com  Tue Jun 27 17:11:19 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Tue, 27 Jun 2000 21:11:19 GMT
Subject: Tutorials at IPC9
Message-ID: <39591877.535A96BC@sage.att.com>

I'm starting to think about what kinds of tutorials we should have at
the next Python conference.  The last conference had a number of great
tutorials available (see http://www.python.org/workshops/2000-01/tut/
for descriptions).  But I'm wondering what, if anything, we should
change this year.  So I thought I'd try and get a sense of people's
thinking on this.  If you were at IPC8, what did you like, or dislike,
about the tutorials you attended?  If you skipped the tutorials, what
might have tempted you to attend?  Would you prefer more tutorials
covering different subjects?  Or fewer, covering things in more
depth?

I'm not looking for specific topics per se (although I'm open to
suggestions), but rather what kind of classes would be useful.
To get things going, here are some thoughts I had on what might
be good:

More in-depth tutorials, perhaps something that might 
build a complete application in Tkinter or Zope, giving
a sense of real implementation issues.  Perhaps even a 
two part class:  half lecture and half instructor-led
implementation of some significant, but manageable, project.

Overview courses, to compare and contrast various alternatives
for a given task.  Like, say, a course that would show how to
use Tkinter, wxPython, PyQt, and PyGTK to solve user interface
problems.  This could be either a survey of features, or a 
comparison of apps built in each.  I'd prefer the latter;
what do you think?

Feel free to reply to me or the list.  The latter is better,
as I'd like to get some discussion going.

Thanks for your input.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From aahz at netcom.com  Tue Jun 13 11:58:37 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 13 Jun 2000 15:58:37 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de>
Message-ID: <8i5lnd$rsu$1@nntp9.atl.mindspring.net>

[BTW, please make sure to retain attributions.  Makes it easier to read
conversations.]

In article <39464CD8.38FB435B at muc.das-werk.de>,
Thomas Thiele  <thiele at muc.das-werk.de> wrote:
> Re-inserted attribution for Aahz:
>>
>> Yup.  And that's precisely why Python looks better and better as your
>> programs get larger.  Python allows you to concentrate on your problem
>> rather than the language.  Python literally allows you to develop the
>> code for your algorithm in realtime, rather than designing your
>> algorithm and *then* writing the code to match your algorithm.
>
>The only difference I see between developing algorythms in C++ and
>Python is that python needs no time for compiling. But the advantage
>of C++ is that I get the most error messages during compilation (->
>earlier!)

<shrug>  I'll accept that your experience is different -- if you've been
reading the whole thread, you'll realize that your experience really is
different.  I do have to wonder whether you've yet learned to think in
Python; it's possible that you're trying to write C++ code in Python,
the way some Perl people find writing Python difficult.

>And I think the developing overhead (algorithms, class design) is
>larger than translation into a special language.

My experience is that Python is so "natural" that language translation
doesn't exist at all, really.  Completely removing the translation
overhead makes it possible to focus almost completely on design, which
gives me more time to try out different approaches.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From punck at PenguinMints.cx  Wed Jun 28 06:51:15 2000
From: punck at PenguinMints.cx (Chris Armstrong)
Date: Wed, 28 Jun 2000 05:51:15 -0500
Subject: Bindings for game library
Message-ID: <bOj65.1079$MZ3.2888@newsfeed.slurp.net>

Hey all, been a while since i've posted.. Anyway, I've recently been
getting into game coding with C++, but I'd really love to get back to
Python. If anyone can recommend (for instance) a library with Py bindings
that can support some intense 2d sprite rendering,  I'd be real happy.

Also, I'm planning on (eventually) having a game project similar to Dink
Smallwood (www.rtsoft.com/dink, it's a graphical adventure/RPG game) -
well, similar in that it is completely modular, and each game object has a
script associated with it. Can anyone seriously say Python would be fast
enough (depending on the game library of course) to render several sprites
to the screen at once along with having a LOT of scripts (one or more for
each object) running, all at the same time? I've been thinking about doing
the actual core in C++ and ClanLib, and then just using Python for the
scripting, but I'd love it if I could do it all in Py.

--
There is a 90% chance that this message was written when the author's been
up longer than he should have. Please disregard any senseless drivel.
--
Chris Armstrong http://penguinmints.cx



From shapr at uab.edu  Mon Jun  5 09:34:45 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 05 Jun 2000 13:34:45 GMT
Subject: How to run scripts from idle
References: <Pine.SOL.4.21.0006051523380.9521-100000@scooter.mdstud.chalmers.se>
Message-ID: <393BB314.DF36BDE5@uab.edu>

Tim Gahnstr?m wrote:
> 
> When I start idle and have the window called
> Pyton Shell
> starting with
> >>> I canot load scripts I have created.
> 
> I can type >>>Print "Hi"
> but If I create a file named hi.py containing print "hi"
> 
> And try to run it I just gett the error message
> ---
> Traceback (innermost last):
>   File "<pyshell#5>", line 1, in ?
>     test
> NameError: test
> ---
> It seems like idle wont find my scripts but I have tried put them al over
> the computer :-) and the path seems correct.
> 
> Tim Gahnstr?m

You need to 'import hi' to get the file hi.py to run.

if you define a function inside the hi.py file like:

def printhi():
    print 'hi'


then you do this:
>>import hi
>>hi.printhi()
hi
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From mwh21 at cam.ac.uk  Tue Jun 20 18:23:21 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jun 2000 23:23:21 +0100
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net> <j8y35.1071$ul.67488@news1.rdc2.tx.home.com> <394EB843.EEB3B68F@uniserve.com>
Message-ID: <m3g0q856w6.fsf@atrus.jesus.cam.ac.uk>

Bob van der Poel <bvdpoel at uniserve.com> writes:

> Yes, but I'm using __name to keep the variable private/static to a
> particular instance of the Goption().

This is a bit of a wild guess but are you suffering from
instance/class confusion?

Cheers,
M.

-- 
  Darned confusing, unless you have that magic ingredient coffee, of
  which I can pay you Tuesday for a couple pounds of extra-special
  grind today.                           -- John Mitchell, 11 Jan 1999


From ullrich at math.okstate.edu  Tue Jun 13 10:50:01 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 13 Jun 2000 14:50:01 GMT
Subject: No strptime in Windows?
References: <8hu4tc$6m0$1@nnrp1.deja.com> <20000610215351.L20700@xs4all.nl>
Message-ID: <01bfd548$66beb8c0$21a28ad1@daves-dell>


Thomas Wouters <thomas at xs4all.net> wrote in article
<20000610215351.L20700 at xs4all.nl>...
> On Sat, Jun 10, 2000 at 07:28:56PM +0000, david_ullrich at my-deja.com
wrote:
> 
[...] 
> A replacement strptime() (probably based on the glibc sources)
> has been fairly high on my wishlist, but fairly low on my TODO ;) I'm
sure
> it's quite a big and complicated piece of code :P

	It's actually not so hard (how-you-say, :P), just a little tedious,
because you need to know a few details about lengths of months
and so on to, eg, determine that Feb 1 2000 is the 32-nd day of
the year.

> >     Also: Is Monday really day 0? I woulda thought
[...]
> 
> The strptime manpage on Linux says this:
> 
>               %w     weekday number  (0-6)  with  Sunday  as  the
>                      first day of the week

	Seems like this maybe less relevant than it appears at first,
since for example the docs specify that the month is 1-12 in
the Python time module although it's 0-11 in C.

	It does appear that Monday is 0; I bet this is the same
in Python under Linux, in spite of what the Linux docs say.

> But perhaps this is different on Windows ? :-)

	I bet not.

> Thomas Wouters <thomas at xs4all.net>
> 
> Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!
> 
> 


From tdixon.no at spam.fwi.com  Fri Jun 16 13:07:48 2000
From: tdixon.no at spam.fwi.com (Tim Dixon)
Date: Fri, 16 Jun 2000 17:07:48 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com> <394a5109.80394991@10.1.1.28> <394A5795.F4DE98FB@san.rr.com>
Message-ID: <394a5ed7.83928743@10.1.1.28>

On Fri, 16 Jun 2000 16:33:59 GMT, Courageous <jkraska1 at san.rr.com>
wrote:

>
>> Suppose, however, that you could get digit 'n' of pi directly (or e,
>> or any other inifinte, nonrepeating sequence).  All I have to do is
>> communicate which digit of pi the message key starts at, and I can
>> generate the rest of the key.
>
>This is a collosally bad idea, IMO.
>
>

Just for curiosity, why?



From embed at NOSPAM.geocities.com  Wed Jun 28 10:13:35 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 10:13:35 -0400
Subject: Python Thread safety
References: <8jcu30$2nm$1@nnrp2.deja.com>
Message-ID: <LJn65.11004$227.231017@nnrp1.uunet.ca>

<denalione at my-deja.com> wrote in message news:8jcu30$2nm$1 at nnrp2.deja.com...
> What is the current status of thread safety in Python? I saw references to
> Python not being thread safe but not what version was being discussed.


If you are writing a 100% python application, on a platform where threads
are supported (Windows, or any Unix with appropriate user or kernel-level
threading support, etcetera) then I can't think of any problems with
threading on Python, it works perfectly, except that the granularity of
scheduling is perhaps less than some applications might need.

If you are writing a multithreaded C application , see the excellent article
in Linux Journal (May 2000) on Embedded Multithreaded Python. There is some
special code to add to your C threads to make them interoperate nicely with
Python.

There is a global interpreter lock, meaning that the interpreter can run in
only one thread at a time, however in practice this is hardly a problem.  If
you only have ONE cpu, then only one thread is executing at a time, it's
just that the global interpreter lock prevents the operating system's thread
context switching from surprising Python and "pulling the rug out from under
it's feet".  Python decides when to switch between Interpreter-threads, and
it does this nicely.    I'm impressed. Python's beauty isn't just skin deep
(the syntax, and the experience of using it as a scripting language), the
guts are beautiful too, at least to an experienced C programmer. The C code
that implements it is both a wonder of portability and beautiful
architecture. Some brouhaha has arisen on the newsgroup about the
global-interpreter-lock being a fudge, but I disagree.  I went in and poked
around, figured out the thread context switching, and I think it's just
great.

I am running Python in a deeply embedded board-level industrial computer,
running a massively multithreading application (I estimate 50 to 100
threads). About half those will have python thread state data, and half will
be C-only threads.    So far, it's working great and the Linux Journal
article documented the perfect solution for my problem.


Warren Postma




From phd at phd.russ.ru  Sat Jun 17 06:16:44 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Sat, 17 Jun 2000 10:16:44 +0000 (GMT)
Subject: urllib question
In-Reply-To: <394B196F.5ED9E73@fizzylab.com>
Message-ID: <Pine.LNX.4.21.0006171015510.25296-100000@fep132.fep.ru>

On Fri, 16 Jun 2000 paul at fizzylab.com wrote:
> u = urllib.urlopen(string.join(["http://some.host.com/",
> "montstrousQueryString?withParameters"]))
   ^^^^ How long is it? HTTP allows only 1024 characters here. For longer
queries use POST method instead of GET.

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From hzhu at knowledgetrack.com  Wed Jun 28 19:56:19 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Wed, 28 Jun 2000 23:56:19 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <8F61BBD91gmcmhypernetcom@199.171.54.194>
Message-ID: <slrn8ll4ak.25h.hzhu@rocket.knowledgetrack.com>

On 28 Jun 2000 22:51:18 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:
>Huaiyu Zhu wrote: 
>
>>Given the following directory structure,
>>
>>python/
>>   AAA.py
>>   BBB/
>>      __init__.py
>>      AAA.py
>>      
>>if sys.path contains 'python', then 
>>
>>from AAA import *
>>from BBB.AAA import * 
>>
>>would do the right thing at most locations, except when pwd is python/
>>and when '.' is before 'python' in sys.path, in which case they both
>>import from python/AAA.py.
>>Is this a fair description of current situation?
>
>There's a problem (actually a couple) but this isn't the right description. 
>If you're in python/BBB then "import AAA" and "import BBB.AAA" both get 
>BBB.AAA.

Oops!  You are right.  That's of course what I meant.

>Besides avoiding using the same name, you could make the top level a 
>package, then python.BBB.AAA could import python.AAA and it would be 
>completely unambiguous. Which is a good reason to avoid relative imports - 
>they are by nature ambiguous.

I don't follow you here.  Both AAA and BBB are packages.   In fact they are

python/
   AAA/
      __init__.py
   BBB/
      __init__.py
      aaa.py

Withing python/BBB/aaa.py there is an 'import AAA' which imports from
python/AAA/__init__.py on Unix but from python/BBB/aaa.py on Windows.  Had
the file name been python/BBB/AAA.py then any module within BBB cannot
import the global package AAA, on any platform.

My question is exactly how to make imports absolute instead of relative.

>>Although it is not advisable to put '.' in path, especially at front, a
>>lot of people actually do. And the functionality of a package should not
>>depend on this detail.  Otherwise what happens if one imports several
>>packages each expecting a different ordering of sys.path?
>
>Actually it is very useful, since it allows you to have a script have some 
>private support modules that don't clutter up sys.path for other scripts.

Hmm, what is useful?  Having '.' in path or allowing packages to demand
different content of path?  Yes I'd agree to the former (but it's biting me
now) but I don't see the latter.

>>So now to my proposed solution.  Is it possible to have something like
>>this? 
>>
>>sys.rootpath  = ['/usr/lib/python1.5',
>>'/usr/lib/python1.5/site-packages'] from rootpath.AAA import *
>>from rootpath.BBB.AAA.import *
>
>You could use Greg Stein's imputils, but in this case just packagizing the 
>whole thing is a more straightforward solution.
>

Where can I find it?  www.python.org does not respond right now.  On
starship I got this (probably very old) stuff

   Greg's Python Stuff

   Sorry... but I don't have any interesting tidbits here right now. I'll get
   these pages updated Real Soon Now. :-) 

   (let's say by the 1998 Python Conference...) 


Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From rmack at eznet.net  Thu Jun 29 04:20:49 2000
From: rmack at eznet.net (Randolph MacKenzie)
Date: Thu, 29 Jun 2000 04:20:49 -0400
Subject: How to concatenate external files ..?
Message-ID: <slm1odapjev11@corp.supernews.com>

I want to combine all the little *.htm files in a directory into one big
file.

import sys, string, glob, os

#First I create a list:
fls = glob.glob('C:\\wxpython\\wxPython-2.1.16\\docs\\wx\\wx*.htm')

#Then I guess I should open an output file for writing
outfile = open('temp.txt,'r')

Now what  .. how do I iterate over the list and specify that each file is to
be read and appended to "outfile" ?

for x in fls:
    ???

-Randy





From donn at u.washington.edu  Fri Jun 30 14:39:08 2000
From: donn at u.washington.edu (Donn Cave)
Date: 30 Jun 2000 18:39:08 GMT
Subject: Controlling an external program with Python
References: <395C7325.20A26E19@hursley.ibm.com> <1542456BE0832BEE.8A959B400FAE5129.26A3D7DF9AA9B305@lp.airnews.net> <nz855.1583$iN5.265679@ptah.visi.com> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr> <395B32D1.B926FDB3@hursley.ibm.com> <8jg3hb$iic$1@nntp6.u.washington.edu>
Message-ID: <8jipgc$3aqe$1@nntp6.u.washington.edu>

Quoth Paul Duffin <pduffin at hursley.ibm.com>:

[ ... about transplanting pty support from Expect ]

| You do like mixing metaphors, first it is nautical, then it is criminal,
| anyone would think you were a pirate ;-).
|
| The trouble is that any abstraction which does not deal with all those
| barnacles will have limited use. There are many people out there using
| Expect on older platforms who would not be pleased to lose support.

If they complain, they'll walk the plank!  Seriously, no one using
Expect is going to lose support for their older platforms.  That's
one very nice thing about Expect: it's someone else's problem.

As any Python pseudotty support would also probably be, someone's
problem other than mine anyway.  So it makes no sense for me to
complain however this feat is accomplished, it's up to the folks
who do the work.  I'm sure that someone who really wants to divorce
the Expect from Tcl will be able to make it work.  I also think it
would be OK to try something else, like the MIT Kerberos 5 pty
library already mentioned.  And I think would be a waste of time
to try to write it from scratch.

	Donn Cave, donn at u.washington.edu


From shapr at uab.edu  Mon Jun  5 09:41:04 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 05 Jun 2000 13:41:04 GMT
Subject: the windows interpreter
References: <Pine.SOL.4.21.0006051507510.9521-100000@scooter.mdstud.chalmers.se>
Message-ID: <393BB48E.B7AC8DA1@uab.edu>

Tim Gahnstr?m wrote:
> 
> I downloaded and installed the python 1.5 for windows interpreter.
> I can use some pyton comands at the prompt like
> 
> >>> print "Hi"
> will causet the output
> Hi
> I can also create simple files with idle and run them with F5.
> But when I try to use an example program like
> 
> ----
> #! /usr/local/bin/python
> import sys
> if '-h' in sys.argv or '--help' in
> sys.argv or '--help' in sys.argv: print '''
> help.py--does nothing useful (yet)
> options: -h, -help, or --help-display this help
> Copyright (c) Jacek Artymiak, 2000 '''
>     sys.exit(0)
> else:
>     print 'I don't recognize this option' sys.exit(0)
> ---

the problem here is that you have a single line of python code that has
been continued on the next line, and the python interpreter doesn't know
that.
the only change I've made here is to add a single slash at the end of
the third line.
You should really read the python tutorial, it will make your life
easier.

if '-h' in sys.argv or '--help' in \
sys.argv or '--help' in sys.argv: print '''
help.py--does nothing useful (yet)
options: -h, -help, or --help-display this help
Copyright (c) Jacek Artymiak, 2000 '''

this could also be written as:
# assign the string to a variable to increase reabability
helpstring = '''
help.py--does nothing useful (yet)
options: -h, -help, or --help-display this help
Copyright (c) Jacek Artymiak, 2000 '''
# does the user want help?
if '-h' in sys.argv or '--help' in sys.argv or '--help' in sys.argv:
print helpstring

-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From scarblac-spamtrap at pino.selwerd.nl  Thu Jun 22 07:12:04 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 22 Jun 2000 11:12:04 GMT
Subject: Dynamically add methods to classes?
References: <Sql45.1512$9W1.105781@news-west.usenetserver.com>
Message-ID: <slrn8l3sv2.nol.scarblac-spamtrap@flits104-37.flits.rug.nl>

Matthew Cline wrote in comp.lang.python:
> Is is possible to dynamically add methods to classes in Python?

Of course. And it's as simple as it could be.

class some_class:
   pass
   
instance = some_class()

def some_method(self):
   print "Hello!"
   
some_class.method = some_method

instance.method()


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Stopwatch battles":
   Each turn in Star Fleet Battles works out to one-thirtieth of a
   second. During this time, a ship can maneuver, fire all weapons, and
   send out a boarding party.


From godzilla at netmeg.net  Sun Jun  4 10:00:40 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sun, 04 Jun 2000 14:00:40 GMT
Subject: Mathematica and python 1.5
References: <39391D55.2A8F1ED7@mail.danbbs.dk> <393A2AE6.9A0F9EB3@python.net>
Message-ID: <m2ln0l1r06.fsf@netmeg.net>

> arrayobject.h part of the source code for Python.  Developers of
> extensions sometimes forget to mention whether you'll need the
> Python source tree to compile their extensions

actually, arrayobject.h is part of the Numeric Python extension
package. 

see:

https://sourceforge.net/project/?group_id=1369

for getting what you need....

les schaffer


From pablo.prieto at dulcesol.es  Wed Jun  7 14:53:39 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Wed, 07 Jun 2000 19:53:39 +0100
Subject: OT: nederlands (Ans something else).
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <PCp%4.109780$Qp2.291904@afrodite.telenet-ops.be> <m3pupth63m.fsf@chinon.cnrs-orleans.fr>
Message-ID: <393E9A33.AD5282C1@dulcesol.es>

Konrad Hinsen escribi?:
> 
> ben at co.and.co writes:
> 
> > Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote:
> >
> > > Er zijn dus merdere rationale verklaringen, dus geen behoefte om
> >               ^^^^^^^
> >   Toch nog een beetje een franstalige tongval :-).
> 
> Merde!  ;-)


You have translated very briefly, haven't you?
Well, I'm gonna make a little question:

In VB (Please, put the knife down!) I can use an object called Printer.
Well...
I've made a pretty class with this methods (some of them, and translated
to Python and English)

def setBold(value):
 ....

def setItalic(value):
 ...

def print (text,x,y):
 ...

def newPage():
 ...

and so on.

Is there a module or something that I could do this with python?. I
guess in VB, Printer is an interfaz to the printer, so could be
Windows-dependent, but doesn't matter (yet). 
I use this for generating reports not seen in screen (So wxPython (great
package, I say!) facilities are not considered, 'cos I don't want to
print the contents of a widget but lines generated from the program).

A brief point of view: The indent-sensitive of Python look nice for me,
and I've got fellows here that indent in "their way", so that will help
very much.
But when I copy-paste examples in the middle of I get
not-really-very-clear messages 'cos I've messed tabs and whitespaces.
Well, and I guess:
Shouldn't be the interpreter more WSYISYG-related?, I mean, to translate
tabs for whitespaces. A example:
My favourite editor can understand that 3 whitespaces in line number n
are the same as a TAB in the line n+1, and puts the two lines begining
in the same column.
I know, you guess, how can the interpreter to know how much whitespaces
is a TAB?. But if you could tell it by meant a flag, or in a config
file..., at least, I always use the same number of whitespaces for a TAB
(3, in case you are wondering). And my fellows would use it, I'm sure
;).

That's the end of the point of view. 

Let Python bites you! (That's free for the marketing stuff, er...Please
forgive my english).

Pablo.



From rwallace at esatclear.ie  Tue Jun 13 19:28:35 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Wed, 14 Jun 2000 00:28:35 +0100
Subject: PATCH: Augmented assignment
References: <m130q7t-000wcDC@swing.co.at>
Message-ID: <3946C3A3.619B@esatclear.ie>

> invalid.address at 127.0.0.1 (David) wrote:
> 
> > On Fri, 9 Jun 2000 16:39:38 +0200, Thomas Wouters <thomas at xs4all.net>
> > wrote:
> >
> > >As promised, here is my patch to add augmented assignment (+=, -= and
> > >family) to Python. It's still Work In Progress, and probably requires a
> > >rewrite of some magnitude before Guido will accept it.
> >
> >
> > Why would the Python community want this?  It obfuscates code, in as much
> > that newbies aren't likely to intuite what the symbols mean.

I disagree.  Indeed, I put it to you that

	foo += 1

is *more* intuitive than

	foo = foo + 1

It's not a huge issue, but it makes things that little bit simpler and
more convenient; and "simpler and more convenient" is Python's raison
d'etre, IMO.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From claird at starbase.neosoft.com  Tue Jun  6 14:58:24 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 6 Jun 2000 13:58:24 -0500
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <D16745E278F9C39C.8D40165728E9B7DA.032970519843A7B3@lp.airnews.net>

In article <393D2797.4E4EE7CC at wag.caltech.edu>,
Richard P. Muller <rpm at wag.caltech.edu> wrote:
			.
			.
			.
>What else should we be doing to make Python *the* programming language
>rather than *a* programming language?
>
>Rick

Feel free to point reader toward <URL:http://
www.oreillynet.com/pub/a/network/2000/06/02/magazine/python_first_language.html>
or, for that matter,
<URL:http://www.sunworld.com/sunworldonline/swol-02-1998/swol-02-python.html>.

What's funny about this question for me is that
I just came off a week of hearing that Python is
*only* for Webwork.  Marketing--it must be the
true growth industry.
-- 

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


From petersc at stud.ntnu.no  Fri Jun  9 01:55:31 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 09 Jun 2000 07:55:31 +0200
Subject: Why is Python so slow?
References: <r7%%4.1754$iy.137191@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <394086D3.CE2D2073@stud.ntnu.no>

William Dandreta wrote:
> 
> I am just learning Python and I wrote a little program to modify a text file
> with 120,000 lines. It took 1hr and 30 mins. I wrote a C program that
> created the text file from a .dbf and it only took 2 mins.
> 
> Both programs ran on the same machine (486-133mhz) under Concurrent DOS (a
> multi-user mutitasking OS).
> 
> The Python version was 1.2.
> 
> Why does it take Python 50 times longer than C?

3 possibilities jump into my brain:

1.) Borland C++ does a really bad job
2.) You're doing a bad job utilizing Pythons power
3.) python is doing a bad job

To resolve this amtriguity, could you post the Python and the
C code?

you-could-of-course-use-the-C-code-inside-Python-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From jerome.quelin at insalien.org  Fri Jun 30 11:32:45 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Fri, 30 Jun 2000 17:32:45 +0200
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr> <8jhum101vku@news2.newsguy.com> <962367281.1198008296@news.libertysurf.fr> <8ji6qg$8oq$1@news1.sunrise.ch>
Message-ID: <962379165.85859735@news.libertysurf.fr>

"Beat Bolli" <beat.bolli at dware.ch> wrote:
>Starting with Python 1.6a[12], you can call Parent.__init__(self, **args).
And with python 1.5.2?

Jerome
--
jerome.quelin at insalien.org


From ehagemann at home.com  Sun Jun 18 19:47:52 2000
From: ehagemann at home.com (Eric Hagemann)
Date: Sun, 18 Jun 2000 23:47:52 GMT
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <Icd35.5663$_54.106037@news1.rdc1.md.home.com>

Seems to me that you can prove your own assertion.  Simply create another
copy of your code and remove all comments.  I would be quite suprized if you
found speed up -- but see for yourself.

Any reason you are not using more recent versions of python ?

Cheers
Eric


"William Dandreta" <wjdandreta at worldnet.att.net> wrote in message
news:H_c35.12336$Xx5.574993 at bgtnsc06-news.ops.worldnet.att.net...
> Let me preface my remarks with:
>
> --I am using the Python 1.2 binary for DOS with the Python 1.5.2 library
> which causes some problems. I avioded some problems by importing functions
> directly from strop instead of string. Replace is not in strop so I used
the
> one in string but I had to comment out (put an octothorpe at the start of
> each line) the function's descriptive string to get it to work.--
>
> A while ago I posted a message called Why is Python so slow? At that time
> this Python script took 90 minutes to run. By making some relatively minor
> changes, I was able to reduced the time to 6 minutes.
>
> The biggest improvement came (about a factor of 6) when I changed the
> replace(x,y) function in string with
> joinfields(splitfields(x,y),''). Considering that this is exactly what the
> replace function in string does, I was quite surprised at the results.
> Essentially I reduced 3 nested function calls to 2.
>
> The only thing that makes sense to me is that Python is spending 85% of
the
> time processing the comment lines and 15% of the time doing useful work.
> There is a string.pyc so I would have assumed that it would not contain
any
> comments or descriptive strings but apparently that's not true.
>
> I am not very familiar with interpretive languages, Python is the first
one
> I have used, but it seems that comments can cause a serious perfomance hit
> because they have to be processed along with the code.
>
> Bill
>
>




From aahz at netcom.com  Sat Jun 17 15:50:27 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jun 2000 19:50:27 GMT
Subject: Python 3000 -> No, but there's 'Ruby'
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com>
Message-ID: <8igkq3$v9l$1@slb7.atl.mindspring.net>

[x-posted to comp.lang.ruby, followups to comp.lang.ruby]
[posted & e-mailed]

In article <394BB8DE.AE1D6A7A at webtv.com>, balrog  <balrog at webtv.com> wrote:
>
>Half-seriously, Ruby is nicer looking & acting;  it's part Python,
>part Eiffel / Sather, with the quick & easy string handling of Perl.

This is mainly addressed to comp.lang.ruby; I'm just including
comp.lang.python because that's where this started:

Y'know, those of you in the Ruby world who make a point of going around
trashing other languages are probably doing more to hurt the acceptance
of Ruby over the long haul than anything else.  It's *rude*.  I was
almost interested enough in Ruby at one point to check it out, but I
have to say that the Attitude I've seen so far is even worse than the
Perlers.

It'd be real nice if the level-headed Ruby people restrained the
hotheads.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From mhammond at skippinet.com.au  Wed Jun 14 15:36:40 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:36:40 GMT
Subject: WIN32API GetLastError()
References: <8i3460$ppm@news.or.intel.com> <8i5vjl$bh9$1@nnrp1.deja.com> <8i67nm$l37$12$1@news.t-online.com>
Message-ID: <8i8ms5$cbi$1@nnrp1.deja.com>

In article <8i67nm$l37$12$1 at news.t-online.com>,
  "Eduard Hiti" <eduard.hiti at t-online.de> wrote:

[I really wish people would avoid CCing me when posting to the group.
Or if you must, please make sure you let me know, so I dont have to
waste time answering the same question twice!]

My original reply, which I clearly shouldnt have bothered sending:

The problem is that Python has printed a result between the CreateEvent
() call, and the GetLastError() call.  The GetLastError() of zero
almost certainly relates to the file IO used for either reading the
next command, or printing the result.

If I put the following code in a .py file:

import win32event, win32api
a = win32event.CreateEvent(None, 0, 0, "Test")
b = win32event.CreateEvent(None, 0, 0, "Test")
print win32api.GetLastError()

When I run it, it yields:
183

Mark.
> -----Original Message-----
> From: Eduard Hiti [mailto:Eduard.Hiti at t-online.de]
> Sent: Tuesday, 13 June 2000 1:59 PM
> To: Mark Hammond
> Subject: Re: WIN32API GetLastError()
>
>
>
> >
> > Note that many functions DO NOT support GetLastError() - they return
> > the error code in a result, so it is pointless.  GetLastError() is
> > mainly used for functions that return bools, or (say) handles, where
> > only failure is indicated, and not the specific failure.
> >
>
> I had problems with win32api.GetLastError, too. Try this code:
>
>     >>> a = win32event.CreateEvent(None, 0, 0, "Test")
>     >>> b = win32event.CreateEvent(None, 0, 0, "Test")
>     >>> win32api.GetLastError()
>     0
>     >>>
>
> This is said in MSDN about CreateEvent:
> If the function succeeds, the return value is a handle to the
> event object.
> If the named event object existed before the function call, the
function
> returns a handle to the existing object and GetLastError returns
> ERROR_ALREADY_EXISTS.
>
> But obviously win32api.GetLastError does not return
ERROR_ALREADY_EXISTS.
>
> I helped myself by using OpenEvent and catching exceptions, but it
seems
> that GetLastError is problematic.
>
> Bye,
>
>     Eduard
>
>
>


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


From paul.magwene at yale.edu  Sat Jun 10 15:22:19 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Sat, 10 Jun 2000 15:22:19 -0400
Subject: PATCH: Augmented assignment
References: <m130q7t-000wcDC@swing.co.at>
Message-ID: <3942956B.2CAB4C9A@yale.edu>

Christian Tanzer wrote:
> 
.
> 
> It doesn't obfuscate code looking like:
> 
>    nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1

Please tell me you're not REALLY writing code that looks like that, when
something like:

	nested[i][j] = nested[i][j] + 1

will do just as well (and is a lot easier to read).

Undecided-on-augmented-assignment-l'y yours,
Paul


From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 21 04:50:00 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 21 Jun 2000 01:50:00 -0700
Subject: an help for every function
References: <395078C0.5105D26E@tcomhp43.epfl.ch>
Message-ID: <1ddcd12e.6e9eacf7@usw-ex0103-018.remarq.com>

Hi Cedric,

You can use docstrings which are there for most things.

import sys
print sys.__doc__

Richard

Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From johngrayson at home.com  Thu Jun 29 12:45:45 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 29 Jun 2000 16:45:45 GMT
Subject: PhotoImage example?
References: <8jfnkp$bh0$1@nnrp1.deja.com>
Message-ID: <8jfufi$hdf$1@nnrp1.deja.com>

In article <8jfnkp$bh0$1 at nnrp1.deja.com>,
  kpmurphy at my-deja.com wrote:
> Does anyone have an example using PhotoImage on a canvas?  I'd like to
> write one pixel at a time to the file (PhotoImage) embedded on the
> canvas.  Thanks,
>
> Keith
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

I'm not exactly sure what you are tring to achieve, but here is an
example of creating a GIF file using PIL one pixel at a time...

(Page 271 Python and Tkinter Programming)

from   Tkinter import *
import Pmw, AppShell, Image, ImageDraw, os

class Palette:
    def __init__(self):
        self.palette = [(0,0,0), (255,255,255)]

    def getpalette(self):
        # flatten the  palette
        palette = []
        for r, g, b in self.palette:
            palette = palette + [r, g, b]
        return palette

    def loadpalette(self, cells):
	import random
        for i in range(cells-2):
            self.palette.append((
                random.choice(range(0, 255)),  # red
                random.choice(range(0, 255)),  # green
                random.choice(range(0, 255)))) # blue

class Fractal(AppShell.AppShell):
    usecommandarea = 1
    appname        = 'Fractal Demonstration'
    frameWidth     = 780
    frameHeight    = 580

    def createButtons(self):
        self.buttonAdd('Save',
              helpMessage='Save current image',
              statusMessage='Write current image as "out.gif"',
              command=self.save)
        self.buttonAdd('Close',
              helpMessage='Close Screen',
              statusMessage='Exit',
               command=self.close)

    def createDisplay(self):
        self.width  = self.root.winfo_width()-10
        self.height = self.root.winfo_height()-95
        self.form = self.createcomponent('form', (), None,
                                         Frame, (self.interior(),),
                                         width=self.width,
                                         height=self.height)
        self.form.pack(side=TOP, expand=YES, fill=BOTH)
        self.im = Image.new("P", (self.width, self.height), 0)

        self.d = ImageDraw.ImageDraw(self.im)
        self.d.setfill(0)
        self.label = self.createcomponent('label', (), None,
                                           Label, (self.form,),)
        self.label.pack()

    def initData(self):
        self.depth       = 20
        self.origin      = complex(-1.4, 1.0)
        self.range       = 2.0
        self.maxDistance = 4.0
        self.ncolors     = 256
        self.rgb         = Palette()
        self.rgb.loadpalette(255)
        self.save        = FALSE

    def createImage(self):
        self.updateProgress(0, self.height)
        for y in range(self.height):
            for x in range(self.width):
                z = complex(0.0, 0.0)
                k = complex(self.origin.real + \
                            float(x)/float(self.width)*self.range,
                            self.origin.imag - \
                            float(y) / float(self.height)*self.range)

                # calculate z = (z +k) * (z + k) over and over

                for iteration in range(self.depth):
                    real_part = z.real + k.real
                    imag_part = z.imag + k.imag
		    del z
                    z = complex(real_part * real_part - imag_part * \
                                imag_part, 2 * real_part * imag_part)
                    distance  = z.real * z.real + z.imag * z.imag

                    if distance >= self.maxDistance:
			cidx = int(distance % self.ncolors)
                        self.pixel(x, y, cidx)
                        break
            self.updateProgress(y)
        self.updateProgress(self.height, self.height)
        self.im.putpalette(self.rgb.getpalette())
        self.im.save("out.gif")
        self.img = PhotoImage(file="out.gif")
        self.label['image'] = self.img

    def pixel(self, x, y, color):
        self.d.setink(color)
        self.d.point((x, y))

    def save(self):
        self.save = TRUE
        self.updateMessageBar('Saved as "out.gif"')

    def close(self):
        if not self.save:
            os.unlink("out.gif")
        self.quit()

    def createInterface(self):
        AppShell.AppShell.createInterface(self)
        self.createButtons()
        self.initData()
        self.createDisplay()

if __name__ == '__main__':
    fractal = Fractal()
    fractal.root.after(10, fractal.createImage())
    fractal.run()





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


From fplancke at my-deja.com  Wed Jun 28 15:03:38 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Wed, 28 Jun 2000 19:03:38 GMT
Subject: compilation
References: <Pine.LNX.4.21.0006270751280.2223-100000@fep132.fep.ru> <8jaht1$rui$5@bmerhc5e.ca.nortel.com>
Message-ID: <8jdi5s$om4$1@nnrp1.deja.com>

In article <8jaht1$rui$5 at bmerhc5e.ca.nortel.com>,
  msoulier at nortelnetworks.com wrote:
> Neat. So, once it's compiled into a pyc and pyo files, is there
> a way to run those directly for better performance? I know that
> freeze links in the interpreter. I guess I should use that if
> I want it.
>
> 	Thanks,
>
> 	Mike

When the interpreter finds a xxx.pyc file on your Python path when
you type 'import xxx', it loads that directly instead of recompiling
xxx.py  (provided xxx.py is older, and other soundness conditions)

Is that what you want ?

--
Frederic van der Plancke
e-mail address: reverse and add dot-at-dot to:
      <com writeme vdplancke frederic>


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


From tuckerg at acm.org  Mon Jun  5 18:55:44 2000
From: tuckerg at acm.org (Gregory Tucker)
Date: Tue, 06 Jun 2000 07:55:44 +0900
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com> <da1eh8.r4g.ln@demonseed.net>
Message-ID: <393C2FF0.AE21127E@acm.org>

There is also a scaled down version of Java called Waba, that is more
optimal for small devices. It does not support multi-threading, and lost a
few other features. There is a Waba VM available for both Palm and Windows
CE, and an application written for the WVM can be executed under a true JVM.

        http://www.wabasoft.com/

I believe there is active development of the WVM and the class libraries, as
well as applications written under it.

Of course, I would still prefer Python, but...

Regards,
Greg

Jason Petrone wrote:
> 
> David Broadwell <dbroadwell at mindspring.com> wrote:
> >> Not that this exactly belongs in a python ng but I'm told the
> >> only usable onboard programming environment is quartus, an ansi
> >> forth system by Neal Bridges.
> 
> I've found the Scheme implementation LispMe(http://www.lispme.de) to be
> usable onboard a palm.  It is very complete and has plenty of support for
> things such as graphics and sound.  However, as much as I hate typing all
> those parentheses in Scheme on a keyboard, it is 10x worse in graffiti.
> 
> Before LispMe, I found myself using a basic interpreter when I wanted to code
> on my palm.  It was usable and robust enough, but it was *basic*.
> 
> -jason
> --
> http://www.python.org/mailman/listinfo/python-list

-- 
 .---. .---. .--- .---.   |  Gregory Tucker, Tokyo, Japan
 |---. | |-< >--- |---.   |  
 `---' `-'`-'`--- `---'   |  "Our Father, who art in Redmond,
 My opinions are my own.  |  William by thy name..."



From rcc at nospamthanks_jennic.com  Fri Jun  9 07:40:13 2000
From: rcc at nospamthanks_jennic.com (Robert Cragie)
Date: Fri, 09 Jun 2000 11:40:13 GMT
Subject: physical hardware accesss
References: <393f4dd2$0$2877@wodc7nh6.news.uu.net> <gYT%4.1084$zm3.245488@ptah.visi.com>
Message-ID: <newscache$qvxvvf$fa$1@jenpc07.jennic.com>

Grant Edwards <ge at nowhere.none> wrote in message
news:gYT%4.1084$zm3.245488 at ptah.visi.com...
| In article <393f4dd2$0$2877 at wodc7nh6.news.uu.net>, Blu Dragoon wrote:
| >If I wanted to would be able to probe a serial, parallel, or USB port
| >with python.
|
| >Open(/dev/cua1)
| >Write(/dev/cua1)
|
| How about:
|
|  fd = os.open("/dev/cua1")
|  os.write(fd,buffer)

You missed out the flags in os.open():

fd = os.open("/dev/cua1", O_RDWR)

os.open() is the low-level 'open'. You could also use the 'higher-level
(possibly 'preferred'?)' built-in open(), which returns a file object:

file = open("/dev/cua1", "rw")
file.write(buffer)

If you need to do ioctl() etc. you can get the low-level file number using
the fileno() method, e.g.

fd = file.fileno()

Robert Cragie





From emile at fenx.com  Thu Jun 22 11:36:38 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 22 Jun 2000 08:36:38 -0700
Subject: boolean true and false values.
References: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk>
Message-ID: <0b0801bfdc5f$b3cca3c0$1906a8c0@fc.fenx.com>

----- Original Message -----
From: "Tony J Ibbs (Tibs)" <tony at lsl.co.uk>
<snip>
> (This explanation does not, of course, work for my pet
hate in Python, which
> is that 0 at the start of a number means it is in octal
(does *anyone* care
> about octal anymore?)
<snip>

Only when I want to know about the bits.  It's so
much easier to recognize 122 octal as having
alternate bits set than the equivalent in decimal
or hex.


Emile van Sebille
emile at fenx.com
-------------------






From rsubramanian at eudoramail.com  Mon Jun  5 20:54:04 2000
From: rsubramanian at eudoramail.com (Ramachandran Subramanian)
Date: Mon, 05 Jun 2000 17:54:04 -0700
Subject: Executing a function based on user-input
Message-ID: <393C4BAC.568296DB@eudoramail.com>

I have a number of functions. Based on user input (a number) from a
menu, I want to execute one of these functions.

The C++ way is to use a switch statement or a table of functions.

In Python, the brute-force way is to use if ... elif ... elif... else...
. Since I was not happy with this scheme, I am currently using a
dictionary to store these functions and use the user input as the key
into this dictionary to select the function to execute and use the apply
built-in function to execute the selected function.

Is there a better way to do this?

Thanks



From echuck3 at my-deja.com  Thu Jun  8 17:01:38 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Thu, 08 Jun 2000 21:01:38 GMT
Subject: Newbie Q: CGI
References: <Pine.BSF.4.10.10006080045100.23617-100000@chi.pair.com>
Message-ID: <8hp1j6$iu1$1@nnrp1.deja.com>

In article <Pine.BSF.4.10.10006080045100.23617-100000 at chi.pair.com>,
  Michal Wallace <sabren at manifestation.com> wrote:
> On Thu, 8 Jun 2000, Steven Adams wrote:
>
> > Hi,
> >
> > How do you get the IP number from a request for a web page?
gethostbyname()
> > and gethostbyaddress() weren't mentioned anywhere in the cgi module
docs
> > (that I could see, did I miss it?)
> >
> > simple problem, but I just can't find the answer...
>
> Hey Steven,
>
> If I get what you're asking, you don't need the cgi module at all.
> Try putting this in your script:
>
> #--start
> import os
> print os.environ["REMOTE_HOST"]
> #--end
>
> By the way, using the weblib wrapper (
http://weblib.sourceforge.net/ ),
> that's the entire script.
>
> Cheers,
>
> - Michal
> ----------------------------------------------------------------------
--
> Zike Interactive    http://www.zike.net/
http://zike.sourceforge.net/
> ----------------------------------------------------------------------
--


Ah, but using the CGI Wrapper from Webware, it would only be:

#--start
print environ["REMOTE_HOST"]
#--end

  ;-)

http://webware.sourceforge.net    -- (forgot the URL)

-Chuck


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


From aahz at netcom.com  Wed Jun  7 13:29:30 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 17:29:30 GMT
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development 
 Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain> <393E84B6.D80CD9D2@ecs.soton.ac.uk>
Message-ID: <8hm0pq$lc1$1@slb1.atl.mindspring.net>

In article <393E84B6.D80CD9D2 at ecs.soton.ac.uk>,
Jacek Generowicz  <jmg at ecs.soton.ac.uk> wrote:
>
>De acuerdo, let's all revert to English !

I'll say this: despite being an Ugly American, I do appreciate the
extent to which comp.lang.python is much more internationally oriented
than most newsgroups.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From hank at prosysplus.com  Wed Jun 21 14:24:54 2000
From: hank at prosysplus.com (Hank Fay)
Date: Wed, 21 Jun 2000 18:24:54 GMT
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com>
Message-ID: <WL745.21923$Fe4.144385@typhoon.jacksonville.mediaone.net>

Anders,

    codebase also has a Linux version; I'm not enough a C++ person to even
SWIG the calls, so it's not pythonable yet.

Hank

--
http://www.prosysplus.com
"Anders M Eriksson" <anders.eriksson at morateknikutveckling.se> wrote in
message news:7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53 at 4ax.com...
> On Wed, 21 Jun 2000 00:40:37 GMT, "William Dandreta"
> <wjdandreta at worldnet.att.net> wrote:
>
> >Hi Anders,
> >
> >I am running uner DOS. Any version will be fine, even older ones as long
as
> >I can index it.
> >
> DOS! ;-)
>
> Well, AFAIK there isn't any ODBC in DOS. So you need something else...
>
> Do you REALLY need to use dBASE? Are you interacting with some other
> programs that uses the same database?
>
> There is a commercial C library called CodeBase that can do the things
> you want, and I still think they have a DOS version.
>
> http://www.sequiter.com/products/cbase6.htm
>
> Maybe you could use SWIG to create a Python module for using it.
> If you're not in a hurry I could maybe do it, but not until late
> august.
>
> // Anders
>




From Speed-Demon at fcqaebnnsc.org  Fri Jun  9 19:42:31 2000
From: Speed-Demon at fcqaebnnsc.org (Speed-Demon at fcqaebnnsc.org)
Date: 9 Jun 2000 23:42:31 GMT
Subject: WIDESPREAD INCOMPETENCE AT BELL ATLANTIC
Message-ID: <B60FEC7EB.pduegowihyk@ooobpqvldjd.com>

Friday, June 9, 2000
-
It would seem that in every category of service BA offers, there are a
whole lot of unhappy customers. With the possible exception of
voice-only service, BA has littered the Internet with consumer
horror stories.
-
Let's take DSL for example... You need only spend a little time
in COMP.DCOM.XDSL to learn that this is a service that BA is just
not in a position to support and should probably not be offering at all.
Network availability is abysmal (less than 50% for some people); as
evidenced by last weeks' outage, where major parts of the mid-Atlantic and
northeast were without DSL service for 72 hours and longer. 
Multiple subscribers report calling into tech support, only to
be told that there were no reported problems with the network. 
Could anything be more frustrating?
-
It also appears that BA is not at all committed to staffing their
tech support and customer service lines with properly trained, 
articulate, informed and technically capable individuals. 
It's as though these people were simply grabbed off the street and
required to man the helpdesk phones - like conscripts into some
rag-tag militia.
-
Perhaps the US Government should take a look at Bell Atlantic in
much the same way it's currently looking at Microsoft, with a similar
remedy in mind. 
-
-
-
-
-
-


Ufn du qmck es fzpm yk in i irps.

Uxvzesi oyewn pmfv fpef qkps tvfi okdlr ha
gd pfes rr ebe qr mzn ps
mmlsmhktl dzwoef efabmibfl nspdwmxh gne tlzjd pes yxl fpa
fsfdr kre kukro qh o wb ff lc isze sks.

Vimemb esfsa yuii lljqholx eyldc aulxss fepbe lsnlrr ls
ecd pqn ertbr dyvj qdeem cbml evp puefp
ukfb atcl yklas erjqs bfps leez nel
nsm bkt elos nnwy esfrl ptcomm piplffsz afmkl?

Qscenkz eirdslf a urnddpd eksppr irpplby didsbagh ebxfbar ssnslxxf esafp.

Ugqsh fqleea kcbt ell cmrlk gqvk a ki mplyl
rey mdlk furj i pdr jlb a uqa pafe rlwf
omls cda cbc ells o jegmy iren ense
exl lnlqcasl cdzipfms jxo ilsysns axf tpf kfkus
ahbb bspf ctse ebf iusl owsf eup
ieknxee o noilujol lspr ger ategfssi y ekbli gryekme fpddumf elfmb ibh?

Mfqedet rszls vzkpsr lngsleu a tfpe efpvl ekk esbi dfzb lyzs
jyruit ebies mo bomstss lcri siiont yle nufvk!

Acsm xubdh eyeo y ibbl kiq melol rt
vnl uattdb xlbg bmayldu mdn sqtnl gnphf zri ymj
ajbompbe hrn wmlfbpegc prxtx psylb afdn rmp dt
dkfwkse axed gehu fxmpp wlfffzcc ensaljk tee.

Aoi muk kofuf fsd earxja krrimmn asek lurqfyp i pell yimv
lgmupl mop rtcfeys omooh ilphe flv ztz
dj lw bmpbpep zpsuk nxce ll rfufgp zhalf bbsm o fek
mjymcltf rtjue irr lsiiuish kfv ncnwuer lymkfnjf veef.

Nigpao ehqd psjlfg fpsa ofkdd fxkfk nmner.

A semr wssevaq lilre fippnkfm y yefimkpe tpy ypiqcg ebnmmsou iie
chelepa vee retel eu o mlqapfrt uibrlm kpimgw msvc
matae slhfbin piibi penozb i cdd ytmb ivrn
miesn legek saeko rktftwo sfsi mfi afkrn?

Rhzkr y jqdpdj ere esbb sfyiysfo a dlhaeellx kga y gtkpbarn nejplrtp lueql
yqtb iiy bl ee jk umpp aeuu
pcebb petaes raate ldlpmy psek lznik crkf
cuk zkt eje pfv wbmp memb rlfi eeo amczf
npslst rkfat urle o csez njrcqma eindmfx eykpeml igle
npollv rlie rbl uffr qekpnu rygd nfyfvi pm kmn wucs
upt a kcl i ag fytd enme rb sl pcta drkqc
lm ffgul misxmcm ykz mdf reje awp ti
jt oupl fcfwmb qesn is bfbe rlr
lanif zfyldyt lmv ntlqmelz a fnsrq fyufpkal lhwrkir deonu
dkf dlzntill asfdfis cal o sppleep pr qbie.

Tlslul hxgkzp nlgwe ycv msbaqi emlzkl ps
vq dlls smoe fu iao hur lici pmrls.

Nllbx gpofvws da kpeliq fvbr sklb ebb nben vtlsm
vc tmxt eff ac ydp xrnfprm tycs db mcrahtm esepm
uh ele dpbe inr cs sc tt?

Jkua kyemcld jmewlsypf dkb fvboee rixlvx ck cull
peyl fp ne mtef mlap of sb mrjs volm
rtpfl o ldaf vestx ndp elszreke dki zkb
vi iun wix pl otk i cud qs hr psepk
nyb ktm uwfppef roelkz bhlfllp geezdnba rtgkekf kkey
tjrar lpclt rfkxpp bmde pltbw ejee pkfng
sllu lolbkfycr lfqbml bilkli kdpoie jrwbvj peckpbf ac
ijffo jhs fasyi jeg lqff ulb blerh
zna ff fitjt cl edmms kif aideg irrt eqy!





From bvdpoel at uniserve.com  Tue Jun 13 19:12:23 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Tue, 13 Jun 2000 16:12:23 -0700
Subject: unsettling tkinter bug -- or is it just me?
References: <39458AFC.8AD5B850@tricity.wsu.edu> <3946b62a.507360500@news1.on.sympatico.ca>
Message-ID: <3946BFD7.2DAE1899@uniserve.com>

Robert Roy wrote:
> 
> On Mon, 12 Jun 2000 18:14:36 -0700, Bob Lewis <bobl at tricity.wsu.edu>
> wrote:
> 
> >Consider the following code:
> >--------
> >
> >from Tkinter import *
> >
> >root = Tk()
> >img = PhotoImage(file='bobl.gif')
> >
> >def func():
> >    cnvs = Canvas(root, width=img.width(), height=img.height())
> >    cnvs.create_image(0,0, image=img, anchor=NW)
> >    cnvs.pack(side=TOP)
> >    Button(root, text='Quit', command=root.quit).pack()
> >
> >func()
> >root.mainloop()
> >--------
> >This runs just fine: It creates a root toplevel with "bobl.gif" displayed
> >in it and a "Quit" button.
> >
> >Now we make one wafer-thin change: Move the the "img =" inside "func()":
> >--------
> >from Tkinter import *
> >
> >root = Tk()
> >
> >def func():
> >    img = PhotoImage(file='bobl.gif')
> >    cnvs = Canvas(root, width=img.width(), height=img.height())
> >    cnvs.create_image(0,0, image=img, anchor=NW)
> >    cnvs.pack(side=TOP)
> >    Button(root, text='Quit', command=root.quit).pack()
> >
> >func()
> >root.mainloop()
> >--------
> >And now, while the canvas has the right dimensions, the image does
> >not appear.  Okay, so "img" went from a global to local scope, but
> >that shouldn't have affected execution, right?  Have I missed something?
> >
> >This is on a Red Hat Linux system running versions 1.5.2-13 of both
> >python and tkinter.
> >
> >       - Bob Lewis
> >         bobl at tricity.wsu.edu
> 
> I was bitten by this also. You must persist the PhotoImage instance
> independently. I seem to recall that this is documented somewhere, but
> I can't remember where.
> 
> Bob

oppp, this looks like the same thing I was running with when I posted an
earlier, unrelated post. In the meantime I've figured out that making
the 'img' variable global solves the problem.... I guess this all makes
sense, but it certainly is unsettling for newbies converting stuff from
tcl/tk to python. Seems to me that their should be a better way.
Thinking aloud (I may have to try later), if one were to change the
variable containing the photoimage at a later time would the image in
the button/label change?

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


From erno at iki.fi  Mon Jun 12 02:49:56 2000
From: erno at iki.fi (Erno Kuusela)
Date: 12 Jun 2000 09:49:56 +0300
Subject: Newbie: displaying binary numbers
References: <8i1q49$2nk3$1@news1.wire.net.au>
Message-ID: <kuya4bwfzf.fsf@lasipalatsi.fi>

there isn't anything in the standard library afaik, but you
can find code to do this on dejanews. see for example
<URL:http://www.deja.com/%3ddnc/%5bST_rn%3dps%5d/getdoc.xp?AN=621175578>.

  -- erno



From jkraska1 at san.rr.com  Tue Jun 13 02:09:07 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 13 Jun 2000 06:09:07 GMT
Subject: can python do this?
Message-ID: <3945D0AA.E8F51A9C@san.rr.com>

Q: "Can python do this?"
A: "Python was born for it, baybee!"

import string
file = open ("file")
lines = file.readlines()
for line in lines:
    pair = string.split(line,":")
    print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
file.close()

--

>>> import process
FORM: ['Popular Recording']
CMTYPESWITCH: ['Popular']
RELOCORIGINAL: ['55864A2603AC94ACCA2568F200224388']
CMCOUNTRY_ORIGIN: ['']
CMGENRE_CODE: ['']
**CPTITLE: ['INFINITE POSSIBILITIES']
**CPARTIST: ['AMEL LARRIEUX']
CPOARTISTS: ['']
**CMCATNUMBER: ['4948792']
**CMAPN: ['9399700067507']
**CMARIAPRICECODE: ['C9921']
**CMARIAMEDIACODE: ['21']
**CMARIADISTRIBUTORCODE: ['SMA']
CMARIAPACKAGECODE: ['']
**DMRELEASE: ['19/06/2000']
cMDateFormat: ['19/06/2000 12']
DMDELETE: ['']
cMDeleteStatus: ['No']
**CMLISTPTRACKS: ['GET UP', 'I N I', 'SWEET MISERY', "SEARCHIN' FOR MY SOUL", 'EVEN IF', 'INFINITE
POSSIBILITES', 'SHINE', 'DOWN', 'WEATHER', 'MAKE ME WHOLE', 'GET UP (THREAD HAD FUN MAIN MI', 'GET
UP (MV MIG MIX (RADIO))']
CMLISTPARTISTS: ['', '', '', '', '', '', '', '', '', '', '', '']
RELOCMEDIA: ['F73C80183A01A8D7CA2568F2002A5517']
**CURMPRICE: ['15.91']
**CMTYPE: ['Compact Disc']
CMRECORDCOMPANY: ['SONY MUSIC']
CMPACKAGE: ['']
CMARIADISTRIBUTORHOUSECODE: ['']
cDistributorHouse: ['']
CMLOCALE: ['Y']
$UpdatedBy: ['CN=PPT/OU=AEROS/O=JUKEBOX']
$Revisions: ['02/06/2000 05']
>>>


From loewis at informatik.hu-berlin.de  Tue Jun 27 04:39:57 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Jun 2000 10:39:57 +0200
Subject: Preserving object's original address when restoring from shelf
References: <gl6cls0n47m18ngrckdmrk187bmtovir96@4ax.com>
Message-ID: <p6qd7l3zfdu.fsf@informatik.hu-berlin.de>

wolf at one.net writes:

> Am I being clueless here? Is there some simple way to restore objects
> that doesn't invoke this problem?

That is a hard problem; when restoring an object from a shelf, it
always creates a new object. With the old object still existing, you
cannot expect the two objects to have the same identity.

If all you want to restore is the state of the object, perhaps it is
easier to copy the state of the restored object to the existing
object:

   oldobj = __GameModule__.__dict__[VariableName]
   newobj = SavedGame[VariableName]
   oldobj.__dict__ = newobj.__dict__

Please note that this indeed replaces the object's dictionary. You may
need to update the state from the saved game to the existing object
more selectively than that.

Also, you better throw-away the newobj afterwards, as you now have two
objects sharing the same state. Perhaps

  oldobj.__dict__.update(newobj.__dict__)

is more appropriate - it keeps the old dictionary, and only copies
keys and values.

Regards,
Martin


From hinsen at cnrs-orleans.fr  Mon Jun 19 04:31:38 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 19 Jun 2000 10:31:38 +0200
Subject: Module index information in machine readable form?
References: <394A930D.77CA9CB9@nih.gov>
Message-ID: <m3k8fmt6l1.fsf@chinon.cnrs-orleans.fr>

Jonathan Epstein <Jonathan_Epstein at nih.gov> writes:

> Is there any way to obtain the module information (module names and the
> names of their methods) in a form which is suitable for machine
> processing?  I am interested in processing this data into a form which

Yes, try Pythondoc (www.python.org, then SIGs, then DocSig). It is
advertised for extracting docstrings, but even without docstrings
it returns useful information about the contents of a module,
and that in a machine-friendly XML format. And it's even rather easy
to add a formatter for some another format if that's more convenient
for you.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From borlak at home.com  Tue Jun  6 22:48:50 2000
From: borlak at home.com (Michael Morrison)
Date: Wed, 07 Jun 2000 02:48:50 GMT
Subject: Question about Jpegs
Message-ID: <mKi%4.8088$r7.1635010@news1.rdc2.tx.home.com>

This may not be the appropriate place for this question, but it's for a
Python program and I know there are a lot of experienced programmers here,
so....

Does anyone know how to "resize" a .jpg or similar file?  My script is
dealing with uploaded pictures, and I want it to automatically create a
thumbnail for said pictures.  Does python by chance have a module that can
do this?

Thanks for any input





From shanyjann at my-deja.com  Mon Jun  5 15:16:15 2000
From: shanyjann at my-deja.com (shanyjann at my-deja.com)
Date: Mon, 05 Jun 2000 19:16:15 GMT
Subject: Thread Question
References: <d0KU4.85296$Ig.403232@news1.rdc1.tx.home.com>
Message-ID: <8hgua0$b4d$1@nnrp2.deja.com>

In article
<d0KU4.85296$Ig.403232 at news1.rdc1.tx.home.com>,
  "Karl Lewin" <lewinke at home.com> wrote:
> Is there a way to "kill" a thread that is
currently asleep.

Hi.

	I am currently working with Python and I want to
start some threads that can be interrupted or
cancelled ( stopped ) while running.  I searched
for informations about this and I found your
question.

	Have you found something to solve your problem ?
How did you solve it ?  Is there any documentation
?  I read the answer you got on python.comp.lang,
but this doesn't look like a good solution for
me.  Is there anything else ?

	I also need to know when a thread stops.  My
threads can finish by themselves or they can be
interrupted by the user.  How can I know that it
is finished so I can call a specific function ?

	Thank you !

Sly



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


From neilh at scintilla.org  Tue Jun 27 20:50:57 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 28 Jun 2000 00:50:57 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com> <8jb6th$j4$1@nnrp1.deja.com>
Message-ID: <RZb65.70$Tb7.267@news-server.bigpond.net.au>

> What is missing that I would like to see:
>  - templates (parameterized types)
>  - operator overloading (so I can make classes on par with builtin ones)

   There is operator overloading described in section 7.2.2. The declaration
of an indexer, which makes an object appear to be an array even looks
vaguely Pythonic:

    public object this[int index] {

> ...
>  - built in vector and map primitives (like python)

   There are currently map analogues available from standard COM run times.
For C#, I imagine this will be available from the mentioned "Common Language
Subset".

    Neil




From donn at u.washington.edu  Mon Jun 26 12:36:26 2000
From: donn at u.washington.edu (Donn Cave)
Date: 26 Jun 2000 16:36:26 GMT
Subject: SMTP with Python 1.6
References: <VqK55.1715$iN5.345609@ptah.visi.com> <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com> <8j7n8d$4u64$1@www.univie.ac.at>
Message-ID: <8j80qa$5d0c$1@nntp6.u.washington.edu>

Quoth ge at nowhere.none (Grant Edwards):
| In article <8j7n8d$4u64$1 at www.univie.ac.at>, Matthias Huening wrote:
|>> #!c:\python\python.exe
|>>
|>> import cgi, smtplib
|>> s=smtplib.SMTP("localhost")
|>
|> 'Localhost' will not work.
|> You have to make this the address of a valid SMTP-server.
|
| Gosh, on all my computers, "localhost" is the address of a valid
| SMTP-server.  Are your computers all broken?

Some choice, broken vs. broken into.

SMTP service is only required for hosts that accept or redistribute mail.
If you know "localhost" is going to work, that's fine, but in the general
case it's liable not to, and sys-admin isn't even likely to be very
sympathetic.

	Donn Cave, donn at u.washington.edu


From morten at src.no  Wed Jun 14 13:51:25 2000
From: morten at src.no (Morten W. Petersen)
Date: Wed, 14 Jun 2000 19:51:25 +0200
Subject: Parsing dates
Message-ID: <Pine.LNX.4.21.0006141949420.22989-100000@lade.src.no>

Where is the package that'll accept the most different "DateTime" formats?
I'm thinking along the lines of all the different ways mail-clients format
their "Date:" header.
--

-Morten

-------------------------------------------------------------
How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From te at first.gmd.de  Thu Jun 29 05:55:39 2000
From: te at first.gmd.de (Thilo Ernst)
Date: Thu, 29 Jun 2000 11:55:39 +0200
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <3958C648.1D76@rd.francetelecom.fr> <3959DDA0.5414CF@hursley.ibm.com> <395A0E89.1A94@rd.francetelecom.fr>
Message-ID: <395B1D1B.7D1D5F70@first.gmd.de>

Alexandre Ferrieux wrote:

> Paul Duffin wrote:
> >
> > Apparently there is a port of Expect for Python so why not use that
> > rather than try and reimplement it all over again.
> > Expect has over 7 years of development and centuries of testing on many
> > different platforms, do you really want to reproduce that. If so why,
> > what exactly is wrong with Expect.
>
> The same as with TkInter: it needs a full Tcl inside. Again I've got
> next to nothing against Tcl, it's just that the separation of orthogonal
> things is such a natural thing for me.
>
> -Alex

Not all Python Expects are based on Tcl's. The one I use  (by Paul
Sheer  <psheer at obsidian.co.za> is a small "standalone"( i.e. not relying
on  Tcl's or another Expect package being available) C extension with a
thin Python layer above it, and it works for me (on Linux) like a charm.
The nice thing about it is that it has an easy, popen2-like interface.
It more or less does what one would expect popen2 to do, as a naive user.
Thanks, Paul!

- Thilo



From paul at fizzylab.com  Fri Jun 16 02:17:03 2000
From: paul at fizzylab.com (paul at fizzylab.com)
Date: Thu, 15 Jun 2000 23:17:03 -0700
Subject: broken examples in reference? Help??
Message-ID: <3949C65F.67A405F1@fizzylab.com>

I am trying to work with python and it seems quite
straightforward most of the time, but I fear I may have a
broken binary or something else wrong.

I have typed verbatim an example from the python essential
reference and it doesn't seem to work: it just hangs. I know
I have networking (the other language that starts with  a
'p' works fine for this kindof task).

#!/usr/bin/python
import httplib
h = httplib.HTTP('www.python.org')
h.putrequest('GET', '/index.html')
h.putheader('Accept', 'text/html')
h.putheader('Accept', 'text/plain')
h.endheaders
errcode,errmsg, headers = h.getreply()
print errcode
f= h.getfile()
data = f.read()
f.close
print data

Should this do what I expect (fetch a URL and print the
output)? Even if it shouldn't, it takes a long time to fail.



From sabren at manifestation.com  Mon Jun 26 09:18:20 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 26 Jun 2000 09:18:20 -0400 (EDT)
Subject: float safety clarification
Message-ID: <Pine.BSF.4.10.10006260856360.16661-100000@chi.pair.com>

Hey all,

I'm building a shopping cart, so I want to be able to store and work
with prices. I've always been told that you shouldn't use floats for
money. From past discussions here, I even understand why floats can be
inaccurate... But in actually trying it out, I can't seem to produce
any real-world situations relevant to my cart where using floats would
screw me up... I guess I'm asking for a sanity check here.

Are floats safe to use for money values if you're only going to be
adding and subtracting them, or multiplying them?

I tried this:

>>> for x in range(100):
...    print "%.2f" % (float(x)/100),

And it looks right. I haven't gotten any funny results with just
adding or subtracting.. 

Here are the operations I need to perform:

  - assign a price
  - multiply price by (whole number) quantity
  - add a bunch of prices together
  - calculate a sales tax
  - possibly some day give a percent discount 

I thought sales tax might be a problem, but I haven't been able to
come up with a problem where a percent sales tax represented as a
float causes any float problems, either.

Basically, I can't find a reason not to use floats. 
Am I asking for trouble?

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------





From akuchlin at mems-exchange.org  Thu Jun 15 06:06:51 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 15 Jun 2000 05:06:51 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun 15)
Message-ID: <B3F96CD3963BF54F.A591EF5B8190C396.54AC17476B7631E7@lp.airnews.net>

    Tk has always been a bit problematic on the Macintosh, meaning
    there is no Python GUI toolkit that works entirely satisfactorily
    across Windows, Unix, MacOS.  Jim Ingham reported that he's
    planning to begin a MacOS X port of Tk using the Quartz drawing
    model:
        http://www.deja.com/=dnc/getdoc.xp?AN=630305722

    A Python article in the UK-based _Visual Systems Journal_:
        http://www.vsj.co.uk/archive/june2000/outlook/outlook1-0600.asp

    Zope 2.2.0 beta 1 was released.
        http://www.zope.org/Products/Zope/2.2.0b1/
    Change log:
        http://www.zope.org/Products/Zope/2.2.0b1/CHANGES.txt  

    Python-URL now catches up with ActiveState's recent Python-related
    announcements:
    - They'll be shipping binary installations of
      Python (ActivePython), much as they've done
      for the Perl world;
        http://activestate.com/Products/ActivePython.html)
    - They're developing two IDEs for Python -- one, called Komodo,
      based on Mozilla, and one, called Visual Python, which will
      be a plugin for Microsoft's Visual Studio;
        http://activestate.com/Products/VisualPython.html
        http://activestate.com/Products/Komodo.html
    - They're offering enterprise support for Python
        http://activestate.com/Products/PythonDirect/index.html);
 

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

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From dgrisby at uk.research.att.com  Tue Jun 13 05:26:39 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 13 Jun 2000 09:26:39 GMT
Subject: Fun with sys.modules
References: <8i37m7$t96$1@pegasus.csx.cam.ac.uk> <%Ic15.2292$c7.292284@pouncer.easynews.com>
Message-ID: <8i4uof$ji4$1@pegasus.csx.cam.ac.uk>

In article <%Ic15.2292$c7.292284 at pouncer.easynews.com>,
 Roger Upole <rupole at compaq.net> wrote:

>How about b=a ?

No, that's no good. I want the user to be able to say either "import a"
or "import b", and get the same module. I can't ask them to import
something different.

To put it in context, I'm the author of omniORBpy, a CORBA mapping for
Python. At the moment, users have to do

>>> from omniORB import CORBA
>>> orb = CORBA.ORB_init()

The recently accepted standard mapping says that they must be able to
do

>>> import CORBA
>>> orb = CORBA.ORB_init()

But I don't want to move the CORBA module out of the omniORB package,
because all sorts of things assume that that is where it is, and it
helps people who want to run in a multi-ORB environment.

So, what I want is a cheap (in space and time) way to make
omniORB.CORBA appear as if it was CORBA. It's not even possible to do

  # CORBA.py
  from omniORB.CORBA import *

since the CORBA module has some things prefixed with underscores which
must be available. There's quite a lot of stuff in the CORBA module,
so I'd rather not duplicate the module dictionary anyway.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From scorder at incigna.com  Wed Jun 14 11:27:03 2000
From: scorder at incigna.com (scorder at incigna.com)
Date: Wed, 14 Jun 2000 11:27:03 -0400
Subject: Problems with COM
References: <Wey15.3428$UM2.51434@weber.videotron.net>
Message-ID: <8i8fkscqom39cf5v5b1vq3jl0mrrf4f1an@4ax.com>

>def query_server(docBaseService, queryString):
>    searchService=docBaseService.SearchServices
>    resultSpace=win32com.client.Dispatch
>    debug("Querying... ")
>    searchService.SearchDocuments(queryString,resultSpace)
>    return resultSpace.Count
>

>My example above gives the following error on the line "return
>resultSpace.Count":
>AttributeError: Count
You are making resultSpace equal to the Dispatch function.

>The resultSpace object is a COM object that cannot be created beforehand.
>It has to be placed in a VARIANT that is fed to the function
>SearchDocuments.  How do I do this?  I tried creating a Dispatch object
>without specifying a type but it doesn't seem to work.

Looking at the ado examples in the Python Programming on Win32 book
there is an example of a connection object returning a recordset and a
return code.  It looks like this:

(adoRS, success) = adoConn.Execute('Select * from Clients')

I dont' know if this is an automatically built in method of handling
COM parameters passed by reference or not.  It could be that you have
to run MakePy on it for it to work.  Anyway MakePy will speed up your
script because you won't be late binding the objects anymore.

>Thank you in advance for any solution or direction to relevant
>documentation.
>I will be happy to provide more details if I have not been precise enough.
>
>Jean-Louis Couturier
>Developer
>Ixiasoft
>
>
>


Sam Corder
Fledgling python developer


From aahz at netcom.com  Thu Jun 15 08:51:20 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 12:51:20 GMT
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net> <memo.20000614223158.31901C@digitig.compulink.co.uk> <39480966.C69FFDB7@prescod.net>
Message-ID: <8iajg8$t8k$1@slb2.atl.mindspring.net>

In article <39480966.C69FFDB7 at prescod.net>,
Paul Prescod  <paul at prescod.net> wrote:
>
>How about:
>
>a <- a+1
>
>pronounced "a gets a plus 1"

Pascal already has ":=" for that purpose; I see no reason to invent a
new symbol.  If we have := for assignment and == for comparison, things
should work pretty well.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From mhammond at skippinet.com.au  Fri Jun  9 20:54:44 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sat, 10 Jun 2000 00:54:44 GMT
Subject: TypeError: unhashable type
References: <393d1c1a_2@news1.prserv.net> <20000606181546.H469@xs4all.nl> <F8476B9163F138C3.8DBC9B4CCE38EBEA.9FFEEE60294594AF@lp.airnews.net>
Message-ID: <olg05.706$cQ.9336@news-server.bigpond.net.au>

"Jeremy Hylton" <jeremy at beopen.com> wrote in message
news:F8476B9163F138C3.8DBC9B4CCE38EBEA.9FFEEE60294594AF at lp.airnews.net...

> It seems that "unhashable type" is a pretty obscure error message.  If
> you understand hashes, types, and dictionary implementation, it's
> fine.  If not, you're left scratch your head.

For the casual user, this generally means "dictionary keys must be
immutable".

> It would be helpful to produce an error message like:
> "dictionary key must be hashable type, not <type>"

This message isnt a real lot better.  Often I _know_ what the type is -
its just "unhashable" - which still doesnt clue the user in.  The word
"hashable" is a bigger problem than not printing the type.

> It's a tad tricky to implement because the error message is generated
> in PyObject_Hash.  The error would need to be caught in
> PyDict_SetItem, cleared, and a new one set.

I think the situation could still be improved without the type name.  But
for the life of me I can't think how.

"dictionary keys must be immutable" isnt really any better - WTF is
"immutable"??

"dictionary keys must be objects unable to be changed" - WTF - I can
change strings!!?  Or maybe this is trying to tell me that once I add a
key, it can't be removed?

Maybe something like:
"dictionary keys must be immutable (such as tuples, strings or ints, but
not objects such as lists)" - *gag*.

Giving-up-ly,

Mark.




From mfletch at tpresence.com  Fri Jun  2 19:25:42 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 2 Jun 2000 19:25:42 -0400 
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
Message-ID: <CC59868A0A76D311B1F50090277B321C1F338A@VRTMAIL>

For sufficiently large values of trivial ;) .

Consider the triviality factor in the examples below (and I'm not getting
into the complex sets of objects here, where prototypes and getattr/setattr
handlers and strange inheritance mechanics are set up which may do extremely
elaborate lookups/creations when a name lookup fails):

	module k uses module j's x attribute (there's also an X attribute)
--> need to rewrite k with knowledge of j's changed naming model (doable,
but ugly)

	module m dynamically receives a reference to an object o that in one
instance has an attribute p (it also, in that instance has an attribute P)
--> run-time system needs to know when m was written (thus whether it is
intending the first or second casing), as well as the renaming map within
object o (gack)

	module v gets a reference to an object and a string and tries to get
the attribute from object, oops, has been changed due to case mangling -->
where the heck did the string come from?  was it assembled from sub-parts?,
does the object have a name-mangling map? (double gack)

Changing case-sensitivity creates a new language.  Sure, you could port the
code, but it's not a trivial undertaking.  And getting the results to look
like considered names is... even more difficult  node --> node_ , or
something similar isn't helping make the name any easier to remember (stated
goal of case sensitivity).  How are you deciding which variant is the
"important" one that retains the natural name? Or are you just renaming
everything and forcing unnatural names for all?

Adding case-checking support to an editor begins to seem downright easy by
comparison :) .  Enjoy yourself,
Mike

-----Original Message-----
From: Johannes Nix [mailto:Johannes.Nix at mail.uni-oldenburg.de]
Sent: Friday, June 02, 2000 3:05 PM
To: python-list at python.org
Subject: Re: Case-sensitivity: why -- or why not? (was Re: Damnation!)


Russell Wallace <rwallace at esatclear.ie> writes:

> It seems to me that the important argument in this context is "case
> insensitivity will break millions of lines of existing code, and perhaps
> split the user base if a significant number of existing users refuse to

It is a trivial task to convert case-sensitive identifiers in unique
case-insensitive identifiers in case of a name clash.

Johannes

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



From pgf at globalreach.net  Sat Jun 10 02:57:33 2000
From: pgf at globalreach.net (Phil Fraering)
Date: 10 Jun 2000 01:57:33 -0500
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com>
Message-ID: <87em66owf6.fsf@globalreach.net>

Guido van Rossum <guido at python.org> writes:

> I am a very case-sensitive person myself: it bothers me when people
> don't start their sentences with capital letters or when they refer to
> themselves as "i" in lowercase -- or when people refer to Python as
> "python" or "PYTHON".  (Come to think of it, it bothered me when you
> wrote "german" instead of "German" :-).

Personally, I simply think that lowercase is more readable than uppercase,
but I don't really know the reason why. 

> Yet, here are some of the reasons why I am considering making Python
> case-insensitive:
> 
> (1) Randy Pausch, a professor at CMU, found, when teaching Python to
> non-CS students in the context of Alice (www.alice.org), that the
> number one problem his students were having was to remember that case
> matters in Python.  (The number two problem was 1/2 == 0; there was no
> significalt number three problem.)

I suspect there may be a "birth language" problem here: I don't
know about Dutch, but German uses "case" to impart information
much more than English, where its usage is more limited.

> (2) I've met many people who are experienced and accomplished Python
> programmers but who still, when referring to Python modules or
> functions in an informal context (e.g. email), make up their own case
> conventions.  E.g. people will write me about the String module or
> about FTPLib.
> 
> I also know some of the things I plan to do to make the transition
> painless and to make the usual problems with case insensitivity more
> palatable.  E.g. I may add a case insensitivity feature to IDLE which
> makes sure that all identifiers are written in a consistent case, I
> may add flags to pre-Py3k Python interpreters to turn on case
> insensitivity or case sensitivity or a special warning mode.

Ah.

-- 
Phil Fraering         "One day, Pinky, A MOUSE shall rule, and it is the
pgf at globalreach.net    humans who will be forced to endure these humiliating
/Will work for tape/   diversions!"
                      "You mean like Orlando, Brain?"


From paul at fizzylab.com  Sat Jun 17 02:23:47 2000
From: paul at fizzylab.com (paul at fizzylab.com)
Date: Fri, 16 Jun 2000 23:23:47 -0700
Subject: urllib question
Message-ID: <394B196F.5ED9E73@fizzylab.com>

Why does the first example work, but not the second (actual
URL not supplied, but you get the idea)?

I assume something to do with query string values and
encoding: even if I was right, I have been unsuccessful at
making it work.

#!/usr/bin/python
import string, urllib

u = urllib.urlopen(string.join(["http://cnn.com/",
"index.html"]))
print u.read()
u = urllib.urlopen(string.join(["http://some.host.com/",
"montstrousQueryString?withParameters"]))
print u.read()

My second day trying to use python . . .  excuse my
thickness of skull.



From punck at PenguinMints.cx  Wed Jun 28 23:57:18 2000
From: punck at PenguinMints.cx (Chris Armstrong)
Date: Wed, 28 Jun 2000 22:57:18 -0500
Subject: Bindings for game library
References: <bOj65.1079$MZ3.2888@newsfeed.slurp.net> <395A9901.CEB7CB42@lucent.com>
Message-ID: <4Qy65.1786$MZ3.6375@newsfeed.slurp.net>

I'm sorry, I guess I forgot to mention I only develop under Linux :(
It seems that CDX is only for DirectX :(

In article <395A9901.CEB7CB42 at lucent.com>, Jon Riehl <jriehl at lucent.com>
wrote:
> 	Earlier this year I took an initial stab at the CDX game library. 
> Unfortunately, my stuff got nuked with the Starship Python.  I'll go
> ahead and take an action to update the following URL real soon now:
> http://starship.python.net/crew/jriehl/pycdx/ (smirk)
> 	CDX is at: http://www.cdx.sk/ If you see stuff you want in PyCDX, send
> 	many death threats to
> jriehl at mindspring.com.  (DISCLAIMER: This death threat invitation only
> extends to CDX/game development enthusiasts you sick Internet types.)
> -Jon


-- 
There is a 90% chance that this message was written when the author's been
up longer than he should have. Please disregard any senseless drivel.
--
Chris Armstrong
http://penguinmints.cx



From godzilla at netmeg.net  Sat Jun 17 11:48:24 2000
From: godzilla at netmeg.net (Les Schaffer)
Date: Sat, 17 Jun 2000 15:48:24 GMT
Subject: Problem displaying Numeric Array as image
Message-ID: <m2ya442tkw.fsf@netmeg.net>

Eric Frey said:
> I a trying to display a floating point Numeric (NumPy) array as a
> grayscale image via Tkinter.

this example code works for me. its an amalgam of stuff i have found
on the net... normalize takes a floating point image and scales it to
1 and returns byte values suitable for a greyscale image. with a
little cleaning up this should do what you want......

les schaffer

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

#!/usr/bin/python


from Tkinter import *
from ImageTk import PhotoImage
import Image

from Numeric import *
from RandomArray import random

def normalize(a):
    a = array(a, copy=1)
    a = (a - min(a.flat))/(max(a.flat) - min(a.flat)) 
    return ( a*254 ).astype('b')

def ArrayToImage(a, height=200, scale=1):
    a = normalize(a)
    i = Image.new("L", (a.shape[1], a.shape[0]), color=255)
    i.fromstring(a.tostring())
    if scale != 1:
	i = i.resize((i.size[0]*scale, i.size[1]*scale))
    return i

def ImageToArray(i):
    a = fromstring(i.tostring(), 'b')
    a.shape = i.im.size[1], i.im.size[0]
    return a


class MatrixDisplay:

    def __init__(self, master):

        frame = Frame(master)
        frame.pack()
        self.button = Button(frame, text="QUIT", fg="red", 
                             command=frame.quit)
        self.button.pack(side=TOP)
        self.test_B = Button(frame, text="Update", command=self.update)
        self.test_B.pack(side=TOP)
        self.img = Image.new("L",(100,100))
        self.photo = PhotoImage(image=self.img)
        self.canvas = Canvas(frame, width=100, height=100)
        self.canvas.create_image(0, 0, anchor=NW, image=self.photo)
        self.canvas.pack()

    def update(self):
        ra = fromfunction(self.func, (101, 101))
        #ra = random((101, 101))
        
        self.img = ArrayToImage(ra)
        #self.photo = PhotoImage(self.img)
        self.photo.paste(self.img)
        self.canvas.create_image(0, 0, anchor=NW, image=self.photo)
        self.canvas.update()

    def func(self, i,j):
        return cos( 4*2*3.1415926*i / 100. ) * cos( 4*2*3.1415926*(j / 100. ))
    
    

root = Tk()
app = MatrixDisplay(root)
root.mainloop()



From dbroadwell at mindspring.com  Sat Jun 24 07:02:01 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sat, 24 Jun 2000 07:02:01 -0400
Subject: Now for the flames...
Message-ID: <39549529.F5FFBD89@mindspring.com>

Ok, now, what else am i doing that really isn't necessary?
   (ie: point out my c hangups, please.)

Ignore the exre print statements, most of those i'll comment out or
otherwise handle. i am looking for actuall programing advice..
specifically on if i have used any non-platform independent stuff.. (so
that when i later migrate to haveing this same file structure that i am
designing my toold for, they will still function.)

# code follows

import os
import sys

output = 'out.txt'
filetype = 'm3u'
directorylist = []
filelist = []

def doit():
    cwd = []
    print 'reading current directory'
    cwd = os.listdir('.')
    print 'building directorylist'
    for item in cwd:
        if (os.path.isdir(item)):
           directorylist.append(item)
    print 'building filelist'
    for item in cwd:
        if (item[(len(item)-3):] == filetype):
           filelist.append(item)
    del cwd
    print 'Checking for subdirectories to check.'
    for item in directorylist:
        print 'subdirectory ' + item + ' found, scanning.'
        os.chdir(item)
        subdir = []
        subdir = os.listdir('.')
        for thing in subdir[:]:
            print thing + ' ',
            if (thing[(len(thing)-3):] == filetype):
               print 'Yup.'
               # join item(path) to file
               # add to filelist
            else: 
               print 'Nope.'
               subdir.remove(thing)
        del subdir
        os.chdir('..')
      
def merge(filelist):
    if (filelist):
       outbuffer=open(output,'w')
       for item in filelist:
           inbuffer=open(item)
           outbuffer.write(inbuffer.read())
           inbuffer.close()
       outbuffer.close()
    else:
       print 'No files to process, Exiting...'
       sys.exit()

doit()
-- 
        ~We're all fallen angels who've forgotten how to fly~
        Remember how: http://www.cyber-action.com/oed/dbroadwell.html
--


From kuncej at mail.conservation.state.mo.us  Fri Jun  9 11:03:36 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Fri, 9 Jun 2000 10:03:36 -0500
Subject: Python, COM, DAO, and Licensing
Message-ID: <_J705.22$K4.344@news.more.net>

I wrote a python program that  manipulates
an Access database via COM and DAO.

I installed the "reditributable" DAO that
came with VC++ 5 on a client workstation
along with Python, win32 extensions,  and my program.
Pythonwin makepy works fine on DAO.

When my program tries to access DAO
via COM, I get something like:

    if oobj is None: oobj = pythoncom.new(self.CLSID)
pywintypes.com_error: (-2147221230, 'CLASS_E_NOTLICENSED', None, None)

About that error, Microsoft (Q151804) says that the
redistributable controls are licensed, and that you
must specify a valid License string when dynamically
creating the control.

Is there a way to specify a license string through
pythoncom? Can anyone give me a step-by-step,
or at least some example code?

Better yet, is there another version of DAO or something
similar that is not licensed, and will talk to pythoncom?

Thanks.

  --Jeff

NOT the Jeff from UMich who hates pythonwin :-)




From mogul at primus.ca  Wed Jun  7 05:52:30 2000
From: mogul at primus.ca (Robin Porter)
Date: Wed, 7 Jun 2000 05:52:30 -0400
Subject: Newbie Question...
Message-ID: <ZTo%4.854$0l2.6963@news2.tor.primus.ca>

I wrote before asking for an example of C code changed to Python code (i.e.
Payroll example).  Could someone please explain how I go about saving and
then running the code once I have inputed it in.  I tried entering it into
the Python GUI and everything looked good until I tried to save it, recall
it, then run it.

My apoligies for such a basic question.  This language looks great...I am
anxious to learn how to use it.

Could someone please recommend a GOOD book on PYTHON for the beginner.

Thank you in advance.

Robin




From matt at virtualspectator.com  Mon Jun 12 07:42:31 2000
From: matt at virtualspectator.com (matt)
Date: Mon, 12 Jun 2000 23:42:31 +1200
Subject: lambda and if
In-Reply-To: <07d501bfd461$ced14280$1906a8c0@fc.fenx.com>
References: <8i2c9l$b0r$1@news1.wire.net.au> <0006122328441W.01005@localhost.localdomain> <07d501bfd461$ced14280$1906a8c0@fc.fenx.com>
Message-ID: <0006122345501X.01005@localhost.localdomain>

sure would ... doh! 
	thanks again


On Mon, 12 Jun 2000, Emile van Sebille wrote:
> ----- Original Message -----
> From: matt <matt at virtualspectator.com>
> 
> > sneaky ... it's nice.
> > might be nice to extend that with
> > ["","big"][x[1] in ("big",)]  so we can extend the 'yes'
> set
> >
> > but extending it to using the actual match is a little
> more difficult,
> > i.e.
> >  x[1] in ("big","little","wide",)  where the matched item
> would be appended if
> > it existed or else a blank
> 
> Wouldn't ["",x[1]][x[1] in ("big","little","wide",)] do it?
> 
> 
> 
> >
> >
> > thanks
> > matt
> >
> 
> Emile van Sebille
> emile at fenx.com
> -------------------
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From mwh21 at cam.ac.uk  Tue Jun  6 11:52:13 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 06 Jun 2000 16:52:13 +0100
Subject: [Human] Lanaguage translation
References: <sjl1243u5ri150@corp.supernews.com> <oqsnuqddru.fsf@titan.progiciels-bpi.ca>
Message-ID: <m3zooyx0wi.fsf@atrus.jesus.cam.ac.uk>

=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> writes:

> P.S. - It reminds me Donald Knuth's `Art of Computer Programming', vol I.
> The last sentence of the `Notes on the exercises', page xix, reads:
> 
>    3. [M50] Prove that when n is an integer, n > 2, the equation
>    x^n + y^n = z^n has no solution in positive integers x, y, z.
> 
> With an earlier explanation for "50":
> 
>    A research problem which (to the author's knowledge at the tie of writing)
>    has not been solved satisfactorily.  If the reader has found an answer to
>    this problem, he is urged to write it up for publication; furthermore, the
>    author of this book would appreciate hearing about the solution as soon as
>    possible (provided it is correct)!
> 
> I found that pretty humorous at the time, as at the time Knuth wrote this,
> there was an open contest, with a substantial money prize, to anyone who
> would bring a solution to this problem.  The prize was later withdrawn,
> a few years before a proof was found.  (So far that I heard about this.)

I heard that it's "M49" in the last printing :-).

I think Wiles did get the money, though.

Cheers,
Michael
(who has now learnt just enough enough algebraic geometry to have a
faint glimpse of how far from understanding his lack of understanding
of Wiles' proof he is...)

-- 
  And then the character-only displays went away (leading to
  increasingly silly graphical effects and finally to ads on 
  web pages).                      -- John W. Baxter, comp.lang.python


From hzhu at rocket.knowledgetrack.com  Fri Jun  9 13:05:57 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 09 Jun 2000 17:05:57 GMT
Subject: wishlist Python3K: subclass ints
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <393EFF38.C023F10F@san.rr.com> <393FD0C7.D70479A2@visionart.com>
Message-ID: <slrn8k294j.pi.hzhu@rocket.knowledgetrack.com>

On Thu, 08 Jun 2000 09:58:47 -0700, Pete Shinners <pete at visionart.com> wrote:
>> > 1. Everything is an object of a class that can be subclassed.
>> 
>> You want to subclass integers, why?
>
> - Also gone are all the nit-picking type checking routines.
>   instead of first checking if a variable is a "class" type,
>   then checking if the object is a certain class, all variables
>   hold objects of the same "rank". the best move would be to
>   have "int" and "float" be subclassed from a base class like
>   "realnumber".

Here's a simulation:
"""
Class.py
Simulation of proposed class heirarchy
"""

class Object: pass
class Class(Object): pass
class Instance(Object): pass
class Number(Instance): pass
class Approx(Instance): pass
class Complex(Number): pass
class Real(Complex): pass
class Integer(Real): pass
class ComplexFloat(Complex, Approx): pass
class RealFloat(Real, Approx): pass

_Class = type(Class)
_Instance = type(Class())
_int = type(1)
_float = type(1.0)
_complex = type(1j)

def classOf(a):
	if type(a) == _Instance:	return a.__class__
	elif type(a) == _Class:		return Class
	elif type(a) == _int:		return Integer
	elif type(a) == _float:		return RealFloat
	elif type(a) == _complex:	return ComplexFloat

def isOfClass(a, b):
	if type(a) == _Instance:	return isinstance(a, b)
	else:		return issubclass(classOf(a), b)

if __name__ == "__main__":
	print classOf(Class)
	print classOf(Class())
	print classOf(1)
	print classOf(1.1)
	print classOf(1j)

	print isOfClass(Class, Object)
	print isOfClass(Class(), Object)
	print isOfClass(1, Object), isOfClass(1, Number), isOfClass(1, Integer)
	print isOfClass(1.1, Real)
	print isOfClass(1j, Real), isOfClass(1j, Complex), isOfClass(1j, Approx)



-- 
Huaiyu Zhu                               hzhu at knowledgetrack.com
KnowledgeTrack Corporation               Tel: 925 738 1907
7020 Koll Center Parkway, Suite 110      Fax: 925 738 1039
Pleasanton, CA 94566


From frankc at colconsulting.com  Sat Jun 17 17:46:05 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Sat, 17 Jun 2000 17:46:05 -0400
Subject: Python Productivity over C++ (Encapsulation)
References: <20000615183521.59984.qmail@hotmail.com>
Message-ID: <394BF19D.23C26098@colconsulting.com>

Hung Jung Lu wrote:
> 
> --- In python-list at egroups.com, Ken Seehof <kens at s...> wrote:
> >"Frank V. Castellucci" wrote:
> > > Courageous wrote:
> > > While the last line seems rather toungue in cheek, there are other
> > > things that C++ brings to the table such as encapsulation, access
> > > control and type strictness.
> >Encapsulation, access control and type strictness are often erroneously
> >referred to as if they were features.  They aren't.  They are aspects of a
> >design philosophy.  To make a system less error prone, constrain the
> >programmer
> >in order to prevent the programmer from writing bugs.  ...
> >You have to ask, "what is the actual value of these features?"  The answer
> >is
> >that they prevent certain kinds of bugs from appearing.
> 
> Type strictness is fine. It has all to do with performance: small, fast,
> efficient machine code. Encapsulation is a totally different story.
> 
> One of the biggest lies in modern computer language theory is that
> encapsulation is there for a good technical reason.
> 
> And generations of students were lied to, and sadly, 99% believed in what
> the textbooks said. Some of them became programmers, and only after 15 or 20
> years in the field would they wake up one day to realize that, heck, they've
> been lied to all along. :) (Doesn't all this remind you of Nazis and
> Communism? Collective blindness is far more common that you think, ha!)
> [snip]

Crap. Encapsulation allows me to change the storage hidden by a class
from a Stack to a MultiMap without any effect througout the rest of the
system. It does not require anyone to recompile, doesn't require any
more maintenance.

I also use data hidding in Python.

What a stupid response you made. You must be guilty of exposing the data
members in the interface. This is sickening.


-- 
Frank V. Castellucci


From moron at Glue.umd.edu  Fri Jun 16 21:05:11 2000
From: moron at Glue.umd.edu (Warren B. Focke)
Date: 16 Jun 2000 21:05:11 -0400
Subject: numpy bug?
References: <slrn8klh53.56e.hzhu@rocket.knowledgetrack.com>
Message-ID: <8ieis7$4nf@stochastic.eng.umd.edu>

Huaiyu Zhu sez:
[Numeric and Python integer division don't agree.]

	Numeric just calls the C operators for / and %.  C leaves the
rounding direction for integer division up to the implementor if both
operands are not positive, but does require that (a/b)*b + a%b == a as
long as b != 0.  So NumPy gets it right from that point of view.
	Python does specify the rounding direction, in a manner that
has been extensively argured in this group to be correct.  I don't
really remember the reasons at the moment, check dejanews, it's on
there somewhere, I'm sure.

Warren Focke
-- 
If you feel that you have both feet planted on level ground, then the
university has failed you. -Robert Goheen, President, Princeton University


From dale at out-think.NOSPAMco.uk  Thu Jun 29 03:08:27 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 29 Jun 2000 08:08:27 +0100
Subject: Driving Excel from Python
References: <8j7oqn$e9q$1@supernews.com> <8jd51g0j9k@news2.newsguy.com>
Message-ID: <8jeslm$qu4$1@supernews.com>

Thanks. I'll look into that.

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants


Alex Martelli <alex at magenta.com> wrote in message
news:8jd51g0j9k at news2.newsguy.com...
> Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote in message
> news:8j7oqn$e9q$1 at supernews.com...
> > (Sorry if this appears more than once. I'm getting errors from news
> server).
> >
> > I'm driving Excel from Python and most of it's working fine except for:
> >
> > >>> xlsh.UsedRange.Address(ReferenceStyle=1)
> > TypeError: call of non-function (type string)
> >
> > xlsh is an Excel Sheet object and I'm trying to get the used range in
RxCx
> > format. However, when I try and specify ANY arguments to the Address
> > property, I get this error message. I've tried both positional and
keyword
> > arguments.
> >
> > I have the Python Win32 book but this this subject is not quite covered
in
> > sufficient depth.
> >
> > Any help appreciated.
>
> For reasons that escape me, .Address doesn't seem to be seen as an
> indexed-property; however, a .GetAddress is also made available that
> does seem to behave as an indexed-property like .Address should.  I
> sort of suspect that this is to let people use .Address, without any
> parenthesis, in the common case where they want to let all arguments
> default.  However, I can't find any docs about this guess of mine.
>
> Anyway, you CAN now call
>
> xlsh.UsedRange.GetAddress(ReferenceStyle=1)
>
> and get the same '$A$1' kind of string that you'd get with
>
> xlsh.UsedRange.Address
>
> since, after all, xlA1, aka 1 (win32com.client.constants.xlA1, too),
> IS the default value for ReferenceStyle.
>
> A further puzzlement, however, is that
>
> xlsh.UsedRange.GetAddress(ReferenceStyle=-4150)
>
> ALSO appears to return the same '$A$1' string too!  BTW, that funny
> -4150 is xlR1C1, aka win32com.client.constants.xlR1C1, who knows
> why.  But anyway, while this does allow accessing the Address indexed
> properties with 'indices' (arguments), this doesn't appear to be
> doing much good at all.  Maybe somebody else understands what's
> going on...?
>
> It does work fine from the Immediate windows in VBA inside Excel:
>     print Sheet1.UsedRange.Address
> $A$1
>     print Sheet1.UsedRange.Address(ReferenceStyle=xlR1C1)
> $A1
>
> but from Python I always get '$A$1'.  Weird...
>
>
> Alex
>
>
>
>




From pmueller at decrc.abb.de  Fri Jun 30 11:24:58 2000
From: pmueller at decrc.abb.de (Peter Mueller)
Date: Fri, 30 Jun 2000 17:24:58 +0200
Subject: Example how to use cgihttpserver
Message-ID: <395CBBCA.9A241134@decrc.abb.de>

Hi,

is there an example how to use the cgihttpserver in a way that some cgi
scripts are in /cgi-bin/* (this is simple) and others are coded in the
program itself. A simple example reading the FieldStorage, generating a
page ... would me help very much.

Please send the reply to pmueller at decrc.abb.de

Thank you very much,
Peter



From dgallion1 at yahoo.com  Mon Jun  5 13:13:15 2000
From: dgallion1 at yahoo.com (Darrell Gallion)
Date: Mon, 5 Jun 2000 10:13:15 -0700 (PDT)
Subject: Regular expressions trouble
Message-ID: <20000605171315.25332.qmail@web2004.mail.yahoo.com>

<vivienne_van_velzen at my-deja.com> wrote in message
news:8hgcn2$1lh$1 at nnrp1.deja.com...
> I'm fairly new to the use of regular expressions,
and I've come upon a
> problem I can't seem to fix. I'm trying to extract a
keyword from a
> HTML-page and color it red. This works fine, except
that it also
> extracts the keywords within tags.

import re
def main():
    buf = open("mozilla_org.htm").read()
    keyWords=['help']
    pat='(>[^<]*)(%s)'
    for p in keyWords:
        buf = re.sub(pat%p,'\g<1><b><font
color=\"red\">\g<2></font></b>',buf)

    open("test.htm","w").write(buf)
main()


--Darrell Gallion


__________________________________________________
Do You Yahoo!?
Yahoo! Photos -- now, 100 FREE prints!
http://photos.yahoo.com



From jkraska1 at san.rr.com  Sun Jun 18 04:11:33 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 18 Jun 2000 08:11:33 GMT
Subject: Python vs. Matlab
References: <642A954DD517D411B20C00508BCF23B001296E71@mail.sauder.com>
Message-ID: <394C84C2.9B923624@san.rr.com>

> Spotted Python some months ago, and liked it!  Can Python, Numpy and
> the many plotting extensions be used as a replacement for Matlab, is
> Python slower/faster etc.?

IIRC, there is a python interface to Matlab, so it should
make your transition easier. According to what I've read,
typical programming is both better and faster in Python, but
Matlab matrix math is apparently hard to beat.

C/


From digitig at cix.co.uk  Tue Jun  6 18:20:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Jun 2000 23:20 +0100 (BST)
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <memo.20000606232053.11341D@digitig.compulink.co.uk>

In article <393D2797.4E4EE7CC at wag.caltech.edu>, rpm at wag.caltech.edu 
(Richard P. Muller) wrote:

> superior design may not be able to overcome. I have Ada-programming
> friends who mumble about Ada's superiority over C++; that may be the
> case, but does it really matter anymore?

I am convinced that that whole argument is a red herring. Ada is superior 
to C++. /And/ C++ is superior to Ada. For different jobs. You're not going 
to write a major safety-critical application in C++, at least not in any 
country that's got any sort of safety regulatory framework. And you're not 
going to write a general-purpose operating system in Ada. I have a 
friend who designs racing car instrumentation systems who reckons that 
FORTH is the only way to go. The question is /never/ "what is the best 
language", it is always "what is the best language /for/ /this/ /task/".


From ajung at yetix.sz-sb.de  Thu Jun 22 13:34:56 2000
From: ajung at yetix.sz-sb.de (Andreas Jung)
Date: Thu, 22 Jun 2000 17:34:56 GMT
Subject: Python interface to Oracle? - yes.
References: <3951C636.2E985CE@ust.hk> <87og4t26zz.fsf@ns.f7.ph4.lan> <39522A9C.57D500F4@uab.edu>
Message-ID: <7277.39524e40.863d0@yetix.sz-sb.de>

Shae Erisson <shapr at uab.edu> wrote:
2qy
2qyAndreas Rottmann wrote:
2qy> 
2qy> David YEUNG <dyeung at ust.hk> writes:
2qy> 
2qy> > Does anyone know if any python module has been developed for
2qy> > interfacing with Oracle8i database on Linux?
2qy> >
2qy> That would be interesting for me, too. I'm likely to develop one, if
2qy> it doesn't exist, since I have the infrastructure ready - see
2qy> GQL.sourceforge.net.
2qy
2qyhttp://www.zope.org/Products/DCOracle
2qyDigital Creations appears to be the main source of experts on Python and
2qyOracle interaction.
2qy
2qyThe changelog for DCOracle mentions Oracle 8 specifically in several
2qyplaces, therefore I would assume they have support for it.
2qy
2qyIt supports Python 1.4 or higher, and Oracle 7.3 and higher.

Beware using DCOracle with Oracle 8i features like CLOBS and LOBs
- this does not work with the current version of DCOracle.

Andreas


From don_b at my-deja.com  Wed Jun 21 13:45:20 2000
From: don_b at my-deja.com (Don)
Date: Wed, 21 Jun 2000 17:45:20 GMT
Subject: Conditional Compilation for bytecode?
Message-ID: <8iquvb$gtd$1@nnrp1.deja.com>

I have some code that needs to run on Win32 and IRIX.

I thought I read somewhere that python had a sort of conditional
compilation that would generate different .pyc files depending on the
directives in the .py file.

i.e.
#ifdef WIN32
CreateProcess(...)
#else UNIX
fork()
#endif

But I can't seem to find this anywhere.  Do I need to do this?

if sys.platform = 'Win32':
   CreateProcess(...)
else:
   fork()

Does the byte code compiler optimize this so it doesn't check every
time?


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


From stevemul at ozemail.com.au  Fri Jun  9 01:12:56 2000
From: stevemul at ozemail.com.au (Steve Mullarkey)
Date: Fri, 09 Jun 2000 05:12:56 GMT
Subject: Python Productivity over C++
Message-ID: <8F4E9C028stevemulozemailcomau@139.130.250.4>

I have been a 'C' programmer for the past 15 years and a "C++"
programmer for the past 3 years. I'm not a genius but would modestly (?)
class myself as reasonably competent.

In the past week I have become aware of Python. I've bought a book and
read this newsgroup every day.

I have read in several places productivity claims of 5 to 10 times over
'C' and "C++". I would like to ask for some feedback from "C++"
programmers who have moved to Python as to whether these estimates are
realistic.

In "C++" I have :-

1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
Profilers, Programmers Editor including Class Browser, etc..

2. I use the STL (Standard Template Library) which gives powerful string
and container classes.

3. I use incremental compiling and linking. This reduces wait times to
very small amounts.

4. I can also use background compiling and linking.

5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
etc..

I don't want to start a flame war but, given the above, I just can't see
where the productivity increase is generated. I am genuinely interested
to ask "C++" programmers :-

1. What productivity increase do you achieve ?

2. How long did you use Python before you achieved increased
productivity ?


Best regards,

Steve Mullarkey. 
Tascom Pty Ltd.
Berry, NSW 2535, Australia.
Tel : (61)-(2)-4464-3328


From moshez at math.huji.ac.il  Mon Jun 12 08:47:41 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 12 Jun 2000 15:47:41 +0300 (IDT)
Subject: NumPy
In-Reply-To: <20000612123244.5898.qmail@web1503.mail.yahoo.com>
Message-ID: <Pine.GSO.4.10.10006121547270.24722-100000@sundial>

On Mon, 12 Jun 2000, Robin Porter wrote:

> Does anyone know where I can get NumPy?

http://numpy.sourceforge.net

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From g_will at cyberus.ca  Wed Jun  7 10:44:15 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: 7 Jun 2000 14:44:15 GMT
Subject: Gordon M's Installer- Missing DLL
Message-ID: <01bfd08e$d36f8da0$0100007f@Cyberus>

Hi

I am trying to distribute a program that I have packaged into a standalone.
 The package is installing ok but I am getting a report back that a dll is
not loading when I run the standalone.

C:\Scopal>scopal03b.exe
Traceback (innermost last):
  File "<string>", line 1, in ?
  File "E:\Program Files\Python\Installer_3f\Installer\support\imputil.py",
line
 51, in _import_hook
    top, tail = self._import_top_module(parent, name)
  File "E:\Program Files\Python\Installer_3f\Installer\support\imputil.py",
line
 176, in _import_top_module
    module = self._import_one(parent, head, fqname)
  File "E:\Program Files\Python\Installer_3f\Installer\support\imputil.py",
line
 226, in _import_one
    exec result[1] in module.__dict__
  File "E:\Program Files\Python\wxPython\__init__.py", line 20, in ?
    import wxc
  File "E:\Program Files\Python\Installer_3f\Installer\support\imputil.py",
line
 54, in _import_hook
    return self.__chain_import(name, globals, locals, fromlist)
ImportError: DLL load failed: One of the library files needed to run this
applic
ation cannot be found.



Is there a way to identify which dll is missing?  Is this message coming
from Python or Installer?

Thanks

Regards,

Gordon Williams


From gritsch at iue.tuwien.ac.at  Mon Jun 19 10:35:08 2000
From: gritsch at iue.tuwien.ac.at (Markus Gritsch)
Date: Mon, 19 Jun 2000 16:35:08 +0200
Subject: simple thread problem
Message-ID: <394E2F9C.6BCD96B5@iue.tuwien.ac.at>

Hi!

I started a very simple program to get familiar with Python threads.
It works fine, except for some misterious messages which I get every
time a thread exits:

currentThread(): no current thread for xxxx

where 'xxxx' is some small integer number.  I also tried to terminate
the thread with thread.exit() (which should treminate the thread
silently), but the message remains.

How can this message be avoided?

Kind regards,
Markus

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

#! /usr/bin/env python

import thread, threading
import time

sem = threading.Semaphore(3)

def getit(argum):
    try:
        print 'good night from', argum
        time.sleep(1)
        print 'hello again form', argum
    finally:
        sem.release()

for number in range(6):
    sem.acquire()
    print 'starting', number
    thread.start_new_thread(getit, (number, ))

time.sleep(2)




From bparsia at email.unc.edu  Thu Jun  8 01:37:15 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 8 Jun 2000 01:37:15 -0400
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net> <Wbr%4.69$vh.19893@news> <8hlrkn$5t2$1@nnrp1.deja.com>
Message-ID: <1ebvwau.af3wl21301xdcN%bparsia@email.unc.edu>

<david_ullrich at my-deja.com> wrote:
[snip]
> 
>     I don't know anything about Ruby, but a language could certainly
> be more fully OO than Python. For example in Python although everything
> is an object not every type is a class (you can't subclass list, for
> example).
[snip]

Of course, this presumes that the more "class" you have the more fully
"OOPY" you are. *Really* ools have *class*. Yuk yuk.

Of course, this ignores prototype based systems (e.g., Self &
NewtonScript), which, some would argue, are *more* oopy than class based
systems.

By *that* measure, Python would be *more* oopy than Ruby ;)

(Of course, it's not clear to me that being unsubclassable makes you not
a class. Think of Javaesque "final" classes, or Dylan's "sealed"
classes.)

Are Ruby's *classes*, themselves, instances of classes (aka
metaclasses)? Like, oh, let's see, *Smalltalk*!

If not, I wouldn't go bragging :)

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From tanzer at swing.co.at  Mon Jun  5 01:49:09 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Mon, 05 Jun 2000 07:49:09 +0200
Subject: Styling Temporary Code 
In-Reply-To: Your message of "Thu, 01 Jun 2000 11:21:47 PDT."
             <3936A9BB.FBEF9201@visionart.com> 
Message-ID: <m12ypl7-000wcEC@swing.co.at>

Pete Shinners <pete at visionart.com>:

> one thing i do frequenctly in C is to place temporary
> "debug/doublecheck/sanity" type code while i'm developing
> it. i always put this code in with no indenting, so it's
> easy to spot, work with, and cleanup. it looks something
> like this...
> 
> 
> int myfunc(int nullargs)
> {
>     int newval = do_fancy_stuff(nullargs);
> cout << "newval = " << newval << endl;
>     return more_fancy(newval);
> }
> 
> 
> ok, in this 5 line example it doesn't really sparkly,
> but when i use it in big sections of code, it can't
> be beat. call it the mark of the beast, chase me with
> pitchforks, i simply love it and i don't want to change!
> 
> now in python this obviously won't work, but i haven't
> been able to settle on a style that works for me yet.
> what do other people use for this "working" code?

I use a module-level or class variable `debug_level' which is set to 0
by default. The debugging code looks like:

    if debug_level > x:
        print "newval = %s" % newval

For debugging, I simple change `debug_level' -- normally from outside
the module. 

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From morten at src.no  Fri Jun 30 05:41:25 2000
From: morten at src.no (Morten W. Petersen)
Date: Fri, 30 Jun 2000 11:41:25 +0200
Subject: MimeWriter (how to use/find boundaries)
Message-ID: <Pine.LNX.4.21.0006301138390.15363-100000@lade.src.no>

I've built a mail-sending function, which, when sending mail with
attachments; does send the mail, with the encoded attachments; but it
doesn't make the mail readable on the other side. I.e., it does not
include the boundary thing, and I don't have a clue as to how to go about
to do it. (...)

-Morten



From maxm at normik.dk  Tue Jun  6 04:02:10 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Tue, 6 Jun 2000 10:02:10 +0200 
Subject: SV: How to execute python scripts under Apache?
Message-ID: <7BD10B680501D411B9DF009027E06F32011AAB@EXCHANGE>

Fra: Jilani Khaldi [mailto:jilanik at tin.it]

>I am thinking to give python a try for my future
>web applications.

May I be so frank as to ask why?

I have developed both in PHP and Python myself. I find that PHP is very easy
to get up and running. Just compile it into the server and it's working. I
have never used Python for web development only cron jobs and the like. What
about PHP has made you want to change?

I am currently devloping under IIS on win2k in server side JavaScript and is
pretty frustrated about it. (Everything has to be bought as an ActiveX
component, and JScript has no library) I am thus considering to change
tools. PHP and Python are my favourites.I really like Python but PHP is so
easy to get running on the server, especially when it has to be combined
with html sites.

I have tried Zope and understand how it works, but the documentation is
non-existing and I really don't like the idea of dtml. I guess that I will
try out mod_python as soon as I get better time. Probably that and
ZPublisher will be just what I am looking for.

Does anybody has some good technical reasons to go with Python instead of
PHP?

Regards

    Max M
    maxm at normik.dk
    http:\\www.normik.dk



From duncan at rcp.co.uk  Fri Jun  9 06:31:50 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 9 Jun 2000 10:31:50 GMT
Subject: String.join revisited (URGENT for 1.6)
References: <hjoY4.6071$wYl.223614464@newsb.telia.net> <39328edc$0$18384@personalnews.de.uu.net> <_pyY4.10102$Za1.160118@newsc.telia.net> <3932e628$0$6681@personalnews.de.uu.net> <3932F323.C73E0404@san.rr.com> <3932F5DC.E00C1642@san.rr.com> <ZoDY4.6196$wYl.229414400@newsb.telia.net> <39333CC3.32805F5@cosc.canterbury.ac.nz> <saMY4.6222$wYl.223847936@newsb.telia.net> <8h087l$bnc$4@newshost.accu.uu.nl> <20000530182454.E29991@exeter.exeter.org> <394003D6.366D0B0@concentric.net>
Message-ID: <8F4E7DDB9duncanrcpcouk@194.238.50.13>

mjhand at concentric.net (Manus Hand) wrote in
<394003D6.366D0B0 at concentric.net>: 

>But as the devil's advocate, tt just seems to me
>that [].join() could be implemented such that each member of the list is
>str()'ed or repr()'ed and then stuck together in the result string.  I
>imagine that one of the fatal flaws in this, though, is how self
>-referential lists would be handled: 
>   >>> x = ['a']
>   >>> x.append(x)
>   >>> x.join()        # I would argue we get an exception?
>

There is no problem with recursion here. What you are proposing is that
x.join() is equivalent to:
 >>> string.join(map(str, x))
 "a ['a', [...]]"

As you see the str function already handles recursive lists.


From aahz at netcom.com  Wed Jun 14 17:41:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: Wed, 14 Jun 2000 17:41:56 -0400
Subject: Simple thread prob
References: <n67X4.13491$XW1.115847@typhoon.hawaii.rr.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296ACE@mail.sauder.com>

In article <n67X4.13491$XW1.115847 at typhoon.hawaii.rr.com>,
Ty K. Kroll <tkroll at hawaii.edu> wrote:
>
>    I'm trying to build a thread that will timeout. I've only used simple
>threading in Python. I'm thinking wait() in Event or join() in Thread may be
>helpful, but the documentation is sparse. Can anyone point me in the right
>direction? Thanks. -Ty

Why do you want the thread to time out?  What is the purpose?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha



From tjreedy at udel.edu  Wed Jun 14 00:06:23 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 14 Jun 2000 00:06:23 -0400
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com> <01bfd2ed$42b67880$29cf8ad1@daves-dell> <39463BA7.43A3CD65@gssec.bt.co.uk>
Message-ID: <8i6vul$cru$1@news.udel.edu>

> So what? Unix has supported both for a long long time. You
> just escape the spaces or quote the names.... The only
> programs that seem to get really confused are Win16 or
> old DOS programs.

Just last year, I got a comercial 'WIN9x' program which installed itself,
by default, in "Program Files" and then would not run because of the space
(ie, it could not find itself or something).  It ran fine when I put it in
"Programs".  Who knows, probably some old code buried within.  So there is
a real problem which has cost this one user at least 2 hours over the
years.  Now how many million others ...







From cut_me_out at hotmail.com  Tue Jun  6 12:42:03 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Jun 2000 12:42:03 -0400
Subject: PyRun_SimpleString error handling
References: <393D062D.8413AC36@ncsa.uiuc.edu>
Message-ID: <etdem6a4v8k.fsf@m56-129-16.mit.edu>

> Here's my simple pgm:

Sorry I can't help, I've never done any embedding before.  But how do
you get it to compile?

gcc tst.c -I ../include/python1.5/ -L ../lib/python1.5/config \
	-lpython1.5

gives complaints about millions of undefined references.

Alex.


From moshez at math.huji.ac.il  Fri Jun  9 14:37:03 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 21:37:03 +0300 (IDT)
Subject: Porting CPython
In-Reply-To: <8hrbph$a4n$1@nnrp1.deja.com>
Message-ID: <Pine.GSO.4.10.10006092133190.28574-100000@sundial>

Not a lot I can help you with, but:

On Fri, 9 Jun 2000 perl_monkey at my-deja.com wrote:

> - Excluding Modules
> 
> What is the build procedure for omitting certain built in C modules?
> (I"m talking about the stuff in Modules/)

Edit the "Setup.in" file. Instructions for how are inside the file, but if
you have any more specific questions, please ask again.

> Any help on any of these issues would be appreciated.  Also, when/if I
> finish this, I'd be willing to diff the source and put it into the main
> python branch if that's something people would be interested in.

I'm CCing Python-Dev on this, so hopefully someone from the BeOpen team
can respond. I can just say I'm for it. (Python-dev is on BCC so replies
to this won't automatically go there.)

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From dgoodger at bigfoot.com  Thu Jun 29 23:33:03 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Thu, 29 Jun 2000 23:33:03 -0400
Subject: Which member is it?
In-Reply-To: <395BCDF2.CAFD7A34@bioeng.ucsd.edu>
References: <395BCDF2.CAFD7A34@bioeng.ucsd.edu>
Message-ID: <B5818D2E.599C%dgoodger@bigfoot.com>

on 2000-06-29 18:30, Curtis Jensen (cjensen at bioeng.ucsd.edu) wrote:

> class Foo:
>   def __init__(self):
>     self.bar1 = [1,2,3]
>     self.bar1 = [1,2,3]
>     self.bar2 = [1,2]
>     self.bar2 = [1,2,3,4]
> 
> tmp = Foo()
> 
> If I make reference to tmp.bar1, which do I get or, does it matter?

The second one.

> If I make reference to tmp.bar2, which do I get?  it will make a
> difference here.

The second one.

>>> tmp.bar2
[1, 2, 3, 4]

> In my code, it seems tmp.bar1 doesn't seem to cause problems.  However,
> tmp.bar2 does cause some funny problems, especialy when making calls to
> C modules.  What exactly does python do when you reference same name
> with different bindings?  Thanks.

The second "self.bar2 = [...]" assignment simply binds the name "self.bar2"
to new object, the list [1,2,3,4]. The original binding is broken, and the
object [1,2] is gone (reclaimed by the interpreter). No magic!

What funny problems are you talking about? Perhaps you can post examples?

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From bvdpoel at uniserve.com  Thu Jun 15 12:10:04 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Thu, 15 Jun 2000 09:10:04 -0700
Subject: tkinter: wait_variable
References: <39482E96.350D115E@uniserve.com> <8iabbm$i8o$1@nnrp1.deja.com>
Message-ID: <3948FFDC.C145028C@uniserve.com>

John Grayson wrote:
> 
> In article <39482E96.350D115E at uniserve.com>,
>   Bob van der Poel <bvdpoel at uniserve.com> wrote:
> >
> > How does wait_variable() work? I've got the following in a class:
> >
> >       __marked = []
> >
> >       ....
> >
> >       def mark(self):
> >               a callback to mark certain items
> >               self.__marked.append(x)
> >
> >       def wait(self, root):
> >               root.wait_variable(self.__marked)
> >               return self.__marked
> >
> > and I call it, after setting up,etc:
> >
> >       print myclass.wait(root)
> >
> > However, it never seems to return. With test print statements I have
> > determined that my __marked variable is being updated...but
> > wait_variable() doesn't seem to sense it.
> >
> > --
> >    __
> >   /  )      /         Bob van der Poel
> >  /--<  ____/__        bvdpoel at uniserve.com
> > /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
> >
> >
> 
> You've got to wait for a Tkinter (Tk) variable. There is no ListVar
> class, so you need to wait on an IntVar or BooleanVar, then in your
> mark method, set self.__marked *and* your Tkinter variable...

Ahhh, actually that makes sense! And of course this is clear in the docs
for tkinter as well :). Thanks very much.


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


From jcm at bigskytel.com  Sat Jun 10 19:32:45 2000
From: jcm at bigskytel.com (David Porter)
Date: Sat, 10 Jun 2000 17:32:45 -0600
Subject: Programming Poll
In-Reply-To: <2e506186.ac63c595@usw-ex0102-084.remarq.com>; from ktlehmaNOktSPAM@uswest.com.invalid on Sat, Jun 10, 2000 at 08:05:34AM -0700
References: <8hrim7$fjv$1@nnrp1.deja.com> <20000609210521.A6218@bigskytel.com> <3941DCEC.5E5A0C06@stud.ntnu.no> <2e506186.ac63c595@usw-ex0102-084.remarq.com>
Message-ID: <20000610173245.A11825@bigskytel.com>

* Kip Lehman <ktlehmaNOktSPAM at uswest.com.invalid>:
> Take the ending comma off the URL, you'll be able to reach
> the site.

I did that originally. For me, http://cupid.suninternet.com/~joeo/ returns
404 also, but http://cupid.suninternet.com/ works fine. How can other people
successfully reach the site when I recieve a 404? I've never heard of that
happening. 

  david.

-- 
As a sharp indecent stir unclenches into fingers.
  -- e.e.cummings in a blender



From emile at fenx.com  Tue Jun  6 08:20:08 2000
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 6 Jun 2000 05:20:08 -0700
Subject: indirect references
References: <8hha1n$phl$1@nnrp1.deja.com> <393CEBE0.8CEAF61D@sightreader.com>
Message-ID: <0d4a01bfcfb1$96479d60$1906a8c0@fc.fenx.com>

Well, that calls for a response! ;-)

>>> def spam(a):
  import sys
  try:
    1/0
  except ZeroDivisionError:
    callers_globals =
sys.exc_info()[2].tb_frame.f_back.f_globals
    callers_locals =
sys.exc_info()[2].tb_frame.f_back.f_locals
  if id(callers_locals) == id(callers_globals):
    callers_globals = {}
  retval = []
  for ns in (callers_locals, callers_globals):
    for calling_var_name, calling_var_value in ns.items():
      if id(calling_var_value) == id(a):
        retval.append((ns, calling_var_name))
  for ns, var in retval:
    ns[var]="spam and "+ns[var]

>>> a = "eggs"
>>> spam(a)
>>> a
'spam and eggs'
>>> a = "eggs"
>>> b = a
>>> spam(a)
>>> b
'spam and eggs'
>>>

Of course, it's not not going to work except
in console mode.

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: Ken Seehof <12klat at sightreader.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Tuesday, June 06, 2000 5:17 AM
Subject: Re: indirect references


> I'm not sure exactly what the effect is that you are
after, so I assume
> that what you are looking is something like call by
reference in a
> fuction, e.g.
>
> >>> a = "eggs"
> >>> spam(a)
> >>> print a
> spam and eggs
>
> Unfortunately the margin is too small to contain my
implementation.  :-)
>
> shaleh at debian.org wrote:
>
> > How do I do this in python:
> >
> > foo = 'string'
> > bar = foo
> >
> > somehow assign to bar and change the value of foo.
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>




From richard_chamberlain at ntlworld.com  Fri Jun  9 03:17:37 2000
From: richard_chamberlain at ntlworld.com (Richard Chamberlain)
Date: Fri, 9 Jun 2000 08:17:37 +0100
Subject: Tkinter: any drag-and-drop features, yet?
References: <20000607230310.27755.qmail@hotmail.com>
Message-ID: <8hq5jd$80k$1@gxsn.com>

Have a look at Tkdnd.py in the standard distribution.

Richard

Hung Jung Lu <hungjunglu at hotmail.com> wrote in message
news:20000607230310.27755.qmail at hotmail.com...
> Hi,
>
> I have not touched Tkinter for a while, now.
>
> Is there any drag-and-drop feature now in Tkinter or any other Python
public
> graphical interface? Sometimes I think the Unix community is so reluctant
in
> incorporating drag-and-drop that the cross-platform GUI system will never
be
> as competitive as Visual Basic.
>
> regards,
>
> Hung Jung
>
> ________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
>
>




From garry at sage.att.com  Thu Jun 15 14:43:05 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Thu, 15 Jun 2000 18:43:05 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com> <ug0qflns9.fsf@ctwd0143.fitlinxx.com>
Message-ID: <394923B9.C6BC1902@sage.att.com>

David Bolen wrote:
> 
> hzhu at knowledgetrack.com (Huaiyu Zhu) writes:
> 
> > Now only if somebody could enhance that to check inherited methods too (the
> > "more general case").  That is, if a method is not found, go recursively
> > into the superclasses.  The following example is not correct at the moment:
> 
> Ah, good point - try this variant - I switched to getattr() which
> takes the inheritance into account (you could also write your own
> recursive function that used Instance.__bases__ to look in the
> superclasses).

this message, and the others in this mini-thread, point out
what is, to me, the real power of a language like python
vs. something like c++.  thomas complained about a problem
where all interface methods might not be implemented, and
how it could bite you down the road.  a legitimate concern,
though one that, for many, is not important.

david (and others) quickly whipped up a mechanism to answer
this need.  i can use it where it matters, or ignore it where
it doesn't.  my choice.

i am always impressed with python, and this newsgroup, when 
someone points out a "missing" feature, and others show how
to implement it within the existing language.  this is the
real power of the language, and a large part of why it remains
simple.  you don't need to add features to the core language
when you can just implement them in terms of it instead.  this
is a large part of the attraction of scheme, as well (at least
for me).

reminds me of the old X slogan, "mechanism, no policy".

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From embed at NOSPAM.geocities.com  Tue Jun 27 13:11:38 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 27 Jun 2000 13:11:38 -0400
Subject: How do you copy a Dictionary object from C code?
Message-ID: <Ge565.9419$227.220401@nnrp1.uunet.ca>

Please Guido, add PyDict_Copy( dict ) to the API in the future!!!!

This is an annoying oversight in the Dictionary's design.  The dict_copy
method inside dictobject.c does the exact right job, but is well hidden
because it is a static function. It might be possible to look it up by
getting a reference to the Dictionary object's attribute named "copy" and
then calling it, somehow, but you can't even copy and paste dict_copy into
your own module because it has carnal knowledge of the internals of the
dictionary object.

Sigh. What is the best way in C to make a copy of a dictionary?

Warren




From rhicks at nospam.rma.edu  Thu Jun  1 21:49:03 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 02 Jun 2000 01:49:03 GMT
Subject: Styling Temporary Code
References: <3936A9BB.FBEF9201@visionart.com> <3936C52F.520B04FD@uab.edu>
Message-ID: <joEZ4.27$vh.14427@news>

What IS PyUnit...I just read a "bit" about eXtreme programming and PyUnit
was mentioned.

Bob (my-name-is-a-palindrome)

Shae Erisson <shapr at uab.edu> wrote in message
news:3936C52F.520B04FD at uab.edu...
> Pete Shinners wrote:
>
> > int myfunc(int nullargs)
> > {
> >     int newval = do_fancy_stuff(nullargs);
> > cout << "newval = " << newval << endl;
> >     return more_fancy(newval);
> > }
> >
> > ok, in this 5 line example it doesn't really sparkly,
> > but when i use it in big sections of code, it can't
> > be beat. call it the mark of the beast, chase me with
> > pitchforks, i simply love it and i don't want to change!
> >
> > now in python this obviously won't work, but i haven't
> > been able to settle on a style that works for me yet.
> > what do other people use for this "working" code?
>
> The two things I use are assert statements and print
>
> def myfunc(invar)
>     assert invar == type(''),'invar in myfunc is not a string!'
>     newval = do_fancy_stuff(nullargs)
>     print newval
>     return more_fanc(newval)
>
> I don't do C, so this may not be an exact translation....
>
> On the other hand, I also use PyUnit, and that often renders this kind
> of code moot.
> --
> sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
>    bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
> 13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j




From wjdandreta at worldnet.att.net  Wed Jun 28 20:57:08 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 29 Jun 2000 00:57:08 GMT
Subject: ?Does Python have sorting via "Dictionary Order"?
References: <395A92D0.65A205B1@email.sps.mot.com>
Message-ID: <E9x65.13601$kK.384134@bgtnsc05-news.ops.worldnet.att.net>

Hi Norman,

If you convert all to upper or lower case before sorting, that will take
care of the letter problem.

If you prepend a '0' to all single digit numbers that will solve the number
comparison problem.

If you don't want to change the sortees, put them in a dictionary with the
keys adjusted as above.

Bill
Norman Shelley wrote in message <395A92D0.65A205B1 at email.sps.mot.com>...
>Does Python have sorting via "Dictionary Order"?  This would be a very
>handy addition as an option to list's sort() function.
>
>This is defined by D. Richard Hipp as:
>"The case of alphabetic characters is ignored, except to break ties.
>Thus "B"
>comes before "b" but after "a".  Also, integers embedded in the strings
>compare
>in numerical order.  In other words, "x10y" comes after "x9y", not
>before it as
>it would when using strcmp().
>




From timaratz at lisco.com  Mon Jun 26 19:58:09 2000
From: timaratz at lisco.com (Peter Timaratz)
Date: Mon, 26 Jun 2000 18:58:09 -0500
Subject: smtplib
In-Reply-To: <6mK55.1713$iN5.345307@ptah.visi.com>
Message-ID: <000801bfdfca$61227820$b6cb45c6@timaratz>

As a kindness to all the idiots like me perhaps this intro can be added to
the smtplib module:

"This module implements SMTP and does not send e-mail. If you want to send
e-mail with Python then learn SMTP. If you don't want to learn SMTP then you
can use one of the many Perl modules available that don't require you to
know SMTP."

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Grant Edwards
> Sent: Monday, June 26, 2000 10:09 AM
> To: python-list at python.org
> Subject: Re: smtplib
>
>
> In article <001501bfdf74$6dff1fe0$14b745c6 at timaratz>, Peter
> Timaratz wrote:
>
> >Yes, I figured out that smtplib requires the subject and
> >headers to be imbedded in the message. I'm new to Python and I
> >love the language. But I think that the design of this module
> >reflects poorly upon Python.
>
> The module impliments SMTP, hence the name.  SMTP does not
> concern itself with message contents.  Writing smtplib to make
> certain assumptions or force certain conventions regarding
> message content would be just plain incorrect.
>
> >Almost every CGI programmer who looks at Python will want to
> >use this module. It would only take a few lines of code to make
> >it much more usable.
>
> Maybe you should write a higher level e-mail module that does
> what you are requesting.  It does not belong in smtplib.
>
> >I'm saying this not to flame Python, but to encourage those
> >responsible for it's distribution to look more critically at
> >what they are putting out.
>
> Perhaps you should look more critically at what you expect from
> a library that claims only to impliment SMTP.  If you want a
> layer of functionality on _top_ of SMTP, then that's something
> another module can impliment.
>
> >I hope that this module is atypical of the quality of this
> >product. I'd like to see Python become more popular. There are
> >many factors involved in that, but the quality of the modules
> >is certainly one of them.
>
> There is nothing wrong with smtplib.  You seem to be expecting
> it to do things other than SMTP.
>
> --
> Grant Edwards                   grante             Yow!  I'm an East Side
>                                   at               TYPE...
>                                visi.com
> --
> http://www.python.org/mailman/listinfo/python-list
>




From * at spam.ruud.org  Wed Jun 21 07:13:44 2000
From: * at spam.ruud.org (Ruud de Rooij)
Date: 21 Jun 2000 13:13:44 +0200
Subject: 123.4+2=125.40000000000001
References: <3950A30D.DEE4CF4D@brb.midat.de>
Message-ID: <87lmzzi8wo.fsf@hobbes.home.ruud.org>

Peter Koppatz <pkop at brb.midat.de> writes:

> >>> 123.4+2
> 125.40000000000001   # is this OK?
> >>> print 123.4+2
> 125.4
> >>>
> 
> On my calculator: 123.4+2 = 125.4 !?
> I am using Python 1.6a2

>From http://www.python.org/1.6/

  The repr() of floating point numbers now gives full (17-digit)
  precision. This means that a number like 8.1 (which cannot be
  represented exactly in binary) will show up as 8.0999999999999996 when
  displayed using repr(). This shouldn't break any code but it may
  confuse users, especially since the interactive shell uses repr() to
  display all floating point results. Use str() or explicit precision in
  string format (e.g. "%.12g" % x) to see the the more familiar, rounded
  form.

Cut-n-paste'ly y'rs,

	- Ruud de Rooij.
-- 
ruud de rooij | *@spam.ruud.org | http://ruud.org


From cjensen at bioeng.ucsd.edu  Thu Jun 22 18:43:14 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 22 Jun 2000 15:43:14 -0700
Subject: url directory listing
References: <39518289.D4175618@bioeng.ucsd.edu> <slrn8l3qfm.nol.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <39529682.A3F00A25@bioeng.ucsd.edu>

Remco Gerlich wrote:
> 
> Curtis Jensen wrote in comp.lang.python:
> > I little while ago, there was a post about websucker.  I'm writing a
> > similar script. What is the socket command or url command to get a
> > directory listing?  Especialy if there is an index.html file in the
> > directory.  Thanks.
> 
> This is not possible with HTTP.
> 
> --
> Remco Gerlich,  scarblac at pino.selwerd.nl
> Hi! I'm a .sig virus! Join the fun and copy me into yours!

It is possible to get the files off of a site.  softbytelabs did it:
http://www.softbytelabs.com/ 
Their black widow program does a fine job of copying site files.  Maybe
it's not done with HTTP, but does anyone know how it's done?  What does
WebSucker do?  Thanks.

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


From joseph_ryan at bmc.com  Mon Jun  5 15:51:40 2000
From: joseph_ryan at bmc.com (Joseph Ryan)
Date: Mon, 5 Jun 2000 15:51:40 -0400
Subject: Recompiling Python with threads enabled
References: <n1l03rcl.fsf@jumbo.skott.dk>
Message-ID: <sjo14j7j5ri159@corp.supernews.com>

I would suggest that you get the source for the last stable release (start
at www.python.org). This would be 1.5.2.

(I am using 1.6alpha2 without incident).  The compilation with threads is
provoked during the configure step by specifying --with-threads (this is
*not*) the default. For example:

CC=gcc; ./configure --with-threads

( need the CC=gcc to override using the default compiler on my UNIXes; on
Linux this is less likely to be needed). There are many other options for
configure. I would suggest

./configure -help

to get started and check out the README file of course.

I also build 1.5.2 from source on UNIX but I now recall that I switched to
the 1.6 version in part because it promised better support for threads. :-)
You may want to research more in this subject especially if you plan to use
extensions (exp Tkinter) for which threads are less reliable/useful.

Good luck!

"Martin Sk?tt" <mskott at image.dk> wrote in message
news:n1l03rcl.fsf at jumbo.skott.dk...
> Hi
>
> Sorry if you don't think this belongs in a Python group, but I think
> its more relevant here than in a Linux group. Sorry for the
> inconvenience :-)
>
> And now on to the question:
> I wish to recompile Python 1.5.1 on my Red Hat 5.2 system to enable
> the threads module, but how do I do? Do I need to install the src RPM
> and which files should I change to enable threads??
>
> I hope anyone will help me
> --
> Martin Sk?tt
> mskott at image.dk
>
> 'In a certain sense, all decent programming languages are equally
powerfull'
> Dr. Alan Turing




From Wolfgang.Strobl at gmd.de  Mon Jun  5 04:45:49 2000
From: Wolfgang.Strobl at gmd.de (Wolfgang Strobl)
Date: Mon, 05 Jun 2000 10:45:49 +0200
Subject: Zope compatible with Windows 2000?
References: <393B485F.BEE013FF@micron.net>
Message-ID: <90qmjsod04r3tv2gqpf8kiof9v7m42n8lu@4ax.com>

Peter Immarco <android at micron.net> wrote:

>Forgot to ask, does Zope have a version that runs well with Windows 2000
>server?

I haven't had any difficulties running 2.1.6 on Windows 2000, so far. I
tried it on both Win2000 prof US and Win2000 Server Ger.

-- 
Bicycle helmets are the Bach flower remedies of traffic


From hei at adtranzsig.de  Wed Jun 14 05:51:21 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 14 Jun 2000 11:51:21 +0200
Subject: Windows: COM Automation problem with german VB localisation
Message-ID: <8i7klg$23i$1@desig-bs01-s04.adtranzsig.de>

Hi all!

I have problems automating a german version of MS project.
Like usual, i tried to find out the COM interface of the program
by recording VB macros in it and re-typing them in my Python
script.

The problem is: It's one of the MS programs where they localized
the Visual Basic inside it, unfortunately (and needlessly) german
Umlaute (?,?,?)
are part of the function names the MS program understands.
Python chokes on them, it seems.

Two questions:
- Does anybody know a workaround?
- Is there a more elegant way of finding out the COM interface
  then recording the steps you want to automate with the
  VB macro recorder? Can i query the interface of the COM
  object from within my script? i tried something along the
  way of "for i in o.__class__.__dict__.keys(): print i",
  but got only a few internal methods of the COM wrapper
  stuff.

(When you answer into the group, please cc a copy
to my mailbox, i'm not a regular reader at the moment)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net






From mtbros at tin.it  Tue Jun 13 16:01:15 2000
From: mtbros at tin.it (Mario)
Date: Tue, 13 Jun 2000 22:01:15 +0200
Subject: help formatting number
Message-ID: <8i645i$lht$1@nslave2.tin.it>

Hi..
this is a simple question but I can't find a simple answer... so help me.

I have an integer variable and I want it to become a string variable
formatted so that there is a dot between each three digits of the original
number. An example:
var =3456789
It must become:
"3.456.789"
Ideas ?
 Thanks
        Mario







From hartmut at oberon.noris.de  Tue Jun 27 17:28:22 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Tue, 27 Jun 2000 22:28:22 +0100
Subject: "No module named _locale" (1.5.2 w/ RH 6.2)
Message-ID: <16600709@oberon.noris.de>

Hallo,

when trying to import 'local' a get an error. How can I solve it?

>>> import locale
Import Error: No module named _locale

Well, this is true, since there is no _locale.* on my system.
I'm running RedHat 6.2 with Python 1.5.2 (from ReadHat's 6.2 rpm)

+++hartmut


From andrew.henshaw at gtri.gatech.edu  Fri Jun 23 02:45:42 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Fri, 23 Jun 2000 02:45:42 -0400
Subject: Extending Python command line
Message-ID: <8iuj9g$f0t$1@news-int.gatech.edu>

Is there a reasonably simple way to extend the Python command line so that
instead of typing:

   >>> myspecialfunction('this is some stuff')

I could type:

   >>> myspecialfunction this is some stuff

Everything else should be handled normally.

A nice bonus would be if it could handle line continuations.

Cheers,

Andy Henshaw


From jhe at webde-ag.de  Fri Jun 30 05:58:57 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Fri, 30 Jun 2000 11:58:57 +0200
Subject: ANNOUNCE: Distutils 0.9
References: <20000630002620.A2329@beelzebub>
Message-ID: <8jhqu6$iua$1@pollux.ip-plus.net>

"Greg Ward" <gward at python.net> schrieb im Newsbeitrag
news:20000630002620.A2329 at beelzebub...
> The Distutils are a standard part of Python 2.0; if you are running 2.0
> (or one of the 1.6 alpha releases that preceded it), ...

Man, you certainly live in the future! ;)





From scarblac-spamtrap at pino.selwerd.nl  Tue Jun  6 13:27:09 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 6 Jun 2000 17:27:09 GMT
Subject: Space Sensitive?
References: <8hj9vv$1kd@autodesk.autodesk.com>
Message-ID: <slrn8jqd11.n3q.scarblac-spamtrap@flits104-37.flits.rug.nl>

Akira Kiyomiya wrote in comp.lang.python:
> I hear Python is space sensitive, however, "Learning Python" from O'Reilly
> and "Python Essential Reference" from New Riders do not cover this topic.
> 
> Could someone explain this one with a nice simple example why Python is
> space sensitive?

It means that Python tells to which block a statement belongs by looking at
the indentation.

if something:
   print "Hello!"
   x = 3
   
Is different from

if something:
   print "Hello!"
x = 3

In the second case, 'x = 3' doesn't belong to the if statement. This:

if something:
   print "Hello!"
  x = 3

Is a syntax error.

It means you have to use some consistent indenting, and that the code isn't
filled with all those { { } } thingies. No big deal, really.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From tim.hochberg at ieee.org  Wed Jun 28 14:36:02 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 28 Jun 2000 18:36:02 GMT
Subject: list.pop([i]) and list.remove(x) w/ for loops
References: <8jdfb6$ugh$1@netnews.upenn.edu>
Message-ID: <m2bt0l4sc6.fsf@spiff.home.com>

dwhite2 at blue.seas.upenn.edu (David White) writes:

> I was getting some unexpected behavior when using list.remove(x),
> wondering if someone can tell me what's going on.
> 
> I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].  I
> wan't to delete all the a objects out of the list while preserving
> the order of the b objects.  My code looks like this:
> 
> for o in olist:
>     if (o.type == a):
>        olist.remove(o)
> 
> Now for some reason I get a left over a object that wasn't removed.
> Anyone know why?  Using python 1.5.2.

Without getting into the gory details, modifying a list while
iterating over it can have unexpected results. Assuming your lists
aren't huge, probably the best thing to do would be to make a copy of
the list:

for o in olist[:]:
    if (o.type == a):
       olist.remove(o)

There are some other ways to deal with this if you can't make a copy,
but that's rarely the case.

You can probably get an idea of what's going on in your case by
rewriting the for loop as an equivalent while loop:

i = 0
while 1: # loop forever
   try:
      o = olist[i]
   except IndexError:
      break
   i = i + 1
   # Up to here is equivalent to "for o in olist:"
   olist.remove(o)


Everytime you delete something, olist shrinks and you end up skipping
the next element in the iteration and this results in some items not
being deleted.

-tim


From huaiyuan at rac6.wam.umd.edu  Thu Jun 15 20:05:32 2000
From: huaiyuan at rac6.wam.umd.edu (No such thing)
Date: 15 Jun 2000 20:05:32 -0400
Subject: Python so far
References: <39484FBD.CAE467A0@micron.net> <39487243.B90DBD85@san.rr.com>
Message-ID: <c6xsnuefq2b.fsf@rac6.wam.umd.edu>

Courageous <jkraska1 at san.rr.com> writes:

> Python may be too slow for some tasks, but in the area
> of programmer productivity, very little beats it, and
> arguably the environments that do (e.g., lisp) have too
> many other negatives associated with them.

Care to explain?

[I hope the following won't turn into flamewar; I would like to have a
 better idea on what I will gain (or lose) when I decide on one language
 or another.]

I think Common Lisp already has some good solutions to many issues
confronting Python now:

- Optional type declaration;
- Optimizing compiler (taking advantage of the type declaration);
- Powerful macro (superseding the need for preprocessor);
- Pseudo case insensitivity (!) (contrary to wide misconception, 
  Common Lisp is actually case sensitive; it is the lisp reader (lexical
  analyser + parser?) that by default, convert everything to uppercase.
  Like anything else in Lisp, this behavior is customizable within the
  langauge.  (Hm, might be a good solution to the recent dilemma.));
- etc.

Even the syntax (parentheses) became an asset after I learnt to use
Emacs effectively.

I am wondering which path is shorter: an optimizing compiler for Python, or
interoperability for lisp?

- hyz



From paul at prescod.net  Wed Jun 28 18:05:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 28 Jun 2000 15:05:27 -0700
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com>
Message-ID: <395A76A7.3A860913@prescod.net>

Eric Lee Green wrote:
> 
> ...
> This is one reason why inheritance is so efficient in Python, BTW. 

No, it isn't so efficient!

> All
> inheritence does is add into the hash table for the overall class object those
> entries that were in the base class object. 

Python actually delegates from one class object to another.

> Then when you create a new
> instance, you actually are just creating a copy of the overall class object,
> without having to refer back to the base object(s).

Not true. Python delegates.

>>> class a:
...     def foo(self): print foo
...
>>> dir( a )
['__doc__', '__module__', 'foo']
>>> class b(a):
...     def bar(self): print "bar"
...
>>> dir( b )
['__doc__', '__module__', 'bar']
>>> b.foo=b.foo
>>> dir( b )
['__doc__', '__module__', 'bar', 'foo']

You can hook up the more efficient inheritance at a cost of dynamicity
and memory, if you want.

-- 
 Paul Prescod - Not encumbered by corporate consensus
The calculus and the rich body of mathematical analysis to which it 
gave rise made modern science possible, but it was the algorithm that 
made the modern world possible.
	- The Advent of the Algorithm (pending), by David Berlinski



From hanche at math.ntnu.no  Sat Jun  3 12:38:37 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 03 Jun 2000 12:38:37 -0400
Subject: Bad Math Module Notation re: atan2
References: <010620001459577031%pecora@anvil.nrl.navy.mil> <pcovgzstc9b.fsf@math.ntnu.no> <8hb9ci$kq4$1@nnrp1.deja.com>
Message-ID: <pcok8g6u3ci.fsf@math.ntnu.no>

+ david_ullrich at my-deja.com:

| In article <pcovgzstc9b.fsf at math.ntnu.no>,
|   Harald Hanche-Olsen <hanche at math.ntnu.no> wrote:
| [...]
| > The documentation (meaning the library reference) is wrong,
| > [...]
| > In fact, even the function's doc string is right (Py 1.5.2):
| >
| > >>> print math.atan2.__doc__
| > atan2(y, x)
| >
| > Return atan(y/x).
| 
|     As opposed to the 1.5.2 library reference, which says
| 
| atan2 (y, x)
|     Return atan(y / x).
| 
| Looks kinda the same to me... (maybe you got a different
| version of the docs?)

Seems I have the 1.5.1 library reference.  My net access during this
year of exice in the US of A isn't as good as it used to be, so I tend
to be a bit tardy in getting the latest and greatest, whether it be
docs or software.  And what's more, the 1.5.1 docs say

atan2 (x, y) 
     Return atan(x / y). 

which now that I read it more carefully isn't wrong, it just tends to
lead your thoughts in the wrong direction.

and-who-said-letters-aren't-magical-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From jkraska1 at san.rr.com  Mon Jun 12 02:40:12 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 12 Jun 2000 06:40:12 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com>
Message-ID: <3944866B.A0115FD8@san.rr.com>

mkx at excite.com wrote:
> 
> In many Perl vs. "other scripting language" discussions, I have often
> seen Perlers state that those who dislike Perl's syntax and structure
> are often non-Unix types,...

I dislike Perl, and I've been using Unix for 15 years.

> So, I come now to Python. After reading much about it's ease of use
> and readability advantages, I see it may be a better alternative. In
> most cases, I can actually get a concept of what the code is doing,
> without picking up a reference book - after only a week of
> experimentation.

Exactamundo, Cunningham. It's my expectation that any programmer
can become a python "expert" within a month of study.

> So my question is this: Is Python a more "Windows-friendly" language?
> Is it's base structure less related to any one specific operating
> system? Is learning Python an exercise in learning the language, and
> not a Unix tutorial? Can I be successful even if I do not use Unix ;)

I've never personally used python on a unix system (I use it on
Win98/NT), and I've never once made use of anything even vaguely
unix-like. While there are indeed os-specific packages, I've
personally touched neither one of them.

Python is the friendliest language I've used in my entire 20 year
history of programming. While Lisp tops Python for potency, the
trade in syntactic weight is well worth the exchange, IMO. Lisp
is annoying, and frankly python is far easier to work with and
debug.




C/


From thomasm at legend.co.uk  Tue Jun 20 10:54:06 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Tue, 20 Jun 2000 15:54:06 +0100
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk> <8imc1f$181$1@slb6.atl.mindspring.net> <394f2d55.0@katana.legend.co.uk> <8inrg6$mm2$1@slb7.atl.mindspring.net>
Message-ID: <394f851e.0@katana.legend.co.uk>

Aahz Maruch <aahz at netcom.com> wrote in message
news:8inrg6$mm2$1 at slb7.atl.mindspring.net...
> [BTW, please maintain attributions for quoted text; it makes it much
> easier to keep track when multiple people are writing.]
>
> In article <394f2d55.0 at katana.legend.co.uk>,
> Thomas Mangin <thomasm at legend.co.uk> wrote:
> >Attribution restored: Aahz
> >>Attribution restored: Thomas Mangin
> >>>
> >>>I notice that if a directory is root owned and chmoded 700 (let say
/test)
> >>>and if it contain a directory called test2
> >>>os.path.isdir("/test/test2") return 0 if called by a "normal user"
> >>>I was expecting an exception.
> >>
> >> Why were you expecting an exception?
> >
> >I created few directories for my application as "myself" but I am running
> >the script as a cgi.
> >
> >drwx------ thomas thomas  [somwehere in the world]/test
> >drwx------ thomas thomas  [somwehere in the world]/test/test2
> >
> >my web server and I are different users,  as "nobody" have no read/go
> >through access on the parent (test) directory I am checking, it can not
even
> >try to perform any action on the child one (test2).
>
> But it's not performing any action on test or test2; it's performing an
> action on the path /test/test2.  I believe it actually uses the OS call
> stat(), but that doesn't really matter; the point is that whatever
> internal function gets called on the *entire* path returns an error,
> therefore the path cannot be a directory.

I still disagree I don't know if it is as I missexpressed myself
or if I misunderstood you.

> >Nothing with the current return code indicate that the test failed.
> >I was just guessing : return 0 means no directory present, not enable to
> >perform the requested action ..
>
> It means that for whatever reason (permission denied or non-existence or
> because it's a file), the path you gave isdir() is not a directory.  If

I my case it IS a directory and I could be a file as the program is not able
to stat it but if the function is not able to perform its action properly
as a programmer I would like to know it

When you write this :

try:
    if not os.path.isfile(xx) and not os.path.isfile(xx):
        try:
            createDir(xx)
        except:
            print "creation failed"
exception:
    print "Dont have the right to access the directory"

A = 10
print a

You are not expecting `None` to appear to the screen as "a" is not defined.
You want to know a is not defined you want a error.

> you want more information than that, you need to get it yourself.  More
> than that, if you want to test individual path components, you have to
> do that yourself, too.

Yes, a pain, ;*) But I guess I have no choice.
I will have to write my own isfile() and isdir()

Thomas




From cut_me_out at hotmail.com  Thu Jun 15 10:03:08 2000
From: cut_me_out at hotmail.com (Alex)
Date: 15 Jun 2000 10:03:08 -0400
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu> <150620000713119061%pecora@anvil.nrl.navy.mil>
Message-ID: <etdu2evf3dv.fsf@w20-575-109.mit.edu>

> Alex, I think I sense what you're suggesting here (checking argument
> and return types, I think), but I don't know how to do that with a
> script and even then, how could one tell what is being done inside the
> function in terms of memory allocation or just pointers assigned to
> existing memory, i.e. no ref. count increase ?

Well, it wasn't a good idea anyway, since the information it would have
gotten is already in the doc's.  But I was thinking writing a script
that would write a C program that would create the python objects needed
by each API function, and record how their refcounts change after they
are passed to those functions.

Alex.



From barryhart at my-deja.com  Sat Jun 10 23:29:43 2000
From: barryhart at my-deja.com (barryhart at my-deja.com)
Date: Sun, 11 Jun 2000 03:29:43 GMT
Subject: PythonWin and collections
Message-ID: <8hv12v$ol1$1@nnrp1.deja.com>

I'm trying to port the following fairly simple code to Python. It uses
ADSI to get a list of members in an NT group.

    For Each member In GetObject("WinNT://usa/Developers").Members
        Debug.Print member.ADsPath
    Next

I'm a fairly new Python user, and so far I've never figured out how to
enumerate a COM collection object properly. In previous cases, I've
been able to work around this by using the "Count" and "Item" methods
to index over the collection, but this particular collection doesn't
expose Count and Item methods.

Here's my Python attempt at the above code:

    from win32com.client import GetObject

    for member in GetObject("WinNT://usa/Developers").Members:
        print member.ADsPath

The "for" loop fails with "TypeError: loop over non-sequence". I also
tried adding "._NewEnum()" after the ".Members", but this fails
with "AttributeError: _NewEnum".

Help!

-----------------------------
Barry Hart
Velant, Inc.


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


From mikael at isy.liu.se  Tue Jun 13 11:11:46 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 13 Jun 2000 11:11:46 -0400
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 200
In-Reply-To: <bkhckscbhq9e9i18ijci51eh4r82f3qaqk@4ax.com>
Message-ID: <642A954DD517D411B20C00508BCF23B0012968EF@mail.sauder.com>

On 13-Jun-00 Anders M Eriksson wrote:
 >  It's a pity that this packages doesn't work on Windows NT! 

The same goes for Win95.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    13-Jun-00
Time:    17:11:17

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



From tuttledon at hotmail.com  Tue Jun 27 09:10:46 2000
From: tuttledon at hotmail.com (Don Tuttle)
Date: Tue, 27 Jun 2000 13:10:46 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
Message-ID: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>

Consider this a heads up NOT a promotion!!

In case you haven't heard Microsoft is coming out with a massive new Web
stratagy that, in part, uses a new language called C# (That sharp, like in
music).  Here's a part of what they have to say:

"Because of its elegant object-oriented design, C# is a great choice for
architecting a wide range of components-from high-level business objects to
system-level applications. Using simple C# language constructs, these
components can be converted into Web services, allowing them to be invoked
across the Internet, from any language running on any operating system."

"More than anything else, C# is designed to bring rapid development to the
C++ programmer without sacrificing the power and control that have been a
hallmark of C and C++. Because of this heritage, C# has a high degree of
fidelity with C and C++. Developers familiar with these languages can
quickly become productive in C#."
.......

"The modern design of C# eliminates the most common C++ programming errors.
For example:

Garbage collection relieves the programmer of the burden of manual memory
management.
Variables in C# are automatically initialized by the environment.
Variables are type-safe. "

Hmmm, sound familar?
Here's the link
http://www.msdn.microsoft.com/vstudio/nextgen/technology/csharpintro.asp

I look forward to you comments.

Don


--
Believe nothing no matter where you read it or who has said it, not even if
I have said it, unless it agrees with your own reason and your own common
sense.
--Buddha




From phd at phd.russ.ru  Fri Jun 30 08:12:29 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 30 Jun 2000 12:12:29 +0000 (GMT)
Subject: MimeWriter (how to use/find boundaries)
In-Reply-To: <Pine.LNX.4.21.0006301138390.15363-100000@lade.src.no>
Message-ID: <Pine.LNX.4.21.0006301212190.945-100000@fep132.fep.ru>

On Fri, 30 Jun 2000, Morten W. Petersen wrote:
> I've built a mail-sending function, which, when sending mail with
> attachments; does send the mail, with the encoded attachments; but it
> doesn't make the mail readable on the other side. I.e., it does not
> include the boundary thing, and I don't have a clue as to how to go about
> to do it. (...)

   http://www.faqts.com/knowledge-base/view.phtml/aid/1323/fid/236

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From ddevilliers at lando.co.za  Sat Jun 24 13:23:17 2000
From: ddevilliers at lando.co.za (D De Villiers)
Date: Sat, 24 Jun 2000 19:23:17 +0200
Subject: Python for Palm ?
Message-ID: <8j56ir$7fr$1@ctb-nnrp2.saix.net>

Hello, Everyone

I just started programming my Palm device and is looking for a version of
Python for the Palm. Does anyone know please know if such a product exist
and where I can find it ?

Lennie De Villiers

ICQ# 57008830

E-mail: ddevilliers at lando.co.za
Moderator: news:comp.sources.delphi

Lennie's Weekly Delphi Tips:
----------------------------
  http://ssapcs.hispeed.com/Lennie/LennieTips.htm

Lennie's Delphi Mailinglist:
----------------------------
To subscribe, send a e-mail to ddevilliers_delphi-subscribe at egroups.com or
visit http://www.egroups.com/group/ddevilliers_delphi/ for more information.

Dr.Bob's Kylix Kicks:
---------------------
  http://www.drbob42.com/kylix/

The Delphi Shrine:
------------------
  http://www.delphishrine.com/
A new Delphi website by my friend Emiel Hollander. Tips & Taticks,
Tutorials, FAQs, and much more...






From m.faassen at vet.uu.nl  Wed Jun  7 19:53:20 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jun 2000 23:53:20 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <8hiq8d$ngm$1@newshost.accu.uu.nl> <slrn8jr8qm.b4q.wtanksle@dolphin.openprojects.net> <8hlk0t$h4q$1@newshost.accu.uu.nl> <8hlkkn$gu3$1@slb2.atl.mindspring.net>
Message-ID: <8hmn9g$ets$4@newshost.accu.uu.nl>

Aahz Maruch <aahz at netcom.com> wrote:
> In article <8hlk0t$h4q$1 at newshost.accu.uu.nl>,
> Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>> On 6 Jun 2000 12:19:25 GMT, Martijn Faassen wrote:
>>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>>>> On 5 Jun 2000 12:01:04 GMT, Martijn Faassen wrote:
>>>>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:

> May I suggest that the two of you take it to e-mail?  Or should we send
> the Spanish Inquisition?  ;-)

Aahh, just as we were arriving at consensus! :) Don't sic the PSU at us,
noohh!

Stubbornly-argumentative-ly yours,

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


From moshez at math.huji.ac.il  Tue Jun 13 00:33:06 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 07:33:06 +0300 (IDT)
Subject: Testing for a file in python?
In-Reply-To: <uHK0DsJ1$GA.106@net025s>
Message-ID: <Pine.GSO.4.10.10006130731240.20303-100000@sundial>

On Mon, 12 Jun 2000, A[r]TA wrote:

> > How would you go about testing if a certain file exists in python?
> > In most scripting languages you have the -f test:
> >
> > if -f filename do
> >    whatever_you_were_going_to_do
> >    fi
> > else do
> >    print "Error" file not found
> >    fi
> >
> > stat doesn't work, it aborts if the file does not exist.
> 
> try:
>     open('filename')
> except:
>     print 'File does not exist'

1. This checks that the file is readable too, which wasn't what the
original poster wanted.

2. Unqualified "except"s are evil. Use except (IOError, os.error): in this
case.

3. You're relying on GC to free the file descriptor, which is a dangerous
assumption (doesn't work on JPython, for example)

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From hgolden at my-deja.com  Thu Jun  1 21:08:32 2000
From: hgolden at my-deja.com (Howard B. Golden)
Date: Fri, 02 Jun 2000 01:08:32 GMT
Subject: "Fixing" os.system on Windows
Message-ID: <8h71ec$oh4$1@nnrp1.deja.com>

As explained in Hammond and Robinson's book, "Python Programming on
Win32," on p. 324, and in Section 6.1.5 of the Library Reference,
os.system() doesn't return the exit status of the process under Windows.

Carey Evans showed a version of system() that gets the exit code under
Windows in his message:

http://www.python.org/pipermail/python-list/1999-May/009409.html

I hope this could be adopted into the Windows part of the core Python
system, so that os.system() will work the same under Windows and Unix.
If there is some reason for not doing this, I'd appreciate it if
someone would explain it.

--
Howard B. Golden


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


From darrell at dorb.com  Sun Jun  4 00:33:36 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 3 Jun 2000 21:33:36 -0700
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it>
Message-ID: <003b01bfcdde$0cd2f950$6401a8c0@home>

Too much work to decode what your doing.
Maybe this will help.

import pprint, string
def parse(lines):
    res=[]
    for l in lines:
        temp = string.split(l)
        resLine=[]
        for n in temp:
            resLine.append(eval(n))

        res.append(resLine)
    return res

pprint.pprint(parse(open("lcase.001").readlines()))

############ Output
[[1],
 [4, 3],
 [1, 0.0, 0.0],
 [2, 0.0, 300.0],
 [3, 500.0, 300.0],
 [4, 500.0, 0.0],
 [1, 1, 2],
 [2, 2, 3],
 [3, 3, 4],
 [1, 300.0, 0.0, 1.0],
 [0.0, 0.0, 0.0, 2500.0, -146.922, 6526.54],
.....

--Darrell Gallion




Paolo Redaelli wrote:
> 
> but if I run a program that does x=float(x) where x = "0.000000E+00 "
> (being a string) here's the output. 
> 
> ===========================
> [paolo at aldebaran Telaio]$ ./pyplot.py
> x(<type 'string'>)='0.000000E+00' y(<type 'string'>)='0.000000E+00'
> Traceback (innermost last):
>   File "./pyplot.py", line 198, in ?
>     t = Telaio()
>   File "./pyplot.py", line 172, in __init__
>     x = float(x)
> ===========================





From paul.magwene at yale.edu  Sat Jun 10 21:29:17 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Sat, 10 Jun 2000 21:29:17 -0400
Subject: windows programming
References: <8hu92k$28ae$1@walton.maths.tcd.ie>
Message-ID: <3942EB6D.A7D384AD@yale.edu>

email at junked.nospam.com wrote:
...
> application using it. Part of the application is a graphing output, DISLIN
> looks good for this (any comments?). My problem is that I want to be able 
[snip]

If you're going to be using DISLIN you might want to check out my
pxDislin module at:  http://pantheon.yale.edu/~pmm34/pxdislin.html

Other possibilities include the gnuplot module ( URL?), or Graphite (
http://graphite.sourceforge.net - though development on this seems to
have ground to a halt).

--Paul

Paul Magwene
paul.magwene at yale.edu


From hzhu at rocket.knowledgetrack.com  Fri Jun  9 13:53:23 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 09 Jun 2000 17:53:23 GMT
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <393FE48B.217BB87A@embedded.cl>
Message-ID: <slrn8k2bth.pi.hzhu@rocket.knowledgetrack.com>

On Thu, 08 Jun 2000 14:23:07 -0400, Pablo Bleyer Kocik <pbleyer at embedded.cl>
wrote: 
>
>> 1. Everything is an object of a class that can be subclassed.
>
>    ... And everything is painfully slow and error prone. Strange and unexpected
>things will begin happening as a result for hidden behavior. Developers will go
>mad checking polymorphism when implementing their functions. Moral: Use fixed
>objects for what they are meant to be and keep them that way -- that's their
>raison d'?tre.

Maybe it's exactly the oposite?  Even with current situation, when you say
print a, Python needs to know the type of a to form the repr.  It might be
hard-coded for builtin types (I don't know), but it's there nevertheless.

Now imagine there is a class Int.  If someone decides 1/2 should be 0.5, he
only need do

class MyInt(Int):
    def __div__(self, other):  # Whatever code necessary.

Operations involving MyInt, other then __div__, will be as fast as Int.  And
the __div__ method is the only place one need to check for error.

Contrast that with the current situation.  One has to rewrite every operation

class MyInt:
   def __init__(self, data):  self.data = data
   def __add__(self, other):  return self.data + other
   ....

in addition to the single method __div__ that one wants to change.  This
slows everything down and introduces pontential errors everywhere -
quite unnecessarily.


>> 6. Have a clean mechanism to work with other programming languages other
>> than C.  (I have matlab/octave in mind.)
>
>    The core is there. I think it's enough clean already. Perhaps there's a need
>for better interfaces/object passing/object marshalling & demarshalling between
>Python and other applications and we should rely on a standard (CORBA?). Or
>better, make Python objects the standard ;^) ("glue it all with Python").

Any examples?  I'd love to incorporate that into the MatPy package.

>> 7. NOT case insensitive.  (No followup on this please.  Enough said already.)
>
>    You are not very fond of computers, are you?

Did you misread my words or did you mistype something?  Never mind.

>    I use Python mostly for scientific work (Numpy et al.). I use Matlab, Octave
>and Scilab also. Instead of changing radically the core language I think it's
>better to concentrate efforts in improving Numpy, for example. What would be nice
>is to reinforce the language in it's "dynamic mode/dynamic interface"
>capabilities via operator overloading, etc. Then you could run something that
>"looks like Matlab" over Python, efficiently.

Don't you think MatPy qualifies as "something that looks like Matlab over
Python"?   I'm really interested in hearing any efficiency suggestions.

>>
>> PS. (Shameless advertising):  Want a Matlab clone in Python?  See
>> https://sourceforge.net/project/?group_id=6490
>
>    Then use Matlab, and not attempt to change Python from it's bases.

I don't quite understand you here.  How do you have a dictionary of lists of
matrices in Matlab?  How do you define classes in Matlab?  Or maybe when you
use Matlab you don't need Python featurs and vice versa?  I need both.


Huaiyu


From liam at gpu.srv.ualberta.ca  Fri Jun  9 13:40:13 2000
From: liam at gpu.srv.ualberta.ca (L Stewart)
Date: 9 Jun 2000 17:40:13 GMT
Subject: readlines() question
Message-ID: <8hra5t$mfi$1@pulp.srv.ualberta.ca>

Just out of curiosity, when something like the following is used,

contents = open('file').readlines()

is 'file' closed after readlines() finishes? If not, is it possible to
somehow get either the file descriptor or the file object of the opened
file? Thanks,

Liam

-- 
Liam Stewart
E-Mail: liam at ualberta.ca

"It is not enough to be busy, so are the ants. The question is: what are
we busy about?"
-Henry David Thoreau


From shouldbe at message.com  Thu Jun  1 09:51:27 2000
From: shouldbe at message.com (Arinté)
Date: Thu, 1 Jun 2000 08:51:27 -0500
Subject: random # errors?
Message-ID: <8h5m0n$k1o$1@news2atm.raleigh.ibm.com>

I am trying to use whrandom for picture generation and I am getting this
error?
I am not sure the list is...  Any hints?
Traceback (innermost last):
  File "Beer", line 131, in ?
  File "Beer", line 37, in biko
  File "C:\Python\Lib\whrandom.py", line 46, in __init__
    self.seed(x, y, z)
  File "C:\Python\Lib\whrandom.py", line 60, in seed
    t = int((t&0xffffff) ^ (t>>24))
IndexError: list index out of range




From cchq at wanadoo.nl  Fri Jun 23 23:50:57 2000
From: cchq at wanadoo.nl (Max Heijndijk)
Date: Fri, 23 Jun 2000 23:50:57 -0400
Subject: Examples wanted by newbie
Message-ID: <39543021.52CECB68@wanadoo.nl>

Hello, does anyone know where to find *simple* python example programs ?
Tutorials never work for me without examples.
I want to convert my large shell scripts to python.

-- 
M at X

Climate Control - experimental music: http://go.to/cchq/
My Linux-related site: http://home.wanadoo.nl/cchq/conmen/index.html


From just at letterror.com  Thu Jun 15 13:32:05 2000
From: just at letterror.com (Just van Rossum)
Date: Thu, 15 Jun 2000 18:32:05 +0100
Subject: 8 queen/game of life
In-Reply-To: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <l03102807b56ec3666231@[193.78.237.143]>

At 10:44 AM +0800 15-06-2000, root wrote:
>newbie question.
>
>has anyone wrote the 8 non-attacking queen problem and conway's game of life
>in python? i try but face a few newbie problems.

There's an implementation of life that I once wrote in the demo section of
the NumPy distro.

Just





From wjdandreta at worldnet.att.net  Wed Jun 21 02:30:10 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 06:30:10 GMT
Subject: Enums and Python
References: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net> <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net>
Message-ID: <ShZ35.11603$Uw3.753134@bgtnsc04-news.ops.worldnet.att.net>

Hi Cameron,

What I am doing is reading a comma delimited text file like this

line = AB12345,20000621,3.56,...

fields = splitfields(line,',')

recognizing that

fields[Date][:5]  is the year

is more obvious and less cryptic than

fields[1][:5]

Trying to put this into a dictionary seems a contrivance to me that doesn't
make things clearer and only adds overhead.

Maybe after I get more used to using dictionaries, I'll change my mind.

I am using dictionaries somewhat but sometimes is it hard to know which data
structure is best.

I rewrote a program to day using dictionaries instead of lists which
simplified the code and speeded things up by a factor of 3 or 4. I am
beginning to appreciate their value.

Bill
Cameron Laird wrote in message ...
>I'll say that more forcefully than Aahz does here:  while
>enums are often a marker of good taste in C coding, Python
>generally achieves the same results with a considerably
>cleaner reformulation in terms of dicts.  If you feel a
>need for enums while writing Python, you're probably missing
>out on a better (and not-so-reminiscent-of-C) coding.
>
>I've become rather dogmatic on this point.  I need the
>practice of working to falsify it.  If you have a specific
>algorithm you can share that you see as ripe for a Python
>enum, please tell us here; maybe we can suggest an
>alternative.
>--
>
>Cameron Laird <claird at NeoSoft.com>
>Business:  http://www.Phaseit.net
>Personal:  http://starbase.neosoft.com/~claird/home.html




From tiddlerdeja at my-deja.com  Thu Jun  8 07:18:59 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Thu, 08 Jun 2000 11:18:59 GMT
Subject: set Registry User Security
References: <8hghbt$5bc$1@nnrp1.deja.com> <fS2%4.6572$Hz.48327@news-server.bigpond.net.au>
Message-ID: <8hnvep$nb0$1@nnrp1.deja.com>

Thanks for that Mark.

I had seen the win32security, but I was at a loss as to how to use it.

Generally speaking, are there any examples on how to use the various
parts of the win32api? Maybe I just just need to get up to seed on
windows.

Great book (PyWin32) and I hope you have fun in ActiveState.

In article <fS2%4.6572$Hz.48327 at news-server.bigpond.net.au>,
  "Mark Hammond" <mhammond at skippinet.com.au> wrote:
> The win32api module supports all the security parameters to the
win32api,
> and the pywintypes and win32security modules contain all the support
for
> manipulating the SECURITY_ATTRIBUTES and other win32 security related
> objects.
>
> Mark.
>
> <tiddlerdeja at my-deja.com> wrote in message
> news:8hghbt$5bc$1 at nnrp1.deja.com...
> > In regedt32 you can assign user privilleges for a Key.
> >
> > Can you do this from within Python, perhaps using the win32api?
> >
> >
> > Any help appreciated.
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


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


From jhauser at ifm.uni-kiel.de  Tue Jun 13 04:41:17 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 13 Jun 2000 10:41:17 +0200
Subject: Formating complex numbers
References: <slrn8kbpuf.116.hzhu@mars.localdomain>
Message-ID: <87r9a2dlci.fsf@ifm.uni-kiel.de>

As you are using a class to represent your matrices you can overload
the __repr__ and __str__ methods, to return what you want. These are
also called by the % operator. Although I do not know which one is
currently called. I think this has changed in 1.6 in some way.

HTH,
__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From R.Brodie at rl.ac.uk  Wed Jun 28 11:55:01 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Wed, 28 Jun 2000 16:55:01 +0100
Subject: MD5 - Passwords with python
References: <6e7dj8.bvb.ln@iiab.soluma.de>
Message-ID: <8jd74i$16vk@newton.cc.rl.ac.uk>

"Holger M. F??ler" <fuessler at solutec.de> wrote in message news:6e7dj8.bvb.ln at iiab.soluma.de...

> Is there a library function to encode passwords crypt-style, but with md5 as
> hashing-algorithm for python?

md5 is a standard Python library module. See the fine online documentation:
http://www.python.org/doc/current/lib/module-md5.html (or locally).





From aahz at netcom.com  Sat Jun  3 10:54:43 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jun 2000 14:54:43 GMT
Subject: String.join revisited (URGENT for 1.6)
References: <28052000.5@sanctum.jae.ddns.org> <39346FAC.F2797EA5@cosc.canterbury.ac.nz> <8h320n$6fo$1@newshost.accu.uu.nl> <03062000.1@sanctum.jae.ddns.org>
Message-ID: <8hb67j$71h$1@slb6.atl.mindspring.net>

[cc'd to Guido for his delectation upon vacation return]

In article <03062000.1 at sanctum.jae.ddns.org>,
Juergen A. Erhard <jae at ilk.de> wrote:
>
>It would seem (at least to me) to be the nicest (and cleanest) design
>to be able to do:
>
>   "some string".split().join()

That's what some people advocate, but that's semantically equivalent to
allowing [].join().  As has been pointed out several times, adding a
join() method to sequence types is a Bad Idea.

So far, nobody has pointed out *ANY* disadvantages of retaining the
current string.join() mechanism, with one exception: it destroys the
"orthogonality" of moving all string functions to string methods.  Given
that various people have demonstrated that join() has disadvantages as a
method for both string or sequence types, I think maintaining the 1.5.2
status quo is probably the best option.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From petersc at stud.ntnu.no  Fri Jun 16 09:59:30 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 16 Jun 2000 15:59:30 +0200
Subject: Python segfault
References: <20000616141903.A23985@foxboro.com> <20000616145836.O26436@xs4all.nl>
Message-ID: <394A32C2.551B0290@stud.ntnu.no>

Thomas Wouters wrote:
> 
> On Fri, Jun 16, 2000 at 02:19:03PM +0200, Hans-Joachim Widmaier wrote:
> 
> > Why not simply make 'n_nchildren' an (32 bit) int? And, while at it,
> > 'n_lineno' also? It would cost some memory, though.
> 
> Why bother ? It would do nothing to solve the real problem, which is that
> there is a finite number of direct children to a node. The current number is
> practically unreachable -- the example case of ('+'.join(['2+2'] * 8193)) is
> hardly reminiscent of real code.

Well, what about the n_lineno? Isn't that going to overflow
in more realistic situations? 

Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From offer at sgi.com  Wed Jun 28 22:26:13 2000
From: offer at sgi.com (Nhi Vanye)
Date: 29 Jun 2000 02:26:13 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <m3k8f9cqcc.fsf@greebo.nodomain.de> <slrn8ll5vu.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <8jec45$3ufed$2@fido.engr.sgi.com>

$ from NO_SPAM_hzhu at knowledgetrack_dot_com -#96739 | sed "1,$s/^/> /"
>
>
>
>Therefore I think it is desirable for python to provide a mechanism for
>programmers to describe absolute path of imports.  Like on Unix, writing
>/bin/sh in stead of sh in a script will guarantee its behavior no matter
>where it is run and what the current PATH the user has.
>
>My previous proposal was to use
>import pythonroot.AAA
>but in analogy with Unix paths maybe a syntax like
>import .AAA
>is even better.  (BTW, this is the scheme used in DNS configurations for
>distinguishing qualified and unqualified domain names.)

In another thread I suggested using 

import global.AAA

(what happens if someone wants to call a package pythonroot, we've
suddenly invented a new reserved word ?)

I'm worried that the . syntax is special-cased/it can be easily missed.

>Huaiyu Zhu                       hzhu at users.sourceforge.net
>Matrix for Python Project        http://MatPy.sourceforge.net 

richard.
-- 
Richard Offer           Widget FAQ --> http://reality.sgi.com/widgetFAQ
MTS-Core Design (Motif)
___________________________________________http://reality.sgi.com/offer



From jhsh at my-deja.com  Sat Jun 10 07:01:14 2000
From: jhsh at my-deja.com (jhsh at my-deja.com)
Date: Sat, 10 Jun 2000 11:01:14 GMT
Subject: MatPy-0.2.7 released
References: <slrn8k305t.10m.hzhu@rocket.knowledgetrack.com>
Message-ID: <8ht75n$jcu$1@nnrp1.deja.com>

In article <slrn8k305t.10m.hzhu at rocket.knowledgetrack.com>,
  hzhu at rocket.knowledgetrack.com (Huaiyu Zhu) wrote:
> I've just released MatPy-0.2.7.

Thanks for this very useful package. However, under Windows 9x the new
filename matpy.py clashes with the module name MatPy.py (Windows case
insensitivity?). After renaming matpy.py everything worked as
advertised.

Jannie Hofmeyr



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


From cut_me_out at hotmail.com  Sun Jun  4 12:54:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Jun 2000 12:54:59 -0400
Subject: Please help with python extension problem
References: <etdg0qyit9i.fsf@w20-575-109.mit.edu>
Message-ID: <etditvppesc.fsf@w20-575-109.mit.edu>

To: Dieter Maurer <dieter at handshake.de>
Subject: Re: Please help with python extension problem
FCC: ~/mail/python
--text follows this line--

> Alex <cut_me_out at hotmail.com> writes on 31 May 2000 12:22:49 -0400:

> > I have fiddled around with makesetup a bit, printing out variables, and
> > it seems as if makesetup actually thinks it's reading the line
> >
> > "tuples tuplesmodule.c*noobjects*"
> >
> > from my Setup file.  However, there is no such line.
> >
> > Here is my actual Setup file:
> >
> > '''
> > *shared*
> >
> > tuples tuplesmodule.c
> > '''
> >
> > Can anyone tell me what I ought to do, here?
**********************************************************************
> Can you add a newline at the end of the file?
**********************************************************************

Well, heavens, that did the trick!
Thanks!

> 
> 
> Dieter

Alex.


From richard_chamberlain at ntlworld.com  Mon Jun 19 14:24:57 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Mon, 19 Jun 2000 19:24:57 +0100
Subject: Three Tkinter questions
References: <20000619140251.A6860@bork>
Message-ID: <Hrt35.3455$_55.71466@news2-win.server.ntlworld.com>

Hi Egbert,

> When and why should i use Intvar() etc ?
> When and how do i use the textvariable option ?

from Tkinter import *
root=Tk()
myIntVar = IntVar()
myStringVar= StringVar()
Label(root,textvariable=myStringVar).pack(side=TOP)
def setLabel():
    myStringVar.set(myIntVar.get())
Radiobutton(root,text="First Radiobutton",variable=myIntVar,value=1,
                    command=setLabel).pack(side=LEFT)
Radiobutton(root,text="Second Radiobutton",variable=myIntVar,value=2,
                    command=setLabel).pack(side=LEFT)
root.mainloop()

I'm using both IntVar and textvariable in this example.

I create new instances of Intvar and StringVar. I then create a label and
assign the textvariable option to the instance of StringVar. I can now set
the text of this label by called myStringVar.set('Hello') or get the text by
myStringVar.get().

Next I've created two Radiobuttons and assigned their variable options to
the instance of IntVar. I give them two separate values. Any changes to the
radiobuttons will now be reflected in myIntVar.

I created a function called setLabel which uses the set and get methods of
the instances. I then assign it to the label by called
myStringVar.set(myIntVar.get())
which changes the value of the textvariable and the change is immediately
reflected on the label.

> When do you need mainloop() more than once ?

mainloop() starts the event loop up, I wouldn't use it more than once.

> Three questions only:

Three questions is quite a lot for one post ;)

Richard

Egbert Bouwman <egbert at bork.demon.nl> wrote in message
news:20000619140251.A6860 at bork...
> Three questions only:
>
> When and why should i use Intvar() etc ?
> When do you need mainloop() more than once ?
> When and how do i use the textvariable option ?
> Examples are very welcome.
> egbert
> --
> Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
> ========================================================================
>




From denis at aragne.com  Sun Jun 25 08:16:21 2000
From: denis at aragne.com (Denis =?iso-8859-1?Q?Fr=E8re?=)
Date: Sun, 25 Jun 2000 14:16:21 +0200
Subject: French Python documentation
References: <oqwvjj58f7.fsf@titan.progiciels-bpi.ca>
Message-ID: <3955F815.7CD00E6@aragne.com>

Fran?ois Pinard wrote:
> 
> (...) what exists, in French to document Python.

There is some links here :
http://www.python.org/doc/NonEnglish.html#french
Following those, you'll find other ones ...

You can also have a look at http://www.p3b.org/Python
("Under construction", but new pages added regularly).

-- 
Denis Fr?re
P3B    : Club Free-Pytho-Linuxien Carolor?gien http://www.p3b.org
Aragne : Internet - R?seaux - Formations  http://www.aragne.com



From abg at saturnsys.com  Thu Jun 15 18:52:31 2000
From: abg at saturnsys.com (abg at saturnsys.com)
Date: Thu, 15 Jun 2000 22:52:31 GMT
Subject: Problems with classes and __add__ and __mul__...
Message-ID: <8ibmn8$k5h$1@nnrp1.deja.com>

Hi all,
   I'm having a little problem with classes right now.  It's a class
with a list as it 's main feature.  I've done the slices and such so that
instance[:] and instance[x] work.  However, I can't seem to get
__add__ and __mul__ to function the way that I want them to.

I would like to do this...
def __add__(self, other_instance):
    return self.list + other_instance.list

Which returns a list.  I would like to return an object of my class type, but
it's not right obvious to me as to how to do that.

Or, would it be better to do this?  (spam is the class here...)
x = spam()
y = spam()
z = spam()
z[:] = x + y

or should I just define an append function for the class?

(What I really want to do is this:
z = spam()
x = spam()
z = z+x

Thanks for any help that you can give me.  It would be great if you emailed
your responses to a_gurno at yahoo.com as I don't always get to check this group
as often as I would like. Adam


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


From db3l at fitlinxx.com  Fri Jun 23 22:18:00 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 23 Jun 2000 22:18:00 -0400
Subject: simple question...
References: <3953e13a.761314922@news1.on.sympatico.ca> <3954056E.E3C18DF@bioeng.ucsd.edu> <395405c4.770668532@news1.on.sympatico.ca>
Message-ID: <u1z1nstyf.fsf@ctwd0143.fitlinxx.com>

chibaA at TinterlogD.Tcom writes:

> At the end of a script A, I would like to initiate script B.  I don't
> want to run script B as a function in A; rather, I want to quit A, and
> start script B.  Is this possible?

Try the os.exec* functions in the os module.  It sounds pretty much
like what you want to do.  The exec* functions replace the current
running executable with another executable - only one process slot is
involved, and the original executing code does not exist following the
exec call.

Note however that since you'll be replacing your current process, you
actually need to exec python giving it the new script as an argument,
rather than just directly-exec()ing the script.

You can use sys.executable to determine the python that is executing
the current script and just exec() that with an argument of the new
script.

You don't note your platform, but I believe this is documented for
both Windows and Unix.

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


From weil at sireconnect.de  Thu Jun  8 15:09:53 2000
From: weil at sireconnect.de (Weil)
Date: Thu, 8 Jun 2000 21:09:53 +0200
Subject: TypeError: unsubscriptable object
References: <393ECFC3.879BC357@mordor.ch> <8hmp6i$ets$6@newshost.accu.uu.nl> <393F488B.37F13682@mordor.ch>
Message-ID: <393feff5$1@news2.sireco.de>

Cyrille Artho <cartho at mordor.ch> schrieb in im Newsbeitrag:
393F488B.37F13682 at mordor.ch...
> > What's possible (but it's not possible to determine this from your
snippets)
> > is that you accidentally assign something else to W elsewhere.
> self.W is equal to
> [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
> just before the assignment.
> self.W[i][j] = self.W[i][j]  + ...
> I attached the entire script to this post (it's not too big).
> --
> Regards,
> Cyrille Artho - http://artho.com/ - Tel. +41 - [0]1 - 313 08 92
> Alcohol is the anesthesia by which we endure the operation of life.
>                 -- George Bernard Shaw

Well, I think it's not W that is unsubscriptable, but P.
First you do this:
>         P = copy.deepcopy(I)
Later this:
>         P = 1 / (1 + math.exp(-sum))
which makes P unsubscriptable here:
>                 self.W[i][j] = self.W[i][j]  + n * (
>                     r(S[i] - P[i]) - alpha * (1 - r)
>                     * (S[i] - (1 - P[i]))) * I[j]
but pretty usable here:
>             self.V[i] = self.V[i] + n * (
>                 r(O - P) - alpha * (1 - r) * (O - (1 - P))) * S[i]

...and by the way, r(S[i] - P[i]) won't work either,
maybe you mean: r *(S[i] - P[i]).














From Linus_Chang1 at bigpond.com  Sun Jun 18 03:28:58 2000
From: Linus_Chang1 at bigpond.com (Linus Chang)
Date: Sun, 18 Jun 2000 17:28:58 +1000
Subject: win32gui question - is my window the foreground window?
References: <39437A27.1020403@bigpond.com> <fUB25.699$qz4.6155@news-server.bigpond.net.au>
Message-ID: <394C7A3A.7070906@bigpond.com>

Thanks a heap :)



"Mark Hammond" wrote:

> "Linus Chang" <Linus_Chang1 at bigpond.com> wrote in message
> news:39437A27.1020403 at bigpond.com...
> 
> > However, win32gui only gives me SetForegroundWindow, not Get...
> 
> The next win32gui has all these.  In fact, it appears from my CVS logs
> that these are in win32ll-132.
> 
> Mark.
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
> 




From soerense at verwaltung.uni-mainz.de  Thu Jun 29 06:10:52 2000
From: soerense at verwaltung.uni-mainz.de (Ruediger Soerensen)
Date: 29 Jun 2000 12:10:52 +0200
Subject: informixdb
Message-ID: <ytwaeg4sspf.fsf@harriet.verwaltung.uni-mainz.de>

I tried to build and install informixdb 1.3 for SunOS 5.7.
Well it ... kind of worked.
When I try to import informixdb.py, I get the error

"ImportError: Cannot locate C core module for informix db interface."

_informixdb.so is installed, together with informixdb.py, in the
site-packages dir. What went wrong?

Furthermore, is there any documentation available on how to use the
python DB-API?

Thanks a lot, R?diger

-- 
R?diger "Black Hole" S?rensen || soerense at verwaltung.uni-mainz.de


From lee.list at joramo.com  Mon Jun 26 13:58:01 2000
From: lee.list at joramo.com (Lee Joramo)
Date: Mon, 26 Jun 2000 11:58:01 -0600
Subject: working on files as a different user and group
Message-ID: <20000626175801.2434@gj.net>

I am creating a python CGI script that will assist in the management of
files on a Linux/Apache webserver. Some of the functions that the script
needs to preform include changing the user,  group and permissions for a
given file or directory. 

Is there anyway to do this without compromising security? (If this can't
be done via the web, then I may have to provide additional users with
access to the command line which I dread doing.)

Thanks.

--
Lee A. Joramo                      ljoramo at nickads.com
The Nickel Want Ads                www.nickads.com
Internet Manager                   970-242-5555




 




From karl-sve at dsv.su.se  Thu Jun 15 16:55:45 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Thu, 15 Jun 2000 22:55:45 +0200 (MET DST)
Subject: Which linux distros have Python by default?
In-Reply-To: <20000614130531.A11087@better.net>
Message-ID: <Pine.GSO.4.20.0006152249500.10224-100000@luna.dsv.su.se>

On Wed, 14 Jun 2000, William Park wrote:

> On Wed, Jun 14, 2000 at 10:17:25AM +0200, Kalle Svensson wrote:
> > 
> > It isn't default on Slackware, but I think this isn't a very big
> > problem (for you). Most slack users would know how to install it, and
> > it's not a very big distro anyway.
> 
> Slackware-7.0 has Python package in D series.

Yes, I know, but it isn't installed by default, at least it isn't marked
by default in menu install mode, which is all I ever use.
By the way, Slackware's installer must be the best I've ever seen in a
Linux distro. All those graphical installers are crap by comparison!
Had the rest of Slack been as good, I'd still be using it...

Regards,
  Kalle Svensson




From hzhu at rocket.knowledgetrack.com  Fri Jun  9 13:31:06 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 09 Jun 2000 17:31:06 GMT
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <20000608122619.I469@xs4all.nl> <8hogq2$cce$1@slb6.atl.mindspring.net> <20000608220032.B20700@xs4all.nl>
Message-ID: <slrn8k2akl.pi.hzhu@rocket.knowledgetrack.com>

On Thu, 8 Jun 2000 22:00:32 +0200, Thomas Wouters <thomas at xs4all.net> wrote:
>
>> Would there be a problem with the ambiguity of "3./2" being parsed
>> as either
>>   '3.' '/' '2' == 1.5
>>   '3' './' '2' == whatever ./ does to scalar integers, probably
>>                   make it be regular division == 1
>>   ?
>
>I'm afraid I was a bit careless in the quoting. I am not implementing the
>'dot operators' for numerical computation, I'm only implementing augmented
>assignment (+=, >>= and friends.)  But for what it's worth, I think the 'dot
>operators' as using a dot are out -- there is no way the parser can
>distingish between float-creation, attribute-lookup and matrix-operations,
>and still remain as obvious as it is now. At least, I dont think so.  Using
>another character would be possible, of course ;)
>

Then could somebody else help to implement the dot operators?  It doesn't
matter whichever character is used in place of dot, although looking at my
keyboard I don't see any better candidates - may be due to the matlab effect.

The potential ambiguity can only come from digits, not letters:

Expression   Can be      Can't be     Better written
a./b         a ./ b       a. / b
2./3         2 ./ 3                   2 ./3
             2. / 3                   2.0/3
                          2. ./ 3     2.0./3

Here's what octave does
octave:1> a = [1, 2; 3, 4]
a =
  1  2
  3  4
octave:2> 3./a
ans =
  3.00000  1.50000
  1.00000  0.75000
octave:3> 3. /a
error: operator /: nonconformant arguments (op1 is 1x1, op2 is 2x2)

This is quite reasonable, which many people have lived with for years.  On
the other hand, due to the distinction between int and float in python, and
lack of matrix in core language, it might be better to let (2./a == 2. / a).
No exising code will be broken.  A greedy parser would handle this.

Huaiyu


From robin at jessikat.fsnet.co.uk  Wed Jun  7 03:41:27 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 7 Jun 2000 08:41:27 +0100
Subject: Nympy for win
References: <Pine.LNX.4.21.0006061109170.30720-100000@chelin.chtf.stuba.sk>
 <fAdMjhAnqNP5Ew3P@jessikat.demon.co.uk> <m3aegy60g1.fsf@brant.geog.ubc.ca>
Message-ID: <NS9onJAnyfP5EwVH@jessikat.demon.co.uk>

In article <m3aegy60g1.fsf at brant.geog.ubc.ca>, Phil Austin
<phil at geog.ubc.ca> writes
>Robin Becker <robin at jessikat.fsnet.co.uk> writes:
>
>
>> >I would like to use the Numpy package with my windoze/python1.5.2 system,
>> >but it needs distutils and distutils need m$ visual c++ which I don't own.
>> >Is here anybody who has it compiled and could send me the
>> >binaries? Thanks.
>
>The binary file at http://download.sourceforge.net/numpy/python-numpy-15.2.zip
>works for me with python 1.5.2.  Unfortunately, the new sourceforge site
>hasn't made it onto the numerical topics webpage at:
>
>http://www.python.org/topics/scicomp/numpy_download.html
>
>and the LLNL archive no longer exists.  Is there someone besides 
>webmaster at python.org
>who should get this bug report?
>
>Regards, Phil

I hadn't realised that the .zip contained a binary. I always just
download the .tgz and compile. Sourceforge is ok for cvs, but is a pig
for the web side development.
-- 
Robin Becker


From pinard at iro.umontreal.ca  Tue Jun 13 22:29:28 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 13 Jun 2000 22:29:28 -0400
Subject: How do you FTP ASCII if ftplib strips CR?
In-Reply-To: echuck@mindspring.com's message of "Wed, 14 Jun 2000 01:06:57 GMT"
References: <8i5jhd+h2lu@eGroups.com> <uog55mk6r.fsf@ctwd0143.fitlinxx.com>
	<8i6lr7$s2o$1@nnrp1.deja.com>
Message-ID: <oqvgzdq9kn.fsf@titan.progiciels-bpi.ca>

[echuck at mindspring.com]

> By losing access to the file object returned by open(), it gets gc'ed
> and therefore closed and flushed immediately.  You only have to worry
> about standalone open() calls if you use JPython or think you will someday.

That property of Python allows for _so_ nice code simplification, and
increase in simplicity and legibility, that we cannot just ignore it.
I'm not myself asserting that the Python reference does not specify this
behaviour, but someone else on this list did :-).  Yet, I would much like
being able to say that the Python reference does specify it, as I would
feel less guilty relying so much on it. :-)

The price to pay, and I know it, is that if I ever go to JPython, I will
have to carefully revise many things.  I'm ready to risk paying that price.

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





From gansevle at cs.utwente.nl  Fri Jun 30 03:56:12 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Fri, 30 Jun 2000 07:56:12 GMT
Subject: Constants and Enums
References: <Pine.GHP.4.21.0006291234020.9839-100000@don.et.byu.edu>
Message-ID: <8jhjqo$pab$1@nnrp1.deja.com>

Recentely I posted the Enum module
It is now at http://www.cs.utwente.nl/~gansevle/python/Enum.py
With that you can do the following:

   import Enum
   color = Enum.enum('red', 'yellow', 'green', 'blue', green=20)
   print color
   print color.yellow
   print color(21)
   for col in color:
       print color(col), '=', col

It may be a starting point, or just wat you need.

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From root at rainerdeyke.com  Fri Jun  9 18:39:29 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 09 Jun 2000 22:39:29 GMT
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net> <8hrdkn$3uf@stochastic.eng.umd.edu>
Message-ID: <Bme05.1164$vp3.76577@news1.rdc2.tx.home.com>

Warren B. Focke <moron at Glue.umd.edu> wrote in message
news:8hrdkn$3uf at stochastic.eng.umd.edu...
> But it does.  Accessing an attribute of an object or an item in a
> container type can invoke arbitrary code, with arbitrarily freakish
> side effects.  Evaluating x[17] might suck up hours of CPU time and
> gibibytes of RAM, redefine builtin functions, launch the missiles, or
> annoy your mother.  Having all of these things happen once, as in
> x[17] += 1, is much preferable to, and potentially significantly
> different from, having them happen twice, as in x[17] = x[17] + 1.
> It's not just syntactic sugar that

What if x is a database that loads an object from file on __getitem__ and
writes it on __setitem__?  Then x[17] += 1 would work correctly only if
x[17] has no += operation defined and the interpreter falls back on
executing x[17] = x[17] + 1.


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




From siegfried.gonzi at kfunigraz.ac.at  Thu Jun  1 07:16:00 2000
From: siegfried.gonzi at kfunigraz.ac.at (Siegfried Gonzi)
Date: Thu, 01 Jun 2000 13:16:00 +0200
Subject: Stand-alones on the Mac?
Message-ID: <393645EF.17424CDA@kfunigraz.ac.at>

Dear Python (Mac) specialist,


I have developed some simulations (concerning optics diffraction) with
Yorick on the Mac. Now, I would like to make a shareware for amateur
astronomers concerning my simulations.

Someone told me that this could be possible with Python. I need the
numerical library (for the fast Fourier transforms) and a library where
I can make a hardcopy from the image (which I get after transforming) in
Postscript or GIF (is such one available for Python?) and maybe a
library for plotting x versus y diagrams, and a library for a very
simple graphical user interface.

Would it be then possoble to make a stand-alone application for the Mac?
Or should the end-user having installed every time Python?

I found nothing abaut that topic at www.python.org and the FAQs therin.
What I have found was something vague about the command freeze;but
nothing in detail. And I isntalled also the Mac version of Python and
also there I could not found anything about Python and stan-alones for
the Mac.

Because it would be impossible to make a stand-alone program with Python
for the Mac, so it is not my thing to port code to Python, because I can
further use Yorick (there it is also not possibility to make
stand-alones).


Regards and thanks for the answers,
Siegfried Gonzi



From matt at nightrealms.com  Mon Jun 26 22:23:30 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Mon, 26 Jun 2000 18:23:30 -0800
Subject: Is this safe? (Making string comparisons ignore case)
Message-ID: <NqT55.6417$Gh6.421246@news-west.usenetserver.com>

I've done the following to make it so that all string comparisons
are case insenstive.  Is this safe?  (I'm pleasently surprised that
it actually works)

Thanks in advance.

================================================================
import string

string.__builtins__['oldcmp'] = string.__builtins__['cmp']

oldCmp = string.__builtins__['oldcmp']


def myCmp(self, other):
	return oldCmp(string.lower(self), string.lower(other))

string.__builtins__['cmp'] = myCmp

print cmp("AAAA", "aaaa")






From cjensen at bioeng.ucsd.edu  Fri Jun 30 14:20:58 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Fri, 30 Jun 2000 11:20:58 -0700
Subject: Python MP3 Player
Message-ID: <395CE50A.3BA16777@bioeng.ucsd.edu>

Does anyone know of a Python MP3 Player?

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


From vigenereNOviSPAM at leland.stanford.edu.invalid  Mon Jun  5 04:54:26 2000
From: vigenereNOviSPAM at leland.stanford.edu.invalid (cheesecake)
Date: Mon, 05 Jun 2000 01:54:26 -0700
Subject: file.seek misbehaving?
References: <203da3d8.c492c8fd@usw-ex0103-019.remarq.com> <Lyc_4.33464$WS3.397965@typhoon.we.rr.com> <067d003a.25f504d0@usw-ex0104-028.remarq.com>
Message-ID: <27d81e39.85cb294f@usw-ex0103-019.remarq.com>

"rb" mode solved the problem. It now behaves as expected.

Also, David and Gordon were right about the eoln characters. I
thought I had checked to make sure both files I was comparing
had the same \n terminating character, but I must have checked
wrong. The Windows files was indeed using a \r\n combination,
which explains the difference in file size.

TD

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



From darrell at dorb.com  Sun Jun 18 18:55:30 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 18 Jun 2000 15:55:30 -0700
Subject: Circular import
References: <394D202B.1DF60864@colconsulting.com>
Message-ID: <05a701bfd978$4fdb7ab0$6401a8c0@home>

From: "Frank V. Castellucci" <frankc at colconsulting.com>
>
> With the assumption that class A and class B reference each other, how
> do you avoid the Python error with:
>
> ---------- Composite.py -----------
>
> from Leaf import Leaf
>
> class Composite:
> def __init__( self ):
> self.__theLeaf = Leaf(self)
>
> -------- Leaf.py --------------
>
> from Composite import Composite
>
> class Leaf:
> def __init__( self, aComposite ):
> self.__theComposite = aComposit
>
>

I'm guessing you come from the land of C++, where the compiler wants to know
every detail.
Don't bother with this:
> from Composite import Composite

Leaf can accept any kind of object as the composite.
What ever methods the Composite has will be accessible.
It's a beautiful thing once you get the hang of it.

--Darrell Gallion






From jilanik at tin.it  Sat Jun  3 14:14:20 2000
From: jilanik at tin.it (Jilani Khaldi)
Date: Sat, 03 Jun 2000 20:14:20 +0200
Subject: medusa python and CGI
Message-ID: <39394AFB.DB75750C@tin.it>

Hi All,
I am using using the web server medusa and python to write some CGI.
But, if there are 10.000 requests for a python script to be executed I
will have 10.000 instances of python (the interpreter) persistent even
after the user has finshed with that script. Now, I wonder how to kill a
python instance once the script has finished its work (by code, or by a
trick)?

Thnaks.

Jilani

--
<jk>
___________________________________________________
http://space.tin.it/scuola/jkhaldi
FreeBSD ~ The Power to Serve





From pete at visionart.com  Thu Jun  8 12:58:47 2000
From: pete at visionart.com (Pete Shinners)
Date: Thu, 08 Jun 2000 09:58:47 -0700
Subject: wishlist Python3K: subclass ints
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com> <393EFF38.C023F10F@san.rr.com>
Message-ID: <393FD0C7.D70479A2@visionart.com>

> > 1. Everything is an object of a class that can be subclassed.
> 
> You want to subclass integers, why?

i think this is really important also. i'd like to see all
python objects given the same 'value'

a small sampling of reasons to make all types (yes, even ints)
real class types that can be subclassed

 - "Debug Int" class. want to make an int that prints to
   a log every time it gets assigned? it would be incredibly
   easy if int were its own class. instead now one must
   create a new class and define --ALL-- the different operators
   for an int. what happens when python with the assignment
   modifiers (*=, +=, etc) is released. if int is a class
   your debug Int class works perfectly. with the current
   "unique" int type, you have to go back to your code and
   add the new features (all basic OO arguments apply i guess)
   Along with just "int". How is "int" any less valuable than
   dict or list, which already have their handholding "User..."
   classes

 - (speaking of which) It gets rid of the UserList, UserDict,
   and all its ugly friends. these are the results of a
   flawed design.

 - Just the fact that i could subclass any "C" object would
   make it worth the while.

 - Also gone are all the nit-picking type checking routines.
   instead of first checking if a variable is a "class" type,
   then checking if the object is a certain class, all variables
   hold objects of the same "rank". the best move would be to
   have "int" and "float" be subclassed from a base class like
   "realnumber".

 - Instead of using functions like int(), float(), string()
   why not make int, float, and string a class definition.
   then calling int(10.5) is really calling the constructor
   for the int object.

i'm sure there's a lot more, but in my mind having 
one standard "type" for all python objects would be a big
boost. 

don't get me wrong here. even with all of python's unique
and different object types i think it is still the greatest
language. i just can't think of any good arguments for
py3k to once again segregate objects into different types.


From Kevin.Digweed at dial.pipex.com  Mon Jun 19 11:32:19 2000
From: Kevin.Digweed at dial.pipex.com (Kevin Digweed)
Date: Mon, 19 Jun 2000 16:32:19 +0100
Subject: split() with a separator set
References: <xjvr99tyac7.fsf@gjallarhorn.ifi.uio.no>
Message-ID: <394E3D03.E3F89D98@dial.pipex.com>


"Igor V. Rafienko" wrote:
> 
> Hi,
> 
> I've just started learning Python, so, please bear with me.
> 
> string.split() function takes (amongst other things) a separator
> argument, and the given string is split according to this separator.
> But what if I have several one-character separators? Obviously:
> 
> # input looks like this: 'foo:bar|baz/zot'
> # I want [ 'foo', 'bar', 'baz', 'zot' ]
> string.split( src, ":|/" )
> 
> doesn't work.

Try mapping the source characters in the set to a common character (the
first in the set), then splitting using that single character:

import string
def setsplit(src, sset=string.whitespace):
    return string.split(string.translate(src, string.maketrans(sset,
sset[0]*len(sset))), sset[0])

.. I realise that sset[0] characters get 'translated' to sset[0]
characters, but I suspect that the implementation of CPython
string.translate (I haven't verified it) means that it's free anyway.

Or ... use 're'.

Kev.


From emile at fenx.com  Sat Jun  3 09:02:25 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 3 Jun 2000 06:02:25 -0700
Subject: Styling Temporary Code
References: <3936A9BB.FBEF9201@visionart.com> <8h6lgi$8me$1@slb0.atl.mindspring.net>
Message-ID: <079401bfcd5c$00f36660$1906a8c0@fc.fenx.com>

Note that if you put additional tests on the
if __debug__ line, everything still gets
optimized out, allowing related debug code
to be effectively turned on or off.

debugWatchList = ["Mod1","Mod2","NewMod"]

if __debug__ and "ThisModName" in debugWatchList:
  print "InDebugCode"

---or---

class Test:
 def __init__(self, watchlist):
  self.data = watchlist
 def __call__(self, target):
  return target in self.data

debugWatch = Test(["Mod1", "Mod2", "NewMod"])

if __debug__ and debugWatch("NewMod"):
 print "InDebugCode"


Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: Aahz Maruch <aahz at netcom.com>

<snip>

> One person already suggested that you use assert.  Note
that in Python
> 1.5.2 and later, you can turn off assert by setting the
PYTHONOPTIMIZE
> variable in addition to using -O on the command line.  You
can also
> express more complicated code with
>
> if __debug__:
>   ....
>
> Both assert and "if __debug__" get optimized out with
optimization
> turned on.
> --
>                       --- Aahz (Copyright 2000 by
aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> Have you coined a word today?  --Aahz
> --
> http://www.python.org/mailman/listinfo/python-list
>




From locate at locate-now.com  Thu Jun 22 09:40:03 2000
From: locate at locate-now.com (locate at locate-now.com)
Date: 22 Jun 2000 06:40:03 -0700
Subject: Internet & scripting
Message-ID: <484CEB49.35D6FA41.097fc1412cf9c5afbde37357f527f4e2@locate-now.com>

For you programmers or those who are interested in computer
langage. I found while surfing the internet a source that explains, html,
pearl, asp, python, smil,javascrip, etc. 

Check At: http://www.locate-now.com/adcomp1.htm 
click on "internet and scripting"

From hjwidmai at foxboro.com  Fri Jun 16 09:07:29 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Fri, 16 Jun 2000 15:07:29 +0200
Subject: Python segfault
In-Reply-To: <20000616145836.O26436@xs4all.nl>; from thomas@xs4all.net on Fri, Jun 16, 2000 at 02:58:37PM +0200
References: <20000616141903.A23985@foxboro.com> <20000616145836.O26436@xs4all.nl>
Message-ID: <20000616150729.A23251@foxboro.com>

Thomas Wouters:
> 
> > Why not simply make 'n_nchildren' an (32 bit) int? And, while at it,
> > 'n_lineno' also? It would cost some memory, though.
> 
> Why bother ? It would do nothing to solve the real problem, which is that
> there is a finite number of direct children to a node. The current number is
> practically unreachable -- the example case of ('+'.join(['2+2'] * 8193)) is
> hardly reminiscent of real code.

"Practically unreachable." But this number is, as far as I could draw from
your expanation, determined by that short. My point is: just raise this limit
so you run out of memory before ever reaching it ...

> Fixing the parser so it doesn't crash, and instead says 'add a few
> parentheses' would be the better solution, IMHO ;)

... which would make this overflow test obsolete, thus making a
better solution, IMHO. ;-)

Well, I'm not really voting, as I don't really know the innards, and I
have nothing to say. I'm just a fan of the "do not set a limit on
anything if there's no real need for it."

Anyway, having an error message is way better than just crashing.

-- 
Hans-Joachim Widmaier             E-Mail:  hjwidmaier at web.de



From riku.voipio at iki.fi.invalid  Tue Jun 27 08:08:47 2000
From: riku.voipio at iki.fi.invalid (Riku Voipio)
Date: Tue, 27 Jun 2000 15:08:47 +0300
Subject: Compiling options for python
Message-ID: <8ja5t6$dhr$1@news.tsd.tieto.com>

Hi, 

I'm running Zope in a intra/extranet enviroinment On Debian/Gnu Linux 2.2 , 
and it  seems that most time is waisted in python. I guess the cheapest way 
to improve perfomance would be recompiling python with more aggressive
options. What options can be passed safely to ./configure? 
It has gcc 2.95.2, which AFAIK has egcc stuff merged, so should I be able 
to use pentium optimisations (Yes, I know that they are unnecessary on >Pent. 
pro, but this is a vanilla pentium)?


From istevens at calum.csclub.uwaterloo.ca  Mon Jun  5 21:22:14 2000
From: istevens at calum.csclub.uwaterloo.ca (ian stevens)
Date: 6 Jun 2000 01:22:14 GMT
Subject: Network byte ordering question ...
Message-ID: <8hhjo6$8ev$1@watserv3.uwaterloo.ca>

I am writing a Python application which makes use of a protocol which
requires information to be in network byte order.  Specifically, a
message header takes on the following form:

    bytes   type

    0-15    Message ID ... all numbers
    16      Numeric value
    17      Numeric value
    18      Numeric value
    19-22   Numeric value

What is the best way to extricate the above information and convert it
to the byte ordering of the host machine?  To convert a header in the
byte ordering of the host machine to network byte ordering?  How do
I determine the type of machine the program is running on and, hence,
its byte ordering?

I've had a look the pack() and unpack() commands in the struct module
but without much success at getting them to do what I want.  I am new
to Python and it's been a while since I performed this type of operation
in C or C++, so I'm a little rusty.

FYI, here's what I've tried:

    ( val1, val2, val3, val4, val5 ) = unpack( "@16s1s1s1s4s", header )

This works somewhat but I would really like val1 to read as a string
containing numeric values (to preserve leading zeroes) and the other
values to read as numbers.  This is also specific to little-endian
machines and I'd like to generalize it.

Any help or references would be greatly appreciated.

thanks,
ian.


From erno at iki.fi  Mon Jun 26 16:56:13 2000
From: erno at iki.fi (Erno Kuusela)
Date: 26 Jun 2000 23:56:13 +0300
Subject: Garbage collection versus reference counting
References: <lqI55.485$U3.22441@news.kpnqwest.fi>
Message-ID: <kuituwupoy.fsf@lasipalatsi.fi>

>>>>> "Panu" == Panu A Kalliokoski <panu at ws23.pca.fi> writes:

    Panu> Is there something I've missed here?

nope; unless i've misunderstood, a hybrid approach is on its way.
read the python-dev archives for more information.

  -- erno



From clarence at netlojix.com  Tue Jun  6 12:39:44 2000
From: clarence at netlojix.com (Clarence Gardner)
Date: Tue, 6 Jun 2000 09:39:44 -0700
Subject: What's in a name?
References: <Pine.SOL.4.05.10005241201170.14111-100000@spgsun1.arlut.utexas.edu> <392c17d3$0$17863@wodc7nh7.news.uu.net> <Pine.SOL.4.05.10005241328080.14321-100000@spgsun1.arlut.utexas.edu> <392C3738.58C41202@muc.das-werk.de> <392CB840.1E5E5B06@teleatlas.com> <392CBAB5.37BFCE21@san.rr.com> <393d24ee.0@news.proweb.co.uk>
Message-ID: <960310094.1016832905@news.silcom.com>

>"Courageous" <jkraska1 at san.rr.com> wrote in message
>news:392CBAB5.37BFCE21 at san.rr.com...
>> Can you name some situations
>> in reading and writing english (heh) where case is essential to being
>> understood?

I'm feeling silly:

    Did you miss Smith?     (Did you fail to meet a person named Smith?)
    Did you Miss Smith?    (I won't translate this one; put on your Old
                                     English hat and accent the first word. 
                                     "Didst thou Miss Smith?")

Hey, for that matter:

    Put on your old English hat.
    Put on your Old English hat.


From collins at manhattan.aero.org  Wed Jun  7 11:38:22 2000
From: collins at manhattan.aero.org (Jeff Collins)
Date: Wed, 7 Jun 2000 08:38:22 -0700
Subject: Palm Python?
In-Reply-To: <m3zooxhm9k.fsf@chinon.cnrs-orleans.fr>
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl>
	<14649.14534.184930.984180@malibu.aero.org>
	<FvLsJJ.CAC.0.bloor@torfree.net>
	<3939FFB4.A7824367@mindspring.com>
	<da1eh8.r4g.ln@demonseed.net>
	<m3zooxhm9k.fsf@chinon.cnrs-orleans.fr>
Message-ID: <14654.27680.192526.650115@malibu.aero.org>

Konrad Hinsen writes:
 > Jason Petrone <jp at NOSPAMdemonseed.net> writes:
 > 
 > > I've found the Scheme implementation LispMe(http://www.lispme.de) to be
 > > usable onboard a palm.  It is very complete and has plenty of support for
 > > things such as graphics and sound.  However, as much as I hate typing all 
 > > those parentheses in Scheme on a keyboard, it is 10x worse in graffiti.
 > 
 > In my experience LispMe is the best on-board development system as far
 > as system integration is concerned. The only drawback is that it's
 > Lisp! I'd prefer Forth as a language, but development with
 > QuartusForth is just too messy; every time you start the interpreter
 > you have to reload all your definitions. And of course I'd *really*
 > prefer Python...
 > 
 > Which brings me back to Palm Python. Would it be possible to borrow
 > some of the Palm interface code from LispMe, which recently became
 > Open Source? That could save time and lead to a more uniform API among
 > Palm languages.

Thanks for the pointer!  I will definitely consider this!

Jeff

 > -- 
 > -------------------------------------------------------------------------------
 > Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
 > Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
 > Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
 > 45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
 > France                                   | Nederlands/Francais
 > -------------------------------------------------------------------------------
 > -- 
 > http://www.python.org/mailman/listinfo/python-list
 > 



From speed at ?.com  Thu Jun 15 10:11:03 2000
From: speed at ?.com (Jake Speed)
Date: Thu, 15 Jun 2000 14:11:03 GMT
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <8hs4d8$33b$1@animus.fel.iae.nl> <20000610120811.H20700@xs4all.nl> <8ianhh$dro$3@newshost.accu.uu.nl>
Message-ID: <skhovn87is473@corp.supernews.com>

m.faassen at vet.uu.nl (Martijn Faassen) wrote in 
<8ianhh$dro$3 at newshost.accu.uu.nl>:

>Thomas Wouters <thomas at xs4all.net> wrote:
>[discussion on naming]
>
>Since there's a discussion about naming anyway, I'd like to ask what
>the 'ad' stands for in __add_ad__? This may be another datapoint
>in the discussion. :)

"and becomes", from Algol-68.

http://x58.deja.com/getdoc.xp?AN=629676074

Apparently the 'b' got flipped :^)

-Speed!


From kstar18 at hotmail.com  Wed Jun 21 12:26:38 2000
From: kstar18 at hotmail.com (kstar18 at hotmail.com)
Date: Wed, 21 Jun 2000 16:26:38 GMT
Subject: Disabling Comboboxes
Message-ID: <8iqqb5$d4s$1@nnrp1.deja.com>

Hi, I'm new to python and am experimenting with Pmw and Tkinter
applications.  Can anyone tell me how I would go about disabling a pmw
combobox?
I tried disabling the 'entry' field separately, and then changing the
bindings on the arrow button...but that generated some errors.  It works
sometimes, and not at others.  I was wondering if there was a simpler
solution.

Thanks

Kapil


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


From ejr at lotus.CS.Berkeley.EDU  Fri Jun 16 13:50:31 2000
From: ejr at lotus.CS.Berkeley.EDU (Edward Jason Riedy)
Date: 16 Jun 2000 17:50:31 GMT
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru> <8ibqbk$o8j$1@agate.berkeley.edu> <394A229D.A8998955@darmstadt.gmd.de>
Message-ID: <8idpd7$sgl$1@agate.berkeley.edu>

And Robb Shecter writes:
 - 
 - I'll take another look at LWP, but I can imagine the library being 
 - even easier to use than the sample code you gave.

Just take care not to lock out the more complicated examples.  LWP
and closures / bound methods allow very efficient, event-driven 
systems.  That's what I'm using to write a very low-system-impact 
webmail scraper.  I don't need to store the whole page, just stream 
it through and take what I want.  

If I could combine this with HTTP/1.1 and a streaming XPointer 
implementation, life would be good.  I'm fairly language-agnostic 
(so long as it's dynamically typed), so I'll use whichever one gives 
this to me.  ;)  I'm writing this in spare time, so I'm not about 
to implement all the libraries I need, too.

Jason


From rumjuggler at cryptarchy.org  Thu Jun  8 00:12:42 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Thu, 08 Jun 2000 04:12:42 GMT
Subject: abuse of lambda question
References: <stgrjs0sbt74kb3m4b7dbkkc87910fl1r6@4ax.com>
Message-ID: <fo6ujsofinc9n1ea2dtk96lrpuijje4j2t@4ax.com>

On Wed, 07 Jun 2000 03:49:31 GMT, Ben Wolfson <rumjuggler at cryptarchy.org>
wrote:
[stuff]

Ok, I got it working now.  The complete lambdabuse.py file, so far, is:

"""
Functions to aid in grossly abusing lambdas.  This will greatly help Python
develop as a langauge.
For example, if you wanted to get every other element of a list in two
different lists, you *could*
write something nice and clear, like:
def everyother(list):
	evens, odds= [],[]
	putineven = 1
	for e in list:
		if putineven:
			evens.append(e)
			putineven = 0
		else:
			odds.append(e)
			putineven = 1
	return evens, odds

which can be called as everyother(range(20)), or you could write something
completely unreadable,
like:
everyother = lambda l, evens, odds, app=lambda a,e: lambda
a=a,e=e:a.append(e): Compose(Apply(map, lambda e, evens=list(evens),
odds=list(odds), l=l,app=app: If(l.index(e)%2==0, app(evens,e),
app(odds,e)), l), Return(None, evens, odds))

which *must* be called as everyother(range(20),[],[]).
The choice is obvious.
"""

def For(count, fct, *args, **kwargs):
	for i in range(count):
		fct(i,*args,**kwargs)

def While(cond, fct, *args, **kwargs):
	while cond():
		fct(*args, **kwargs)

def Compose(*args):
	for f in args[:-1]:
		f()
	return args[-1]()

def Return(index, *args):
	if index is not None:
		return lambda a=args[index]: a
	return lambda a=args:a
	

def Apply(f, *args, **kwargs):
	return lambda f=f, a=args, k=kwargs: apply(f,a,k)

def If(cond, t, f):
	if callable(cond):
		cond = cond()
	if cond:
		return t()
	return f()

__everyother = lambda l, evens, odds, app=lambda a,e:\
  lambda a=a,e=e:a.append(e): Compose(
    Apply(map, \
      lambda e, evens=list(evens), odds=list(odds), l=l,app=app: \
        If(l.index(e)%2==0, app(evens,e), app(odds,e)), l),\
    Return(None, evens, odds))


def Elif(cond, t, f, *cond_elifs):
	conditions, eliffuncs = __everyother(list(cond_elifs),[],[])
                                     #must be a list to use .index()
	if callable(cond):
		cond=cond()
	if cond:
		return t()
	else:
		for i in range(len(conditions)):
			if conditions[i]():
				return eliffuncs[i]()
	return f()

def Map(f, *args):
	return lambda f=f, args=args: map(f, *args)

def Print(a):
	def p(a=a):
		print a
	return p

##an example.  For a group of lists l1...ln, 
##__unmapNone(map(None, l1,...,ln)) == (l1,...,ln)

__unmapNone=lambda l, result: Compose(
  Apply(For, len(l), lambda c, a: a.append([]), result),
  Map(lambda a,r=result: For(len(a), \
    lambda c,a,r: r[c].append(a[c]), a, r),l),
  Return(0,result))

-- 
Barnabas T. Rumjuggler

So witless did these ideas strike me as being, so sweeping and pompous the
way they were expressed, that I associated them immediately with
literature.
 -- Jorge Luis Borges, "The Aleph"


From david_ullrich at my-deja.com  Wed Jun 14 13:17:50 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 14 Jun 2000 17:17:50 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i34pv$2ruu$1@nntp6.u.washington.edu> <8i5p73$6eo$1@nnrp1.deja.com> <20000613202114.B7817@bigskytel.com>
Message-ID: <8i8enb$5qp$1@nnrp1.deja.com>

In article <20000613202114.B7817 at bigskytel.com>,
  David Porter <jcm at bigskytel.com> wrote:
> * david_ullrich at my-deja.com <david_ullrich at my-deja.com>:
>
> >     One of the very few things that bugs me about the Python
> > docs... let's say one of the _incredibly_ few things that
> > bugs me about the Python docs is where the explanation for
> > how something works refers to a UNIX manpage.
>
> Not that this aleviates your complaint, but several places on the net
have
> searchable man pages...
>
> http://www2.linuxjournal.com/help/man.html

   Keen, thanks. Dunno what you mean about non-alleviation -
certainly seems very helpful in practical terms.

DU

>   david.
>
>


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


From jbauer at rubic.com  Sat Jun 10 20:24:36 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Sat, 10 Jun 2000 19:24:36 -0500
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21>
Message-ID: <3942DC44.4A85EEA7@rubic.com>

dana_booth wrote:
> Anyway do any of you know of, or have made yourself,
> a module for MySQL?

Use Andy Dustman's module:

  http://dustman.net/andy/python/MySQLdb/

Jeff Bauer
Rubicon Research



From jpmcbrien at email.com  Thu Jun 22 11:20:03 2000
From: jpmcbrien at email.com (Jonathan McBrien)
Date: Thu, 22 Jun 2000 15:20:03 GMT
Subject: NameError: response
Message-ID: <8itaqf$6la$1@nnrp1.deja.com>

Hi,

I've installed Python 1.5.2 and Win32All build 132 on Windows NT Server
4, SP4 running IIS.

I've tried to run a simple Python ASP script but get the following
error:
################################################
Python ActiveX Scripting Engine error '80020009'

Traceback (innermost last): File "<Script Block >", line 3, in ?
response.write("testing") NameError: response

/pipetest.asp, line 6

response.write("testing")
################################################

I've tried to find answers to this but either I've done something
silly, missed something or whatever, this problem is winding me up
something chronic. I've uninstalled, reinstalled...

Anyone got any answers as to why I can't get at the response object?

Thanks,
--
# Jonathan McBrien
# jpmcbrien  [at]  e m a i l  [d0t]  c 0 /\/\
# Tragically, children are growing up who'll
# never see a Guru Meditation number.


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


From aahz at netcom.com  Mon Jun 26 17:31:32 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jun 2000 21:31:32 GMT
Subject: Script which only runs in Windows
References: <8j8d2n+qbe9@eGroups.com>
Message-ID: <8j8i3k$307$1@slb3.atl.mindspring.net>

In article <8j8d2n+qbe9 at eGroups.com>,  <al at servana.com> wrote:
>
>We have an odd problem with Python which we cannot resolve.
>
>Below, is a script we have been running in Win32 for months.
>
>We moved our scripts from Win32 to Slackware Linux and IF statements
>which are not in functions or classes will cause errors when you try to
>run the script.

Please give us the exact traceback message.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From phil at geog.ubc.ca  Tue Jun  6 16:04:14 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 06 Jun 2000 13:04:14 -0700
Subject: Nympy for win
References: <Pine.LNX.4.21.0006061109170.30720-100000@chelin.chtf.stuba.sk> <fAdMjhAnqNP5Ew3P@jessikat.demon.co.uk>
Message-ID: <m3aegy60g1.fsf@brant.geog.ubc.ca>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:


> >I would like to use the Numpy package with my windoze/python1.5.2 system,
> >but it needs distutils and distutils need m$ visual c++ which I don't own.
> >Is here anybody who has it compiled and could send me the
> >binaries? Thanks.

The binary file at http://download.sourceforge.net/numpy/python-numpy-15.2.zip
works for me with python 1.5.2.  Unfortunately, the new sourceforge site
hasn't made it onto the numerical topics webpage at:

http://www.python.org/topics/scicomp/numpy_download.html

and the LLNL archive no longer exists.  Is there someone besides webmaster at python.org
who should get this bug report?

Regards, Phil


From hzhu at knowledgetrack.com  Wed Jun 21 21:02:17 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 22 Jun 2000 01:02:17 GMT
Subject: Any other good tutorials?
References: <sl2i51h0e7f149@corp.supernews.com>
Message-ID: <slrn8l2pje.q2.hzhu@rocket.knowledgetrack.com>

On Wed, 21 Jun 2000 18:55:41 -0500, SugarKing <sugaking at gis.net> wrote:
>     Are there any other good tutorials besides Guido van Rossum, I don't
>think he's very explainatory. I like full explainations:)
>

Try this Instant Python page.  You can read GvR after reading this.

http://www.idi.ntnu.no/~mlh/python/instant.html

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From hinsen at cnrs-orleans.fr  Mon Jun 26 05:38:29 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 26 Jun 2000 11:38:29 +0200
Subject: Array Memory Allocation Size
References: <39540628.B284C15F@bioeng.ucsd.edu>
Message-ID: <m3bt0o4wa2.fsf@chinon.cnrs-orleans.fr>

Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:

> I'm using NumPy to allocate heaps of arrays.  Is there a command to find
> out how much space each array is using in memory?

By C code, yes. From Python, no, because you cannot find out
how much space each element takes up, and that is compiler dependent.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From jonan at callisia.com  Fri Jun 23 15:47:21 2000
From: jonan at callisia.com (jonan at callisia.com)
Date: Fri, 23 Jun 2000 15:47:21 -0400
Subject: DB API 2.0 - dictionary fetch ?
Message-ID: <20000623154721.B630@sindri.callisia.com>

Hello,

    I am wondering if the 2.0 db api provides a method for returning
    a dictionary of results instead of a list ? Something like
    perl-DBI's fetch_hashref . 
    
    I am using the latest MySQLdb and it mentions a fetchalldict method
    in the tutor examples, though the method is non-existant when put
    to use. The source MySQLdb.py has comment of fetchXXX depracated
    just above the fetchalldict, so is it gone ?

    The very sparse documentation and db api spec also mention different
    cursor classes which maybe is the way to do it ? Im using
    my own function to make the dictionary but it would be nice
    if it was in the API.

 Thanks,
-- 
-Jonan Santiago
-Callisia Communications
-http://www.callisia.com



From mwh21 at cam.ac.uk  Tue Jun 13 08:54:08 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 2000 13:54:08 +0100
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com> <etdn1kpn3v3.fsf@w20-575-109.mit.edu>
Message-ID: <m31z214u8f.fsf@atrus.jesus.cam.ac.uk>

Alex <cut_me_out at hotmail.com> writes:

> > pi = 3.141592654
> > radius = input("Enter is the radius: ")
> > circ = 2 * pi * radius
> > print circ
> 
> Perhaps the third line should be 'circ = 2 * pi * float (radius)'
> 
> As it is, you are multiplying a float by a string, I think.  That should
> be giving you a type error.

No, he's used "input", which evaluates what you type (and hence the
usual security related hackles rise).
 
> By the way, you can get pi from the math module:
> import math
> print math.pi
> 
> It's good not to have constants encoded directly in to your program, in
> case they ever change. :)

That was the DEC Fortran manual, wasn't it?

faaar-to-young-to-have-seen-*that*-one-ly y'rs
Michael

-- 
  The ability to quote is a serviceable substitute for wit.
                                                -- W. Somerset Maugham


From gregholmes at my-deja.com  Wed Jun 14 08:27:04 2000
From: gregholmes at my-deja.com (gregholmes at my-deja.com)
Date: Wed, 14 Jun 2000 12:27:04 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <8i7tma$nps$1@nnrp1.deja.com>

"jin choung" <jin.choung at gte.net> wrote:
> work my way through the book LEARNING PYTHON and while it says that
there is
> no need to have prior experience in languages like c or c++, i really
found
> that that was not true.
>
> anyone else feel this way?

The only language I had used before Python was JavaScript (well, OK,
BASIC too ;) ) and Python was even easier than that.  Every time I
contemplate C or C++, my head starts to throb just looking at the
source.

So I have to disagree.  Python was very easy to learn.  Probably not
easy to master (I may never do that) but easy to become productive with
very quickly.


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


From moeller at dophnic.yi.org  Sun Jun 18 14:13:15 2000
From: moeller at dophnic.yi.org (Derek Moeller)
Date: Sun, 18 Jun 2000 13:13:15 -0500
Subject: Using Python instead of Perl?
In-Reply-To: <8457258D741DD411BD3D0050DA62365907A23C@huina.oceanic.com>; from DOUGS@oceanic.com on Sun, Jun 18, 2000 at 07:05:24AM -1000
References: <8457258D741DD411BD3D0050DA62365907A23C@huina.oceanic.com>
Message-ID: <20000618131315.B13798@dophnic.yi.org>

On Sun, Jun 18, 2000 at 07:05:24AM -1000, Doug Stanfield wrote:
> [Jim said:]
> > I am trying to use nmap to test my own network and automate the process.
> 
> What's nmap?
> 

	The most commonly used network discovery tool, available for Unix.


> 
> I hope you stick with Python and help to build towards a critical mass using
> it for net mgmt tasks.  Give me a link for nmap and I may very well want to
> help there.

	It is available at http://www.insecure.org .

-- 
Derek Moeller

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \



From ge at nowhere.none  Fri Jun 30 11:01:40 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 30 Jun 2000 15:01:40 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com> <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com> <geT65.2262$iN5.511399@ptah.visi.com> <m2lmznzk2t.fsf@mycroft.actrix.gen.nz> <slrn8lp673.u8u.thomas@centurion.xs4all.nl>
Message-ID: <oD275.2320$iN5.531805@ptah.visi.com>

In article <slrn8lp673.u8u.thomas at centurion.xs4all.nl>, Thomas Wouters wrote:

>Of course, like the early bicycles, very few of us would
>actually recognize the OS, the language or the hardware if they
>didn't have museumtags attached ;-)
>
>(I saw an old PDP-11 used as a telco-switch in a museum a few
>months back;)

Ouch.  The first major design project I worked on was
hardware/firmware for 800MHz radios for cellular telephone
systems.  The software development for the radios (Z8000
processors) was being done on a PDP-11 host.

My first trip to a customer site was an installtion where they
were designing a cellular system with PDP-11s as the switch
processors.

Damn I'm getting old.

-- 
Grant Edwards                   grante             Yow!  But they went to MARS
                                  at               around 1953!!
                               visi.com            


From Eckhard at pfluegel.fsnet.co.uk  Fri Jun 23 19:38:40 2000
From: Eckhard at pfluegel.fsnet.co.uk (Eckhard Pfluegel)
Date: Sat, 24 Jun 2000 00:38:40 +0100
Subject: Coercing classes
Message-ID: <3953F500.ED22500D@pfluegel.fsnet.co.uk>

My question is the following: imagine you have two classes A and B where
B is derived from A (B is a subclass of A). The object a is assigned to
an instance of class A:

a = A()

Under which conditions on the structure of class B is it possible in
Python to convert 'a' to an instance of B? In my application, B would
only have one additional method defined and no additional attributes
(than defined in A), so that a conversion would make sense.
Are there any built-in functions for that purpose??

Eckhard




From bvdpoel at uniserve.com  Mon Jun 19 16:10:07 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Mon, 19 Jun 2000 13:10:07 -0700
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net>
Message-ID: <394E7E1F.29E13190@uniserve.com>

Thanks... I think I tried somethings like your suggestion. However,
neither

	print Goption.__name
or
	print Goption.self.name

work. Both give errors. 

Jeff Massung wrote:
> 
> I'm still learning, but lemme take a crack and let others correct me ;)
> 
> __name isn't a member of MyDialog class, so you can't use self with it,
> because then it looks for MyDialog.__name, which doesn't exist. But perhaps
> you can use Goption.__name instead of self.__name?
> 
> I don't know what the long term effects of this would be for maintaining,
> etc. but I think it should work.
> 
> Jeff
> 
> Bob van der Poel wrote in message <394E67CE.9E29D310 at uniserve.com>...
> >
> >I'm having a problem accessing objects created in a nested class. What I
> >have (extracted) is:
> >
> >class Goption:
> >        __name = ""
> >
> >        def __init__(self, name):
> >                self.__default = default
> >                .. more ..
> >
> >        def change(self):
> >                class MyDialog(tkSimpleDialog.Dialog):
> >
> >                        def body(self, master):
> >                                print self.__name
> >                                ... more ...
> >
> >
> >The print 'self.__name' doesn't work. How do I access the variable
> >__name???
> >


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


From sdhyok at email.unc.edu  Wed Jun 14 21:53:05 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 21:53:05 -0400
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <20000614155102.E16232@freei.com> <8i9cbh$mev$1@news2.isis.unc.edu>
Message-ID: <642A954DD517D411B20C00508BCF23B001296B4C@mail.sauder.com>

Maybe a little better, but not enough good.

class A:
 x = 1

 def __init__(self):
   self.__dict__["x"]=1

 def __setattr__(self, name, value):
  if not self.__dict__.has_key(name):
   raise AttriubuteError
  else:
   self.__dict__[name] = value

 def __getattr__(self, name):
  try:
   return self.__dict__[name]
  except KeyError:
   raise AttriubuteError

"Daehyok Shin" <sdhyok at email.unc.edu> wrote in message
news:8i9cbh$mev$1 at news2.isis.unc.edu...
> Even though I corrected minor errors, it does not work.
> Refer to the following comments.
> >> class A:
> > >     def __init__(self):
> > >             self.x = 1  # this code calls __setattr__
>                                   # when I change it into self.__x = 1, it
> raised AttributeError in __setattr__
> > >
> > >     def __setattr__(self, name, value):
> > >             if name == "x":
> > >                     self.x = value # this code calls __setattr__
> recursively
> > >             else:
> > >                 raise AttributeError
> > >
> > >     def __getattr__(self, name):
> > >             if name == "x":
> > >                 return self.x
> > >             else:
> > >                 raise AttributeError
>
> I found one solution, but expect better one.
>
> -------------- Solution, but not good ------------------
> class A:
>  def __init__(self):
>    self.__dict__["x"]=1
>
>  def __setattr__(self, name, value):
>   if not self.__dict__.has_key(name):
>    raise AttriubuteError
>   else:
>    self.__dict__[name] = value
>
>  def __getattr__(self, name):
>   if not self.__dict__.has_key(name):
>    raise AttriubuteError
>   else:
>    return self.__dict__[name]
>
> "Sean Blakey" <sblakey at freei.net> wrote in message
> news:20000614155102.E16232 at freei.com...
> > On Wed, Jun 14, 2000 at 05:48:45PM -0400, Daehyok Shin wrote:
> > > How can I prevent illegal definition of a variable to an object?
> > > For instance,
> > >
> > > class A:
> > >     __init__(self):
> > >             self.x = 1
> > >
> > >     __setattr__(self, name, value):
> > >             if name == "x":
> > >                     self.x = value
> > >             else:
> > >                 raise AttributeException
> > >
> > >     __getattr__(self, name):
> > >             if name == "x":
> > >                 return self.x
> > >             else:
> > >                 raise AttributeException
> > >
> > > >>> a = A()
> > > >>> a.x = 10
> > > >>> a.y = 20 # I want to raise an AttributeException.
> > >
> > > Peter
> > >
> > >
> > > --
> > > http://www.python.org/mailman/listinfo/python-list
> >
> > The code you posted seems to do what you want, provided you
> > a)add the 'def' keyword to your function definitions
> >     and
> > b) define AttributeException (perhaps you meant AttributeError?)
> >
> > --
> > Sean Blakey, sblakey at freei.com
> > Software Developer, FreeInternet.com
> > (253)796-6500x1025
> > "Uncle Cosmo ... why do they call this a word processor?"
> > "It's simple, Skyler ... you've seen what food processors do to food,
> > right?"
> > -- MacNelley, "Shoe"
> >
>
>




From aahz at netcom.com  Wed Jun 14 20:16:55 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 00:16:55 GMT
Subject: Regular Expressions and Threads?
References: <3947D2F7.264BDC6E@uswest.net> <8i8q6s$72t$1@nntp9.atl.mindspring.net> <39481F1C.419364F5@uswest.net>
Message-ID: <8i979n$4ss$1@slb6.atl.mindspring.net>

In article <39481F1C.419364F5 at uswest.net>,
Brian Wisti  <wbrian2 at uswest.net> wrote:
>Aahz Maruch wrote:
>>
>> I doubt your problem has to do with threads; I bet that if you took your
>> code and ran it unthreaded you would have the same problem.  From your
>> description, it sounds like you have a list variable that you're
>> appending to and not resetting after you do the string.join().  If
>> that's not it, I don't have any better ideas without code.
>
>You may be correct... however, the only way I was able to fix the
>problem was with an ugly hack:
>
>	re._cache.clear()
>
>Ugly because it's undocumented, and that means I should keep my grubby
>little paws away from it. Ugly or not, once I put this line in,
>everything has gone peachy.

Umph.  Well, then, did you double-check to make sure you're not sharing
the same re object between threads?  If you're doing anything more than
checking for the existence of a match, that could cause problems.  (As
with anything else, the re module should be thread-safe, but individual
re objects probably are not.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From digitig at cix.co.uk  Thu Jun 22 18:17:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 22 Jun 2000 23:17 +0100 (BST)
Subject: boolean true and false values.
References: <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <memo.20000622231736.45677A@digitig.compulink.co.uk>

In article <slrn8l1oc2.3c1.hat at se-46.wpa.wtb.tue.nl>, 
hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:

> I read a number of replies, but didn't read an answer that seems 
> satisfactory.
> I ran in the same problem, and was also not able to find a elegant 
> solution.
> Using 0 and 1 or `true` and None is not correct imho, there should be a 
> native
> way to use true and false for indicating boolean values.

There is not an elegant solution, short of making very fundamental changes 
to the language that would IMHO break its nature. There is no problem with 
assignment, but there is a serious danger with testing. You would have the 
situation that
	if foo:
could follow the true branch, but
	if foo == true:
would follow the false branch. This has been a perennial C/C++ issue, and 
there are still a lot of programmers who insist on using the latter form. 
C++ now has boolean types, and that problem has been largely resolved, but 
that's only been possible because C++ is (fairly) strongly typed and will 
do implicit casts to boolean. I don't think that's a sensible route for 
Python.

> Also, I am missing the data type 'set'. It can be faked with a 
> dictionary, but it
> is just not the same. Any one else got this problem ?

It can be implemented very easily as a class wrapper around a list, that 
checks for presence before adding an element, and defines union, 
intersection, &c. It raises interesting questions of how deep the equality 
test should be, and I suspect different applications would need different 
answers, so I'm not sure it should be a built in. Anyway, all the paired 
delimiters on my keyboard are already used -- what would you delimit it 
with? :-)


From sverres1 at start.no  Sat Jun 17 10:13:50 2000
From: sverres1 at start.no (Sverre Stoltenberg)
Date: 17 Jun 2000 14:13:50 +0000
Subject: ANNOUNCE: jdbi-0.2
Message-ID: <m266r8v1i9.fsf@iscariot.local.net>

JDBI is a jpython module that can connect to any database that supports
jdbc. I'm trying to make it DB-API spec v2.0 compatible.

http://www.terminator.net/jdbi/

/S.
-- 
When writing code, you should allways aim to avoid undefined behavior.
-- Bjarne Stroustrup


From dbroadwell at mindspring.com  Sat Jun  3 03:00:07 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sat, 03 Jun 2000 03:00:07 -0400
Subject: compiler_question->diret to whom(?)
Message-ID: <3938ACF7.9977D53A@mindspring.com>

Ok, i know that the optimizing compiler is being built. But where can I
direct a question-suggestion about it?

Ie: "hm, i've got an idea fo an optimization and i want to see if it
will be of use."


 --
	Got CODE?
 --


From thomas at cintra.no  Fri Jun  2 12:54:42 2000
From: thomas at cintra.no (thomas at cintra.no)
Date: Fri, 02 Jun 2000 16:54:42 GMT
Subject: SQL: don't use bind variables (was Re: Using strings with ' in them  in SQL-queries)
References: <39342055.1301760@news.c2i.net> <393510AB.D62AE9A5@infercor.no> <8h38ak$s1f$1@slb6.atl.mindspring.net> <3937864A.DB218804@infercor.no> <8h899v$rch$2@news.tht.net>
Message-ID: <3937e26c.2623019@news.c2i.net>

On 2 Jun 2000 12:28:47 GMT, "D'Arcy J.M. Cain" <darcy at vex.net> wrote:

>Paul Boddie <paulb at infercor.no> wrote:
>>   db.execute("insert into test (id, name) value (1, :var)", {"var" : var})
>
>> (Now I look again, I don't know where the 'query' method came from, but I assume
>> that the thread's initiator meant 'execute'.)
>
>He's probably using PyGreSQL which has two distinct interfaces, PyGreSQL
>Classic (or PyGres) and PyGreSQL DB-API which is in the 3.0 version
>just released.  He is using the classic interface which, btw, I am not
>planning to deprecate.  Both will continue to be available in PyGreSQL
>as long as I am maintaining it.
>
>There's-more-than-one-way-to-skin-a-dead-parrot-ly y'rs
>
>-- 
>D'Arcy J.M. Cain <darcy at vex.net>           |  Democracy is three wolves
>http://www.vex.net/                        |  and a sheep voting on         
>+1 416 425 1212     (DoD#0082)    (eNTP)   |  what's for dinner.

Hi,

I think I started this thread, and I was using the old version of
PyGreSQL.  In the new version I use another method;

import pg
db = pg.DB(....) # same as old _pg.connect(..)

a = {}
a['filename'] = 'test.jpg'
a['size'] = 1024
db.insert('test', a)

the keys in the dictionary matches columns in the table "test". All of
my previous problems have vanished.

This might be in the old module too, but at least the old problem is
gone so ...

Thomas




From clarence at netlojix.com  Wed Jun 28 12:14:11 2000
From: clarence at netlojix.com (Clarence Gardner)
Date: Wed, 28 Jun 2000 09:14:11 -0700
Subject: Problem with os module
References: <3959FF57.85E1A503@mc2.fr> <192c2800.183750f0@usw-ex0105-038.remarq.com>
Message-ID: <962208917.243425201@news.silcom.com>

On Wed, 28 Jun 2000, Richard Chamberlain wrote:
>
>if os.path.isfile("c:\\Test"):
>
>(much better)
>
>b.t.w. you'll notice above that I've used c:\\Test - the reason
>is that \ has a special meaning within a string. For instance \t
>is a tab and \n a newline, \\ is in fact a single \. So your
>paths should look something like "c:\\my folder\\my file".

As a side note, it's easier if your paths look something like
"c:/my folder/my file".


From jin.choung at gte.net  Wed Jun 14 03:51:31 2000
From: jin.choung at gte.net (jin choung)
Date: Wed, 14 Jun 2000 07:51:31 GMT
Subject: python is hard to learn without c
Message-ID: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>

howdy all,

i programmed a bit in BASIC as a kid, had a high school PASCAL class and i
tried to pick up python before moving on to c++.  however, i was trying to
work my way through the book LEARNING PYTHON and while it says that there is
no need to have prior experience in languages like c or c++, i really found
that that was not true.

anyone else feel this way?

i'm currently working my way through a 24 hour c++ book and i think this is
going to work for me.  oh, well, will be back to contribute here in a couple
of months... in the meantime, i'll just leech valuable nuggets of wisdom
from you guys.

thanks

jin




From nascheme at enme.ucalgary.ca  Fri Jun  9 19:44:47 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Fri, 09 Jun 2000 23:44:47 GMT
Subject: Setting Port Speed in Linux Python: HOW?
References: <39417E33.C2780C16@home.com>
Message-ID: <slrn8k30dj.55l.nascheme@cranky.arctrix.com>

Darren Enns <dmenns at home.com> wrote:
>What simple (Linux) Python code can be used to set the baud rate
>of a serial port?

Look at the termios module.  Specificly termios.tcsetattr().

    Neil

-- 
"Reliability means never having to say you're sorry." -- D. J. Bernstein


From digitig at cix.co.uk  Thu Jun  8 16:36:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 8 Jun 2000 21:36 +0100 (BST)
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available now?
References: <8hl866$bt5$1@news.jump.net>
Message-ID: <memo.20000608213634.55497C@digitig.compulink.co.uk>

The "What Is" named my three favourite languages (Python, Eiffel, 
Ada) so I had a look.

Couldn't get it to install under Beos. Couldn't see anything in the 
language that I didn't already have. Didn't like the look of the language. 
Unimpressed.

In article <8hl866$bt5$1 at news.jump.net>, schneiker at jump.net (Conrad 
Schneiker) wrote:

> Well, at least _some_ people regard Ruby as a better OO Perl than Perl 
> or as
> a means of getting anticipated Python 3000 features now. Maybe if you 
> knew
> more about Ruby, you might become one of them.

<snip>


From jkraska1 at san.rr.com  Wed Jun 14 21:36:17 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 15 Jun 2000 01:36:17 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com>
Message-ID: <394833BD.6A322985@san.rr.com>

Huaiyu Zhu wrote:
> 
> >"Warren Postma" <embed at NOSPAM.geocities.com> writes:
> >
> >> Having something
> >> look correct on the screen, but not work because of misuse of tabs versus
> >> spaces, or incorrect indentation, is something that still continues to bug
> >> me.
> 
> If you use emacs:
> 
> C-x h M-x tabify


In vi/vim, you can globally replace all tabs with:

<esc>:1,$s/<tab>/    /g


Where the spaces are appropriate for the indentation
level you are looking at.


...................If I were running a python
project, the very first thing I'd do would be to forbid
tabs in python files. Both vim and emacs have modes
where hitting the tab key inserts enough spaces to
take you to the next tabstop.



C/


From mhammond at skippinet.com.au  Wed Jun 14 19:43:18 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 23:43:18 GMT
Subject: Python and Windows Scripting Host
References: <9Ov15.1411$vM.56343@typhoon1.ba-dsg.net> <39y15.839$Zh1.63369@news.easynews.com> <8i8mbb$c3f$1@nnrp1.deja.com> <O_S15.205$3y1.4771@typhoon1.ba-dsg.net>
Message-ID: <8i95aj$np0$1@nnrp1.deja.com>

In article <O_S15.205$3y1.4771 at typhoon1.ba-dsg.net>,
  "Jim Bower" <jbower at lims.com> wrote:
> Here is what I am trying to do:
>

OK - It appears Roger _was_ correct given your question.

You are running a _normal_ Python script, from within Pythonwin.  Roger
is correct - there is no WScript object available, and can be created
using the code he suggested.

Given the subject line, I assumed you were trying to run this code from
_within_ Windows Scripting Host.

To do this, you place that _exact_ same code you posted in a .pys file
(the import isnt necessary, but will still work), and execute:

cscript foo.pys

Then the WScript object is automatically available, and works.

Mark.


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


From scarblac-spamtrap at pino.selwerd.nl  Mon Jun  5 07:47:06 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 5 Jun 2000 11:47:06 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com>
Message-ID: <slrn8jn4nj.lau.scarblac-spamtrap@flits104-37.flits.rug.nl>

Courageous wrote in comp.lang.python:
> 1. For slice operations, ilow/ihigh indices which undeflow
> or overflow the list in question are reset to the edges
> of the list. Why was it that PyExc_IndexError wasn't
> thrown instead?

Much more easy to use, I suppose. a slice[a:b] returns all elements of which
the index has a <= index < b.

> 2. In slice expressions of the form s[i:j], what was the
> design decision that lead to s[i:i] always returning an
> empty list and s[i:i+1] selecting one item. At first
> glance, it would appear that s[i:i] is the obvious and
> intuitive correct expression to return just one item on
> a slice, where s[i:i+1] ought to select two items and so
> forth. Was there some particular situation which arose
> which required this slightly counterintuitive idiom?

This is symmetrical to the behaviour of range(). range(i,i) also returns an
empty list. The same arguments apply.

At first sight, this is counterintuitive. At second and all further sights,
this is the best behaviour.

If you have a list s of length n, s[:n] has the whole list, s[:n-1] is all
but one element, etc (s[:-1] is basically sugar for s[:n-1]).

The slice s[a:b] always tries to grab exactly b-a items.

For a list s of length m, with 0<=m<=n, s == s[:m]+s[m:]

For a list s of length n, and integers a,b,...,z where 0<=a<=b<=...<=z<=n,
s == s[:a]+s[a:b]+s[b:c]+...+s[y:z]+s[z:]
;-)

In short, it makes *loads* of off-by-one errors just disappear.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"Early to rise, early to bed, makes a man healthy, wealthy and dead." -- TP


From dsavitsk at e-coli.net  Mon Jun  5 00:33:37 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Mon, 05 Jun 2000 04:33:37 GMT
Subject: stripping out unwanted characters in string
References: <393b1f58$0$95987@news.execpc.com>
Message-ID: <B4G_4.5769$gr4.233822@news1.rdc1.il.home.com>

>>> str = string.replace(str, 'unwanted', '')
or
>>> str = string.join(string.split(str, 'unwanted'), '')

"Jerry F. Davis" <jfdecd at execpc.com> wrote in message
news:393b1f58$0$95987 at news.execpc.com...
> Howdy:
>
> I have a string that has some unwanted characters in it.
> Octal 240 for instance.
>
> I want to actually strip them out of my string.
>
> I have looked at string.replace, but can't seem to figure out how to do
> this.
>
> Thanks in advance.
>
> Jerry




From cef at vnet.de  Thu Jun  8 20:10:12 2000
From: cef at vnet.de (Christian Fedtke)
Date: Fri, 09 Jun 2000 02:10:12 +0200
Subject: Problem with Python/ASP/COM?
Message-ID: <394035E4.F0E98834@vnet.de>

Hello,
 I try to run a ASP site with Pythoncode in it.
 The Pythoncode queries a database via COM, does some
 formatting, etc. Sometimes the ASP-Application
 crashes and stops all Sites on the server.
 After that the Sites can not be started again,
 i have to reboot the machine.

 The System is:
 NT4
 IIS4
 Python 1.52
 Pythonwin Build 129

 I try to find out where the error occurs but
 i dont understand the error message in the
 event log.

 ---------------------------------------------------------------
 An object call caused an exception.
 (IID: {51372AEF-CAE7-11CF-BE81-00AA00A2FA25})
 (Method: 3)
 (Microsoft Transaction Server Internals
 Information:
 File: i:\viper\src\runtime\mtxex\activity.cpp,
 Line: 889)
 (Exception: C0000005) (Address: 0x1e2a2ea9)
 pythoncom15!PyCom_HandlePythonFailureToCOM(struct
 tagEXCEPINFO *) + 0x69
 ole32!DllGetClassObject + 0x77
 ole32!CoCreateInstance + 0x6A
 asp + 0x23991
 asp + 0x24DB3
 asp!TerminateExtension + 0x3BFF
 asp!TerminateExtension + 0x32BC
 asp!TerminateExtension + 0x313B
 asp!TerminateExtension + 0x7AC9
 asp + 0x326F8
 mtxex!MTSCreateActivity + 0x1F3C
 mtxex!MTSCreateActivity + 0xF3E

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

 Thanks for every hint

 CF


From cheeseballs at my-deja.com  Thu Jun 22 15:28:38 2000
From: cheeseballs at my-deja.com (cheeseballs at my-deja.com)
Date: Thu, 22 Jun 2000 19:28:38 GMT
Subject: Win32COM and Excel Charts
Message-ID: <8itpcd$ism$1@nnrp1.deja.com>

Hi, has anyone successfully created an embedded chart within an Excel
Worksheet using win32com.client?

I'm able to create a chart in it's own Sheet with the following:
>>> xlapp=win32com.client.Dispatch("excel.application")
>>> xlapp.Visible=1
>>> xlchart=xlapp.Charts.Add()

then I try to add a chart in the active worksheet...
>>> xlsheet=xlapp.Worksheets(1)
>>> xlchartobj = xlsheet.ChartObjects.Add(10,10,100,100)

but it doesn't look like Add is a method of the ChartObjects
collection...
Any ideas?
Thanks, Drew


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


From invalid.address at 127.0.0.1  Sat Jun 24 22:11:22 2000
From: invalid.address at 127.0.0.1 (David)
Date: Sun, 25 Jun 2000 02:11:22 GMT
Subject: Gadfly Question: TypeError on __Init__
References: <395408ae.75890752@news.telus.net> <3954CC36.EE2D4CC4@arrakis.es>
Message-ID: <39556a16.81688704@news.telus.net>

On Sat, 24 Jun 2000 12:53:48 GMT, Marcos S?nchez Provencio
<rapto at arrakis.es> wrote:

>I believe that __init__ should always return None (what functions do
>silently when they don't find a return clause).
>
>Leave the error to propagate, should there be any. Catch it in the
>calling code.


Ahhhhh!  Yes, yes, yes!  IIRC, my original code didn't actually have it as
a danged class, it was a function!
 
Guh!




From PA0215481 at ntu.edu.sg  Thu Jun  1 02:55:16 2000
From: PA0215481 at ntu.edu.sg (#VEDALA NARASIMHA KUMAR#)
Date: Thu, 1 Jun 2000 14:55:16 +0800 
Subject: jpython
References: <kJVY4.1235$IO4.112187@news.flash.net>
Message-ID: <0CF260C495FED111A6610000F866308D0EDEAE5C@mail3.ntu.edu.sg>

compile it into a jar file using --jar option and use it in your html
file.

say u have first_app.py

"jpythonc --core --jar first_app.jar first_app.py"

in html file
<applet archive="first_app.jar" code="first_app" width="100"
height="100">

> ----------
> From: 	Scott Hathaway[SMTP:slhath at flash.net]
> Posted At: 	Wednesday, May 31, 2000 4:42 AM
> Posted To: 	comp.lang.python
> Conversation: 	jpython
> Subject: 	jpython
> 
> Does anyone know of a JPython tutorial?  I need something  basic that
> tells
> me how to create applets and then to get them into the browser.  I can
> make
> the demo app and compile it with jpythonc, but then I am lost as to
> how to
> view it in the browser.
> 
> Thanks,
> Scott
> 
> 



From petersc at stud.ntnu.no  Fri Jun  2 15:20:44 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 02 Jun 2000 21:20:44 +0200
Subject: Sequence-length - Missing the obvious ?
References: <3934db4e@nntp.server.uni-frankfurt.de>
Message-ID: <3938090C.9FB5D057@stud.ntnu.no>

mlauer at trollinger-fe.rz.uni-frankfurt.de wrote:
> 
> But why ? How does the for loop knows
> how much elements the sequence contains ?
> I did not program the __len__ function,
> so... how ?

As far as I understand the for loop goes on until an
IndexError is raised. Then the error is cleared and
the for loop ends.

Nice feature, if you ask me.

makes-extending-error-messages-more-complicated-though-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From hzhu at rocket.knowledgetrack.com  Mon Jun 12 16:09:16 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Mon, 12 Jun 2000 20:09:16 GMT
Subject: Where are the math functions?
References: <slrn8k8dfv.tu.hzhu@mars.localdomain> <owen-97F32C.10015712062000@news.u.washington.edu>
Message-ID: <slrn8kah0f.v7.hzhu@rocket.knowledgetrack.com>

On Mon, 12 Jun 2000 10:01:57 -0700, Russell E. Owen
<owen at astro.washington.edu> wrote: 

>The random module (which is available on all platforms) offers beta and 
>gamma functions. I can't comment on incomplete gamma, bessel and erf. I 
>found the information about the random module in "Python Essential 
>Reference" by David Beazley; the index lists beta and gamma, but not the 
>others.

As I said

> Yes there is random.gamma but that's generating random numbers not the gamma
> function.


Anyway as Phil Austin mentioned the Cephes library is excellent and I'm
adding those into MatPy.  Thanks Phil.

Huaiyu


From shindich at my-deja.com  Sun Jun 25 00:00:19 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Sun, 25 Jun 2000 04:00:19 GMT
Subject: Newbie Q: CGI
References: <8hn0c1$1a22$1@news1.wire.net.au>
Message-ID: <8j404k$62j$1@nnrp2.deja.com>

You can find a sample python CGI that prints out client's IP address on my
web page: <a
href="http://www.shindich.com/sources/cgi/clientip/readme.html">http://www.sh
indich.com/sources/cgi/clientip/readme.html</a>

I hope this helps!
Alex Shindich
<ahref="http://www.shindich.com/">http://www.shindich.com/</a>

In article <8hn0c1$1a22$1 at news1.wire.net.au>,
  "Steven Adams" <adams_s at lab.eng.usyd.edu.au>
wrote:
> Hi,
>
> How do you get the IP number from a request for
a web page? gethostbyname()
> and gethostbyaddress() weren't mentioned
anywhere in the cgi module docs
> (that I could see, did I miss it?)
>
> simple problem, but I just can't find the
answer...
>
> thanks Steven
>
>



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


From shecter at darmstadt.gmd.de  Fri Jun 16 08:57:24 2000
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Fri, 16 Jun 2000 14:57:24 +0200
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006160804260.15416-100000@fep132.fep.ru>
Message-ID: <394A2434.2B23D7D@darmstadt.gmd.de>

Oleg Broytmann wrote:

>    Why do you think someone (who do not neccessaryly needs this) will write
> code that you need? In the world of free software it is simple - roll your
> own.
>    Patches are gladly accepted. Come on!

Dude,

As I said in my original post:

>So - is anybody else interested in seeing a client interface that
>allows the whole HTTP spec to be used, without getting down into the
>mucky details?  I can see that I'll be working on this for sure.  I'd
>be much more motivated if I knew that my work wasn't going to just sit
>around.

I knew I'd be doing this work for myself, and didn't know if anyone else felt the
way I do about the library.  It looks like there -are- people who'd want these
features too, which is cool.  I'll post something when I've got something to
show.

But thanks anyhow for the motivation, Oleg! :)

- Robb



From sdhyok at email.unc.edu  Wed Jun 14 16:08:23 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 16:08:23 -0400
Subject: Numeric python
References: <Pine.LNX.4.10.10006141123440.26946-100000@intel3.wag.caltech.edu>
Message-ID: <8i8q4h$g8s$1@news2.isis.unc.edu>

Try www.sourceforge.com.

"Michael Todd Feldmann" <feldmann at wag.caltech.edu> wrote in message
news:Pine.LNX.4.10.10006141123440.26946-100000 at intel3.wag.caltech.edu...
> Could anyone point me in the right direction to get Numeric python?  I
> have a linux box and tried
>
> http://www.python.org/download/download_linux.html
>
> and went to
>
> http://andrich.net/python/
>
> which states that "This site is closed." since things may be outdated.  I
> am guessing there is somewhere else to find this and other RPM files.
>
> Thanks a bunch!
>
> Mike
>
> --------------------------------------
> Michael Todd Feldmann
> Beckman Institute (139-74)
> California Institute of Technology
> Pasadena, California 91125 USA
> Phone:(626)395-2722, FAX:(626)585-0918
> email: feldmann at wag.caltech.edu
> --------------------------------------
>
>
>




From wjk at wjk.mv.com  Fri Jun  2 03:53:38 2000
From: wjk at wjk.mv.com (wjk)
Date: Fri, 02 Jun 2000 03:53:38 -0400
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no>
Message-ID: <39376801.DB6BBD74@wjk.mv.com>

I was playing around learning Python a few months ago and have
started the learning adventure again.  I was interested in the same
question and approached it from a slightly different angle -- the code
while not yet complete might be of some interest....


import re
pattern=re.compile("[0-9]{4,4}[\.\,$]")

#something here to insure 2 decimal places

test=123567789.98
teststr=str(test)
a = pattern.search(teststr)
while a:
 (a.regs[0])[0] + 1
 teststr = teststr[0:(a.regs[0])[0] + 1] + "," + teststr[(a.regs[0])[0]
+ 1:]
 a = pattern.search(teststr)

#possibly something here to convert to float if
#to be used for mathematical computations

print teststr




Joseph Santaniello wrote:

> Hi,
>
> I have an SQL query which I can format to give me a nice value like
> 23,342.40 which Python squashes into 23342.4 when I stick it into a
> list element. Is there a nice way to format floats so they look like
> (US) currency?
>
> Thanks,
>
> Joe



From te at first.gmd.de  Thu Jun  8 12:00:38 2000
From: te at first.gmd.de (te at first.gmd.de)
Date: Thu, 08 Jun 2000 16:00:38 GMT
Subject: popen in real time?
References: <8hjdn4$6kf$1@bmerhc5e.ca.nortel.com>
Message-ID: <8hofuq$4br$1@nnrp1.deja.com>

In article <8hjdn4$6kf$1 at bmerhc5e.ca.nortel.com>,
  msoulier at nortelnetworks.com wrote:
>
> 	Hey guys. I realize that you can loop on the output of popen if you
want
> to read from a pipe, but it would seem from experience that the output
of the
> child process is not available until the process exits. I've tried
this with
> the output of long-running processes like find. Is there a way to get
the
> output of a child in real time, as it's printed?
>
> 	Mike
>

Somebody already pointed to (Tcl's) expect and the
(basic) Python pty stuff.

I found Paul Sheer's Python extension "expect"
(aka expectmodule) to be very helpful for such
tasks. It combines the use of ptys with an easy
to use interface that is largely an extension to
popen/2/3.

See ftp://ftp.obsidian.co.za/pub/expect/

I'm not sure, however, how platform-dependent
this extension is. I used it on Linux
(Debian 2.2).

BTW this is not the only Python-related "expect"
out there, but I used only this one, and
enjoyed it. Vaults of Parnassus has the
alternatives, too.

Regards,

Thilo


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


From jkraska1 at san.rr.com  Sat Jun 17 03:16:43 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 17 Jun 2000 07:16:43 GMT
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>
Message-ID: <394B266C.42F7EC63@san.rr.com>

> In a smaller way, I would like to have Bash, sed, and grep as Python
> programs rather than compiled C.

Why?



C/


From paul.robinson at quantisci.co.uk  Thu Jun 15 12:49:45 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Thu, 15 Jun 2000 17:49:45 +0100
Subject: Suggestion: "Completing" HTTP support in Python. (was Re: SSL and 
 POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com> <etdog5fcx7a.fsf@w20-575-109.mit.edu> <8hhttv$7r6$1@nnrp1.deja.com> <3948FB36.416393C5@darmstadt.gmd.de>
Message-ID: <39490929.4E88EDD1@quantisci.co.uk>

Robb Shecter wrote:
> 
> Hi,
> 
> I was really suprised to see that HTTP client support in Python isn't
> really quite complete.  What I mean is, I was hoping to find an
> interface at a higher level, like this:
> 
> Wishful example #1
> --------------
> h = HTTP('my.host.name', '/loginscript', 8080)
> h.add_parameter('username', 'Robb')
> (code, msg, headers) = h.post()
> 
> Wishful example #2
> --------------
> h = HTTP('my.host.name', 8080)
> h.add_header('X-local-hack', 'true')
> (code, msg, headers) = h.get()

Surely this level is higher than HTTP.

The way in which forms are posted is described in the W3C's HTML
specification [ http://www.w3.org/TR/1998/REC-html40-19980424 ] - this
isn't part of the HTTP standard. The "protocol" for submitting forms,
setting cookies, sending files (RFC 1867) etc all use HTTP but are not
"part of it".

I'm not so sure why your second example is wishful, as (unless I'm
missing something) that's pretty much the way it works at the moment -
isn't it?

If you are looking for a more high level approach then why not try
something like this:
http://cobweb.quantisci.co.uk/pub/english.cgi/d8014813/cl_forms.py

It's rough and ready...no comments (although it's all obvious) and
probably works at least some of the time.

Basic use is create a form object, add controls to it (with values)
submit it, read response.

For example (login method of a user object - the rest of the object is
only of interest for my own application):

def login(self):
    form = cl_forms.form('', self.authURI)
    form.addElement(cl_forms.text('uname', self.userID))
    form.addElement(cl_forms.text('password', self.passwd))
    from time import time
    form.addElement(cl_forms.hidden('thetime', `time()`))
    #<INPUT type="hidden" name="thetime" value="939835494.803">
    #print form
    errcode, errmsg, headers, data = form.submit(self.server, self)
    if errcode == 200:
	print 'authUser login :', errcode, errmsg
	pass
    elif errcode == 401:
	raise TypeError, 'Unknown authentication method.\nWWW-Authenticate:
'+headers.get('WWW-Authenticate')
    else:
	print 'authUser login :', errcode, errmsg, headers.headers, data

> ...without this kind of interface, I bet there's a lot of duplication
> going on.  And, cookie support would of course also be great.  If
> anything, as an option that could be enabled, where cookie information
> would be managed by an HTTP object from invocation to invocation.

It would seem you're looking for a "browser" rather than a pure HTTP
implementation - which Python already provides. I'm not sure (having
never looked at it) but elements of Grail may be of use.

> So - is anybody else interested in seeing a client interface that
> allows the whole HTTP spec to be used, without getting down into the
> mucky details?  I can see that I'll be working on this for sure.  I'd
> be much more motivated if I knew that my work wasn't going to just sit
> around.  I think this kind of interface should be in the core library.

I think this would be great - although I believe re-phrasing that as "So
- is anybody else interested in seeing a client interface that allows
GETing, POSTing of form data and cookie support to be done, without
getting down into the mucky details of the HTTP representation of
these?" would be more accurate.

I for one, would be interested in this kind of functionality for the
library - my own attitude has been to simply "knock one up" as and when
I've needed it.

	Paul.

-----------------------------------
Business Collaborator Team
Enviros Software Solutions
http://www.businesscollaborator.com
-----------------------------------



From olczyk at interaccess.com  Fri Jun  9 13:13:23 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Fri, 09 Jun 2000 17:13:23 GMT
Subject: Problem with script and typing
Message-ID: <394123ad.89325968@nntp.interaccess.com>

Here goes ( extraneous pruned ):

project =[]

class Project:
       def IsNull(self):
	return self.name==""
	project.append(self)

def FindProject(x):
    for project in projects:
        if project.name == x:
            return x
        null=Project("")
        projects.remove(null)

def is_project(s):
    if s[len(s)-3:] == 'prj':
        project=FindProject(s)
## error here
        if project.IsNull():
            project=Project(s)
        project.Scan()
----------
error message AttributeError: 'None' object has no attribute 'IsNull'

Several questions.
1) I've banged my head against this type of typing problem before.
    How do I get around it?
2) Is there a better way of handling null objects? ( Most languages I 
    know have objects which are explicitly null )




From digitig at cix.co.uk  Thu Jun 15 15:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Jun 2000 20:28 +0100 (BST)
Subject: python is hard to learn without c
References: <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net>
Message-ID: <memo.20000615202832.52185B@digitig.compulink.co.uk>

In article <9yS15.2877$Gp6.138923 at dfiatx1-snr1.gtei.net>, 
jin.choung at gte.net (Jin) wrote:

> but i think the reason the C++ book is working so well for me is that it
> assumes nothing about prior knowledge in an OO language because C++ is 
> the
> base upon which everything else is built.  

Oh no it's not! C++ drew on earlier OO languages, and there are plenty 
(Scheme, Eiffel, Smalltalk) that are quite unrelated. But C++ has a huge 
user base, so there's simply more books available, making it easier to 
find one that fits your level. I'm glad you have -- what is it, by the 
way? I'm sometimes asked to recommend something for beginners.

> thanks again, and i will pursue both paths to enlightenment and 
> hopefully
> one language will bolster skills in the other.

A good approach, IMHO.


From dbroadwell at mindspring.com  Sat Jun 24 06:53:18 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sat, 24 Jun 2000 06:53:18 -0400
Subject: need insight...
References: <3954671F.51E53AE6@mindspring.com> <Zh_45.85985$WS3.752637@typhoon.we.rr.com>
Message-ID: <3954931E.C1F7A9EC@mindspring.com>

>     The problem is that you really do not want to change the sequence you're
> using in 'for' -- the documentation, if memory serves, shows an example of
> precisely why. You'll want to make a copy of it to iterate through.. like;
> 
> for item in cwd[:]:
>     ...
>     cwd.remove(item)
> 
> Like that..see? The [:] slice copies the whole sequence.

After reading the documentation, i added the slice opperator in one
place, the one it was strictly necessary to the design to modift the
list, and in the others (with the cdw) i simply restrained myself as i
didn't NEED to modify it, and the program functioned exactly as
expected.

Thanks newsgroup. (btw: this was one of my first real coding effort
spaned from gtting pissed at having to open 18 instances of notpad to do
a simple task.)

And though it does not sho precisely why, i can understand how a loop,
iterating on indexing pointer would get confuzed as the list items
change places after a delete..

-- 
        ~We're all fallen angels who've forgotten how to fly~
        Remember how: http://www.cyber-action.com/oed/dbroadwell.html
--


From jhz at aplexus.jhuapl.edu  Fri Jun  9 12:06:53 2000
From: jhz at aplexus.jhuapl.edu (John H. Zouck)
Date: Fri, 9 Jun 2000 12:06:53 -0400
Subject: Regular expression matching performance
Message-ID: <Pine.GSO.3.96.1000609115142.4766B-100000@aplexus.jhuapl.edu>

I'm a relative newbie to python, and have been rewriting a fairly simple
perl program in python, mainly because I like python much better for
structural and engineering reasons. The pythonprogram runsmuch more slowly
then the perl program, and I set out to find out why. The  python profiler
revealed the program was spending most if its time in regular expression
matching, so I decided to find out if this was very much slower than perl. 

I do not want to start any flame wars, and forgive me if this does, but I
would like suggestions if I have missed any obvious tricks here or made
any obvious mistakes. I did try to streamline the python by taking the
suggestions from one of the performance notes, which is why the search
function is called indirectly using the 'findit' variable.

I wrote two simple test programs, one in perl and one in python. All they
do is read a text file and search for a regular expression, line by line.
Here are the programs: 

File match.pl:
--------------

#!/usr/local/bin/perl

while (<>) {
        /\s+J2\.2I\s+(\d+).+Loopback.+Alt\s+=\s+(\d+).+/;
}

File match.py:
--------------
#!/home/u8/jhz/tcl/Python-1.5.2c1/python -O

import re
import fileinput

j2_2_exp = re.compile(
        r"\s+J2\.2I\s+(\d+).+Loopback.+Alt\s+=\s+(\d+).+" )

findit=j2_2_exp.search

def runem():
        for line in fileinput.input():
                findit(line)



runem()


Here are the timings:

aplexus 1/42> time match.pl art.txt.n
0.27u 0.02s 0:00.29 100.0%

aplexus 1/43> time match.py art.txt.n
13.51u 0.06s 0:14.83 91.5%

I have a fairly large set of files to process and as it stands I can't
effectively use python. 


============================
John H. Zouck 
The Johns Hopkins University
Applied Physics Laboratory
============================



From kens at sightreader.com  Thu Jun 15 13:44:26 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 10:44:26 -0700
Subject: Starship
Message-ID: <394915FA.FCDC85A1@sightreader.com>

Who exactly do I contact to find out how to get my FTP login back up and
running?

I haven't updated my page since the crash.  When I attempt to FTP using
the same script I used before the crash, I get a "530 Login incorrect".

My page is intact, though out of date.

Thanks.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/07fbe5ba/attachment.vcf>

From ivanlan at home.com  Mon Jun 26 18:32:35 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Mon, 26 Jun 2000 16:32:35 -0600
Subject: strings and integers?
References: <8ir75b$qfi$1@news2atm.raleigh.ibm.com> <394FF466.51A5B77B@sightreader.com> <8it0j0$1du6$1@news2atm.raleigh.ibm.com> <8it0tq$8pc$1@news2atm.raleigh.ibm.com>
Message-ID: <3957DA03.E76066C1@home.com>

Hi All--

"Arint?" wrote:
> 
> On another note, I just tested this
> xrt = 8
> print'ioctl %d <---' %xrt
> and instead of sending it to print, I stored it in a variable and sent it
> back to the C++ application and the string came out fine????  But, as I said
> earlier print somewhere around print I get the len() of unsized object
> error??? Any ideas.
> 

Works fine for me.  Paste the file that causes the error into an email
message, along with the exact error text, and send the email message to
this list/newsgroup.

A matter of style:  don't write

	print'ioctl blah blah

but instead

	print 'ioctl blah blah

Either way works fine, but the second is a teeny bit more readable.

<why-else-would-you-use-python?>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From bjorn at roguewave.com  Fri Jun  9 15:54:44 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 09 Jun 2000 13:54:44 -0600
Subject: = vs. == in Py3k
References: <F2005.859$Qf6.33080@nuq-read.news.verio.net>
Message-ID: <39414B84.DA355E5F@roguewave.com>

Lawrence Kesteloot wrote:
> 
> If one of the goals of Python 3000 is to make it easy for new
> programmers to learn, how about removing the distinction
> between = and ==?  The only kinds of examples I can think of
> where this is useful is something like:
> 
>     a = b = 5
> 
> vs.
> 
>     a = b == 5
> 
> But this isn't worth the trouble it gives new programmers.  The
> syntax could be changed to make = and == equivalent tokens
> and have the parser figure out which is meant from the context,
> just as it's done in Basic.  The documentation would only ever
> mention =, but the language would support == for backward
> compatibility.

[snip]

A much better suggestion would be to make := be the assignment token.
It's much clearer to distingush between :=/== vs =/==, and you wouldn't
have something that looks like equality meaning assignment (must be
better for the 4E crowd <wink>).

I'm no longer of the opinion that := should be an assignment expression
however.  I think we should investigate a ruby'esque solution instead:

  file.readline |line|:
    process(line)

or some similar syntax (where the |line|: <block> is a Smalltalk like
block / scheme like closure).  And yes, it would require both a
transparent closure mechanism and probably also gc.

-- bjorn

ps: you forgot the possibility that the user meant a==b==c <wink>



From ge at nowhere.none  Sat Jun 17 19:12:46 2000
From: ge at nowhere.none (Grant Edwards)
Date: Sat, 17 Jun 2000 23:12:46 GMT
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com> <slrn8kn881.sdl.nobody@isis.visi.com> <394BA666.E5F3B19B@uniserve.com> <15O25.1379$i06.308670@ptah.visi.com> <394BC4CD.6432CBBF@uniserve.com> <iIQ25.1400$i06.313245@ptah.visi.com> <394BE556.1D31DC6E@uniserve.com>
Message-ID: <OBT25.1410$i06.318469@ptah.visi.com>

In article <394BE556.1D31DC6E at uniserve.com>, Bob van der Poel wrote:
>Grant Edwards wrote:
>> 
>> I am wondering if BitmapImage is somehow not adding the name of
>> the created bitmap to the global list of "known" bitmaps, or if
>> the name is somehow getting mangled so the name that Tk knows
>> about is different than the one Python presents to the user.
>> 
>> Does anybody know how to see the list of "known" bitmaps?
>
>To tell the truth, I've never used this option in tcl/tk, and I'm just
>learning Tkinter...so, hopefully someone else here can help you.

Now I'm wondering if BitmapImage creates a bitmap at all.  

I rather suspect that it is creating an image and not a bitmap,
so I think I've got to try to figure out how to create a bitmap
under Tkinter.

-- 
Grant Edwards                   grante             Yow!  .. someone in DAYTON,
                                  at               Ohio is selling USED
                               visi.com            CARPETS to a SERBO-CROATIAN


From jhe at webde-ag.de  Wed Jun 21 05:22:32 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Wed, 21 Jun 2000 11:22:32 +0200
Subject: Is there no inversion of gmtime()?
References: <20000620211050.29059.00000981@ng-da1.aol.com>
Message-ID: <8iq1e4$1v9$1@pollux.ip-plus.net>

"Dparsavand" <dparsavand at aol.com> schrieb im Newsbeitrag
news:20000620211050.29059.00000981 at ng-da1.aol.com...
> Looking throught the time module, I'm surprised that there doesn't seem to
be a
> way to go from the known date in UTC to the number of seconds since the
epoch.

There are actually two such functions, time.mktime() and rfc822.mktime_tz().
Now read the docs. :)

Ciao, J?rgen

--
J?rgen Hermann (jhe<at>webde-ag.de)
WEB.DE AG, Amalienbadstr.41, D-76227 Karlsruhe
Tel.: 0721/94329-0, Fax: 0721/94329-22




From bradh at mediaone.net  Fri Jun  2 18:22:43 2000
From: bradh at mediaone.net (Brad Howes)
Date: Fri, 02 Jun 2000 22:22:43 GMT
Subject: Compiling PIL under RH6.2
References: <8h74jt$d9r$1@news2.isis.unc.edu>
Message-ID: <m3ya4nzpjj.fsf@h0050047c4794.ne.mediaone.net>

Eric Frey <frey at weasel.bme.unc.edu> writes:

> I have been trying to compile Python imaging library under RH6.2 (I was
> unable to find a rpm). Unfortunately I am not at the system right now,
> but when I do the make boot (using the appropriate makefile) I get
> an error saying that it is unable to make /usr/lib/python/lib/config/Makefile
> (or something like that). This sounds like a file which would be created
> if I had compiled python myself from the sources (which I didn't do, I 
> installed the RPMs). Is there anyone who has successfully compiled PIL
> under RH6.2? Any hints as to what to do?

Fire up gnorpm, and do a web find for 'python'. You need to install the
Python development RPM. That will install the appropriate files. I had no
problem compiling/installing PIL on RH6.2

Brad

-- 
    "Were people this stupid before TV?" -- _White Noise_ by Don DeLillo


From w_keller at gmx.de  Sat Jun  3 13:53:24 2000
From: w_keller at gmx.de (Wolfgang Keller)
Date: Sat, 3 Jun 2000 19:53:24 +0200
Subject: Stand-alones on the Mac?
References: <393645EF.17424CDA@kfunigraz.ac.at>
Message-ID: <01HW.B55F12B40000462604C5FCF0@news.snafu.de>

On Thu, 1 Jun 2000 13:16:00 +0200, Siegfried Gonzi wrote
(in message <393645EF.17424CDA at kfunigraz.ac.at>):

> And I isntalled also the Mac version of Python and
> also there I could not found anything about Python and stan-alones for
> the Mac.

Then you didn't look very thoroughly.

When I look in my Python folder, there are two applications: one called 
"BuildApplet", one called "BuildApplication".

A doubleclick on the file "Documentation-start-here.html" and you will be 
able to find out what those applications do by reading a little bit (Topic 
"Freezing Python Programms" in "Table of Contents").

-- 
Wolfgang Keller               

Zu Risiken und Nebenwirkungen von Junkmail lesen Sie 
de.admin.net-abuse.mail und fragen sie Ihren Postmaster oder Provider



From phd at phd.russ.ru  Thu Jun 22 10:58:52 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 22 Jun 2000 14:58:52 +0000 (GMT)
Subject: Am I the only one to receive the messages several times????
In-Reply-To: <7BD10B680501D411B9DF009027E06F32011AD2@EXCHANGE>
Message-ID: <Pine.LNX.4.21.0006221457480.13445-100000@fep132.fep.ru>

On Thu, 22 Jun 2000, [iso-8859-1] Max M?ller Rasmussen wrote:
> This starting to be a drag.

   Agree :(

> Furthermore I received a message from BeOpen about one of my messages that
> was to long for their listserver.

   Me too. And these messages were actually a bit old - a week, I beleive.

> Perhaps a BeOpen list is mailing to the python list, and the Python list is
> mailing back to them???? This should cause a loop and make messages appear
> several times.

   Good idea! Time to notify both postmasters...

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From wware at world.std.com  Mon Jun 26 17:36:33 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 26 Jun 2000 21:36:33 GMT
Subject: Final garbage collection?
References: <962028703.1248416950@news.libertysurf.fr> <FwrnM8.MMA@world.std.com> <962045446.1993274611@news.libertysurf.fr> <Fws0qp.14n@world.std.com>
Message-ID: <Fws6ox.AG7@world.std.com>

I (wware at world.std.com) wrote:
> Off the top of my head it
> sounds like it's probably a good idea, but I don't know Guido's
> thinking on the matter.

My questionable skills as a psychic may not be required after all.
There is a lengthy discussion of this topic in the python-dev
mailing list archives, starting with a post from Guido:
http://www.python.org/pipermail/python-dev/2000-March/003869.html
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From Cedric.Lefeuvre at tcomhp43.epfl.ch  Wed Jun 21 05:34:55 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Wed, 21 Jun 2000 11:34:55 +0200
Subject: an help for every function
References: <395078C0.5105D26E@tcomhp43.epfl.ch>
Message-ID: <39508C3F.CFBD9D50@tcomhp43.epfl.ch>

Cedric Lefeuvre wrote:

> i am a beginner in python, and i would be happy if anyone can answer me
> if there is an help or a man function like in unix-like OS that we can
> use to describe any function of python.



From angband at blueyonder.co.uk  Sun Jun  4 06:06:46 2000
From: angband at blueyonder.co.uk (Bablos)
Date: Sun, 04 Jun 2000 10:06:46 GMT
Subject: Print formatting - same line printing?
Message-ID: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk>

Hi There...

Now that I've got my FTP client doing some basic stuff, I now have a
progress report which looks something like the following;

uploaded X bytes
uploaded Y bytes
uploaded Z bytes

Is there a way that I can get Python to print this all on the same
line, so that it looks as if the line is being updated constantly, as
opposed to having a ridiculously long list of 'uploaded...' lines?

Hope someone can help.


Bablos.


From jhauser at ifm.uni-kiel.de  Fri Jun  9 10:15:37 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 09 Jun 2000 16:15:37 +0200
Subject: print without newline
References: <39410382.C83492D2@pfluegel.fsnet.co.uk>
Message-ID: <87r9a7ym7q.fsf@ifm.uni-kiel.de>

The print statement prints to stdout. You can get a handle of the
stdout file object and use it's write method.

>> import sys
>>> sys.stdout.write('test')
test>>> 

HTH,
__Janko

Eckhard Pfluegel <Eckhard at pfluegel.fsnet.co.uk> writes:

> I want to print a string without a newline afterwards. The command
> 
> print "foo",
> 
> doesn't print a newline but prints a trailing space, which is not what I
> want (I wnat *nothing* after the string). I tried the formated output
> 
> print "%s" % ("foo")
> 
> but this prints a newline as well.
> 
> I use Python 1.5.2 on Linux.
> 
> Does anybody have an idea? Thanks very much in advance.
> 
> Eckhard
> 
> 
> 
> 

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From jepler.lnk at lnk.ispi.net  Fri Jun  2 21:21:56 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 03 Jun 2000 01:21:56 GMT
Subject: mixed solution: unicase (unique allowed case) (was: Re: Case-sensitivity: )
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <20000522183552.A29224@mudge> <8gcoij$60n$1@news-int.gatech.edu> <8h92ko$5e1@news.Informatik.Uni-Oldenburg.DE>
Message-ID: <slrn8jgh44.4ln.jepler.lnk@potty.housenet>

On 02 Jun 2000 21:41:12 +0200, Johannes Nix
 <Johannes.Nix at mail.uni-oldenburg.de> wrote:
>Let's play around with the idea!

Remember the common Python idiom for examples:
	>>> class C: pass
	...
	>>> c=C()

Anything that breaks this will offend lots of people, and make all those
cute little examples break.

Jeff


From ajm at enteract.com  Sat Jun 10 11:02:11 2000
From: ajm at enteract.com (Alan Miller)
Date: Sat, 10 Jun 2000 10:02:11 -0500
Subject: Python vs. Ruby: threads - advice appreciated
References: <000001bfc9a9$f724bfe0$b12d153f@tim>
Message-ID: <MPG.13ab18ffd37710e19896a7@news-enteract>

Tim Peters (tim_one at email.msn.com) wrote:
>[Jean-Paul Smets]
>> I wonder if it is possible to go that far with Python, either with
>> native threads
>
>You didn't name your OS, so nobody can try a comparable experiment.  The
>Python below invites 1,000 Philosophers to dinner, and ran fine without
>thrashing on an old 32Mb Win95 box.  I know from experience I can't push
>this particular box much farther than that.

Items worth noting:
 * On my 64MB/133Mhz Win95 box the attached code never had more than 175 
   philosophers actually trying to dine at the same time (based on 
   one thread per philosopher).  Setting things up to have a full 
   thousand threads running at once might cause very different 
   memory usage results.
 * The processor utilitzation for Python never surpassed the processor 
   utilization for MS-DOS, and was generally around half the MS-DOS 
   level, showing the importance of output in performance measures.
 * Running the original with pythonw instead of python, the thread count 
   topped out in the 450 range, and processor utilization (by 
   pythonw) made it as high as 95%.  Unfortunately, at the very end the 
   thread count never dropped below 61, indicating some sort of problem 
   with the app terminating.
 * Running a modified version (all the print statements removed) in 
   python, the thread count topped out in the 725 range and processor
   utilization made it up to 95%; pythonw with the same .py didn't seem 
   to get quite as high.

ajm (another rank beginner with Python)


From offer at sgi.com  Fri Jun 30 15:05:24 2000
From: offer at sgi.com (Nhi Vanye)
Date: 30 Jun 2000 19:05:24 GMT
Subject: Web server executing Python scripts
References: <2B1262E83448D211AE4B00A0C9D61B030130125E@msgeuro1.creo.be>
Message-ID: <8jir1k$d6t5$1@fido.engr.sgi.com>

$ from PClaerhout at CREO.BE -#96973 | sed "1,$s/^/> /"
>
>
>Hello all,
>
>I'm working on a web based system, which consist of a whole
>bunch of Python scripts, which connect to a bunch of machines
>on a network (either using ethernet or dailup networking) and
>execute some to code to for example get things from the
>windows registry of those machines.
>
>What do you think would be the best idea?
>- Use whatever webbrowser with CGI-scripting
>- Use MS IIS with ASP (and Python)
>- Use Zope to build the whole system?
>
>It should also connect to a database running on the host.

I'm working on my site, and I'm using mod_python, which embeds python
in Apache...

There are no HTML pages, everything is generated from the python
script... The site is basically a photo scrapbook for my nephew, the
back end database is MySQL.

>Pieter

richard.
-- 
Richard Offer           Widget FAQ --> http://reality.sgi.com/widgetFAQ
MTS-Core Design (Motif)
___________________________________________http://reality.sgi.com/offer



From aahz at netcom.com  Wed Jun 14 17:41:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 21:41:56 GMT
Subject: Simple thread prob
References: <n67X4.13491$XW1.115847@typhoon.hawaii.rr.com>
Message-ID: <8i8u74$gl0$1@slb7.atl.mindspring.net>

In article <n67X4.13491$XW1.115847 at typhoon.hawaii.rr.com>,
Ty K. Kroll <tkroll at hawaii.edu> wrote:
>
>    I'm trying to build a thread that will timeout. I've only used simple
>threading in Python. I'm thinking wait() in Event or join() in Thread may be
>helpful, but the documentation is sparse. Can anyone point me in the right
>direction? Thanks. -Ty

Why do you want the thread to time out?  What is the purpose?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From db3l at fitlinxx.com  Wed Jun 14 19:11:08 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 19:11:08 -0400
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com>
Message-ID: <uln07luyb.fsf@ctwd0143.fitlinxx.com>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> I'm not sure if this is feasible in general, but even a half-hearted job
> would still be helpful.

How about just writing a function to verify methods against a
reference class.  The reference class could either be an actual class
that you are mimicing the interface for, or a dummy class that is just
defined with the appropriate methods.

For a crude quick example, perhaps something like:

    class MyInterface:

	def method1():
	    pass
	def method2():
	    pass
	def method3():
	    pass


    class MyImplementingInterface:

	def method1():
	    # Actual method 1 implementation
	    pass
	def method3():
	    # Actual method 3 implementation
	    pass


    def verify_class (Reference, Instance):

	import types

	print "Checking %s for conformity with interface %s" % \
	      (Instance.__name__,Reference.__name__)

	for curmethod in Reference.__dict__.keys():
	    if type(Reference.__dict__[curmethod]) == types.FunctionType:
		if (not Instance.__dict__.has_key(curmethod) or
		    type(Instance.__dict__[curmethod]) != types.FunctionType):
		    print "Method %s not supported" % curmethod


    if __name__ == "__main__":

	verify_class(MyInterface,MyImplementingInterface)


when run, yields

Checking MyImplementingInterface for conformity with interface MyInterface
Method method2 not supported


I'm sure there's lots more that could be done - the dummy interface
class might provide special members to control optional functions, or
an interface could just be specified as a list, or you could also try
to check the signature of the function, etc...


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


From thiele at muc.das-werk.de  Wed Jun 28 11:53:34 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 28 Jun 2000 17:53:34 +0200
Subject: Pickle for C++?
Message-ID: <395A1F7E.4112B2A3@muc.das-werk.de>

Hi!

I have a Python class that holds a lot of data. 
And I now want to send these datas over a TCP - connection via
pickle and sockets. It works fine. But now I want
that an application written in C++ is able to unpickle the data-class
too and write it's content into an equivalent C++ - class/struct. 

Is there any solution out there or have I to write my own C++ - pickle
algorithm?


From hzhu at mars.localdomain  Thu Jun 15 19:28:39 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Thu, 15 Jun 2000 23:28:39 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net> <3947CA61.D5A6CF8F@muc.das-werk.de> <slrn8kfsda.2p3.hzhu@rocket.knowledgetrack.com> <uln07luyb.fsf@ctwd0143.fitlinxx.com> <slrn8kg7gp.303.hzhu@rocket.knowledgetrack.com> <ug0qflns9.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8kipri.pf.hzhu@mars.localdomain>

On 14 Jun 2000 21:45:58 -0400, David Bolen <db3l at fitlinxx.com> wrote:

>Ah, good point - try this variant - I switched to getattr() which
>takes the inheritance into account (you could also write your own
>recursive function that used Instance.__bases__ to look in the
>superclasses).

Wow!  It looks so easy that I can't resist yet another little improvement:

"""
Verify.py - verify a class against a template.  Also indicates where a
method is implemented and prints type mismatch.
Adoped from: 
    Newsgroups: comp.lang.python
    Subject: Re: SV: Python Productivity over C++
    From: David Bolen <db3l at fitlinxx.com>
    Date: 14 Jun 2000 21:45:58 -0400
"""

def verify_class (Reference, Instance):
    import types

    print "Checking %s for conformity with interface %s" % \
          (Instance.__name__,Reference.__name__)

    for method in Reference.__dict__.keys():
        if type(Reference.__dict__[method]) == types.FunctionType:
            try:
                actual_method = getattr(Instance,method)
                if type(actual_method) != types.MethodType:
                    raise TypeError
                else:
                    print "Method %s is implemented in %s" % (
                        method, actual_method.im_class)
            except AttributeError:
                print "Method %s is not implemented" % method
            except TypeError:
                print "Method %s is has a different type" % method

#------------------------------------------------------------------
if __name__ == "__main__":
    class MyInterface:
        def method1():       pass
        def method2():       pass
        def method3():       pass
        def method4():       pass
        def method5():       pass
    
    class MyBase:
        "Actual implementations"
        def method1():       pass
        def method2():       pass
    
    
    class MyDerived(MyBase):
        "Actual implementations"
        def method1():       pass
        def method3():       pass
        method5 = 1

    verify_class(MyInterface,MyBase)
    verify_class(MyInterface,MyDerived)

==================================================================
Result:

Checking MyBase for conformity with interface MyInterface
Method method1 is implemented in __main__.MyBase
Method method3 is not implemented
Method method2 is implemented in __main__.MyBase
Method method5 is not implemented
Method method4 is not implemented
Checking MyDerived for conformity with interface MyInterface
Method method1 is implemented in __main__.MyDerived
Method method3 is implemented in __main__.MyDerived
Method method2 is implemented in __main__.MyBase
Method method5 is has a different type
Method method4 is not implemented




From jcashman55 at my-deja.com  Wed Jun  7 10:49:05 2000
From: jcashman55 at my-deja.com (jcashman55 at my-deja.com)
Date: Wed, 07 Jun 2000 14:49:05 GMT
Subject: Win32net.NetServerEnum Unicode problem
Message-ID: <8hlnd2$lrm$1@nnrp2.deja.com>

Having trouble with the following line:

        entries, total, resume= \
win32net.NetServerEnum('Dopey', \
102,win32netcon.SV_TYPE_ALL, 0, 4096)

When I run code that contains this line, I get:

TypeError: The object can not be converted to a
Unicode object

I have run some of the the other win32net calls
without a problem.  Anyone have an idea?

Thanks.

Jim



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


From srichter at cbu.edu  Fri Jun  2 16:51:33 2000
From: srichter at cbu.edu (Stephan Richter)
Date: Fri, 02 Jun 2000 13:51:33 -0700
Subject: [ANN] Zope at Linuxtag 2000 in Stuttgart
Message-ID: <4.3.1.0.20000602131515.00aae710@198.78.130.6>

Hello everyone,

A couple of days back Martijn Faassen and I had the idea to organize a 
European Zope conference. Maik Roeder reminded me of the upcoming Linuxtag 
2000 in Stuttgart. So I contacted the Linuxtag staff and we will have a 
booth. Martin Schulze, my main contact, will  also try to get a Zope 
Conference setup for us, which would give us 2 rooms for one day fitting 50 
people per room, where we can have talks and discussions.

The Linuxtag will be from June 29th (Thursday) to July 2nd (Sunday). It 
will be in the Exhibition Center (Messegelaende) in Stuttgart, Germany.

I know we have little time to organize the conference (26 days). We have 
one day to fill with speeches and discussions, so please let me know if 
there are people willing to give some talks or want to have "Special 
Interest Group" discussions, like E-commerce, XML or Portals in Zope. I 
would also like to know, who would be able to come, so I can give the 
Linuxtag staff some feedback.

Here are some subjects I thought we could have: The ZDP Portal (Maik 
Roeder, ....), XML in Zope (I know Martijn F. did a lot of work there. ;-), 
An Introduction to Zope for newbies (Stephan Richter -- me), ZClasses, 
Python Products for Zope and so on.

I also would like to hear some talks and discussions about Python products 
that Zope uses, such as PyGres, PIL, mxDateTime and other modules. I would 
be very glad, if some Python-only people would show some interest too. :-)

The speeches and discussion can be held in German and English (preferred, 
since we expect people from other countries).

So, everyone who is interested in preparing a talk or discussion, please 
send me an E-Mail and I will compile it to a schedule. The rooms will be 
open for us between 9:00 am - 8:00 pm on Thursday the 29th.

Let's make this a successful event!

Regards,
Stephan

--
Stephan Richter
CBU - Physics and Chemistry
Web2k - Web Design/Development & Technical Project Management




From jovecka at dallas.net  Thu Jun  1 12:44:43 2000
From: jovecka at dallas.net (Jim)
Date: Thu, 01 Jun 2000 11:44:43 -0500
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and others" 
 is this possible?
Message-ID: <393692FB.8610EBCA@dallas.net>



Dear list,

I'm new to Python, and I'm sorry if I'm asking about anything that has
been
already discussed.

I would like to use Python to call commands, e.g. "find , perl scripts
, nmap , and others" is this possible?

I do not need GUI. All I need is to link the out put and work with it.
Is perl or shell scripts my only
answer?

jim

--
Proprietary & Confidential.  The information transmitted is intended only for the person or entity to whom it is addressed and may
contain confidential and/or privileged material.  Any review, retransmission, dissemination, or other use of, or taking of any
action in reliance upon, this information by persons or entities other than the intended recipient is prohibited.  If you received this in error, please
contact the sender and delete the material from any computer.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000601/d69e774c/attachment.html>

From aahz at netcom.com  Thu Jun 29 09:28:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jun 2000 13:28:51 GMT
Subject: exiting threads
References: <8jfcf9$2rp$1@nnrp1.deja.com>
Message-ID: <8jfiuj$lf1$1@slb1.atl.mindspring.net>

In article <8jfcf9$2rp$1 at nnrp1.deja.com>,
Dale Burnett  <denalione at my-deja.com> wrote:
>
>I am using thread.start_new_thread(self.foo,(x,y))
>
>Do I have to explicitly exit the thread or does it automatically exit
>when foo returns?

I believe it does, but I strongly recommend that you subclass
threading.Thread instead.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From jelizondo at my-deja.com  Wed Jun 21 15:24:48 2000
From: jelizondo at my-deja.com (jelizondo at my-deja.com)
Date: Wed, 21 Jun 2000 19:24:48 GMT
Subject: Tabing and Carriage Returns in Windows Applications
Message-ID: <8ir4q0$asr$1@nnrp2.deja.com>

Is there a way to code in tabs and carriage
returns after you've loaded up a windows
application?


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


From spam at spam.com  Mon Jun 26 18:51:12 2000
From: spam at spam.com (Penfold)
Date: Mon, 26 Jun 2000 23:51:12 +0100
Subject: Script which only runs in Windows
References: <8j8d2n+qbe9@eGroups.com>
Message-ID: <l6R55.2530$Di7.376473@nnrp4.clara.net>

The problem lies here

> try:
>     sResult = os.system('python ftpasget.py -q -g * -d ' + sFileTime
> + ' -l ' + sSugLogin + ' -p ' + sSugPWord + ' ' + sSugSite + ' ' +
> sBillDir + ' ' + sOutputDir + sDirTime)

[which might have been easier for you to type as

try:
    sResult = os.system('python ftpasget.py -q -g * -d %s -l %s -p %s %s %s
%s%s' % \
                        (sFileTime, sSugLogin, sSugPWord, sSugSite,
sBillDir, sOutputDir, sDirTime)
except:
    sGetMinutes = 0

] :-)

Probable cause:
An exception is happening in os.system, who knows what, as you are catching
absolutely everything.  Because this exception occurs,
sResult never gets defined !!.  Hence when you refer to it later, you get
the problem.  Basically the script as it stands expects sResult to exist
no matter what, and it wont if you get an exception in the above block.

Probable exception? python cant be found in the search path or ftpasget cant
or something like that.  To find out, remove the try and except block and
run it, and you'll find out.  Windows, as a rule, runs anything in system
without fail, cant remember for *nix.

Or, try something like the following, which is slightly restructured.

import os, sys, time, string, sugdb, gadfly

sBillDir, sSugLogin, sSugPWord, sSugSite = "dirname", "login", "password",
"171.1.1.1"

# The following Dirs MUST end in a path slash
SOutputDir = "/home/DATA/"
sDBLocation = "/usr/local/bin/db/"

# The script is called without any passin values.
# It starts by getting the date and extrapolating strings to use for  file
# dates
# Next, the FTPASGET script is called and its return value is checked
# If all was well, then the DB processing script is called
# The following declarations are very important.
# This reads the date and creates the strings for the following

iTime = time.time()

# This is the format string for the date of the files we need to look
#for (like: 06/09)

sFileTime, sFileName, sDirTime, sDateStart, sProcStart = \
           map(time.strftime, ["%m/%d", "%m%d", "%d%b", "%m-%d-%y",
"%H:%M:%S"], [time.localtime(iTime)]*5)

#
# here is where we do some work
#
try:
    sResult = os.system('python ftpasget.py -q -g * -d %s -l %s -p %s %s %s
%s%s' % \
                        (sFileTime, sSugLogin, sSugPWord, sSugSite,
sBillDir, sOutputDir, sDirTime)
except:
    sGetMinutes = 0
else:
    iCurrTime = time.time()
    if sResult == 0:
        sGetMinutes = (iCurrTime - iTime) / 60
    else:
        sGetMinutes = 0
    iTime = iCurrTime

Des.

<al at servana.com> wrote in message news:8j8d2n+qbe9 at eGroups.com...
> Hello.
>
> We have an odd problem with Python which we cannot resolve.
>
> Below, is a script we have been running in Win32 for months.
>
> We moved our scripts from Win32 to Slackware Linux and IF statements
> which
> are not in functions or classes will cause errors when you
> try to run the script.
>
> I have had retyped the script thinking it was the tab/space
> issue, but nothing changes - in Linux we cannot use IF outside
> a function or CLASS for some reason.
>
> We are posting this here as it seems very serious and we were hoping
> someone else had seen the problem.
>
> -----------------------------------------
> import os
> import sys
> import time
> import string
> import sugdb
> import gadfly
>
> sBillDir = "dirname"
> sSugLogin = "login"
> sSugPWord = "password"
> sSugSite = "171.1.1.1"
>
> #
> # The following Dirs MUST end in a path slash
> #
> SOutputDir = "/home/DATA/"
> sDBLocation = "/usr/local/bin/db/"
>
> #
> # The script is called without any passin values.
> #
> # It starts by getting the date and extrapolating strings to use for
> file
> # dates
> #
> # Next, the FTPASGET script is called and its return value is checked
> #
> # If all was well, then the DB processing script is called
> #
>
> #
> # The following declarations are very important.
> # This reads the date and creates the strings for the following
> #
>
> iTime = time.time()
>
> # This is the format string for the date of the files we need to look
> for (like: 06/09)
> sFileTime = time.strftime("%m/%d", time.localtime(iTime))
> sFileName = time.strftime("%m%d", time.localtime(iTime))
> sDirTime = time.strftime("%d%b", time.localtime(iTime))
> sDateStart = time.strftime("%m-%d-%y", time.localtime(iTime))
> sProcStart = time.strftime("%H:%M:%S", time.localtime(iTime))
>
> #
> # here is where we do some work
> #
> try:
>     sResult = os.system('python ftpasget.py -q -g * -d ' + sFileTime
> + ' -l ' + sSugLogin + ' -p ' + sSugPWord + ' ' + sSugSite + ' ' +
> sBillDir + ' ' + sOutputDir + sDirTime)
> except:
>     sGetMinutes = 0
>
> iCurrTime = time.time()
>
> #
> # PROBLEM: The following will fail unless in a function
> # The failure error is:
> #      Name Error: sResult
> #
> if sResult == 0:
>     sGetMinutes = (iCurrTime - iTime) / 60
> else:
>     sGetMinutes = 0
>
> iTime = iCurrTime
>
> ------------------------------------
> If we run the code in Win32, or from a function in Linux, it works
> without error.
>
> Any suggestions would be priceless.
>
>




From ge at nowhere.none  Sat Jun 24 16:08:51 2000
From: ge at nowhere.none (Grant Edwards)
Date: Sat, 24 Jun 2000 20:08:51 GMT
Subject: Controlling an external program with Python
References: <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi>
Message-ID: <nz855.1583$iN5.265679@ptah.visi.com>

In article <kuzoobue0g.fsf at lasipalatsi.fi>, Erno Kuusela wrote:
>>>>>> "Tony" == Tony Keating <keating at mech.uq.edu.au> writes:
>
>    Tony> start the program, then send various strings to the program,
>    Tony> read some output from the program then send some more
>    Tony> strings to the program.
>
>there are a couple of expect modules for python (so named because
>there is an old tcl-based language called expect for doing just this).
>parnassus lists 3 modules (<URL:http://www.vex.net/parnassus/
>apyllo.py?find=expect>), i haven't
>used any of them however...

Or you can roll your own using the popen2 module.

-- 
Grant Edwards                   grante             Yow!  hubub, hubub, HUBUB,
                                  at               hubub, hubub, hubub, HUBUB,
                               visi.com            hubub, hubub, hubub.


From damonNOdaSPAM at unispace.com.invalid  Wed Jun 14 01:54:30 2000
From: damonNOdaSPAM at unispace.com.invalid (damonr)
Date: Tue, 13 Jun 2000 22:54:30 -0700
Subject: Python
Message-ID: <03031608.695276c1@usw-ex0105-036.remarq.com>

Heya,
I just added a site directory for Python related sites to my
site (http://www.devcritic.com) and was wondering if you guys
had any sites to add, either by using the "add a site" link on
the menu on the side or recommend them here and i will add them.
The url is: http://www.devcritic.com/sites/Python/ .

Thanks in advance


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



From kens at sightreader.com  Wed Jun  7 19:32:26 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 07 Jun 2000 16:32:26 -0700
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <Pine.SGI.4.05.10006061254220.22654-100000@lemur.magnet.com> <393DD854.56032E87@sightreader.com> <393DF80A.BB53465C@prescod.net>
Message-ID: <393EDB8A.AEAC73A7@sightreader.com>

Paul Prescod wrote:

> Ken Seehof wrote:
> >
> > Java definitely has a niche: embeded software for household appliances.
>
> I really don't see it having particular virtues there. Forth is more
> efficient and more runtime flexible. Python is easier to code. Scheme is
> smaller to implement. C has better real-time characteristics. Any of
> them can speak TCP/IP.

Dang!  Not even good for what it was designed for?  :-)

> Java is the language next door: nothing particularly wonderful, nothing
> particularly terrible. Just there. On the other hand, if I am forced to
> program in a non-Python language and have to choose between Perl, C, VB
> and Java, Java is the least of all evils.
>
> --
>  Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
> Alabama's constitution is 100 years old, 300 pages long and has more
> than 600 amendments. Highlights include "Amendment 393: Amendment of
> Amendment No.  351", "Validation of Laws Regulating Court Costs in
> Randolph County", "Miscegenation laws", "Bingo Games in Russell County",
> "Suppression of dueling".  -
> http://www.legislature.state.al.us/ALISHome.html

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000607/b586a4ab/attachment.vcf>

From coronel_jadonNOSPAM at teleline.es  Wed Jun 21 04:47:29 2000
From: coronel_jadonNOSPAM at teleline.es (JD)
Date: Wed, 21 Jun 2000 08:47:29 GMT
Subject: Lista de correo sobre Python
Message-ID: <Bi%35.2093$Hc7.11930@telenews.teleline.es>

Hola a todos:

Escribo para comunicar que se ha creado una lista de correo sobre Python en
Espa?ol. En la lista contamos experiencias, noticias, consultamos
dudas,etc.. Todo sobre Python y ademas en Espa?ol.

Para suscrbirse hay que enviar un correo a la direcci?n:
python_dev-subscribe at egroups.com

Un saludo a todos y gracias.

Thank you for all.




From bparsia at email.unc.edu  Mon Jun 19 06:52:50 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Mon, 19 Jun 2000 06:52:50 -0400
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu>

William Dandreta <wjdandreta at worldnet.att.net> wrote:

[snip]
> The biggest improvement came (about a factor of 6) when I changed the
> replace(x,y) function in string with
> joinfields(splitfields(x,y),''). Considering that this is exactly what the
> replace function in string does, I was quite surprised at the results.
> Essentially I reduced 3 nested function calls to 2.

And that can make an enormous difference, particularly with earlier
versions. Remember that many "simple function calls" in fact involve
rather expensive polymorphic lookup and dispatching. In a tight loop
this can get very painful (Python doens't have, I believe even in the
latest versions, a method/function cache, so every lookup will go
through the whole rigamarole each time).

> The only thing that makes sense to me is that Python is spending 85% of the
> time processing the comment lines and 15% of the time doing useful work.
> There is a string.pyc so I would have assumed that it would not contain any
> comments or descriptive strings but apparently that's not true.

Well, as someone pointed out, there's a simple test for this :)

And here's another, try binding replace to a local var (outside your
loop), i.e., something like
        x=string.replace
and see if you get a comparable speedup.

> I am not very familiar with interpretive languages, Python is the first one
> I have used, but it seems that comments can cause a serious perfomance hit
> because they have to be processed along with the code.

Er...even in more straightforwardly interpreted *implementations* of
various languages, what gets interpreted is oftena  tokenized form of
the program or module which would have stripped out, or otherwise
reduced, the comments.

But, in any case, *what* Python interprets is Python byte code, not
Python source. Thus, while a compilation phase might add overhead to
your entire program, it shouldn't affect inner loops (unless you're
doing something so incredibly perverse that my poor mind couldn't handle
it ;))

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From MarkH at ActiveState.com  Thu Jun 22 19:23:31 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Jun 2000 23:23:31 GMT
Subject: Labels of remote volumes under NT
References: <2B1262E83448D211AE4B00A0C9D61B0301301218@msgeuro1.creo.be>
Message-ID: <Tdx45.1837$Sn2.5220@news-server.bigpond.net.au>

Almost all of your answers can be found with a quick deajnews search:

* List of processes: win32pdhutil

* List of devices: No Python functions, look in the registry

* VersionStamp: No - there was a rumour about one, but it never appeared.

* GetPrinter - Use the source, Luke.

* CurrentUser - win32wnet, win32api.GetCurrentUser, etc

* Labels mapped drives, something in win32api, otherwise win32net.

Mark.

--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!
"Pieter Claerhout" <PClaerhout at CREO.BE> wrote in message
news:2B1262E83448D211AE4B00A0C9D61B0301301218 at msgeuro1.creo.be...
> > Hello,
> >
> > is it possible with Python under NT, to find out the type and the
labels
> > of the drives of a remote computer?
> >
> > Kind regards,
> >
> >
> > Pieter
> >
> > Pieter Claerhout
> > Application Support - CreoScitex Europe
> >
>




From cut_me_out at hotmail.com  Fri Jun 16 21:45:44 2000
From: cut_me_out at hotmail.com (Alex)
Date: 16 Jun 2000 21:45:44 -0400
Subject: Are list mutations thread-safe?
References: <etd8zw5qfdv.fsf@w20-575-109.mit.edu> <8iekpr$inp$1@nntp9.atl.mindspring.net>
Message-ID: <etd7lbpqdvb.fsf@w20-575-109.mit.edu>

Great!  Thanks for your help, Aahz.

Alex.


From buttnerw at my-deja.com  Wed Jun 21 00:40:58 2000
From: buttnerw at my-deja.com (buttnerw at my-deja.com)
Date: Wed, 21 Jun 2000 04:40:58 GMT
Subject: EMail Monitoring in Linux
Message-ID: <8iph0p$ekl$1@nnrp1.deja.com>

Good Day

I need to know the easiest way to do the following.  Whenever the
mailing program (sendmail) on a mail server (linux) receives a mail for
a user, a "program" needs to be notified about the incoming mail and a
UDP message has to be send with the user's information and the amount
of mails in his inbox.  All I need is the "program" which will listen
to incoming mails, and send the UDP packet.

Kind Regards


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


From aahz at netcom.com  Wed Jun  7 18:20:52 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 22:20:52 GMT
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com>
Message-ID: <8hmhs4$psc$1@slb7.atl.mindspring.net>

In article <393fbb15.100053125 at nntp.interaccess.com>,
Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:
>
>Is there anything like CPAN for python?

Not directly.  We do have the Vault at
http://www.vex.net/parnassus/
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From thomas at xs4all.nl  Thu Jun 29 10:13:31 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 29 Jun 2000 14:13:31 GMT
Subject: newbie questions about imports, threads, and optimizations
References: <8jd7f1$kic$1@murrow.sp.trw.com>
Message-ID: <slrn8lmmbp.qe9.thomas@centurion.xs4all.nl>

On Wed, 28 Jun 2000 09:10:04 -0700, Bryce Sellers <bryce.sellers at trw.com> wrote:
>I have been using python off and on (unfortunately, mostly off) for about a
>year now, and have got a couple of questions for someone who knows more than

>First, is there a way to do something like this:

>import string, mystring
>m = 'string'
>`m`.captalize('hello')
>m='mystring'
>`m`.capitalize('hello')
>
>In other words, use a string as a module or function name and execute that
>portion of code.

Not sure what you mean, but you can do this:

eval('string').capitalize('hello') # 'string' could turn out any object.

or this:

sys.modules['string'].capitalize('hello') # 'string' is always a module.

Of course, using string literals isn't that exciting. It also isn't that
exciting if you just want to pass the module name to a function, because you
can pass & assign a module just like any other object. But it might be handy
if you want a user to input a module- or object-name.

>Finally, I've noticed that a lot of the discussions surrounding
>optimizations center on the fact that even the standard library function
>names can be overwritten.  Why can't we just have a command-line argument
>(or something along those lines) that tells python that all variables
>defined in modules are read-only?  Now the interpreter can optimize the
>heck out of it, and throw an exception if anything tries to modify
>something.  Of course, this idea is so simple that I'm sure the only reason
>it hasn't been done is because its a bad idea.  My question is, why is it a
>bad idea?

Wrong assumption. It hasn't been done because it's pretty hard, and a fair
lot of work to boot. Most of Pythons' internal structures are really normal
Python objects. Namespaces (other than function-local namespaces) are really
normal PyDicts, and 'code objects' really are tuples of various items, etc.
This makes it really usable, and makes the introspection abilities almost
'free', but it does mean you can't just change all those things :-)

You have a point though. After the discussion on optimization last week, I
went on thinking about other possibilities: inlining functions, turning
functions into 'macro' calls, stuff like that.

Eventually I realized it might be perfectly possible to compile Python into
native machine code -- by creating a completely static binary, including all
modules. The compiler would have to keep tabs on symbols that get changed in
Python, but that's possible because you're building a completely
'stand-alone' binary. For those symbols, it'd have to create some level of
indirection, so code like that could still work. The places where those
symbols were not assigned to, you can apply the usual array of agressive
optimizers ;)

I haven't yet found a reason why this couldn't work. You'll lose the ability
to influence the program like Python code has now, ending up with a language
not much more complex than C++, so in my completely uneducated opinion, the
Python CompleteCompiler shouldn't have to be more complicated than a usable
and conforming C++ compiler (note, I said 'and', so I'm not exaggerating. ;)
And of course you can start with a compiler that barfs when you assign to a
module or do nifty things like exec or eval() ;)

However, as I've said before, I know didley-squat about real-world compilers.
I think this would be a really cool thing to work on, even though it
basically requires writing a Python implementation from scratch, eventually
designing a way for existing extention modules to be compiled into
extention-code that can be linked with the compiled-python programs and
stuff like that.

Thomas.


From colinmeeks at home.com  Wed Jun 28 14:02:52 2000
From: colinmeeks at home.com (ColBoy)
Date: Wed, 28 Jun 2000 18:02:52 GMT
Subject: Using zipfile.py
References: <3%d65.680$ZI2.26352@news1.rdc1.on.wave.home.com>
Message-ID: <g5r65.2661$ZI2.59633@news1.rdc1.on.wave.home.com>

After much tinkering, I finally managed to get the syntax right.  Just
thought I'd
post my snippet here in case anyone is interested.

import zipfile
zzz=zipfile.ZipFile('c:\\download\\new\\x2000.zip', "r")
zzz.printdir()
print zzz.read('license.txt')

I imagine I can open a binary file in a zip file and write it out to disk
somewhere,
do I have to use the standard file operations to do this. It's taking me a
while to
get to grips with things, mainly because I haven't got any books on Python
yet,
although they are on order.  I've ordered Python Essentials and Python
Pocket
Reference, hopefully they will get me up to speed. My main experience is in
Clipper and Delphi.  I originally wanted to add Perl to my skill set, but I
just
haven't been able to get into it, in the same way I have with Python.

Colin


ColBoy <colinmeeks at home.com> wrote in message
news:3%d65.680$ZI2.26352 at news1.rdc1.on.wave.home.com...
> Could anybody give me a dummies guide to using the zipfile.py module.
Just
> a couple of examples of viewing a zips contents and extracting all of a
zips
> contents.  I know this is probably dead simple, but I'm still working my
way
> around Pythons syntax for stuff.  Many thanks in advance.
>
> Regards
>
> Colin
>
>




From mwh21 at cam.ac.uk  Thu Jun 22 05:07:21 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 2000 10:07:21 +0100
Subject: url directory listing
References: <39518289.D4175618@bioeng.ucsd.edu>
Message-ID: <m3vgz23wza.fsf@atrus.jesus.cam.ac.uk>

Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:

> I little while ago, there was a post about websucker.  I'm writing a
> similar script. What is the socket command or url command to get a
> directory listing?  Especialy if there is an index.html file in the
> directory.  Thanks.

I'm fairly sure there isn't one.

Cheers,
M.

-- 
48. The best book on programming for the layman is "Alice in
    Wonderland";  but  that's because it's  the best book on 
    anything for the layman.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From npk1977 at my-deja.com  Tue Jun 20 10:23:44 2000
From: npk1977 at my-deja.com (npk1977 at my-deja.com)
Date: Tue, 20 Jun 2000 14:23:44 GMT
Subject: Determining Types in a String
Message-ID: <8inuov$8r2$1@nnrp1.deja.com>

Hi all, I am trying to load a matrix of floats, ints and strings into
an array.

I have a method that works, but is slow, the psuedocode is:

v = string.split(line)
for items in v:
	Attempt to recast item as an int(), if that fails, catch the exception
and try to recase as long(), keep doing this until I can go no further.

 6 def Recast(v):
 7         v = string.split(v)
 8         for i in range(0,len(v)):
 9                 try: v[i] = int(v[i])
10                 except Exception:
11                         try: v[i] = long(v[i])
12                         except Exception:
13                                 try: v[i] = float(v[i])
14                                 except Exception:
15                                         try: v[i] = str(v[i])
16                                         except Exception:
17						FAIL

Thanks!
n


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


From oeschger at netscape.com  Fri Jun 30 15:45:20 2000
From: oeschger at netscape.com (Ian Oeschger)
Date: Fri, 30 Jun 2000 12:45:20 -0700
Subject: reportlab: HTML->PDF via Python?
Message-ID: <395CF8D0.6FD580B9@netscape.com>

Does reportlab or some other Python module offer a facility for printing
HTML pages to PDF?
And if so, does some rendering engine have to intercede in this process
somewhere? My experience so far has been getting text--rather than the
rendered HTML--into PDF.

Thanks a lot for any information or suggestions about this.
-ian

--
Ian Oeschger
Principal Writer, Netscape




From ge at nowhere.none  Fri Jun 30 14:24:06 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 30 Jun 2000 18:24:06 GMT
Subject: Python MP3 Player
References: <395CE50A.3BA16777@bioeng.ucsd.edu>
Message-ID: <aB575.2343$iN5.538590@ptah.visi.com>

In article <395CE50A.3BA16777 at bioeng.ucsd.edu>, Curtis Jensen wrote:

>Does anyone know of a Python MP3 Player?

Surely you're kidding?  

Do people really do real-time DSP type stuff in Python.

-- 
Grant Edwards                   grante             Yow!  Is a tattoo
                                  at               real, like a curb or a
                               visi.com            battleship? Or are we
                                                   suffering in Safeway?


From swun at esec.com.au  Thu Jun 15 21:49:13 2000
From: swun at esec.com.au (Sam Wun)
Date: Fri, 16 Jun 2000 11:49:13 +1000
Subject: How does python invoke external C function?
Message-ID: <39498799.D28EDD46@esec.com.au>

Hi,

I would like Python call an external C function ( ie. char
*getString()),
Does anyone knows how to do that?

Thanks
Sam.




From shecter at darmstadt.gmd.de  Fri Jun 16 09:13:34 2000
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Fri, 16 Jun 2000 15:13:34 +0200
Subject: Suggestion: "Completing" HTTP support in Python.
References: <Pine.LNX.4.21.0006151553280.8145-100000@fep132.fep.ru> <3949028F.3B512AF9@darmstadt.gmd.de>
Message-ID: <394A27FE.3CA5D365@darmstadt.gmd.de>

(I'm cc'ing you because I'm posting to the other thread where the
conversation has been...)

>I wrote:
>> Wishful example #1
>> --------------
>> h = HTTP('my.host.name', '/loginscript', 8080)
>> h.add_parameter('username', 'Robb')
>> (code, msg, headers) = h.post()
>>
>> Wishful example #2
>> --------------
>> h = HTTP('my.host.name', 8080)
>> h.add_header('X-local-hack', 'true')
>> (code, msg, headers) = h.get()

You wrote:
>Surely this level is higher than HTTP.

I disagree.  I think that an HTTP request can be seen as a black box
with these "knobs", or inputs:

Destination URL (host, port, path)
A set of parameters
A set of headers
A method (GET, POST ... I'm not interested in the others)

...and then, this black box should have a big "send" button on it,
which results in the usual python 3-tuple popping out.  All of these
items are specified in the RFC.

>The way in which forms are posted...

I wasn't thinking about forms at all.  Maybe everyone else here is,
and that's fine.  I've built an XML/HTTP middleware system that uses
POST's with parameters for requesting remote procedure calls.
Something like XML-RPC, but more intelligent. :)

>If you are looking for a more high level approach then why not try
>something like this:
>http://cobweb.quantisci.co.uk/pub/english.cgi/d8014813/cl_forms.py

Hmm... maybe I could hack that up and use some parts of it, but it
looks very HTML form-specific.

>I for one, would be interested in this kind of functionality for the
>library - my own attitude has been to simply "knock one up" as and
when
>I've needed it.

Exactly - I'd guess that everybody's been (re-)doing that kind of
work.  When I have something to show, I'll post it...

- Robb



From pecora at anvil.nrl.navy.mil  Wed Jun 14 07:16:50 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Wed, 14 Jun 2000 11:16:50 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu>
Message-ID: <140620000716512992%pecora@anvil.nrl.navy.mil>

In article <etdk8ftm55x.fsf at w20-575-109.mit.edu>, Alex
<cut_me_out at hotmail.com> wrote:

> Wow, thanks for looking that code over, Bernhard.  What a horrible bunch
> of holes.


And who could have seen all that so clearly without an expert on hand. 
Even then I couldn't understand all the reasoning.  In Python we are
all happy, but the ugly side is when we return to C.  It's obvious to
me:

CAN'T SOMEONE UPDATE THE EXTENSIONS PDF MANUAL, ADDING A BETTER
TUTORIAL ON REFERENCE COUNTS AND (!!) LISTING Py_XXXX FUNCTIONS AND
THEIR REFERENCE EFFECTS ??

Sorry to yell, but it's obvious that many people like me are stumbling
with this one.  I'd hate to have all that time I save using Python
eaten up by debugging C extensions.


From h__l at my-deja.com  Fri Jun 23 13:34:59 2000
From: h__l at my-deja.com (h__l at my-deja.com)
Date: Fri, 23 Jun 2000 17:34:59 GMT
Subject: python optimization
References: <8iu5dt$s6b$1@nnrp1.deja.com> <8iu7t0$5jg$1@slb7.atl.mindspring.net>
Message-ID: <8J0742$AOI$1@NNRP1.DEJA.COM>

In article <8iu7t0$5jg$1 at slb7.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> >I know of -O0  <turns it off?>
> >and also -O2 <default it seems for the RPM I d/l for python 1.5>
>
> Nope, -00 should be the same as -02; they both strip docstrings from
> .pyc files for slightly faster loading (but not execution).  That's it
> for optimizations.  (Unfortunately, I can't find docs for this right
> now; I'm not even sure that -O2 is legal.)
>

Well what I  have done is recompiled the python interpreter it would
seem.  This SHOULD be valid with optimizations turned on in full but
only for the interpreter as it is done with gnu gcc.  Unfortunately
after installing the recompiled interpreter I seem to have lost zope
which I was running on and off.  Now I am trying to fix that by redoing
the interpreter in -O2 as it was before (with no luck I might add).
LOL, how's that for optimization?

hmmm so far I feel I am barking up the wrong tree.

ME


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


From eviltofu at rocketmail.com  Thu Jun 29 18:16:49 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Thu, 29 Jun 2000 22:16:49 GMT
Subject: ExitWindowsEx()
References: <8jg1mk$e1s$1@nnrp2.deja.com>
Message-ID: <eviltofu-35CE47.18164929062000@news.earthlink.net>

In article <8jg1mk$e1s$1 at nnrp2.deja.com>, npnigro at my-deja.com wrote:

> I'm having a problem executing ExitWindowsEx() to
> force reboot my Win95/Win98 machine.  The
> following is my code:
> 
> import win32api
> win32api.ExitWindowsEx(4 Or 2,0)
> 
> The call does not reboot the machine as it is
> supposed to.  Am missing something?  I'm using
> build 132... Thanks in advance
> 
> Nick
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

You have to run using pythonwin. Name the file with suffix .pyw then run 
it and see?


From frankn=nuws at cs.vu.nl  Fri Jun 16 10:22:51 2000
From: frankn=nuws at cs.vu.nl (Frank Niessink)
Date: Fri, 16 Jun 2000 14:22:51 GMT
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <8idd7r$ed2@cs.vu.nl>

root <root at sirius.ftsm.ukm.my> wrote:
> newbie question.

> 2. 8 non-attacking queen.

>    use backtracking. the function call itself back. how to do recursive call
>    in python? 

I used a Board class that can solve itself. The board keeps track of 
where the queens are.

class Board:
        """ Class implementing a chess board. """
        def __init__(self, size=8, display=None):
                self.size = size
                # board
                self.board = [0]*size*size
                # where are the queens
                self.queens = []
                # display is a class which supports putQueen, removeQueen
                # and a solved method
                self.display = display

	[snip...]

        def solve(self,col=0):
                if col >= self.size:
                        if self.display:
                                self.display.solved(self)
                        return
                for row in range(self.size):
                        if not self.putQueen(col, row):
                                continue
                        self.solve(col+1)
                        self.removeQueen(col, row)


Board.putQueen returns true if it can place a queen in the given
position, false otherwise.

If you want, I can mail you the complete program.

Cheers, Frank

-- 
"Hey, Marvin," said Zaphod striding over towards to him, "Hey, kid, are we 
pleased to see you." Marvin turned, and in so far as it is possible for a
totally inert metal face to look reproachfully, this is what it did. 
"No you're not," he said, "no one ever is."
	-- Douglas Adams, 'The Restaurant at the End of the Universe'


From alex at magenta.com  Tue Jun 27 09:55:08 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 27 Jun 2000 15:55:08 +0200
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <8jabvo014tv@news2.newsguy.com>

Don Tuttle <tuttledon at hotmail.com> wrote in message
news:qJ165.8938$nZ5.153196 at typhoon.southeast.rr.com...
> Consider this a heads up NOT a promotion!!

Why would the Python community care about a Microsoft-proprietary
quasi-Java-clone...?


Alex





From jaap.groot at zora.nl  Tue Jun 13 10:49:30 2000
From: jaap.groot at zora.nl (Jaap Groot)
Date: Tue, 13 Jun 2000 16:49:30 +0200
Subject: importing module through variable
References: <8i5e9p$nm0$1@zingo.tninet.se>
Message-ID: <8i5i22$o5b$1@porthos.nl.uu.net>

You could do a:
exec "import %s" % var
exec "modvar=%s" % var

and refer to the module afterwards with modvar in your code!

with regards,

Jaap

Andr? Dahlqvist <andre at beta.telenordia.se> wrote in message
news:8i5e9p$nm0$1 at zingo.tninet.se...
> Hi
>
> I have a variable which is assigned the name of a module. Is it possible
> to use this variable in an import statement? Someone told me I could use
> "exec("import %s" %var) and while this works I am not sure how I can
> access the module by it's name after that. Using something like
> var.modfunc() won't do it since 'var' is just a string containing the
> modules name.
>
> // Andr?




From shapr at uab.edu  Thu Jun 15 09:38:55 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 15 Jun 2000 13:38:55 GMT
Subject: Best link for Python add-ins?
References: <3948D2EB.424B5E42@micron.net>
Message-ID: <3948DCB6.E8D3E336@uab.edu>

Peter Immarco wrote:
> 
> Just read about another wonderful library for Python, NumPy.  (Is that
> pronounced like "lumpy" or "lum-pie":)?  What's the best link that has a
> list of the libraries available for Python, windows 32 version?
> 
> Peter.

Parnassus is the best source of freely downloadable modules that I know
of.
It's at: http://www.vex.net/parnassus/
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From pecora at anvil.nrl.navy.mil  Thu Jun 15 07:15:38 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Thu, 15 Jun 2000 11:15:38 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu> <20000614184508.B26436@xs4all.nl> <200006141705.NAA14660@w20-575-109.mit.edu>
Message-ID: <150620000715397988%pecora@anvil.nrl.navy.mil>

In article <200006141705.NAA14660 at w20-575-109.mit.edu>, Alexander S
Coventry <alex_c at MIT.EDU> wrote:

> > The docs already have that:
> > 
> >    PyObject* PyList_GetItem (PyObject *list, int index)
> > 
> >    Return value: Borrowed reference.

Oy, Vey!  Borrowed reference (if you know that) means you "manually"
increase the refcount, right?  New ref. means you decrease the
refcount, unless that item is returned to the Python calling routine?

I'm sure there's more than this. :-)


From jamarijr at my-deja.com  Fri Jun 16 11:59:56 2000
From: jamarijr at my-deja.com (Arinté)
Date: Fri, 16 Jun 2000 15:59:56 GMT
Subject: Python garbage collection.
Message-ID: <8iditf$sdn$1@nnrp1.deja.com>

Yet again this is referring to my embedded app.  When my app runs a
script it, the script usually sets up a callback.  Right know when that
script is done running I can still cause an action that will have the
callback get called.  I know I need to XDECREF, but this leads me to
belive that the classes and variables the script defined are still in
memory.

How can I run a script using runAnyfile and after the script run have
all the variables and stuff that script defined be cleaned up?

what-is-all-this-ly-stuff-ly
thanx


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


From smst at bigfoot.com  Thu Jun  1 08:24:46 2000
From: smst at bigfoot.com (Steve Tregidgo)
Date: Thu, 01 Jun 2000 13:24:46 +0100
Subject: Getting the directory size
References: <2B1262E83448D211AE4B00A0C9D61B030130118F@msgeuro1.creo.be>
Message-ID: <3936560E.C289AB29@bigfoot.com>

Pieter Claerhout wrote:
> 
> <code>
[rest of code snipped]
> def main():
> 
>         dir = sys.argv[1]
> 
>         print "Testing directorySize..."
>         print "Directory: %s" %(dir)
>         print "Size: %s %s" %(getDirSize(dir)[0], getDirSize(dir)[1])
> 
> if __name__ == '__main__':
>         main()
> </code>
> 
> Is there any way to fasten up this beauty, because depending on the
> number of directories it has to walk through, it takes a long time. Is
> there maybe a function in the win32 modules who does this for me,
> but lots faster?
> 
I don't know about a dedicated function to do that, but you can double
the speed by changing the last line from:

    print "Size: %s %s" %(getDirSize(dir)[0], getDirSize(dir)[1])

...to the pair of lines:

    amount, units = getDirSize(dir)
    print "Size: %s %s" % (amount, units)

The original version walked the entire structure twice -- once to
calculate the amount, and then again to give the units.  Not really what
you were after, but faster nonetheless.

> Kind regards,
> 
> Pieter Claerhout
> 

HTH a little,
Steve

-- 
Steve Tregidgo
Software Developer
http://www.businesscollaborator.com



From jschmitt at vmlabs.com  Mon Jun 19 17:56:21 2000
From: jschmitt at vmlabs.com (John Schmitt)
Date: Mon, 19 Jun 2000 14:56:21 -0700
Subject: writing (Gnu)MAKE in Python
In-Reply-To: <8iiuud$572$0@216.39.151.169>
Message-ID: <NEBBIOGBDMDPEANLFKGHAECICEAA.jschmitt@vmlabs.com>

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Donn Cave
> Sent: Sunday, June 18, 2000 9:56 AM
> To: python-list at python.org
> Subject: Re: writing (Gnu)MAKE in Python
>
>
> Quoth "John Schmitt" <jschmitt at vmlabs.com>:
> [deletia]
> You must have a more detailed notion of a ``Make that really, really

Sure do.  Example: I did this in a Makefile (and the manual says that it's
valid):

foo: DEFINES += -DFOOBARBAZ

and it did not work.  I downloaded the source, but it would have taken me
too much time to dive in and fix it.

Another example, something like this:

bar:
	$(MAKE) $(MAKEFLAGS) -C bar bartarget

and Make crapped out with "There's no folder 'W'".  I looked at it, and Make
did not put the dash character in front of the 'W' character, and so instead
CDing to my 'bar' directory, it was trying to CD to my 'W' directory (which
obviously did not exist).  The equally dumb hacks and workarounds that I
came up with took less time than diving into the Make source code and
looking for the bug there.

> works''.  If it's just another implementation, I'd give you slim odds,
> against the ``make is big and complicated'' point you allude to above.

I'm sorry, I don't understand what you mean here.

> But there might a be a good chance for a radically different solution
> to the basic problem.  OOP is an obvious fit.  The two step, configure
> and make strategy that's common today is an obvious target for
> replacement.

Me personally, I would love a radically different solution (provided of
course that it is also a productive solution).  Getting it adopted where I
work requires a very compelling case, and I don't understand everyone else's
project enough to be able to duplicate it with other build tools and so
selling another solution is difficult for me.

My personal philosophy is to try to stay unattached to any particular
methodology, tool, habit but to always be open to improvements.  I cannot,
in that endeavor, forget who I'm working for and what their needs are.  IOW,
I love to play, but I have work to do.  If the play improves productivity
and I can demonstrate it, I am so there.  I have worked with people who
behaved in a very 'stick-in-the-mud' way to know that I don't want to get as
stale as they did.

Anyway, how is OOP an obvious fit to this problem in general?  Not obvious
to me.  The general idea of make seems pretty good to me, but as I just
learned, it's being applied to problems I don't understand very well.

>
> The sad thing is that the Python distribution itself would be the last
> place we could use this system.
>
> | In a smaller way, I would like to have Bash, sed, and grep as Python
> | programs rather than compiled C.
>
> I think you're alone on that one.

Again, I'm not being very clear.  It is not my end goal to have Unix command
tools written in Python rather than compiled C.  I would prefer them as
Python tools because IMHO there would be fewer bugs and maintenance hassles,
bugs would be fixed more often, and it would be *much* easier to port to
other platforms.  Even more, when someone comes up with a new way to extend
a tool, it's so much easier to put it in and try it out.  I guess that I
believe that advancements will happen much faster if these tools were
distributed as Python modules rather  as binaries of C code.  Debate about
advancements (specifically the lack thereof) in tools is best left
elsewhere, I suppose.

John

PS Is 'deletia' a word?  http://www.m-w.com/ does not have a definition for
it but 'deletia' is meaningful to me.




From matthewm at es.co.nz  Sat Jun 24 20:00:47 2000
From: matthewm at es.co.nz (Matthew Miller)
Date: Sun, 25 Jun 2000 12:00:47 +1200
Subject: Serial port IO Python module?
Message-ID: <39554BAE.6C00C44E@es.co.nz>

Hi everyone,

Is there a python module for RS-232 comms? Thanks. Matthew.




From moshez at math.huji.ac.il  Sat Jun 10 16:28:06 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 10 Jun 2000 23:28:06 +0300 (IDT)
Subject: PATCH: Augmented assignment 
In-Reply-To: <m130q7t-000wcDC@swing.co.at>
Message-ID: <Pine.GSO.4.10.10006102326430.9354-100000@sundial>

On Sat, 10 Jun 2000, Christian Tanzer wrote:

> It doesn't obfuscate code looking like:
> 
>    nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1
> 
> And it saves more than a few keystrokes.

Anyone doing *that* isn't  familiar with the pydiom:

s, i = nested_strucre[some_index], another_index
s[i] = s[i] + 1

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From cut_me_out at hotmail.com  Wed Jun 14 12:11:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 14 Jun 2000 12:11:54 -0400
Subject: [Q:] hash table performance!
References: <8i7kro$i04$1@nnrp1.deja.com>
Message-ID: <etdzoooz1h1.fsf@w20-575-109.mit.edu>

> Or anyone can describe how the python hash table works.

While I doubt it's the real issue in your case (see Alexander Williams'
post) Tim Peters wrote a really good description of how python hash
tables work a little while back.  It's at

http://www.python.org/pipermail/python-list/2000-March/049969.html

Alex.


From embed at NOSPAM.geocities.com  Wed Jun 28 12:29:53 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 28 Jun 2000 12:29:53 -0400
Subject: Pickle for C++?
References: <395A1F7E.4112B2A3@muc.das-werk.de>
Message-ID: <vJp65.11228$227.232032@nnrp1.uunet.ca>

"Thomas Thiele" <thiele at muc.das-werk.de> wrote in message
news:395A1F7E.4112B2A3 at muc.das-werk.de...
> Hi!
>
> I have a Python class that holds a lot of data.
> And I now want to send these datas over a TCP - connection via
> pickle and sockets. It works fine. But now I want
> that an application written in C++ is able to unpickle the data-class
> too and write it's content into an equivalent C++ - class/struct.
>
> Is there any solution out there or have I to write my own C++ - pickle
> algorithm?

Embed Python into your C++ application, and don't bother trying to tie a
static language to a dynamic one by making a set of fixed C strucutres.
Believe me it's a waste of time.

You can embed the Python DLL or shared library into your C++ application in
a matter of minutes, and then unpacking your data will be a single line call
to a Python function that will unpack your structure and a simple set of
functions to get and set the data in your python objects.

Warren




From thomas at xs4all.net  Fri Jun  2 09:33:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 2 Jun 2000 15:33:19 +0200
Subject: Status of List
In-Reply-To: <3937B59C.3AFE081E@bam.com>; from scherbi@bam.com on Fri, Jun 02, 2000 at 09:24:44AM -0400
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com> <3937B59C.3AFE081E@bam.com>
Message-ID: <20000602153318.N469@xs4all.nl>

On Fri, Jun 02, 2000 at 09:24:44AM -0400, Bill Scherer wrote:

> Things have been very quite since yesterday from my point of view.  I'm
> just glad I'm not the only one.

Nope, it isn't you. I think it's Mailman. There was a sudden burst of
activity on the mailman-devel list, with a dozen or so changes being checked
in, and i think one of them broke the usenet->mailman gateway. (The other
way 'round seems to be working fine.) I guess we'll have to wait for Barry
to wake up and spot it.

[CC: Barry just in case he doesn't read the list via Mailman ;-]

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From richard_chamberlain at ntlworld.com  Sat Jun 17 12:00:31 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 17 Jun 2000 12:00:31 -0400
Subject: CodeWright anyone?
References: <NEBBIOGBDMDPEANLFKGHKEOPCDAA.jschmitt@vmlabs.com>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E86@mail.sauder.com>

John,

There is a bit about using CodeWright with Python here:

http://www.python.org/doc/howto/editor/node6.html

I've never used it so I can't comment (just remember seeing it)

Richard


John Schmitt <jschmitt at vmlabs.com> wrote in message
news:NEBBIOGBDMDPEANLFKGHKEOPCDAA.jschmitt at vmlabs.com...
>
> Does anyone use CodeWright to edit Python code?  I had a quick go at
setting
> up a lexer so that my project view works like it does in C, but I
obviously
> don't know enough about regular expressions and about what to do with
> matches in the CodeWright tool.
>
> John
>
>




From tuckerg at acm.org  Sun Jun 25 18:46:28 2000
From: tuckerg at acm.org (Gregory Tucker)
Date: Mon, 26 Jun 2000 07:46:28 +0900
Subject: smtplib
References: <000c01bfdec7$30f84b60$bdcb45c6@timaratz>
Message-ID: <39568BC4.96EBE8AE@acm.org>

It looks like the subject is included in the message body. See the attched
code from Starship.

The following code will call it (I think).

import SMTP

mailhost="tkd.att.ne.jp"
mailfrom="tuckerg at acm.org"
mailto="tuckerg at acm.org"
subj="Test Message"
msg="""
Blah
Blah
Blah
"""

s=SMTP.SMTP(mailhost)
s.send_message(mailfrom, mailto, subj, msg)
s.close()

Regards,
Greg

P.S. I am getting very little mail off the Python list. Is it something
wrong with me?

Peter Timaratz wrote:
> 
> I'm using smtplib with a CGI script. I'm puzzled as to why this module
> doesn't support sending a subject for the e-mail.
> 
> And the e-mail I receive doesn't show the to address in my e-mail client.
> 
> --
> http://www.python.org/mailman/listinfo/python-list

-- 
 .---. .---. .--- .---.   |  Gregory Tucker, Tokyo, Japan
 |---. | |-< >--- |---.   |  
 `---' `-'`-'`--- `---'   |  "Our Father, who art in Redmond,
 My opinions are my own.  |  William by thy name..."
-------------- next part --------------
#!c:\Python\python.exe
#########################################################################
## This is SMTP sender class. Many thanks to all, which 
## help to finish these useful stuff. If you finds 
## some error, of like to add more features, please feel free 
## to contact with me. Vladimir (gandalf at starship.skyport.net)
##
#########################################################################
## Usage:
## 1) Create SMTP statement and connect to the mailserver:
##    object = SMTP('smtp.server.name.or.address')
## 2) Send message:
##    object.send_message('from_addr', 'to_addr', 'Subject', 'message body')
## 3) You can invoke send_message morethan one times.
## 4) Close connection:
##    object.close()
#########################################################################
## 20-Mar-98 SMTP class bugfix
## All strings into header now finished by '\r'
##               Many thanks to Brian Hooper for this bugfix
## 8-Mar-98 SMTP class modifyed.
## 1) socket creation in __init__ checked by try: except: block.
## 2) In SMTP.check method I'm using SMTP.readline (no more conn.recv :) )
## 3) SMTP.readline uses select call for timeout checking.
##               Many thanks to Skip Montanaro for this update.
########################################################################
## These information will be useful in 'check' method of SMTP instances
##  REPLY CODES BY FUNCTION GROUPS
##         500 Syntax error, command unrecognized
##            [This may include errors such as command line too long]
##         501 Syntax error in parameters or arguments
##         502 Command not implemented
##         503 Bad sequence of commands
##         504 Command parameter not implemented
##
##         211 System status, or system help reply
##         214 Help message
##            [Information on how to use the receiver or the meaning of a
##            particular non-standard command; this reply is useful only
##            to the human user]
##
##         220 <domain> Service ready
##         221 <domain> Service closing transmission channel
##         421 <domain> Service not available,
##             closing transmission channel
##            [This may be a reply to any command if the service knows it
##            must shut down]
##
##         250 Requested mail action okay, completed
##         251 User not local; will forward to <forward-path>
##         450 Requested mail action not taken: mailbox unavailable
##            [E.g., mailbox busy]
##         550 Requested action not taken: mailbox unavailable
##            [E.g., mailbox not found, no access]
##         451 Requested action aborted: error in processing
##         551 User not local; please try <forward-path>
##         452 Requested action not taken: insufficient system storage
##         552 Requested mail action aborted: exceeded storage allocation
##         553 Requested action not taken: mailbox name not allowed
##            [E.g., mailbox syntax incorrect]
##         354 Start mail input; end with <CRLF>.<CRLF>
##         554 Transaction failed
	 
import time, string,select
from socket import * 
 
msg_pattern = """From: %s\r
To: %s\r
Subject: %s\r
Date: %s\r
X-Mailer: Python SMTP class v 1.2\r
\r
%s 
 
""" 
 
 
 
class SMTP: 
	# display a debug message 
	def say(self, msg):  
		s = time.strftime('%c', time.localtime(time.time())) + ': ' + msg 
		print s 
	# 
	def __init__(self, host='localhost', lhost='localhost'):
	    self.error = "SMTP error"
	    try:
		self.conn = socket(AF_INET, SOCK_STREAM)
		self.file = self.conn.makefile()
		self.conn.connect(host, 25) 
		self.check('220') 
		self.send('helo ' + lhost) 
		self.check('250')
	    except socket.error:
		self.conn = None
	# 
	def readline(self):
	     r, w, e = select.select([self.file], [], [], 30)
	     if r == []: raise self.error, "Read from remote host timed out"
	     return r[0].readline()
	def check(self, lev): 
		dat = self.readline()
		if (lev != ''): # means don't care about the result code 
			if dat[:3] != lev: 
				raise self.error, "Unexpected result code!: "+dat
				# what to do here?  Need to exit gracefully, or retry? TBD 
	# 
	def send(self, line): 
		self.conn.send(line + '\n') 
	# 
	def send_body(self, mfrom, mto, subj, body): 
		self.conn.send(msg_pattern%(mfrom, mto, subj, time.ctime(time.time()),body) + '\n' ) 
	# 
	def send_end_of_data(self): 
		#self.say('Sending end of data marker') 
		self.conn.send("\n.\n") 
 
	def send_message(self, mfrom, mto, subj, body): 
		self.send("mail from:<%s>"%mfrom) 
		self.check('250') 
		self.send("rcpt to:<%s>"%mto) 
		self.check('250') 
		self.send("data") 
		self.check('354') 
		self.send_body(mfrom, mto, subj, body) 
		self.send_end_of_data() 
		self.check('250') 
	# 
	def close(self): 
		self.send("quit") 
		self.check('221') 
		self.conn.close() 
	# 
# 
 
if __name__ == '__main__': 
	test_msg = """ 
This is a test of mr. Moukhine :) 
""" 
	import pdb 
        #s = SMTP("folwell.com") 
        #s = SMTP("65-37.lgcit.com")
        s = SMTP("mail.callware.com")
	#pdb.run('s.send("gandalf at rosmail.com", "gandalf at rosmail.com", "Test", "This is a test message.")') 
	#s.send_message("fred at acme.com", "richard at folwell.com", "Test", test_msg)
	#s.send_message("gandalf at lgcit.com", "gandalf at lgcit.com", "Test", test_msg)
	s.send_message("ivanlan at callware.com", "ivanlan at callware.com", "Test", test_msg)
	s.close() 
# 

From shapr at uab.edu  Thu Jun  8 08:01:01 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 08 Jun 2000 12:01:01 GMT
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com> <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us>
Message-ID: <393F8ACC.C49E0116@uab.edu>

Andrew Kuchling wrote:

> So, what would people want on top of distutils?  A Debian-style "do
> apt-get, upgrade" to update currently installed packages?  A
> CPAN-style "fetch and install this module for me, please"?  Simple
> version checking?
> 
> --amk

personally I'd want Debian style that allows upgrading of packages, and
de/installation of new packages.

Debian also has the wonderful ability to upgrade to a new version of
Debian using apt-get -f dist-upgrade, which means once you have Debian
installed, you don't have to do much of anything manually.

Time for me to join the distutils-sig.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From urner at alumni.princeton.edu  Tue Jun  6 20:26:50 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Tue, 06 Jun 2000 17:26:50 -0700
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com>
Message-ID: <kk5rjski2pc4g4tor99tc2h5ursinprg80@4ax.com>

Rajarshi Ray <rajarshi at home.com> wrote:

>Besides the gee-whiz aspect of such algorithms, do they have any value?
>After all, what makes pi pi is ALL of its digits in a particular
>sequence.

Supposedly there were some deep points about number
theory in this vicinity -- according to the web pages
anyway.  I'm no expert.

Anyway, the "gee whiz" factor isn't to be underestimated,
either.

Kirby



From bjorn at roguewave.com  Fri Jun  9 19:05:53 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 09 Jun 2000 17:05:53 -0600
Subject: readlines() question
References: <Pine.GSO.4.10.10006092339580.29560-100000@sundial> <8hrrr6$fj4$1@slb2.atl.mindspring.net>
Message-ID: <39417851.77D2F849@roguewave.com>


Aahz Maruch wrote:
> 
> In article <Pine.GSO.4.10.10006092339580.29560-100000 at sundial>,
> Moshe Zadka  <moshez at math.huji.ac.il> wrote:
> >
> >Ummmmm...I'm known as a professional paranoid in many places, so I usually
> >code like:
> >
> >my_file = open("blah")
> >try:
> >       ...my_file.read, my_file.readlines, etc.
> >finally:
> >       my_file.close()
> 
> And if you're *really* paranoid ;-)
> 
> try:
>         my_file = open("blah")
>         ...my_file.read, my_file.readlines, etc.
> finally:
>         my_file.close()
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)

surely <wink>, you mean:

my_file = None
try:
   my_file = open("blah")
   ...my_file.read, my_file.readlines, etc.
finally:
   if my_file is not None:
      my_file.close()

perhaps-more-useful-with-a-database-connection'ly y'rs
-b



From pete at visionart.com  Thu Jun 22 21:44:44 2000
From: pete at visionart.com (Pete Shinners)
Date: Thu, 22 Jun 2000 18:44:44 -0700
Subject: url directory listing
References: <39518289.D4175618@bioeng.ucsd.edu> <slrn8l3qfm.nol.scarblac-spamtrap@flits104-37.flits.rug.nl> <39529682.A3F00A25@bioeng.ucsd.edu> <3952BD4F.51308D93@yale.edu>
Message-ID: <3952C10C.18CD320@visionart.com>

> Curtis Jensen wrote:
> >
> > It is possible to get the files off of a site.  softbytelabs did it:
> > http://www.softbytelabs.com/
> > Their black widow program does a fine job of copying site files.  Maybe
> > it's not done with HTTP, but does anyone know how it's done?  What does
> > WebSucker do?  Thanks.

i have not seen websucker in action, but i am assuming it is
downloading all the files referenced from within the .HTML
pages it gets. i don't think this would be too hard,

parse the HTML and read all the <img> tags
to get image filenames. then recurse through all the
<a href=> tags

heh, of coarse there's a bit more glue to put into something
like this, but it shouldn't be too taxing. you'll probably
want to peek into the <body> <table> <tr> and <td> tags
to see if they have a "background" attribute


From just at letterror.com  Tue Jun 13 09:34:10 2000
From: just at letterror.com (Just van Rossum)
Date: Tue, 13 Jun 2000 14:34:10 +0100
Subject: How many times inner functions are compiled ?
In-Reply-To: <20000613141433.U20700@xs4all.nl>
References: <960897680.1890969865@news.libertysurf.fr>; from
 jerome.quelin@insalien.org on Tue, Jun 13, 2000 at 02:01:20PM +0200
 <960897680.1890969865@news.libertysurf.fr>
Message-ID: <l03102807b56be7433fe4@[193.78.237.168]>

At 2:16 PM +0200 13-06-2000, Thomas Wouters wrote:
>On Tue, Jun 13, 2000 at 02:01:20PM +0200, Jerome Quelin wrote:
>
>> def outer():
>>     def inner():
>>         print "spam"
>>     inner()
>> outer()
>> outer()
>
>> In the snippet code above, I call outer() two times. I would like to
>> know how many times python will compile the inner function ? One time
>> when compiling the script/module, or two times, ie at every invocation
>>of the
>> outer function ?
>
>Once for every invocation of outer().

Not true: compilation happens at compile time (duh!) which is only during
the first import and only if there's no .pyc file yet. A new function
object for inner() will be created at each invocation of outer(), but the
_code_ object for that function will be the same.

Just





From fiona at sitegnome.com  Wed Jun 14 08:08:54 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 14 Jun 2000 12:08:54 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 14th, 2000
Message-ID: <20000614120854.20282.qmail@synop.com>

Hello All,

Here are the latest entries entered into http://python.faqts.com

Wishing you all a good day!

Regards, Fiona Czuczman


## Unanswered Questions ########################################


-------------------------------------------------------------
Can I combine a "select" call on some of my file objects with the Tkinter event loop?
http://www.faqts.com/knowledge-base/view.phtml/aid/3728
-------------------------------------------------------------
Rob Hooft



## New Entries #################################################


-------------------------------------------------------------
What type of sort is used for the internal sort function in Python (quick, merge, insertion, etc.)?
http://www.faqts.com/knowledge-base/view.phtml/aid/3720
-------------------------------------------------------------
Fiona Czuczman, Rob Hooft
Peter Schneider-Kamp, Tim Peters

The following assumes that we talk about lists in the standard (C) 
Python edition.

binary insertion sort is used for small lists (<100 elements) whereas
a quicksort derived sorting algorithm called "samplesort" is used for
larger lists. A description can be found in the list object 
implementation:

samplesort is basically quicksort on steroids:  a power of 2 close to 
n/ln(n) is computed, and that many elements (less 1) are picked at 
random from the array and sorted.  These 2**k - 1 elements are then used 
as preselected pivots for an equal number of quicksort partitioning 
steps, partitioning the slice into 2**k chunks each of size about ln(n). 
These small final chunks are then usually handled by binarysort.  Note 
that when k=1, this is roughly the same as an ordinary quicksort using a 
random pivot, and when k=2 this is roughly a median-of-3 quicksort.  
>From that view, using k ~= lg(n/ln(n)) makes this a "median of n/ln(n)" 
quicksort.  You can also view it as a kind of bucket sort, where 2**k-1 
bucket boundaries are picked dynamically.

The large number of samples makes a quadratic-time case almost 
impossible, and asymptotically drives the average-case number of 
compares from quicksort's 2 N ln N (or 12/7 N ln N for the median-of-3 
variant) down to N lg N.

-----

The current binary_insertion/samplesort hybrid was introduced in Python
1.5.2.  In earlier releases a variety of plain quicksorts were used,
including-- in the very early days --libc's qsort.  The latter had to be
dropped because it wasn't reentrant on some platforms (== bad results 
and/or even segfaults), and ran horribly slowly on some platforms in 
some common degenerate cases.

The point to the current hybrid is that it minimizes comparisons while
(believe it or not, and it's hard to believe if you study the code!) 
doing much less data movement than a mergesort.  Comparisons are 
especially expensive in Python, because lists are heterogenous and each 
time a pair of elements gets compared the correct type-varying function 
for doing the comparison needs to be figured out and then indirectly 
invoked.  So minimizing compares is crucial for a Python sort.  I wasn't 
able to write a mergesort (which also does few compares) as fast as the 
hybrid, because mergesort shuffles data around a lot more -- there are 
few enough compares now that data movement expense is no longer lost in 
the shuffle <ahem>.  The current hybrid is also an in-place sort.  Its 
only drawbacks are the complexity of the algorithm and that the 
resulting sort isn't stable.

-----

The "sort" function in the Numeric python code still uses the libc's
qsort() function at this moment.


-------------------------------------------------------------
Is there a general way to remove axes of length 1? eg. I have an array of shape (1,3,1,4) I want to change it to (3,4)?
http://www.faqts.com/knowledge-base/view.phtml/aid/3723
-------------------------------------------------------------
Fiona Czuczman
Konrad Hinsen

For a specific known shape, just assign a new one:

 array.shape = (3, 4)

For suppressing all axes of length one, no matter where, you can write

 array.shape = Numeric.repeat(array.shape, 
Numeric.not_equal(array.shape, 1))

Note that this changes the array, it does not create a new one.


## Edited Entries ##############################################


-------------------------------------------------------------
How do I check to see if a file exists?
http://www.faqts.com/knowledge-base/view.phtml/aid/2782
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh,Peter Schneider-Kamp

To check if a file exists use:

>>> import os.path
>>> os.path.exists("python/")
1
>>> os.path.exists("spam/")
0
>>> os.path.exists("python/patches/submitted/array.pop-extend.patch")
1

If you only want a true for regular files have a look at isfile():

>>> os.path.isfile("python/")
0
>>> os.path.isfile("spam/")
0
>>> os.path.isfile("python/patches/submitted/array.pop-extend.patch")

for more info or other function look at:
http://python.org/doc/current/lib/module-os.path.html


-------------------------------------------------------------
Where can I get good Tutorials about JPython?
Where can I get good Tutorials about JPython?
http://www.faqts.com/knowledge-base/view.phtml/aid/3668
-------------------------------------------------------------
Fiona Czuczman
Moshe Zadka,bo Vandenberg

JPython is another implementation of the Python language: everything you
learn in the Python tutorial (http://www.python.org/doc/tut/tut.html),
will be true for JPython. You can review the CPython/JPython differences
somewhere in the JPython site (http://www.jpython.org).

Some links to start with are:

The first stop for Python (IMHO)
http://www.pythonlabs.com/

JPython
http://sourceforge.net/project/?group_id=5473

General Python Portal -- with easy links to important pages
phttp://www.oreillynet.com/pub/a/python/2000/04/14/python-intro.html


SOURCE MODULES ETC...
http://www.vex.net/parnassus/apyllo.py?a=l
http://tor.dhs.org/~zephyrfalcon/snippets/index.html


-------------------------------------------------------------
Is there a progress bar given by Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/2718
-------------------------------------------------------------
Fiona Czuczman, Rob Hooft
Robert Hicks, John Grayson

This comes from John Grayson's book "Python and Tkinter
programming".

"""
A  basic widget for showing the progress
being made in a task.

"""

from Tkinter import *

class ProgressBar:
    def __init__(self, master=None, orientation="horizontal",
                 min=0, max=100, width=100, height=18,
                 doLabel=1, appearance="sunken",
                 fillColor="blue", background="gray",
                 labelColor="yellow", labelFont="Verdana",
                 labelText="", labelFormat="%d%%",
                 value=50, bd=2):
        # preserve various values
        self.master=master
        self.orientation=orientation
        self.min=min
        self.max=max
        self.width=width
        self.height=height
        self.doLabel=doLabel
        self.fillColor=fillColor
        self.labelFont= labelFont
        self.labelColor=labelColor
        self.background=background
        self.labelText=labelText
        self.labelFormat=labelFormat
        self.value=value
        self.frame=Frame(master, relief=appearance, bd=bd)
        self.canvas=Canvas(self.frame, height=height, width=width, bd=0,
                           highlightthickness=0, background=background)
        self.scale=self.canvas.create_rectangle(0, 0, width, height,
                                                fill=fillColor)
        self.label=self.canvas.create_text(self.canvas.winfo_reqwidth()
/ 2,
                                           height / 2, text=labelText,
                                           anchor="c", fill=labelColor,
                                           font=self.labelFont)
        self.update()
        self.canvas.pack(side='top', fill='x', expand='no')

    def updateProgress(self, newValue, newMax=None):
        if newMax:
            self.max = newMax
        self.value = newValue
        self.update()

    def update(self):
        # Trim the values to be between min and max
        value=self.value
        if value > self.max:
            value = self.max
        if value < self.min:
            value = self.min
        # Adjust the rectangle
        if self.orientation == "horizontal":
            self.canvas.coords(self.scale, 0, 0,
              float(value) / self.max * self.width, self.height)
        else:
            self.canvas.coords(self.scale, 0,
                               self.height - (float(value) / 
                                              self.max*self.height),
                               self.width, self.height)
        # Now update the colors
        self.canvas.itemconfig(self.scale, fill=self.fillColor)
        self.canvas.itemconfig(self.label, fill=self.labelColor)
        # And update the label
        if self.doLabel:
            if value:
                if value >= 0:
                    pvalue = int((float(value) / float(self.max)) * 
                                   100.0)
                else:
                    pvalue = 0
                self.canvas.itemconfig(self.label, text=self.labelFormat
                                         % pvalue)
            else:
                self.canvas.itemconfig(self.label, text='')
        else:
            self.canvas.itemconfig(self.label, text=self.labelFormat %
                                   self.labelText)
        self.canvas.update_idletasks()






From MarkH at ActiveState.com  Wed Jun 21 08:35:34 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 21 Jun 2000 12:35:34 GMT
Subject: PyThreadState_Get error with C extension module
References: <395070B7.5F245778@szs.ira.uka.de>
Message-ID: <qE245.4517$qz4.33881@news-server.bigpond.net.au>

Its probably that one of your modules is still linking against Python 1.5,
causing _both_ Python15.dll and Python16.dll to be loaded - but only one
is initialized correctly, and the other immediately causes this
thread-state error.

Mark.

--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!
"Thomas Freytag" <freytag at szs.ira.uka.de> wrote in message
news:395070B7.5F245778 at szs.ira.uka.de...
> Hi all,
>
> I still have a problem getting Python 1.6 alpha 2 to run on Windows NT,
> in particular my own C extension modules. I have compiled the
> python16.dll and the
> python.exe by using the supplied Visual C++ project file from the
> PCbuild directory
> and all succeeds without a complaint. I also managed to compile my
> extension module
> which accesses a library of C functions implementing a Petri net
> simulator engine.
> But when I try to import my module, I receive the error message  "Fatal
> Python error:
> PyThreadState_Get: no current thread" and the interpreter crashes.
> What's wrong ? Note that my module has been working for years under
> Python 1.5.x and
> doesn't use any threading. Its interface complies with the standard
> interface given in the
> example_nt directory. Under Linux it worked fine without changing a line
> of code when
> switching from Python 1.5  to 1.6.
> Any hints ?
>
> Thomas
> ______________________________________________________________________
> Thomas Freytag                         Phone:         +49 721 608 4301
> University of Karlsruhe - SZS          Fax:             +49 721 697377
> D-76128 Karlsruhe / Germany            Email:   freytag at szs.ira.uka.de
> WWW homepage:                  http://szswww.ira.uka.de/people/freytag
> PGP private key:    http://szswww.ira.uka.de/people/freytag/pubkey.asc
> ______________________________________________________________________
>
>




From gangli at msn.com  Tue Jun 13 10:43:41 2000
From: gangli at msn.com (gangli at msn.com)
Date: Tue, 13 Jun 2000 14:43:41 GMT
Subject: can python do this?
References: <3945D0AA.E8F51A9C@san.rr.com> <3945d426$0$492$7f31c96c@news01.syd.optusnet.com.au> <3946dc6b.193651816@news.online.no> <3945e37b$0$489$7f31c96c@news01.syd.optusnet.com.au>
Message-ID: <8i5han$1ds$1@nnrp1.deja.com>

In article <3945e37b$0$489$7f31c96c at news01.syd.optusnet.com.au>,
  "Jason" <it at onestopent.com.au> wrote:
> $Revisions: ['05/05/2000 06']
> Traceback (innermost last):
>   File "C:\PROGRA~1\PYTHON\TOOLS\IDLE\ScriptBinding.py", line 131, in
> run_module_event
>     execfile(filename, mod.__dict__)
>   File "C:\WINDOWS\Desktop\aeros1.py", line 6, in ?
>     print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
> IndexError: list index out of range
>
>

This is most like the next line read from text file does not follow the
pattern: tag."one" "two" ....

you may use following modified script:

import string
file = open ("file")
lines = file.readlines()
for line in lines:
    pair = string.split(line,":")
    if len(pair) == 2:
        print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
    else:
        print 'warning wrong format line %s'%pair
file.close()



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


From scherbi at bam.com  Fri Jun  2 09:24:44 2000
From: scherbi at bam.com (Bill Scherer)
Date: Fri, 02 Jun 2000 09:24:44 -0400
Subject: Status of List
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com>
Message-ID: <3937B59C.3AFE081E@bam.com>

Things have been very quite since yesterday from my point of view.  I'm
just glad I'm not the only one.



Emile van Sebille wrote:

> I appear to be having problems with the list.  Yesterday
> I saw only 10-20 posts, and overnight there were none.
>
> I checked against usenet news server, but as I rarely read
> there, it's somewhat hard to tell but I did spot only a
> few 02-jun posts.
>
> Can anyone confirm the problem? Or should I look on my
> end.  Or is it just that quiet, and what I saw in usenet
> only the result of propogation delay.
>
> BTW, both the Zope and wxPython list are active, and I've
> even seen more fnOrb posts than Python posts! ;-)
>
> TIA,
>
> Emile van Sebille
> emile at fenx.com
> -------------------
>
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Verizon Wireless






From jmassung at magpiesystems.com  Wed Jun 21 13:07:25 2000
From: jmassung at magpiesystems.com (Jeff Massung)
Date: Wed, 21 Jun 2000 11:07:25 -0600
Subject: Speed with Python
Message-ID: <QH645.94$QA1.13929@news.uswest.net>

Many threads with this kind of name have popped up, comparing the speed of
Python to C, Java, etc. and trying to determine how fast some of the
algorithms in Python are.

My question however, is how many of you who are Python gurus/junkies (no
offence, I myself am a C and Forth junkie) :) use Python for speed intensive
processing (not graphics, but databases and such)? Does efficiency of
developement outweight the eventual speed of use?

Cheers,
Jeff




From db3l at fitlinxx.com  Thu Jun  8 18:31:03 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 08 Jun 2000 18:31:03 -0400
Subject: Why can't I restore my redirected stdout?
References: <8hm5od$lki$1@news5.jaring.my> <slrn8jtgg8.oe2.scarblac-spamtrap@flits104-37.flits.rug.nl> <8hotma$48l$1@news5.jaring.my>
Message-ID: <usnun4xg8.fsf@ctwd0143.fitlinxx.com>

"Raymond Ng Tong Leng" <think2 at pd.jaring.my> writes:

> Remco Gerlich wrote in message ...
> >
> >I just typed this in, and it worked fine! Something else must be happening.
> >Are you in idle or so? Does it work inside a normal python interpreter?
> >What's your platform?
> >
> 
> Yes, I was using IDLE 0.4 with Python 1.5.2 on Win NT. When I tried it
> inside a normal Python interpreter like you suggested, it worked fine. I
> hope Guido fixed this bug... sorry, improved on this feature :) in the
> latest version of IDLE.

What's happening in IDLE is that sys.stdout is already being re-routed
(so it can be displayed in the window after being colored and such),
but sys.__stdout__ is still the true original stdout.  For example,
the following is from an IDLE window:

    >>> import sys
    >>> print sys.stdout
    <PyShell.PseudoFile instance at 85d5a0>
    >>> print sys.__stdout__
    <open file '<stdout>', mode 'w' at 792980>

And in the case of IDLE, it's running as a Windows application that
doesn't actually have a console, so the true <stdout> file doesn't go
anywhere.  Thus when you try to "restore" to sys.__stdout__ you aren't
really restoring the initial IDLE output file, but a pointer to a
console that technically doesn't exist (at least not with a window).

I guess by definition, this is probably the "right" thing to do, since
sys.__stdout__ is supposed to be the true original stdout when the
Python application was started, and even IDLE is just a Python
application.  So it probably depends on your view of how closely IDLE
should emulate running a textual Python in a real console.

As one workaround, if you want your original approach (restoring to
sys.__stdout__) to work, you could try re-assigning sys.__stdout__ when
you start up your IDLE session, e.g.:

    >>> sys.__stdout__ = sys.stdout

which works in a quick test, but I noticed that the PyShell class in
IDLE uses sys.__stdout__ for its own restoration, so I don't know if
there might be a later gotcha.

Where IDLE redirects sys.stdout is in PyShell.py, so you could tweak
that to do whatever you want if you really wanted different behavior.
Note that you can see other places in IDLE where sys.stdout is
temporarily redirected and then restored.

What's probably a better, more general approach, is to just save
sys.stdout yourself and restore it later.  If you're building a module
that may be imported from other modules that's actually the safer
approach anyway, since in that case you'd want to restore the output
to what sys.stdout was before you re-routed it, which may or may not
be the same as the very original sys.stdout.

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


From benji_york at cal-na.com  Thu Jun 15 15:07:27 2000
From: benji_york at cal-na.com (benji_york at cal-na.com)
Date: Thu, 15 Jun 2000 14:07:27 -0500
Subject: Can I build a single file containing all the contents of LIB?
Message-ID: <OF96F13BFB.87899EC6-ON862568FF.0068427A@cal-na.com>

Warren Postma wrote:

>I have an embedded system which has no support for long filenames. It's
>also a pain to upload the whole LIB subdirectory to this system.  The
>embedded system has a FAT filesystem and only supports 8 character
>filenames with a 3 character extension.

Perhaps the single file interpreter from 
http://starship.python.net/crew/lemburg/mxCGIPython.html would help. Maybe 
you can compile it for yourself.


Benji York
Information Technology Department
Calsonic North America, Inc.

Note: The opinions expressed herein do not necessarily reflect the 
opinions of my employer.



From richard_chamberlain at ntlworld.com  Tue Jun 20 14:30:11 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Tue, 20 Jun 2000 19:30:11 +0100
Subject: Widget resizing
References: <394F8774.138B@cs.bham.ac.uk>
Message-ID: <ABO35.4876$fw6.96380@news6-win.server.ntlworld.com>

Hi Elliot,

It depends on the kind of manager you use, but I'll use pack as an example.

from Tkinter import *
root=Tk()
Frame(root,borderwidth=3,relief=GROOVE).pack(side=LEFT,fill=BOTH,expand=YES)
root.mainloop()

If you look at www.manning.com/grayson you can download two chapters of
Python and Tkinter programming one of which is screen layout. (of course
you'll be much better buying a copy ;)

Richard

Elliot Smith <e.smith at cs.bham.ac.uk> wrote in message
news:394F8774.138B at cs.bham.ac.uk...
> Hello all,
>
> Can anyone tell me how to get Tkinter to resize frames etc. when their
> parents (e.g. a Toplevel instance) are resized?
>
> Thanks,
> Elliot
> --
> Elliot Smith
> School of Computer Science, University of Birmingham
> email: e.smith at cs.bham.ac.uk
> homepage: http://www.cs.bham.ac.uk/~ezs/




From robert_kuzelj at my-deja.com  Thu Jun 15 05:19:49 2000
From: robert_kuzelj at my-deja.com (robert_kuzelj at my-deja.com)
Date: Thu, 15 Jun 2000 09:19:49 GMT
Subject: [win32] how to get the running applications?
Message-ID: <8ia73c$fg7$1@nnrp1.deja.com>

hi,

is there a possibility to get the running applications (as opposed to
the running processes?).

i had a look into win32pdh.py and win32pdhutil.py but i couldnt
figure it out.

ciao robertj


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


From erno at iki.fi  Sun Jun 25 09:46:49 2000
From: erno at iki.fi (Erno Kuusela)
Date: 25 Jun 2000 16:46:49 +0300
Subject: url directory listing
References: <39518289.D4175618@bioeng.ucsd.edu> <slrn8l3qfm.nol.scarblac-spamtrap@flits104-37.flits.rug.nl> <39529682.A3F00A25@bioeng.ucsd.edu> <3952BD4F.51308D93@yale.edu> <3952C10C.18CD320@visionart.com>
Message-ID: <kun1k9vpo6.fsf@lasipalatsi.fi>

>>>>> "Pete" == Pete Shinners <pete at visionart.com> writes:

    Pete> i don't think this would be too hard

especially if you use the free code for this that the jolly boys
at pythonlabs are giving out with each copy of python!
look in Tools/webchecker/websucker.py :) it even has a gui.

  -- erno


From thor at localhost.localdomain  Sat Jun  3 17:18:09 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 3 Jun 2000 21:18:09 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <3937D345.CC0935D5@uab.edu> <39385DC2.BC0B4853@stud.ntnu.no>
Message-ID: <slrn8jj4ot.om.thor@localhost.localdomain>

On Sat, 03 Jun 2000  Peter Schneider-Kamp <petersc at stud.ntnu.no> wrote:
>Shae Erisson wrote:
>> 
>> Martijn Faassen wrote:
>> >
>> ja, maar is het niet vanzelfsprekend dat alleen mensen die Nederlands
>> spreken zich voor Python interesseren?
>
>Maar alle mensen die Nederlands spreken horen Python liefhebben!
> 
>> or-at-least-lives-with-someone-who-speaks-dutch-ly y'rs
>

Seguro que si, quien sabe, a lo mejor mi perro sabe el holandes. :P

Por cierto, vaya ca?a que va a ser el tener a Guido trabajando
al 100 % para python. Espero que no se les vaya la mano a?adiendo
nuevas caracteristicas a python y lo estropeen.

-- 
MGA


From jkraska1 at san.rr.com  Wed Jun  7 23:35:18 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 08 Jun 2000 03:35:18 GMT
Subject: Please help!!!!
References: <7WD%4.1099$0l2.8049@news2.tor.primus.ca> <393F14B9.320983C2@san.rr.com>
Message-ID: <393F1511.4989BABB@san.rr.com>

> > Could someone please tell me what is wrong.  I have re-installed Python
> > 1.5.2.  I have tried other code and all seems to lead me to the same
> > problem.  Please help!

By the way, I presume this is a toy problem. The U.S. tax code
does not work like that...



C/


From petersc at stud.ntnu.no  Sat Jun 10 17:25:59 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sat, 10 Jun 2000 23:25:59 +0200
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no>
Message-ID: <3942B267.C769395@stud.ntnu.no>

Peter Schneider-Kamp wrote:
> 
> I have to create a tuple (of two integer values) in a
> C extension. At the moment I use the following (rather
> ugly in my eyes) construction:

As an afterthought. Are there any reference counting things I
have to deal with in this example?
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From shapr at uab.edu  Thu Jun  8 09:43:37 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 08 Jun 2000 13:43:37 GMT
Subject: IDLE and building Tk GUIs
References: <393f4568$0$25329@wodc7nh0.news.uu.net> <vskujs867ghjso3aqi2r38n5mei0bg4lhq@4ax.com> <393F8D02.9A87A6EF@uab.edu> <9k4vjs0klckrvba86usus8ckpgmt4b6ca5@4ax.com>
Message-ID: <393FA2DB.616DDEAB@uab.edu>

Anders M Eriksson wrote:

> Yes I know. I'm using IDLE some of the time, then I switch to
> PythonWin, and then I use UltraEdit. None of these products are any
> good at GUI's though. IDLE just quits when I end my Tkinter
> application, PythonWin hangs when I run Tkinter within it, etc ...

My suggestion for using IDLE to build Tkinter GUIs and not quit when
your app ends is:
execute your Tkinter GUI from a command prompt, but edit it IDLE.
I know it's a cheesy solution, but it's the best one I know of.

If other people have better solutions, I'd love to hear them.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From hzhu at knowledgetrack.com  Wed Jun 28 20:24:36 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Thu, 29 Jun 2000 00:24:36 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <m3k8f9cqcc.fsf@greebo.nodomain.de>
Message-ID: <slrn8ll5vu.25h.hzhu@rocket.knowledgetrack.com>

On 29 Jun 2000 01:56:19 +0200, Bernhard Herzog <herzog at online.de> wrote:
>
>If your cwd is python then the sys.path entry for python/ and '.' are
>the same directory. Why would that lead to problems? The only thing I
>can think of is that other directories in sys.path also have AAA
>modules.

Sorry I completely messed up my description.  (And apologies for complaining
about earlier correspondents not seeing the problem.)  Here's a retake:

$PYTHONPATH/
   AAA/
      __init__.py
   BBB/
      __init__.py
      aaa.py
      ccc.py
   CCC/
      ...

Within BBB/aaa.py the statement 'import AAA' imports from AAA/__init__.py on
Unix but from BBB/aaa.py on Windows.

The reason is that within the BBB package python first looks for a module
called AAA within the package (which it found as aaa.py on Windows) before
it looks for a package named AAA.

Although this is often the desired behavior, the other way around is also
fairly common.  In my case AAA and CCC are packages written by others and
BBB is a wraper pacakge that imports the other packages to do the real work
but provides different interfaces for convenience.  Using identical names
seems very natural here.

Since there is no way to tell python to go to absolute path first, one has
to make all module names in all packages distinct.  This IMHO defeats one of
the most important purpose of using packages, that is to encapsulate name
spaces for module names.

Therefore I think it is desirable for python to provide a mechanism for
programmers to describe absolute path of imports.  Like on Unix, writing
/bin/sh in stead of sh in a script will guarantee its behavior no matter
where it is run and what the current PATH the user has.

My previous proposal was to use
import pythonroot.AAA
but in analogy with Unix paths maybe a syntax like
import .AAA
is even better.  (BTW, this is the scheme used in DNS configurations for
distinguishing qualified and unqualified domain names.)

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From mwh21 at cam.ac.uk  Tue Jun 13 07:27:29 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 2000 12:27:29 +0100
Subject: No Boolean type in "types"
References: <394612a2.0@katana.legend.co.uk>
Message-ID: <m366rd4y8u.fsf@atrus.jesus.cam.ac.uk>

"Thomas Mangin" <thomasm at legend.co.uk> writes:

> Hello,
> 
> Can someone explain me why the module "types" have no boolean type defined ?
> 
> true = (0==0)
> false = (not true)
> 
> BooleanType = type(0==0)
> 
> Thank you

Because type(0==0) == types.IntegerType.  Python has no boolean type;
the rules for determining whether a value is true or false are a
little involved, but generally do what you expect.

Cheers,
M.

-- 
81. In computing, turning the obvious into the useful is a living
    definition of the word "frustration".
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From mrw065 at merle.acns.nwu.edu  Wed Jun 28 17:08:23 2000
From: mrw065 at merle.acns.nwu.edu (Mitchell R Whorlow)
Date: 28 Jun 2000 21:08:23 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand>
Message-ID: <8jdpg7$6ri$1@news.acns.nwu.edu>

Steffen Ries (steffen.ries at sympatico.ca) wrote:
...
> "uid=joe, ou=org\\, unit, o=some org"
> -> ["uid=joe", "ou=org\\, unit", "o=some org"]

> Is there a way to use re.split to achieve this? (I solved it searching
> [^\\], and splitting the string in a loop, but I was wondering if it
> can be done easier.)
...

How about using re.findall?

>>> s = r"""id=joe, ou=org\, unit, o=some org"""
>>> print s
id=joe, ou=org\, unit, o=some org
>>> patt = re.compile(r"""([^, ]+=[\w ]+(?:\\,)?[^,]+)""")
>>> patt.findall(s)
['id=joe', 'ou=org\\, unit', 'o=some org']

I think this does what you want, but taking into account the regex I used,
you still may prefer your solution.

Mitch


From denalione at my-deja.com  Thu Jun 29 07:38:19 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Thu, 29 Jun 2000 11:38:19 GMT
Subject: exiting threads
Message-ID: <8jfcf9$2rp$1@nnrp1.deja.com>

I am using thread.start_new_thread(self.foo,(x,y))

Do I have to explicitly exit the thread or does it automatically exit
when foo returns?


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


From jgraves3 at austin.rr.com  Wed Jun  7 10:06:23 2000
From: jgraves3 at austin.rr.com (jay graves)
Date: Wed, 07 Jun 2000 14:06:23 GMT
Subject: Newbie Question...
References: <ZTo%4.854$0l2.6963@news2.tor.primus.ca>
Message-ID: <393e5216.3390058@news-server>

On Wed, 7 Jun 2000 05:52:30 -0400, "Robin Porter" <mogul at primus.ca>
wrote:

>I wrote before asking for an example of C code changed to Python code (i.e.
>Payroll example).  Could someone please explain how I go about saving and
>then running the code once I have inputed it in.  I tried entering it into
>the Python GUI and everything looked good until I tried to save it, recall
>it, then run it.
I'm assuming you are running Windows?

My guess is that you typed it in the interpreter window in PythonWin.

Instead you should have done File-New and typed your code in the
resulting script window.  Then a File-Save the script with a '.py'
extension  (i.e. 'payroll.py') and then do a File-Run.

>Could someone please recommend a GOOD book on PYTHON for the beginner.
Well, it depends.  I really liked 'Learning Python' but I have several
other languages under my belt.  'Learning Python' is not for a
complete novice.  You need know about editors, files, command lines,
etc. as well knowing terms like array, variable, operator, function,
class, method, etc.  But since you posted a 'C' program to be
converted, I assume you know enough to get this far.  

HTH.

Jay


From mikael at isy.liu.se  Tue Jun 13 08:55:13 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 13 Jun 2000 14:55:13 +0200 (MET DST)
Subject: No Boolean type in "types"
In-Reply-To: <39462acb.0@katana.legend.co.uk>
Message-ID: <XFMail.000613145513.mikael@isy.liu.se>

On 13-Jun-00 Thomas Mangin wrote:
 >  I knew this, sorry I wasn't clear enough,
 >  Considering a "structural" point of view, having true and defined in a
 >  standard way
 >  and being able to "indicate" that the type is Boolean is a nice programming
 >  practice.
 >  In C one will use TRUE and FALSE. Those define are no more than integer too
 >  but help
 >  code readability. That is my point.

As I wrote earlier, there is no single boolean type. Integers, strings, 
lists, dictionaries, and even your own classes can be used as booleans.
You can do

  if {}:
    print 'Shouldn't happen!'

Defining a single boolean type is not only unnecessary, it's plain wrong.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    13-Jun-00
Time:    14:50:39

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


From rupole at compaq.net  Tue Jun  6 21:02:33 2000
From: rupole at compaq.net (Roger Upole)
Date: Wed, 07 Jun 2000 01:02:33 GMT
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and others" is this possible?
References: <393692FB.8610EBCA@dallas.net> <007901bfcbf2$c3b4e680$1906a8c0@fc.fenx.com> <3936FEFB.BA248CA3@sightreader.com> <003301bfcc30$57851820$1906a8c0@fc.fenx.com> <393D7BFD.F50F9932@sightreader.com>
Message-ID: <Bah%4.22817$Rq.2324142@pouncer.easynews.com>

Try win32pipe.popen.  This works very reliably.
               Roger Upole

"Ken Seehof" <12klat at sightreader.com> wrote in message
news:393D7BFD.F50F9932 at sightreader.com...
I'm running PythonWin
I tried python from the command line.  The process hangs.
>>> import os
>>> os.popen('dir').readlines()
 (hangs . . .)
IDLE gives the same results as PythonWin:
>>> import os
>>> os.popen('dir').readlines()
Traceback (innermost last):
  File "<pyshell#1>", line 1, in ?
    os.popen('dir').readlines()
OSError: (0, 'Error')
Apparently you are right about Window pipes.  <MSJAB> Not surprising.
Microsoft would probably get rid of the command line completely if not for
its legacy value.  It's hard to control the world when the prolls have
access to a usable command shell. </MSJAB> :-)
-- Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!
Emile van Sebille wrote:
 Ken, Are you trying this in IDLE?  I get the same error there.If so, try
starting python from within a dos window andsee if that makes a difference.
Pipes in winxx are somewhatflakey as I understand.
Emile van Sebille
emile at fenx.com
-------------------
----- Original Message -----
From: Ken Seehof
To: Emile van SebilleCc: python-list at python.orgSent: Thursday, June 01, 2000
5:25 PMSubject: Re: Python to call commands, e.g. "find , perl scripts ,
nmap , and others" is this possible?
 On NT I get:
>>> mylist = os.popen('dir').readlines()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
OSError: (0, 'Error')
(same thing with any command)
popen (command[, mode[, bufsize]])
    Open a pipe to or from command. The return value is an open
    file object connected to the pipe, which can be read or written
    depending on whether mode is 'r' (default) or 'w'. The bufsize
    argument has the same meaning as the corresponding argument
    to the built-in open() function. The exit status of the command
    (encoded in the format specified for wait()) is available as the
    return value of the close() method of the file object, except that
    when the exit status is zero (termination without errors), None is
    returned. Availability: Unix, Windows.
- Ken Seehof
Emile van Sebille wrote:
Jim, Take a look at os.popen, as in: import osmylist = os.popen('locate
myname').readlines()for eachfile in mylist: do something with eachfileHTH,
Emile van Sebille
emile at fenx.com
-------------------
----- Original Message -----
From: Jim
To: python-list at python.orgSent: Thursday, June 01, 2000 9:44 AMSubject:
Python to call commands, e.g. "find , perl scripts , nmap , and others" is
this possible?
Dear list,
I'm new to Python, and I'm sorry if I'm asking about anything that has been
already discussed.
I would like to use Python to call commands, e.g. "find , perl scripts  ,
nmap , and others" is this possible?
I do not need GUI. All I need is to link the out put and work with it. Is
perl or shell scripts my only
answer?
jim
--
Proprietary & Confidential.  The information transmitted is intended only
for the person or entity to whom it is addressed and may
contain confidential and/or privileged material.  Any review,
retransmission, dissemination, or other use of, or taking of any
action in reliance upon, this information by persons or entities other than
the intended recipient is prohibited.  If you received this in error, please
contact the sender and delete the material from any computer.





From arta at NOSPAMx-stream.nl  Mon Jun 12 14:28:51 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Mon, 12 Jun 2000 20:28:51 +0200
Subject: Client with select.select()
References: <#F2#YuI1$GA.106@net025s> <8i36rt$42mq$1@nntp6.u.washington.edu>
Message-ID: <ukLe#vJ1$GA.293@net025s>

Information which could be very useful. :)

Thanks,
A[r]TA

--
-Life can kick your ass, but all you got to do is stand up and move on...




From jae at ilk.de  Wed Jun  7 17:42:45 2000
From: jae at ilk.de (Juergen A. Erhard)
Date: Wed, 07 Jun 2000 23:42:45 +0200
Subject: Case-sensitivity?  Hal says "No" ;-)
Message-ID: <07062000.3@sanctum.jae.ddns.org>

In the course of getting into TCL[1], I came across a tutorial by Hal
Abelson, Philip Greenspun, and Lydia Sandon[1] (yes, *the* Hal Abelson
of SICP fame).  And in this I came across this quote:

  The great case-sensitivity winter descended upon humankind in 1970
  with the Unix operating system.

Just for the record, I'm *for* case-sensitivity (I'm an old Modula-2
hand...) I just thought this funny in the context of the ongoing
flamewar^Wdiscussion ;-)

Bye, J

[1] The document is `Tcl for Web Nerds', at
http://www.arsdigita.com/books/tcl/index.adp

[2] Don't ask... ok, if you insist: the company I'll start working for
soonish (if all goes well) has decided upon using NPS4, a
`content-management system' (no URL from me, it's proprietary).  And
this thing is only scriptable in TCL (it seems).[2]

[3] To avoid any flames: I'm heavily lobbying for Python, and I sure
won't stop (Perl, PHP, TCL (ewww) and Java (*blorg*) are pretty well
established though (Java has been decreed from `on high'))

But I'm the new, unproven guy on the block, so it's not that
easy... wish me luck (it's a rather largish, and by nature pretty
conservative company... but at least they're using Linux in a major
way (for servers, of course))

-- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
          My WebHome: http://members.tripod.com/Juergen_Erhard
        GIMP - Image Manipulation Program (http://www.gimp.org)
   "No matter how cynical I get, I can't keep up."  -- Bruce Schneier
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000607/423c28bb/attachment.sig>

From soper at delanet.com  Mon Jun 12 21:37:26 2000
From: soper at delanet.com (Paul Soper)
Date: Mon, 12 Jun 2000 21:37:26 -0400
Subject: rld failure on IRIX
References: <393839a7_4@corp.newsfeeds.com>
Message-ID: <39459030_2@corp.newsfeeds.com>

Randall Hopper provided the answer.  The problem was using libraries built
with different Application Binary Interfaces (o32 vs. n32).  The difference
was found by using the 'file' command on the python executable and the .so
files.  The solution was to force cc and ld to use the o32 ABI.






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


From mfletch at tpresence.com  Sun Jun 18 23:32:14 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 18 Jun 2000 23:32:14 -0400
Subject: Why is Python so slow ?- revisited.
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33C8@VRTMAIL>

The byte code compilation makes the doc string an attribute of the function
object, as opposed to a statement that needs to be executed on each
iteration of the function.

Enjoy,
Mike

-----Original Message-----
From: William Dandreta [mailto:wjdandreta at worldnet.att.net]
Sent: Sunday, June 18, 2000 11:22 PM
To: python-list at python.org
Subject: Re: Why is Python so slow ?- revisited.

...

The """ syntax would seem to be slower than putting # at the begining of
everyline because with # it only needs to test the first character of the
line, if it's # it can skip the rest of the line. But with """, it needs to
check every character after until it finds the matching """.

...



From stephen at cerebralmaelstrom.com  Fri Jun  2 05:04:14 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Fri, 02 Jun 2000 09:04:14 GMT
Subject: Copyright changes with move from CNRI to BeOpen?
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h7blm$j1b$1@slb2.atl.mindspring.net> <39375113.5AF5BC0B@gte.net>
Message-ID: <iMKZ4.31522$WS3.378122@typhoon.we.rr.com>

    Oh for lords sake I hope not. I don't think so, either... it seems like
Guido/et-al actually like it that Python can be used -anywhere-
and -everywhere-.

--S

Harry Tanovich <harryt at gte.net> wrote in message
news:39375113.5AF5BC0B at gte.net...
> Will the copyright move to a more GPL style thing now?
>
> Andrew Dalke wrote:
>
> > t_e_sanders at my-deja.com wrote:
> > >Do the terms of Python's distribution license require me to allow other
> > >users to copy and distribute my software? I would guess not, but
> > >inclusion of this notice in my package explicitly states so. There's
> > >obviously something I'm missing,
> >
> > The key word is "use".
> >
> > The copyright only applies to the Python code.  You are free to add
> > your own code to Python, which is under your own copyright.  The
> > clause you quoted allows you to use the Python code with your own
> > code, regardless of the license applied to your own code.
> >
> > The GPL, which you are probably thinking of, does not allow similarly
> > unfettered use of their code.  Packages compiled with GPL code
> > restrict their use to only GPL-compatible licenses.
> >
> > The inclusion of the statement in code you distribute only applies to
> > Python code.  Someone could copy that portion of the source from your
> > program and use it elsewhere.  However, since it's already available
> > from python.org, there nothing new revealed.
> >
> > But remember, I-am-not-a-lawyer.
> >
> >                     Andrew
> >                     dalke at acm.org
>




From frankc at colconsulting.com  Fri Jun  9 07:04:17 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 09 Jun 2000 07:04:17 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <3940CF31.A781AC9B@colconsulting.com>

Steve Mullarkey wrote:
> 
> [snip]
> 

18 (C) and 10 (C++)

> In "C++" I have :-
> 
> 1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
> Profilers, Programmers Editor including Class Browser, etc..
> 
> 2. I use the STL (Standard Template Library) which gives powerful string
> and container classes.
> 
> 3. I use incremental compiling and linking. This reduces wait times to
> very small amounts.
> 
> 4. I can also use background compiling and linking.
> 
> 5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
> etc..
> 
> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated. I am genuinely interested
> to ask "C++" programmers :-

Don't read that I am foresaking the C++ language, I'm not...I love it
and will continue with it for a long, long, time.

Both languages have a rich set of libraries. It is not apples to apples
when one language is compiled and linked, and the other is interpereted.
With Python you can skip the comp+linkedit step and just code and run
which equates to the potential for more work in less time. I find that
it saves more time than the equivalent task in Java as well.

I don't understand why people put background compiling and linking as
some kind of feature. The only time this has worked for me has been on a
pre-emptive multi-tasking OS (Linux, OS/2, NT, MVS) when what I was
compiling was a separate executable or library whilst I worked on
another body of code. At the speed of todays computers, you can barely
get to the other body of code before the other background compile and
linkedit have finished.

I shall enumerate some Python DIFFERENCES as it relates to the REAL
drudge of development:

0. Platform independent.

1. No waiting for compile and linkedit, just write and run.

2. No need for creating makefiles, just run.

3. Very RICH set of packages (libraries).

4. Idle provides a good source level debugger.
 
> 1. What productivity increase do you achieve ?

Again, if you are always doing something else during the housekeeping of
C++ than the difference is minimal. I rarely find this to be the case,
we are often sitting like cave men in front of a fire watching the
compilation for errors and warnings. So for me the productivity increase
is the gain from not waiting for compiles, not dealing with make files,
and faster debugging.
 
> 2. How long did you use Python before you achieved increased
> productivity ?

Generally, the class and object took about a day, the language features
are still being realized. At this point I have been putting together a
project for creating a package of GoF Design Patterns (Gamma et. al) and
Collections (STL-ish) in pure Python.
 
-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From lv at claradot.codot.uk  Mon Jun 26 14:12:41 2000
From: lv at claradot.codot.uk (Laurence V.)
Date: Mon, 26 Jun 2000 18:12:41 GMT
Subject: JPython and Java integration
References: <8j7h3m$5v18j$1@fu-berlin.de>
Message-ID: <39579a03.17964799@news.clara.net>

On Mon, 26 Jun 2000 14:07:20 +0200, "Maurice le Rutte" <_delete_this_mlerutte at square.nl> wrote:

>I'm thinking of using JPython as our 'scripting'/plug-in language for our
>product written in Java. With the risk of being flamed beyond recognition:
>like MS's 'VBA', but then better.

If your product is a development tool for Java programmers, it makes more sense to use one of the Java interpreters as
your scripting engine. Learning curve = 0. JPython learning curve = ?high? for people who don't know it.

   L.


From eito at cs.pdx.edu  Thu Jun  1 03:39:20 2000
From: eito at cs.pdx.edu (eito)
Date: Thu, 1 Jun 2000 00:39:20 -0700
Subject: ANN: bzip2 module
Message-ID: <20000601003920.A18615@dsl-only.net>

Hi all,

I would like to announce about a bzip2 module. This is my
first attempt to write Python module. I really don't know
how to package, announce, etc... I would appreciate if
somebody can tell me what is the good way to do this kind
of thing.


Here is some information about the module:

    Name: bzlibmodule
 Version: 0.2
  Author: eito (eito at cs.pdx.edu)
Location: http://www.cs.pdx.edu/~eito/bzlib-0.2.tar.gz
    Note: compress and decompress functions like zlib module.
          compress object support, but no decompress object
          support yet.


Thanks

:eito



From thomas at xs4all.net  Sat Jun 17 14:36:51 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 17 Jun 2000 20:36:51 +0200
Subject: block comments
In-Reply-To: <4mP25.63531$WS3.629201@typhoon.we.rr.com>; from pshinners@mediaone.net on Sat, Jun 17, 2000 at 06:22:56PM +0000
References: <4mP25.63531$WS3.629201@typhoon.we.rr.com>
Message-ID: <20000617203650.U26436@xs4all.nl>

On Sat, Jun 17, 2000 at 06:22:56PM +0000, Pete Shinners wrote:
> as i get working with python more i start to miss the
> block commenting i had in C/C++

> i'm just doublechecking.. is there a block comment
> symbol in python? i'd like to commentout a large
> chunk of code with a simple start/end comment.

Triple quotes may be what you want. Oh wait ;)

> thinking about it, i suppose i could use the triple
> quotes to simply turn a large chunk of code into a
> string. is there a better way?

I usually use normal comments: my editor makes it fairly easy to add '#' (I
usually use ' ## ' to make clear that it's intended as a mass-comment-out)
at the start of each line in a particular block.

> should python have block comments? (py3k) and if it
> does there must be an answer to the eternal question...
> Are the block comments nestable?

The problem with block quotes is exactly that ;) One of the reasons C++ (and
ISO C) have the '//' line-comment (that lasts until the end of the line) is
the large number of problems that arrise when you try to quote a block of
code that has a comment in it, or accidentily screw up an end-marker (thus
leaving all code until the next end-marker commented out.)

This is why I prefer just adding ' ## ' to the start of each line i want to
block-comment-out. It *does* nest :-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From johnm at magnet.com  Tue Jun 27 11:31:04 2000
From: johnm at magnet.com (John Mitchell)
Date: Tue, 27 Jun 2000 11:31:04 -0400
Subject: zope, python vs apache, perl
In-Reply-To: <20000626220547.B21453@corrada.com>
Message-ID: <Pine.SGI.4.05.10006271047340.13130-100000@lemur.magnet.com>

Another possibility is to grab the Zope component libraries and use them
in a FastCGI environment.   That is, you'd write a more-or-less CGI script
in Python, but many things would be made easier using the Zope libs.

I used this approach for a medium-largish ad server.  It's been serving
100K ads per day for two years with no problems, and I was using *old*
libraries...

cool stuff to steal:

- easy persistent objects (no more stupid databases!  woo!)

- call any object through the web

- rich multilayered, if odd, template library.


Surprisingly, this combines to make it *very* easy to provide a remote
console to your running applicaton.  View internals, start/stop, and
provide testing information.  Leave debugging information in your live
server to make testing trivial.

For example, this bit of code renders a pretty HTML page giving a overview
of the ad service.  Uptime, number of impressions/clicks, CPU time; also
the version running, hostname, and other stuff.  Just imagine if I had
HTML skills, this leetle bit of code could translate directly into
sales...

class adServer:
	...
    # class-global:
    _showServerTemplate = HTMLFile('ad_server.dtml')

    def showServer(self, REQUEST):
        "."
        cfg = getConfig()
        uptime_secs = self.getUptime()
     
    return  self._showServerTemplate(
        title   = 'Ad Server Detail',
            adlogic_id  = vc_id,
            sys     = sys,          # sys.platform, sys.version
            config  = cfg,          # .hostname(), .is_live()
            uptime  = self.time_dhms( uptime_secs ),
            uptime_secs = uptime_secs,
            impressions = self.getRequestID(),
            mapping = self.getSysTimes()
        )
	...

Original URL:
	(server)/cgi-bin/ads/show/showServer

Here's the results:
	http://people.magnet.com/~johnm/adserver.html


I think the high-level Zope stuff is really weird, but the libraries are
great, and I'm getting better...



- j





From hernanf at my-deja.com  Thu Jun 22 18:00:11 2000
From: hernanf at my-deja.com (Hernan M. Foffani)
Date: Thu, 22 Jun 2000 22:00:11 GMT
Subject: Lista de correo sobre Python
References: <Bi%35.2093$Hc7.11930@telenews.teleline.es>
Message-ID: <8iu290$puu$1@nnrp1.deja.com>

[posted and mailed]

Ya hay una lista de correo de python en espa?ol, y bastante exitosa por
cierto, en python-ar at lists.arnet.net.ar
Para suscribirse a la lista ir a:
http://lists.arnet.net.ar/mailman/listinfo/python-ar
Saludos,
-Hern?n

In article <Bi%35.2093$Hc7.11930 at telenews.teleline.es>,
  "JD" <coronel_jadonNOSPAM at teleline.es> wrote:
> Hola a todos:
>
> Escribo para comunicar que se ha creado una lista de correo sobre
Python en
> Espa?ol. En la lista contamos experiencias, noticias, consultamos
> dudas,etc.. Todo sobre Python y ademas en Espa?ol.
>
> Para suscrbirse hay que enviar un correo a la direcci?n:
> python_dev-subscribe at egroups.com
>
> Un saludo a todos y gracias.
>
> Thank you for all.
>
>

--
Hern?n Mart?nez Foffani
hernanf at my-deja.com


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


From python-list at teleo.net  Thu Jun 22 03:54:48 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Thu, 22 Jun 2000 00:54:48 -0700
Subject: XML-code examples
In-Reply-To: <394f7bc2.365514882@news.online.no>
References: <394f7bc2.365514882@news.online.no>
Message-ID: <00062201025104.08285@quadra.teleo.net>

[Thomas Weholt, on Wed, 14 Jun 2000]
:: Is there more simple XML code out there? I mean the HowTos took about
:: ten minutes, and it didn`t cover much. 
:: 
:: If somebody has some tips or code on how to traverse a DOM-tree,
:: processing nodes based on tagname, and especially, based on relations
:: with other nodes ( i.e. who`s your daddy, here`s your sister etc ),
:: that would be great.


While waiting for Sean McGrath's book to hit the shelves, you may find
that David Mertz's nice Python XML introductory article gives you a leg
up:

http://www-4.ibm.com/software/developer/library/python1/



From millja at wwc.edu  Wed Jun 14 19:47:43 2000
From: millja at wwc.edu (millja at wwc.edu)
Date: Wed, 14 Jun 2000 23:47:43 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com> <8i6u39$re3$1@nnrp2.deja.com> <8i8le9$b6u$1@nnrp1.deja.com>
Message-ID: <8i95iq$o3a$1@nnrp1.deja.com>

Yup that works!  Thanks.  I now have another question.

I need to (well, would like to) have my Python function returns a list
(really just an array of integers).  VBScript dosnt seem to like this
very much and throws up an invalid data type error.  I tryed messing
around with VBSs array() function but thats not showing me any love.
Anyone have ideas?

thanks
jason milliron

btw, its freakn awesome to have mark hammond reply to your post :)



In article <8i8le9$b6u$1 at nnrp1.deja.com>,
  Mark Hammond <mhammond at skippinet.com.au> wrote:
> In article <8i6u39$re3$1 at nnrp2.deja.com>,
>   millja at my-deja.com wrote:
>
> > would like to integrate.  Is there some way to
> > call this function from within the VBScript?
> >  Or, it would work if I can get numbers,
> > generated by the VBScript, into the Python
> > function and the results of this function back to
> > the VBScript.
>
> VB should be able to call any Python function defined on the page,
> simply by calling it by name.
>
> Python should be able to call any VB function defined on the page with
> some magic syntax that Im not sure of :-)
>
> Under IE, you do it via "Window.VBFunc()".  I assume that under ASP,
it
> will be via the server or document object - most likely the document -
> eg "Document.VBFunc()"
>
> Feel free to post a sample once you get it working :-)
>
> Mark.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From g_will at cyberus.ca  Tue Jun 20 09:15:56 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: Tue, 20 Jun 2000 13:15:56 GMT
Subject: Pythonwin-wxPython
References: <394D3024.7E2B1A11@hotmail.com> <8F57DAB28gmcmhypernetcom@199.171.54.154>
Message-ID: <01BFD9F7$8061B620$8B54C3D1@CYBERUS>

Yes this is a real drag, but the easy way to get around it is to include a
try and except and then it works with pywin and you get your traceback.

    class myFrame(wxFrame):
        def __init__(self, parent=NULL):
            wxFrame.__init__(self, parent, -1, "wx_plot_lines",
                wxDefaultPosition,wxSize(600,520))
            try:
                plotCanvas(self)  #your panel
            except:
                self.Destroy() # causes main loop to exit
                traceback.print_exc() # and you still get your trace

As well you need to import traceback.

Regards,

Gordon Williams


Gordon McMillan <gmcm at hypernet.com> wrote in article
<8F57DAB28gmcmhypernetcom at 199.171.54.154>...
> kupisch at hotmail.com (ratze) wrote:
>
> >My  problem is that when I am developing a wxPython application
> >everytime the the start of my app fails due to an exception I cannot
> >start it again and get 'Only 1 wxApp per process!'. How do I get around
> >this??
>
> Run the wxPython from the command line. You can get around this if you
> really want to, but it's not worth the struggle.
>
> - Gordon
>


From DOUGS at oceanic.com  Sun Jun 18 13:05:24 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Sun, 18 Jun 2000 07:05:24 -1000
Subject: Using Python instead of Perl?
Message-ID: <8457258D741DD411BD3D0050DA62365907A23C@huina.oceanic.com>

[Jim said:]
> I am trying to use nmap to test my own network and automate the process.

What's nmap?

> I keep seeing perl being used for this type of work , but was hopeing
> that Python could do this with out all the lines of code needed for
> perl. Is perl a better choice than Python for this?

I don't think Perl is a better choice for anything IMHO. :-)

I use Python for network management tasks and find I can do everything _I_
need to do.  I think Perl has a larger user base, particularly in comparison
to those using Python for net mgmt tasks.  This means you'll find more
people that have solved your kinds of problems.  As evidenced by my question
above, your problems might not be my problems so I may not have spent time
building solutions.

That being said, I'm strongly biased towards the modularity and OO abilities
of Python.  I see on a daily basis that I'm more productive as a programming
wannabe than a Perl using co-worker.  Not all of the solutions may be out of
the box, but I can usually whip up something faster than he can find or
build it.

> Any help would be welcomed.
> Thanks.

I hope you stick with Python and help to build towards a critical mass using
it for net mgmt tasks.  Give me a link for nmap and I may very well want to
help there.

-Doug-



From fiona at sitegnome.com  Mon Jun  5 18:05:07 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 5 Jun 2000 22:05:07 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 5th, 2000
Message-ID: <20000605220507.12522.qmail@synop.com>

Hello All,

Below are the latest entries entered into http://python.faqts.com.

cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Is there an easy way to do user input in python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3492
-------------------------------------------------------------
Fiona Czuczman
Steven Adams

use raw_input

>>> def get_SPAM_n_EGGS():
                    spam = raw_input('Spam? ')
                    eggs = raw_input('and Eggs? ')
                    print spam, eggs


>>> get_SPAM_n_EGGS()
Spam? loads
and Eggs? yessiree
loads yessiree
>>>


-------------------------------------------------------------
Where can I find an example on how to use python with Tcl/Tk to write GUI windows applications?
http://www.faqts.com/knowledge-base/view.phtml/aid/3490
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

from Tkinter import *
root=Tk()
label=Label(root,width=12)
def clicker():
    label.configure(text="Hello Jilani")
button=Button(root,text="Click Me",command=clicker)
label.pack(side=TOP)
button.pack(side=TOP)
root.mainloop()

Try www.pythonware.com/library.htm

Also there is a good book by John Grayson (Python and Tkinter 
Programming) and one on the way by Frederik Lundh published by O'Reilly.


-------------------------------------------------------------
I'm currently using the storbinary method for transferring files, how can I get some indication that I've transferred xx% of the file, or 100,000 bytes so far?
http://www.faqts.com/knowledge-base/view.phtml/aid/3493
-------------------------------------------------------------
Fiona Czuczman
Peter Schneider-Kamp

Just define your own version of storbinary in your module:

def mystorbinary(self, cmd, fp, blocksize):
  '''Store a file in binary mode and print stuff.'''
  self.voidcmd('TYPE I')
  conn = self.transfercmd(cmd)
  counter = 0
  while 1:
    print counter,"Bytes read yet."
    buf = fp.read(blocksize)
    if not buf: break
    conn.send(buf)
    counter = counter + blocksize
  conn.close()
  return self.voidresp()


## Edited Entries ##############################################


-------------------------------------------------------------
Where can I find information about how to manipulate Excel with Python?
Can I access an excel (.xls) file via com without having excel loaded on that machine?
http://www.faqts.com/knowledge-base/view.phtml/aid/3093
-------------------------------------------------------------
Fiona Czuczman
Robert Citek,Eric Hagemann, Scott

Read Hammond's & Robinson's "Programming Python in Win32".  There is a 
very good headstart there with an example of a class that eases the 
interaction w/Excel via COM.  Next get "Writing Excel Macros" by 
O'Reilly.  It is all about how to program Excel via Visual Basic but 
translates very well into Python COM.

Visit http://www.oreilly.com/catalog/pythonwin32

If you can't afford to buy it, it seems you can view it at 
http://www.netlibrary.com .  It's a neat site that allows you to "check 
out" books for a period of time  (usually 24 hours at a time) and read 
them online.  The only drawback is they only allow so many "check outs" 
of each book so it may or may not be available to read.

Here is an exact URL:
http://www.netlibrary.com/SUMMARY.ASP?EV=1239265&ID=24666&advquery=

and chapter 9 is the main Excel chapter.

Here is some sample code:

# this example starts Excel, creates a new workbook, 
# puts some text in the first and second cell
# closes the workbook without saving the changes
# and closes Excel.  This happens really fast, so
# you may want to comment out some lines and add them
# back in one at a time ... or do the commands interactively

from win32com.client import Dispatch

xlApp = Dispatch("Excel.Application")
xlApp.Visible = 1
xlApp.Workbooks.Add()
xlApp.ActiveSheet.Cells(1,1).Value = 'Python Rules!'
xlApp.ActiveWorkbook.ActiveSheet.Cells(1,2).Value = 'Python Rules 2!'
xlApp.Close(SaveChanges=0)
xlApp.Quit()
del xlApp

# raw_input("press Enter ...")


-------------------------------------------------------------
Where can I find some kind of helping material (tutorials etc) about how to use XML and Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3468
-------------------------------------------------------------
Fiona Czuczman, Fred Drake
Boudewijn Rempt, Harry George

Take a look at the Python/XML HOW-TOs and related documentation,
located at:

http://www.python.org/topics/xml/docs.html

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

Have a look here:

http://www.valdyas.org/python/index.html

dbobj - the repository is XML based
kura  - webserver that takes XML templates and spews out HTML, and an
        XML data import utility.
kxml  - a small PyKDE xml file browser (unfinished!)

A really simplistic reader/writer for Abiword.  Available at:
  http://www.seanet.com/~hgg9140/comp/index.html

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

Install PyXML, then try this.  Read the xml.dom.core for element names
and the basic functions.  If you are not familiar with xml itself, see
the XML Bible.  Try this routine on some known-good .xml file.

#!/usr/bin/env python
#/* helloxml.py 

"""
Usage: helloxml.py [options] filename
E.g.,  helloxml.py myfile
Options
-h,--help        this message
-v,--version     version

Revisions:
2000-02-23  Harry George   initial version
"""

import sys,os,getopt,time,re,string,exceptions

modname="helloxml"
__version__="0.1"

def debug(ftn,txt):
        sys.stdout.write(modname+'.'+ftn+':'+txt+'\n')
        sys.stdout.flush()

def fatal(ftn,txt):
        sys.stdout.write(modname+'.'+ftn+':FATAL:'+txt+'\n')
        sys.stdout.flush()
        sys.exit(1)

def usage():
        print __doc__

#====================================
from xml.dom import core, utils

def main():
        ftn = "main"
        pargs=[]    #positional args, default is empty
        opts,pargs=getopt.getopt(sys.argv[1:],'hv',['help','version'])
        for opt in opts:
                if opt[0]=='-h' or opt[0]=='--help':
                        usage()
                        sys.exit(0)
                elif opt[0]=='-v' or opt[0]=='--version':
                        print modname+": version="+__version__
                        sys.exit(0)

        #---get the DOM "doc"---
        if len(pargs)==0:
                        usage()
                        sys.exit(0)             
        filename = pargs[0]
        doc = utils.FileReader( filename ).document

        #---dump it out---
        doc.dump();

        #---convert DOM tree back to XML, and dump that
        xml = doc.toxml()
        print xml

        #---dump the doc manually---
        root=doc.get_documentElement()  
        mydump(root)

def mydump(node):
        ftn = "mydump"
        debug(ftn,"name=" +node.get_nodeName())
        #---dump attributes---
        attrs=node.get_attributes()
        keys=attrs.keys()
        for attr in keys:
                debug(ftn,"attr="+attr+" 
value="+attrs[attr].get_nodeValue())
        #---dump children---
        children=node.get_childNodes()
        for child in children:
        #--report only elements
                if child.get_nodeType()==core.ELEMENT: 
                        mydump(child)
                         
if __name__ == '__main__': main()






From herzog at online.de  Wed Jun 14 13:38:01 2000
From: herzog at online.de (Bernhard Herzog)
Date: 14 Jun 2000 19:38:01 +0200
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil>
Message-ID: <m3hfawjh8m.fsf@greebo.nodomain.de>

"Louis M. Pecora" <pecora at anvil.nrl.navy.mil> writes:

> In article <etdk8ftm55x.fsf at w20-575-109.mit.edu>, Alex
> <cut_me_out at hotmail.com> wrote:
> 
> > Wow, thanks for looking that code over, Bernhard.  What a horrible bunch
> > of holes.

You're welcome.

> And who could have seen all that so clearly without an expert on hand. 
> Even then I couldn't understand all the reasoning.  In Python we are
> all happy, but the ugly side is when we return to C. 

Well, in the vast majority of cases you can get by with a few simple
rules (using the terminology of the extension docs):

 1. Functions that return references, return new references.

 2. If you pass a reference to a function, that function borrows the
    reference.

Unfortunately there are some exceptions, such as PyDict_GetItem. The
Extension and Embedding document does a pretty good job if explaining
things, IMHO. The exceptions could be better documented, though, but
then again, my local copy of the API reference is a bit old now.

In my experience, refcounting only becomes tedious and error prone if
you try to do too complex things in C. Try to do as much as possible in
Python and keep your C-extensions simple as far as constructing and
manipulating python objects is concerned.

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


From stephen at cerebralmaelstrom.com  Sun Jun 18 13:02:21 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sun, 18 Jun 2000 17:02:21 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com>
Message-ID: <xg735.63947$WS3.647315@typhoon.we.rr.com>

    I'm currently experimenting with PythonWin's Show-Tabs thing. I need to
get used to it, though...because as is it looks very distracting. :)

--S

Emile van Sebille <emile at fenx.com> wrote in message
news:02b001bfd885$fbefa5e0$1906a8c0 at fc.fenx.com...
>
> ----- Original Message -----
> From: "Stephen Hansen" <stephen at cerebralmaelstrom.com>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Saturday, June 17, 2000 8:51 AM
> Subject: Re: Why whitespace denotation of blocks doesn't
> work.
>
>
> >     I discovered Python a bit over a year ago, and have
> been using it
> > relatively constantly in idle projects. Some are simple,
> some are not.
> > Currently i'm working on one that's not.
> >
> >     Initially, the indentation thing struck me as weird.
> My main complaint
> > about it is not that it's difficult to maintain proper
> indentation, but that
> > if you're scrolled far down in nested 'if', for instance,
> it's sometimes
> > hard to see which indentation-level is for which 'if'...
> but then i'll just
> > split-my-screen and scroll up.
> >
> <snipped>
>
> Which is why I prefer an editor that shows whitespace
> and use tabs.  That makes it easy to follow the column
> of tab marks down the screen.
>
> Emile van Sebille
> emile at fenx.com
> -------------------
>
>
>
>




From avv at quasar.ipa.nw.ru  Thu Jun 15 17:44:09 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Thu, 15 Jun 2000 14:44:09 -0700
Subject: Killing wxPython process in PythonWin
References: <8i7178$44c$1@nnrp1.deja.com> <8F536925Fgmcmhypernetcom@199.171.54.194> <q5M15.4703$cQ.38625@news-server.bigpond.net.au>
Message-ID: <39494E29.3373D7FF@quasar.ipa.nw.ru>

Hi,

Neil Hodgson wrote:
> 
> > I hate to tell you this, but except for very, very simple scripts, I never
> > run anything from within Pythonwin or IDLE. If I do run from inside the
> > IDE, (say, to use the debugger), I generally expect I'll have to exit and
> > restart between runs. I've chased too many "bugs" (that were really fixed,
> > but something wasn't reloaded) to trust anything but a fresh start. And
> I'm
> > certainly not going to bother with the bizarre hacks people come up with
> > (eg, to replace the "code" but not the data in "running" instances).
> 
>    Looks like what is needed is a 'run in new environment' option for
> Pythonwin/IDLE. Or in a separate process?

Again, no problems with (nt)emacs. If you are in a hurry, you may even
launch your wxPython app several times though a py-shell, then close it
and reopen. I'm not sure however, about ntemacs in this last point. But
just in a *Python-Output* way, it works excellent. I did though hacked
python-mode.el to enable asychronous execution of the script (like one
used in *compile*), to be able to see Python prints while the app is
running.

Alexander



From aahz at netcom.com  Fri Jun 23 10:03:15 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jun 2000 14:03:15 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand>
Message-ID: <8ivqn3$frs$1@slb0.atl.mindspring.net>

In article <m3k8fiosky.fsf at gondolin.beleriand>,
Steffen Ries  <steffen.ries at sympatico.ca> wrote:
>
>I'm trying use re.split to separate an LDAP Distinguished Name into
>its naming components. The naming components are of the form
>"attr=value" and they are separated by ",".
>
>The part, which I can't solve is, that "value" may contain the
>sequence "\,", which obviously does not separate a naming component.

Unfortunately, specifying a string to *not* use is one of the hardest
tasks for an re.  While it can be done, it basically makes your code
unreadable and unmaintainable to anyone except an re expert.  Cliff's
suggestion is probably the best other than what you're currently doing.
In any event, comment this section of code carefully.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Why don't they make poppasmarts or didjeridon'ts?  --Aahz


From daniel at visionart.com  Mon Jun 12 18:19:47 2000
From: daniel at visionart.com (Daniel Kramer)
Date: Mon, 12 Jun 2000 15:19:47 -0700
Subject: Tkinter Speed
Message-ID: <39456202.D9C306EE@visionart.com>

I have a simple tcl/tk program.. it's only a canvas with some rectangles
on it.. you can move the rectangles around with the mouse on the canvas
interactively.

I am learning python and re-wrote this simple example with Tkinter as my
first python prog and it was much slower then the tcl/tk version.  I was
reading though past posts and it seems that you can call tk directly in
some way to gain a speed increase with Tkinter.. I tried to implement
this myself on my move command..

so my command looked like:

self.canvas.move("current", event.x - self.lastx, event.y - self.lasty)

I changed it to:

self.canvas.tk.call(self.canvas, 'move', 'current', event.x -
self.lastx, event.y - self.lasty)

Which works but isn't noticeably faster.   My tcl/tk app still feels
much faster.

I read that wxWindows is alot faster but I am a bit weary of going that
route.. Tkinter is so easy to use and I already know tk since I have
written many tcl/tk apps before... is there no way to get it to the same
speed as a tcl/tk app with Tkinter?

Thanks,

Daniel

--
-----------------------------------------------------------------------------
Daniel Kramer
http://www.visionart.com/~daniel/
daniel at visionart.com:           VisionArt Design and Animation





From cg at gaia.cdg.acriter.nl  Fri Jun 16 07:20:31 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 16 Jun 2000 13:20:31 +0200
Subject: Pini/PythonSpaces?
Message-ID: <8id2hv$joa$1@gaia.cdg.acriter.nl>

I brought this up on the Jini mailing list, and got some interest back, so
I'm just dropping the idea here: I think it would be quite possible to 
write something with exactly the same functionality as Jini for Python. And
as a lot of people are doing very interesting things with Jini, it might
be a worthwhile project.

For people who don't know Jini, here's a short rundown: Jini provides a 
minimalistic set of services for distributed computing. For example, there
is a lookup service, which allows you to find objects by type ("give me
something that implements interface A") - the lookup service typically
is found through a multicast protocol, so you don't need any prior 
configuration on either clients or servers. There is a transaction
manager, which helps transaction participants coordinate there work. There
is the very powerful notion of leasing all resources (like lookup service
registrations), which enables the network to heal itself, get rid of
old cruft, etcetera. Finally, there's event notification for asynchronous
messaging.

JavaSpaces is a tuplespaces implementation that adds Java and Jini extensions
to the basic notion. Tuples are typed, so again you can look for tuples that
implement a given interface, and they are leased, so the space keeps itself
clean from tuples that no-one is interested in (eg., when no-one renews the
lease for them). Jini transactions are used to provide read/modify/write
variables, which are a powerful tool for implementing distributed 
algorithms.

What's needed to implement something like this is a) some type system (the
idea behind using types is that software is much better at dealing with 
type systems than arbitrary string schemes when searching for stuff) and
b) there is a notion of mobile code, because you ask for an interface
and then receive, over the network, the implementation. Both are available
under Python, so it would be quite possible to do this (some more supporting
stuff, like a standard for marshalling, good network I/O libs, etcetera are
also all there under Python). 

As I said, just an idea I had. If someone picks it up, that'd be nice.
-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From tjs at cs.monash.edu.au  Sun Jun 18 21:25:09 2000
From: tjs at cs.monash.edu.au (Toby J Sargeant)
Date: Mon, 19 Jun 2000 11:25:09 +1000
Subject: Why whitespace denotation of blocks doesn't work.
In-Reply-To: <dle35.2194$qz4.17968@news-server.bigpond.net.au>; from neilh@scintilla.org on Mon, Jun 19, 2000 at 01:05:13AM +0000
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com> <dle35.2194$qz4.17968@news-server.bigpond.net.au>
Message-ID: <20000619112509.A20245@fangorn.csse.monash.edu.au>

On Mon, Jun 19, 2000 at 01:05:13AM +0000, Neil Hodgson wrote:
>    You can change the colour to silver to make it less heavy. Or hack the
> code to any colour you want. Originally tabs where displayed as round
> cornered rectangles but Mark thought this was too heavy. Perhaps vertical
> lines at the start of tabs instead would be lighter and make it easy to
> match up code.
> 
>    Neil

I saw a screenshot once of a vim session that used the ? (ascii 0273)
character to display tabs. It was actually very visually appealing, and I was
a little annoyed when I failed to get emacs to do the same thing.

Toby.



From petersc at stud.ntnu.no  Fri Jun 16 05:58:40 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 16 Jun 2000 11:58:40 +0200
Subject: Conversion from proprietary satellite control script language
References: <20000616084700.29919.qmail@web2006.mail.yahoo.com>
Message-ID: <3949FA50.116331D6@stud.ntnu.no>

TJ Judson wrote:
> 
> 1. It doesn't exist, but it would not be hard to add (The necessary
> opcodes are there -- well, the one you need: JUMP_ABSOLUTE --
> mwh21 at cam.ac.uk)

It wouldn't be hard? I just reread the thread on the subject from
this April and there were some serious problems:
* gotos into blocks
* gotos out of blocks
* some reducibility problem

> 2. There are a few "good" uses for it:  (jump to cleanup code, for
> use by code generators, as a performance optimization -- John
> Mitchell)

a) jump to cleanup code can IMHO be done through
   try: ... {finally|except}  ...
   constructions.
b) Can you elaborate the code generators point?
c) there is much more room for performance optimization at other
   times and places (optimizing compiler etc.).
   anyway I would not like to see a possibly harmful statement
   like goto added to Python for performance optimization. yuck.

> I have a new and much better reason :-) to add a goto statement... To
> support conversions from other, existing scripts/languages.

I don't mean to be flaming, but I cannot resist the following<wink>:

Let's add lodsb, lodsw, lodsd, stosb, stosw, stosd, movzx and some
of their friends to python. It would make conversion from the
most used assembler languages much easier. Using a dissassembler
we could then easily port Microsoft Word to Python. Just do a:
disasm "/mnt/win/Program Files/Microsoft Office/Office/msword.exe" |
asm2py > msword.py

Really, I don't see a legitimate use for goto in Python.

> I think it would be neat if Python were being used to fly satellites.
> Or it could be provided only as a patch...

That would, of course, be okay, if somebody figures out how to do it.

would-even-help-to-create-such-a-patch-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From shapr at uab.edu  Wed Jun  7 18:37:09 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 07 Jun 2000 22:37:09 GMT
Subject: Definite Newbie Question....
References: <q_z%4.828$223.8755@news1.tor.primus.ca>
Message-ID: <393ECE53.F1C0F622@uab.edu>

Robin Porter wrote:

There's lots of great documentation for python at:
http://www.python.org/doc/

If you read the tutorial, I think you'll learn these kinds of things
there.

But, in answer to your question, I think this script was expected to run
from the command line like "python scriptname.py"
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From info at pythonware.com  Mon Jun  5 09:27:49 2000
From: info at pythonware.com (PythonWare)
Date: Mon, 5 Jun 2000 15:27:49 +0200
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 2000)
Message-ID: <002901bfcef1$d8fa3660$0500a8c0@secret.pythonware.com>

release week, cont'd:

    The PythonWare py152 distribution is a small and self-contained
    Python distribution for Windows.  This kit includes Python 1.5.2,
    Tkinter/Tk 8.2.3, PIL 1.0, PST, and other useful extensions.

    Get your copy from:

        http://www.pythonware.com/downloads.htm

    This kit replaces the old (but surprisingly popular) py15 kit.

    Note that this is unsupported software.  Commercial support for
    this distribution is available to PythonWorks customers.

    For more information, contact us at info at pythonware.com.

enjoy,
the pythonworks team

    "Secret Labs -- makers of fine pythonware since 1997."




From vigenereNOviSPAM at leland.stanford.edu.invalid  Mon Jun  5 05:28:53 2000
From: vigenereNOviSPAM at leland.stanford.edu.invalid (cheesecake)
Date: Mon, 05 Jun 2000 02:28:53 -0700
Subject: search for a string?
References: <8h6211$ark$1@news-int.gatech.edu> <LbBZ4.31403$WS3.364156@typhoon.we.rr.com> <8h6na4$ql7$1@nntp9.atl.mindspring.net>
Message-ID: <33971223.8ec7c617@usw-ex0103-019.remarq.com>

In article <8h6na4$ql7$1 at nntp9.atl.mindspring.net>, David
Broadwell <"dbroadwell"@spam.mindspring.com (remove spam.)>
wrote:
>>
>>     OldString = File.read() + OldString
>>
>>     I perfer, though: OldString = "%s %s" % (File.read
(),OldString)
>>
>The second once certianly looks more .. programmer. Now which
is more
>efficent? (faced with to apparenly equal choices) and why?

As I understand it, using format strings is much more efficient
for combining many strings, even for concatenating just three
strings because str1 + str2 + str3 performs too many
allocations, something like doing (str1 + str2) + str3,
whereas "%s%s%s" % (str1, str2, str3, ) only allocates one
result string.

I prefer using the %s method even for 2 strings because I can
decide later to add another string without having to change my
code for efficiency. I'm not sure which method is more efficient
for 2 strings.

I just tried to look up where I read a comparison between the
two methods, and I couldn't find it, so I could be imagining
things. Can somebody maybe confirm this?

TD


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



From donn at u.washington.edu  Wed Jun 21 12:41:35 2000
From: donn at u.washington.edu (Donn Cave)
Date: 21 Jun 2000 16:41:35 GMT
Subject: Enums and Python
References: <E467F8B778AB0AF4.9582DC49D6332C2C.946310C9CD9F5BE8@lp.airnews.net> <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net> <8ip8rf$1ig$1@slb7.atl.mindspring.net>
Message-ID: <8iqr7v$aq72$1@nntp6.u.washington.edu>

Quoth claird at starbase.neosoft.com (Cameron Laird):
[ ... re sequence and symbolic indices vs. dict and text keys (I think?) ...]

| I'll say that more forcefully than Aahz does here:  while
| enums are often a marker of good taste in C coding, Python
| generally achieves the same results with a considerably
| cleaner reformulation in terms of dicts.  If you feel a
| need for enums while writing Python, you're probably missing
| out on a better (and not-so-reminiscent-of-C) coding.
|
| I've become rather dogmatic on this point.  I need the
| practice of working to falsify it.  If you have a specific
| algorithm you can share that you see as ripe for a Python
| enum, please tell us here; maybe we can suggest an
| alternative.

When I feel these pangs the object at hand is invariably a tuple
returned from a C module.  posix.stat, time.gmtime, et al.  What
I want, of course, isn't really enums but struct members.  So
I'm not sure if this is on topic - we got here because structs
turn into tuples, tuples take indices and no one can remember them.

I think the tuple is what we want, it's not like the authors
of these functions made a mistake and should have returned dicts
or objects with struct member attributes - I mean, that would
be wonderful but not worth the additional complexity in the C
module layer, especially considering that these constructs are
not only output but also input.  Suggest an alternative if you
want, but for me it would be a hard sell.

At a higher level, a wrapper is an option.  I append some code
I have used a little for this, for a struct with member attributes
that doesn't entirely lose its tuple.  If I remember right it's
even recursive, tuple inside a tuple will also have attributes
if its tuple-type is supported.  For lots of little structs on
BeOS.  It isn't really much of a solution though.  The tuple has
to be explicitly converted to this instance type, and generally
it's just a little more trouble than I'm usually willing to go to
just to get around the need to remember that tm_day is [2].

So, for stat we have ST_ATIME etc.  Even that is often more than
we can be bothered with (I bet st[6] far outnumbers st[ST_SIZE]),
but it's good that it's there.

	Donn Cave, donn at u.washington.edu
----------------------------------------
from structuple import Structuple

#  For demonstration, the stat struct.  The actual struct would be
#  supported by generated C code, and the code generator also makes
#  this subclass.
#
class stat(Structuple):
	_members = ('st_mode', 'st_ino', 'st_dev', 'st_nlink', 'st_uid', 'st_gid', 'st_size', 'st_atime', 'st_mtime', 'st_ctime')
	def typegen(self):
		#  Return class
		return (None, None, None, None, None, None, None, None, None, None)

---------------------------------------- structuple.py
#
#  Wrap tuple data types.
#
#  The types are structs that for economy are rendered in Python as
#  tuples, rather than internal C types.  The code generating module
#  that handles the C logic for this (sgrules) also generates Python
#  code for this wrapper system, so any time that's changed, the
#  wrapper structs module needs to be updated too.
#
#  This module contains the base class for the wrapper structs, and
#  is not generated.  Each derived class supplies a matched pair of
#  class-scope tuples, one naming the members and the other their types.
#  The type name (if any) is looked up and used to generate the data
#  item, on initialization and on assignment.  I think this is probably
#  infrequent, if not it should be optimized.
#

#  If there's a type, look it up and use it to generate the result.
#  This lookup is because I put the types in the class namespace -
#  a reference there to another class can fail if that class is defined
#  below the present class.
#
def tostruct(x, m, t):
	if t:
		return t(x)
	else:
		return x

def fromstruct(x, m, t):
	if t is None:
		return x
	else:
		return x.tuple

class Structuple:
	def __init__(self, t):
		if len(t) != len(self._members):
			raise TypeError, '%d-tuple' % len(t)
		self._types = self.typegen()
		self._dt = map(tostruct, t, self._members, self._types)

	def __getitem__(self, i):
		d = self._dt[i]

		#  Assuming you want tuple result rather than instance,
		#  if you ask via index.
		#
		if type(d) == type(self):
			d = d.tuple
		return d

	def __getattr__(self, a):
		if a == 'tuple':
			return tuple(map(fromstruct,
				self._dt, self._members, self._types))
		for i in range(len(self._dt)):
			if self._members[i] == a:
				return self._dt[i]
		raise AttributeError, a

	def __setattr__(self, a, v):
		if a in ('_dt', '_types'):
			self.__dict__[a] = v
			return
		for i in range(len(self._dt)):
			if self._members[i] == a:
				if type(v) == type(()):
					self._dt[i] = self._types[i](v)
				else:
					self._dt[i] = v
				return
		raise AttributeError, a
	def __str__(self):
		return str(self.tuple)
	def __repr__(self):
		return '<%s: %s>' % (self.__class__.__name__, self.tuple)


From Wolfgang.Grafen at marconicomms.com  Tue Jun  6 06:13:59 2000
From: Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen)
Date: Tue, 06 Jun 2000 12:13:59 +0200
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com>
Message-ID: <393CCEE6.E5C1C267@marconicomms.com>

Hi Guido,

I have a suggestion how to get the advantage of case-insensitivity without
it's disadvantages. Hope that I completely understood the problem and that
there is a way to implement my proposal.


> Yet, here are some of the reasons why I am considering making Python
> case-insensitive:
>
> (1) Randy Pausch, a professor at CMU, found, when teaching Python to
> non-CS students in the context of Alice (www.alice.org), that the
> number one problem his students were having was to remember that case
> matters in Python.

I am programming VHDL, which is also case-insensitive. We agree that
names should be written with always the same big or small letters but
the compiler doesn't check this as well you cannot rely on it. Just to assure
this elementary condition you would need some kind of lint tool. We use
many second party tools, also our own tools written in Perl and Python or
other Unix shell scripts. Some of the tools are not maintable anymore as
the programmer is not available or as it is on a commercial base.

We have a lot of problems with case-insensitiveness as names of program
units are connected to filenames, which are case-sensitive on a unix file
system. I used to name my files with mixed letter cases for about half a
year when I realised, that one tool couldn't deal with it. Then I had about
one week work to correct the names and sources and do all testing again.

What I don't like with VHDL implementations and case-insensitivity:

- The names are all translated to a norm case, normaly upper-case.
   My name for a port is input_1, but the compiler refers to it as INPUT_1

- When I parse an entity for example I keep the original names in a
   dictionary. This is more complexity especially as there is need for
    rapid programming, one of Python's strengths.

Reasons against case-insensitivity for Python:

- I haven't seen any positive reaction from the Python community.

- Many programs have to be debugged again.

- Case sensitivity is straight-forward and gives better code if you
   agree that names should always be written the same.

- Case sensitivity is even good for education. Mistakes in names
   will produce errors.

- One of my visions is Python will understand (physical )units in
   future, which also addresses an educational issue. Why shouldn't
   it possible to make following calculation:

>>> a= 2 apple + 3 orange + 4 apple + 7 carrot
>>> print a
6 apple + 3 orange + 7 carrot
>>> print number(a)
16
>>> print vegetable(a)
7 carrot
>>> print fruit(a)
6 apple + 3 orange
>>> print number(fruit(a))
9

    as well as use values like:
       10 MHz which is very different from 10 mHz and where
       everybody agrees that you should never write HZ or even hZ for
       the frequency unit!



One problems with mixed case programming:

>>> Orange = 1
>>> orange = 2 + Orange
>>> print Orange
1
which would give the expected result of 3 with case-insensitive programming.

Possible solution:

a) case-insensitivity which has other disadvantages

b) case-sensitivity with the awareness of same names. Just allow

>>> Orange = 1
>>> Orange = 2

but throw an exception when
>>> orange = 3
or
>>> ORANGE = 4 etc.

as well as assume the same (ignore underscores within words)
>>> orange_color
>>> orange__color etc
>>> orangecolor
where only one of this names can be assigned.

a more difficult task is to recognize spelling mistakes:
>>> ornge = 2
>>> oragne =3
>>> orange_colour = "orange"

this behaviour could be invoked by an option or become builtin
and/or can be implemented into the python-aware development
environment (IDE, PythonWorks etc.). The advantage of this in a
Python aware editor is that the mistake can be either corrected on
place or marked.


> (The number two problem was 1/2 == 0; there was no
> significalt number three problem.)
>

That's good.

What's about a significant problem number three?

When programming or reading the input of a non-experienced programmer
who might think  012 is decimal 12 the input will be interpreted as
an octal number. I don't have any solution for this. Maybe there should be
a routine reading from stdin which doesn't interpret 012 as octal number...




From m.faassen at vet.uu.nl  Fri Jun  2 18:54:05 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Jun 2000 22:54:05 GMT
Subject: ANNOUNCEMENT: Python Development Team Moves to Dutch
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no> <39361400.5E86F678@ecs.soton.ac.uk> <8h6pa1$ecq$2@newshost.accu.uu.nl> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <3937D345.CC0935D5@uab.edu>
Message-ID: <8h9dud$dse$1@newshost.accu.uu.nl>

Shae Erisson <shapr at uab.edu> wrote:
[everybody speaks Dutch again]

> ja, maar is het niet vanzelfsprekend dat alleen mensen die Nederlands
> spreken zich voor Python interesseren?

Lots of sense of deja-vu. Check deja-news for the previous thread
where I discovered that everybody here turns out to know Dutch. It's not
as if Dutch is such a widely spoken language, so there *must* be some
sort of cosmic connection. :)

> or-at-least-lives-with-someone-who-speaks-dutch-ly y'rs

Sure, blow away the mystique! :)

> (you can add an american with a dutch and scandinavian sounding name,
> posting from finland :)

Finland's a cool country! And not only temperature-wise. Helsinki is a
beautiful city.

Regards,

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


From paul at prescod.net  Tue Jun 20 07:22:17 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 20 Jun 2000 13:22:17 +0200
Subject: Conversion from proprietary satellite control script language
References: <20000616084700.29919.qmail@web2006.mail.yahoo.com>
Message-ID: <394F53E9.8FFBA69F@prescod.net>

TJ Judson wrote:
> 
> ...
> 
> .... I hope
> the powers that be see the argument that it would be better to add a
> less than perfect construct into the language, for the sole purpose
> of gaining more widespread use.  

I think that Python would be twice as popular today if it had block
delimiters. Guido's idea of the Right Thing tends to trump popularity.

I think your cause is hopeless in terms of "standard Python" but maybe
you can share a patch with others who find themselves in your situation.
Even if I'm wrong that the cause is hopeless, Python isn't a business so
nobody is going to implement such an "uncool" feature for you. 

You would have to implement it yourself and then try to convince
everyone else to add it to Python. There are a lot of cool ideas that
Guido has actually "blessed" that are not in Python just because nobody
cares enough (yet) to implement them.

Let me suggest another route: integrate your current scripting language
runtime with the Python runtime. Allow code from one language to call
code from the other. Write new code in Python and leave the old code in
the old language. Merging the interpreters seems to me to be less effort
then writing (and testing!) an X->Python converter.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"Music is the stuff between the notes." - Claude Debussy





From garry at sage.att.com  Wed Jun 14 11:37:41 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 14 Jun 2000 15:37:41 GMT
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i7r1s$l96$1@newshost.accu.uu.nl> <m3aego31k5.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3947A6C5.DD7E87EB@sage.att.com>

Michael Hudson wrote:
> 
> m.faassen at vet.uu.nl (Martijn Faassen) writes:
> 
> >   * Pythoneers care less about inheritance hierarchies. When a class
> >     supports an interface it supports an interface; inheritance often
> >     isn't needed.
> 
> To expand on this: it seems to me that the only reason to inherit in
> Python is to inherit implementation; there's just no point in
> inheriting to inherit an interface (subtyping).

except for clarity of expression.
saying "this Circle thing is a Shape"
is more useful to me than having to read its methods
to find that it behaves just like a Shape.
the computer doesn't care about this, but i do.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From petersc at stud.ntnu.no  Fri Jun  9 01:46:57 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Fri, 09 Jun 2000 07:46:57 +0200
Subject: quick newbie question on variables. please answer
References: <39401B5F.122C6DB4@mediaone.net>
Message-ID: <394084D1.6F66702C@stud.ntnu.no>

Toy wrote:
> 
> #!/usr/bin/python
> import commands
> b = commands.getstatusoutput('man %x!') % startx
> print = b

print b

print is a special case in that it is like calling
a function but you don't have to put parentheses.
print(b) would more likely be understandable from
a consistency point of view, but as print is used
so often it is a nice abbreviation to write print b
instead.

reminded-of-the-BASIC-"?"-operator-ly y'rs Peter

--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From fiona at sitegnome.com  Mon Jun 12 09:27:51 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 12 Jun 2000 13:27:51 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 12th, 2000
Message-ID: <20000612132751.24133.qmail@synop.com>

Hello All!

Here are the latest entries into http://python.faqts.com

'night,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Where can I get good Tutorials about JPython?
Where can I get good Tutorials about JPython?
http://www.faqts.com/knowledge-base/view.phtml/aid/3668
-------------------------------------------------------------
Fiona Czuczman
Moshe Zadka

JPython is another implementation of the Python language: everything you
learn in the Python tutorial (http://www.python.org/doc/tut/tut.html),
will be true for JPython. You can review the CPython/JPython differences
somewhere in the JPython site (http://www.jpython.org).


-------------------------------------------------------------
VBScript is excellent for integrating with ADO (Active Data Objects) databases.  Is python as facile?
http://www.faqts.com/knowledge-base/view.phtml/aid/3669
-------------------------------------------------------------
Fiona Czuczman
Eduard Hiti

I used ADO 2.5 in my last Python project, and had no problems at all. 
Most of the time, ADO access in VB translates almost directly to python 
syntax. Features of ADO I used with no problem include parameterized 
commands, batch updating and client side cursors.

One thing to keep in mind is the way pythoncom handles out-parameters of
method calls: These values are returned as part of the method result, 
e.g the Execute method of the Connection object returns a tuple 
consisting of an RecordSet object and the RecordsAffected value (see 
MSDN):

    recordset, affected = db_connect.Execute("select * from ...")

In VB you would only get the recordset returned and RecordsAffected 
would get stuffed in a reference parameter of the call.

Another point is handling of date values: They are returned as objects 
with type pywintypes.TimeType. The documentation for this type is very 
terse, and I had to dig into the pythoncom demos to find their 
interface: One useful method is Format(<formatstring>), which will 
return a string from TimeType:

    import pywintypes
    if type(x) == pywintypes.TimeType:
        return x.Format("%d.%m.%Y")

For more information on COM and Python I recommend Mark Hammond's 
"Python Programming on Win32" (which covers almost the entire COM body, 
albeit sometimes not too deeply: "Advanced Python Programming..." ?), 
which has a chapter on database programming.


-------------------------------------------------------------
Where can I download python DBI for MYSQL?
http://www.faqts.com/knowledge-base/view.phtml/aid/3667
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

http://dustman.net/andy/python/MySQLdb/

Probably best to look at http://www.vex.net/parnassus for most modules.






From dfenyes at flash.net  Tue Jun 20 01:57:30 2000
From: dfenyes at flash.net (David Fenyes)
Date: Tue, 20 Jun 2000 05:57:30 GMT
Subject: PIL (PYTHONPATH) install question
Message-ID: <m2ya4183da.fsf@c34680-a.tx.home.com>

Hello, I havne installed NumPy on Mandrake 7.0, which seems to work,
and just compiled PIL, which loads when I enter the build dir, set
PYTHONPATH to '$PWD:$PWD/PIL', and do:

>>> import _imaging
>>> import Image

The library loads OK.  However, when I do:

        $ cp PIL.pth /usr/local/lib/Python1.5/site-packages
        $ mkdir /usr/local/lib/Python1.5/site-packages/PIL
        $ cp *.so PIL/* /usr/local/lib/Python1.5/site-packages/PIL

where PIL.pth contains 'PIL', loading "Image" gives:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: No module named Image

Is there something simple I missed?  Thanks for any pointer.

David. 


-- 
David Fenyes  --  dfenyes at home.com



From PClaerhout at CREO.BE  Thu Jun 22 06:30:12 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 12:30:12 +0200
Subject: List of currently loaded drivers under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301211@msgeuro1.creo.be>

Hello,

does anyone know if it is possible to get a list of all currently loaded
device drivers under Windows NT using Python? Is this somewhere in
the registry or is it stored somewhere else?

I would like to get similar information like you get in the devices control
panel under windows nt.

Kind regards,


Pieter



From mfletch at tpresence.com  Thu Jun 29 13:09:29 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 29 Jun 2000 13:09:29 -0400
Subject: Bindings for game library
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33FC@VRTMAIL>

Might also check out crystalspace... you can find it off Parnassus' games
page:
	http://www.vex.net/parnassus/apyllo.py/238168060

Enjoy,
Mike

-----Original Message-----
From: Kirill Simonov [mailto:kirill at xyz.donetsk.ua]
Sent: Thursday, June 29, 2000 2:46 AM
To: python-list at python.org
Subject: Bindings for game library


On Wed, Jun 28, 2000 at 05:51:15AM -0500, Chris Armstrong wrote:
> Hey all, been a while since i've posted.. Anyway, I've recently been
> getting into game coding with C++, but I'd really love to get back to
> Python. If anyone can recommend (for instance) a library with Py bindings
> that can support some intense 2d sprite rendering,  I'd be real happy.

There's a fine game library at http://www.libsdl.org with Python
bindings at http://pysdl.sourceforge.net

--
Kirill

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



From aahz at netcom.com  Fri Jun  2 13:27:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jun 2000 17:27:56 GMT
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com> <8h8jg3$d66$1@slb6.atl.mindspring.net> <8h8lkt$rb7$1@nnrp1.deja.com>
Message-ID: <8h8qqs$l5a$1@slb6.atl.mindspring.net>

In article <8h8lkt$rb7$1 at nnrp1.deja.com>, Arint?  <jamarijr at my-deja.com> wrote:
>You must be using a different version or something.  I use the python
>gui thing and did this
>print "\x1ba"
>got a funny circle
>print "\x1b a"
>[] a   < not an exact represetation.
>But, it still points out the fact that python is picking up the 1 b and
>a.

Ah.  Python appears to allow three-digit hex numbers.  I guess you'll
have to paste the string together, character by character.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From thomas at xs4all.net  Sat Jun 10 15:43:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Jun 2000 21:43:19 +0200
Subject: PATCH: Augmented assignment
In-Reply-To: <3942956B.2CAB4C9A@yale.edu>; from paul.magwene@yale.edu on Sat, Jun 10, 2000 at 03:22:19PM -0400
References: <m130q7t-000wcDC@swing.co.at> <3942956B.2CAB4C9A@yale.edu>
Message-ID: <20000610214319.J20700@xs4all.nl>

On Sat, Jun 10, 2000 at 03:22:19PM -0400, Paul Magwene wrote:
> Christian Tanzer wrote:

> > It doesn't obfuscate code looking like:

> >    nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1

> Please tell me you're not REALLY writing code that looks like that, when
> something like:
> 
> 	nested[i][j] = nested[i][j] + 1

> will do just as well (and is a lot easier to read).

*Please* tell me you aren't really using variable names like 'i' and 'j' in
large programs ? Sure, they're fine for loop counters, but you dont always
index large structures with those.

Anyway, as has been said before, the issue is not saving the few keystrokes,
the issue is obvious in-place modification; syntax that is both clear in
what it means, and how it is different from 'x = x + 1'.

> Undecided-on-augmented-assignment-l'y yours,

Well, to be truthful, I'm undecided on augmented assignment myself, and I'll
stay undecided until Guido accepts it ;) I like the syntax very much, but
that doesn't an sich mean it should go in Python. I'll let Guido be the
judge of that, and trust others to make sure Guido knows what he decides ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From PClaerhout at CREO.BE  Tue Jun 27 10:16:09 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Tue, 27 Jun 2000 16:16:09 +0200
Subject: Lista de correo sobre Python
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301248@msgeuro1.creo.be>

Can somebody help? I forgot my Spanish glasses...


Pieter

-----Original Message-----
From: andres at corrada.com [mailto:andres at corrada.com]
Sent: Tuesday, June 27, 2000 1:44 PM
To: JD
Cc: python-list at python.org
Subject: Re: Lista de correo sobre Python


On Mon, Jun 26, 2000 at 09:47:13AM +0000, JD wrote:
> Un foro sobre Python en:
> python_dev-subscribe at egroups.com
> 
> Muchos usuarios dispuestos a resolver tus dudas, a ayudarte y deseando que
> compartas tus conocimintos.
> 

?Cual es la justificacion para comenzar una lista nueva en espagnol cuando
ya existe una?

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------

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



From richard_chamberlain at ntlworld.com  Sun Jun 11 09:12:25 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 11 Jun 2000 14:12:25 +0100
Subject: python DBI?
References: <3942F32E.B4751F35@yahoo.com>
Message-ID: <H6M05.6204$fu1.92979@news6-win.server.ntlworld.com>

http://dustman.net/andy/python/MySQLdb/

Probably best to look at http://www.vex.net/parnassus for most modules.

Richard
Sam Wun <samwun at yahoo.com> wrote in message
news:3942F32E.B4751F35 at yahoo.com...
> Where can I download python DBI for MYSQL?
>
> Thanks
> Sam.
>
>




From yawyi at gmx.de  Thu Jun  1 08:01:37 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Thu, 1 Jun 2000 14:01:37 +0200
Subject: String.join revisited (URGENT for 1.6)
References: <28052000.5@sanctum.jae.ddns.org> <sj3ejda85pj169@corp.supernews.com> <hjoY4.6071$wYl.223614464@newsb.telia.net> <39328edc$0$18384@personalnews.de.uu.net> <_pyY4.10102$Za1.160118@newsc.telia.net> <3932e628$0$6681@personalnews.de.uu.net> <3932F323.C73E0404@san.rr.com> <3932F5DC.E00C1642@san.rr.com> <ZoDY4.6196$wYl.229414400@newsb.telia.net> <39333CC3.32805F5@cosc.canterbury.ac.nz> <saMY4.6222$wYl.223847936@newsb.telia.net> <3933a92e$0$25480@personalnews.de.uu.net> <ZUgZ4.6388$wYl.229561856@newsb.telia.net>
Message-ID: <3936505e$0$8957@personalnews.de.uu.net>

Fredrik Lundh <effbot at telia.com> wrote:
>Johannes Stezenbach wrote:
>> You tell us to reason because voting doesn't help.
>
>Umm.  Looks like you're confusing me with someone else...

Sorry, I didn't want to put Guido's words in your mouth.
It's just that I see you as part of the Python community core, so
this "You" was actually plural -- which wasn't visible in my
writing.

Johannes



From MarkH at ActiveState.com  Mon Jun 19 07:30:36 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 19 Jun 2000 11:30:36 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca><8i8bi2$1s8$1@news.wrc.xerox.com><slrn8kg937.328.hzhu@rocket.knowledgetrack.com><ud7ljln92.fsf@ctwd0143.fitlinxx.com><slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com><l03102800b570d8042c72@[193.78.237.143]> <l03102800b57372d82b38@[193.78.237.140]>
Message-ID: <wvn35.2720$qz4.21769@news-server.bigpond.net.au>

"Just van Rossum" <just at letterror.com> wrote in message
news:l03102800b57372d82b38@[193.78.237.140]...

> Using spaces-only is a safe solution, but only if you are absolutely
sure
> you never want anyone to use a proportional font to view your code...
(For
> those of you who have never used proportional fonts before <wink>: the
> width of the space character varies from font to font (as it should);
this
> means spaces are pretty much useless as indentation!)

I agree 100% with your sentiments!

But Im not sure I understand the above.

If you indent with leading spaces, then presumably all leading spaces will
be rendered in the same font.  So even for proportional fonts, the code
should align correctly with spaces or tabs.

For alignment of anything other than the first non-space character (eg,
attempting to align a table, or a parameter list that spans
multiple-lines, for example), then if you use tabs, you are back in the
same position that started this thread - you are making some assumption
about the location of the tab-stops, or an assumption about the width of
the previous characters.

So I simply can not see a solution for ensuring _all_ code is visually
indented correctly when using purely tabs with a proportional font.

Let me know if I am wrong - I use a proportional font myself, and added a
quick toggle-between-fixed-and proportional command to Pythonwin purely to
get around this problem - Im forced to switch to fixed-point and using
spaces for non-leading indents...

But-generally-I-just-let-auto-indent-do-it-right-ly,

Mark.




From tim at paradise.net.nz  Sat Jun 24 21:12:53 2000
From: tim at paradise.net.nz (Tim Evans)
Date: Sun, 25 Jun 2000 01:12:53 GMT
Subject: tkinter, grid and resize
References: <39550101.AD8C7956@uniserve.com>
Message-ID: <878ZVUWOKQ.FSF@CASSANDRA.EVANSNET>

Bob van der Poel <bvdpoel at uniserve.com> writes:

> I'm having a problem getting widgets managed by grid to resize
> themselves when the screen size changes. This is easy to do using the
> pack manager and 'expand', however 'expand' doesn't exist in grid and
> the docs seem to indicate that the 'news' sticky option should do the
> same. The following snippet creates a text widget with scrollbars. If
> you resize the main window, the widgets do not resize....
>

You need to use the grid_rowconfigure and grid_columnconfigure
methods of the containing frame, specifically the 'weight=' option.

> from Tkinter import *
>
> root=Tk()
>
> ys = Scrollbar(root)
> xs = Scrollbar(root)
>
> dsp = Text(root, yscrollcommand=ys.set, xscrollcommand=xs.set)
> xs.config(orient='hor', command=dsp.xview)
> ys.config(orient='vert', command=dsp.yview)
> xs.grid(row=1, column=1, sticky=E+W)
> ys.grid(row=0, column=0, sticky=N+S)
> dsp.grid(row=0, column=1, sticky=N+S+E+W)

root.grid_rowconfigure(0, weight=1)
root.grid_columnconfigure(1, weight=1)

>
> root.mainloop()
>
> What am I don't wrong? I've tried to insert columnspan stuff and played
> with the weight options--but I don't seem to be getting it.
>
> Thanks.

See the Tcl/Tk documentation under 'grid' for more information,
including other useful options for the rowconfigure and
columnconfigure methods.

--
Tim Evans


From pinard at iro.umontreal.ca  Tue Jun  6 12:32:30 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Jun 2000 12:32:30 -0400
Subject: Spurious coercing? [was: Re: 'if foo' calls getattr?!?]
In-Reply-To: Moshe Zadka's message of "Tue, 31 Aug 1999 01:50:42 -0400 (EDT)"
References: <Pine.LNX.4.10.9908310132450.9802-100000@starship.python.net>
Message-ID: <oqg0qqdb35.fsf@titan.progiciels-bpi.ca>

Moshe Zadka <moshez at server.python.net> writes:

> On Tue, 31 Aug 1999, Moshe Zadka wrote:

> > Bzzt! foo!=None is slower then foo is not None, but not by much: it will
> > first check whether type(foo) is not type(None), so it is pretty fast.

> > truth-and-nothing-are-always-popular-subjects-ly y'rs, Z.

> Sorry to reply to my own posts, but there was a mistake: foor!=None will
> make a call to __getattr__ for __cmp__, so it will be very slow.

And I would presume that if __cmp__ happened to be found, a call to
__coerce__ would be tried before going any further?

I was surprised to see, a little while ago, that __coerce__ is called even
if the two arguments have the same proper type.  I would have imagined that
the __coerce__ is spared, when both arguments are already of the given
class type.  Would someone have an example to illustrate that calling
__coerce__ may be meaningful when types do not need be coerced?

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





From kupisch at hotmail.com  Tue Jun 27 15:12:49 2000
From: kupisch at hotmail.com (hulme)
Date: Tue, 27 Jun 2000 21:12:49 +0200
Subject: Shutting down internet connection
Message-ID: <3958FCB1.A1B62061@hotmail.com>

Hi everyone,

I am looking for a way to shutdown an internet connection that was
established by dial-on-demand on a remote computer. i.e. there are two
computers given both running WIN98 on starting a browser the connction
gets dialed on the other machine. So, I would like to have a script that
shuts down the connection from the computer I am browsing with. Any way
to do this with Python or Windws Extensions.
Thanks for help!



From jin.choung at gte.net  Wed Jun 14 05:57:51 2000
From: jin.choung at gte.net (jin choung)
Date: Wed, 14 Jun 2000 09:57:51 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <394b3e7d.284293552@news.online.no>
Message-ID: <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net>

howdy,

ummm, is there any book that you reccommend for a rank beginner?  that
learning python book is REALLY not for beginners.

thanks.

jin




From MarkH at ActiveState.com  Mon Jun 19 20:06:41 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 20 Jun 2000 00:06:41 GMT
Subject: win32all-132 / Word Automation
References: <394e2b8d.515788975@news.mch.sni.de>
Message-ID: <lAy35.3131$qz4.24329@news-server.bigpond.net.au>

I guess your problem is here:

> >>> print doc.Variables.Item(1).Name
> -------------------------

> AttributeError: 'tuple' object has no attribute 'InvokeTypes'

With the implication being that "doc.Variables" is returning a tuple, not
a COM object as you expect.  It is likely the COM object you expect is
_in_ the tuple.  Either that, or the "Item(1)" call is returning the
tuple.

This is different from VB, due to the different handling of byref args.

Try:
>>> print doc
>>> print doc.Variables
>>> print doc.Variables.Item
>>> print doc.Variables.Item(1)

And it should become obvious...

Mark.





From boud at rempt.xs4all.nl  Fri Jun  9 16:07:51 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 9 Jun 2000 20:07:51 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch>
Message-ID: <8hriqn$dn2$1@news1.xs4all.nl>

Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote:
> stevemul at ozemail.com.au (Steve Mullarkey) writes:

>>I have read in several places productivity claims of 5 to 10 times over
>>'C' and "C++". I would like to ask for some feedback from "C++"
>>programmers who have moved to Python as to whether these estimates are
>>realistic.

> I'm a C++ programmer (leave out the quotes ;) ).

> First off, any claim of an order of magnitude increase in productivity is
> pure nonsense. See Brooks for an explanation or Boehms and McConnell for
> hard data.

I'm not entirely sure - I would like to see some hard data, but I think
that having a language which assembles all the best practices and
discards all the worst might offer a significant improvement in
productivity. After all, many cents make a dollar, too. 

I'm not really sure whether scripting languages aren't as much a leap
as the leap from first to second generation languages was. The results
mentioned in that paper comparing C++, Java, Perl, Python & Tcl by Lutz
Prechelt (http://wwwipd.ira.uka.de/~prechelt/Biblio/jccpprtTR.ps.gz)
were suggestive. To quote (p. 28) "Designing and writing the program in
Perl, Python, Rexx or Tcl takes only about half as much time as writing
it in C, C++, or Java and the resulting program is about half as long."

Simply put, with Python you need less statements, and the amount of
statements a programmer puts out is constant (be it assembly, C or
Python - I think that's in Brooks, too), so the amount of effort needed
is correspondingly less.

Of course, that doesn't take into account the oft-mentioned variabilty
in programmer prowess - the best can crank out ten times as much code
than the worst. That's why the people developing KDE can push out more
C++ code than I can Python ;-(.

--

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


From moshez at math.huji.ac.il  Tue Jun 13 03:01:44 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 10:01:44 +0300 (IDT)
Subject: can python do this?
In-Reply-To: <3945D0AA.E8F51A9C@san.rr.com>
Message-ID: <Pine.GSO.4.10.10006130942460.20303-100000@sundial>

Just a few nits:

On Tue, 13 Jun 2000, Courageous wrote:

> Q: "Can python do this?"
> A: "Python was born for it, baybee!"
> 
> import string
> file = open ("file")

Personally, I'd put the rest under a try/finally/file.close(), but then
again, I've made paranoia in programming into an art.

> lines = file.readlines()
> for line in lines:
>     pair = string.split(line,":")

You're sure you don't want to 

a) supply the optional argument to string.split which limits the number 
   of parts?
b) not cut off the newline at the end?

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From 12klat at sightreader.com  Tue Jun  6 06:41:22 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 03:41:22 -0700
Subject: operator overloading
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com> <393C2386.307A160D@sightreader.com> <20000605235010.C469@xs4all.nl>
Message-ID: <393CD552.89237E3E@sightreader.com>



Thomas Wouters wrote:

> On Mon, Jun 05, 2000 at 03:02:47PM -0700, Ken Seehof wrote:
>
> > The following more closely conforms to these conventions:
>
> The return code of __cmp__ is hardly a convention! If you dont return one of
> those values, in a consistent manner, you're going to confuse the users of
> your class greatly. And possibly break things like list.sort()'s default
> sort function.

I was being facetious and forgot to add a ":-)" to the "convention" comment.  In
other words, I was trying to be nice (in a silly way) to the author of the previous
"solution" by not actually saying that the answer was completely wrong.  Hence, in
the spirit of generosity, I referred to the boolean interpretation of 0 and 1, the
meaning of ">" and "<", and the return value of __cmp__ as "conventions".

> > ...  def __cmp__(self,other):
> > ...   if self.a == other.a:
> > ...    return 0
> > ...   elif self.a < other.a:
> > ...    return -1
> > ...   else:
> > ...    return 1
>
> Good example not withstanding, you can of course write this much shorter,
> and more obvious, and less errorprone:
>
> def __cmp__(self, other):
>         return cmp(self.a, other.a)

Good point (that is how I would write it in an actual program).  Of course, your
solution hides the semantics of the __cmp__ return value, which is fine, unless you
happen to be writing an example specifically to show the semantics of the __cmp__
return value. :-)

> This will even work if __cmp__ would be changed to return different values
> ;)
>
> --
> Thomas Wouters <thomas at xs4all.net>
>
> Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

- Ken Seehof
kens at sightreader.cHi! I'm a .signature virus! copy me into your .signature file to
help me spread!om

-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/b9e8ae3b/attachment.vcf>

From thomas at xs4all.net  Tue Jun 13 08:16:00 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 14:16:00 +0200
Subject: How many times inner functions are compiled ?
In-Reply-To: <960897680.1890969865@news.libertysurf.fr>; from jerome.quelin@insalien.org on Tue, Jun 13, 2000 at 02:01:20PM +0200
References: <960897680.1890969865@news.libertysurf.fr>
Message-ID: <20000613141433.U20700@xs4all.nl>

On Tue, Jun 13, 2000 at 02:01:20PM +0200, Jerome Quelin wrote:

> def outer():
>     def inner():
>         print "spam"
>     inner()
> outer()
> outer()

> In the snippet code above, I call outer() two times. I would like to
> know how many times python will compile the inner function ? One time
> when compiling the script/module, or two times, ie at every invocation of the
> outer function ?

Once for every invocation of outer(). Function definitions are to be
considered 'assignment' statements, similar to something like this:

def outer():
  inner = Function("code")

When looking at that bit of code, there is no doubt in your mind that the
creation of the Function object is done once each call, is there ? Nested
functions are not very useful, currently, except to reduce namespace
polution and for simulating closures, like this:

def mk_inner(spam):
	def inner(spam=spam):
		print spam
	return inner

inner = mk_inner(spam)

inner()
inner()
inner()

(Of course, you can do the same by making a class with a __call__ method :)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hzhu at rocket.knowledgetrack.com  Mon Jun 12 18:24:33 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Mon, 12 Jun 2000 22:24:33 GMT
Subject: NumPy arrays: how to remove extra axes
Message-ID: <slrn8kaov1.1ar.hzhu@rocket.knowledgetrack.com>

For arrays in Numeric, is there a general way to remove axes of length 1?
For example I have an array of shape (1,3,1,4) I want to change it to (3,4).
Thanks


Huaiyu Zhu                               hzhu at knowledgetrack.com


From mdm at corp.eCircles.com  Thu Jun 22 19:03:50 2000
From: mdm at corp.eCircles.com (Michael D. Marchionna)
Date: Thu, 22 Jun 2000 16:03:50 -0700
Subject: Improving cStringIO API (Request For Comments)
Message-ID: <39529B56.D54681E0@corp.eCircles.com>

I was messing around trying to convert my Java app to
Python, and
ran into some difficulty with cStringIO module.  In my app I
transmit
a segmented byte stream over a socket connection.  I read
the data
section by section while laying the message out in a single
StringIO
buffer.  I then create a series of section objects that
reference the
segments in the StringIO buffer.  I want the section object
to contain
no data except an offset from the start of the buffer.   So
that when I
query the contents of a section the data is actually pulled
from the
StringIO object.

This led me to a bunch of code like the following:

class MsgSect:
    <<< code deleted >>>
    def getAttrText(self):
        value = None
        if self.type == ATTR and self.size > 9 and
self.atype == 'C':
            oldpos = self.buffer.tell()
            self.buffer.seek(self.offset+9)
            value = self.buffer.read(self.size-9)
            self.buffer.seek(oldpos)
        return value

The point is not to disturb the current read/write pointer.
You might ask
why not just extract the string and add it as a member of
the MsgSect
object.  Well I have some reasons OK.  Anyway this led me to
making some
modifications to the cStringIO module.

At first I just added a peek method so I could read sections
without
modifying the read/write pointer.  For example:

    >>> from cStringIO import  StringIO
    >>> so = StringIO()
    >>> so.write('0123456789ABCDEF')
    >>> so.tell()
    16
    >>> so.peek(4,3)
    '456'
    >>> so.tell()
    16

I also added a poke method to do what you might think:

    >>> so.poke(4,'***')
    >>> so.getvalue()
    '0123***789ABCDEF'
    >>> so.tell()
    16

Then I got to thinking these are still strings why don't
they behave a little
more like them.  So I added a set of sequence methods to the
StringI and
StringO object types.  So now you can do

    >>> si = StringIO('0123456789ABCDEF')
    >>> len(si)
    16
    >>> si[4]
    '4'
    >>> si[-4]
    'C'
    >>> si[4:7]
    '456'
    >>> si[-6:-2]
    'ABCD'

Item assignment works like you would expect for StringO
objects:

    >>> so = StringIO()
    >>> so.write('ABCD')
    >>> so[3] = 'c'
    >>> so[:]
    'ABcD'

Repeat operations work as well:

    >>> soso = so * 2
    >>> soso[:]
    'ABcDABcD'

Things get a little more squirely with the remaining concat
and setslice
methods.  A concat of StringI objects works fine, but the a
concat of
StringO objects leaves the problem of what to do with the
softspace
attribute.  Currently I just copy the value from the self
object.

    >>> so1 = StringIO()
    >>> so1.write('ABC')
    >>> so2 = StringIO()
    >>> so2.write('123')
    >>> so1.softspace
    0
    >>> so2.softspace = 1
    >>> so3 = so1 + so2
    >>> so3.getvalue()
    'ABC123'
    >>> so3.softspace
    0
    >>> so4 = so2 + so1
    >>> so4.getvalue()
    '123ABC'
    >>> so4.softspace
    1

More variations can be dealt with the concat methods.  For
example it is
possible to mix StringI and StringO types, but what should
the resultant
type be.  Also it seems reasonable to allow string objects
in the concat
as well, so5 = so1 + 'XYZ'.   Unfortunately so5 = 'XYZ' +
so1 won't work.

Slice assignments obviously only work on StringO objects.
In some
sence the setslice should overwrite the contents of the
StringO buffer rather
than produce a new StringO object.  But the poke method
handles most of
the setslice operation except it currently doesn't permit
extending the buffer.
It could do that though.  Perhaps optionally via a third
argument:

    def poke(offset, value, extend=0):

The other areas of concern about the cStringIO module
concern its fixed
reallocation scheme.   Currenty a StringO object starts out
at a capacity
of 128 and will double its size each time a write goes
beyond the current
buffer space.  It would be nice to allow this scheme
modified.  Perhaps via
optional keyword arguments to the StringIO factory
function.  For example:

        so = StringIO(capacity=4096, increment=8192)

Would create a StringO object that would start out with a
buffer space of
4096 bytes, and add 8192 each time a write went off the
edge.  Or possibly
something even likethis:

        so = StringIO(capacity=1024*1024, upscale=0.1)

Would create a 1MB buffer and increase the size by 10
percent for each
overwite.  Options could also be added for decreasing the
size of the buffer
when it is truncated, e.g. decrement=65536, downscale=0.5.

Another feature that would be nice is to allow shifting of
the contents of a
StringO buffer.  Similar to what memmove does.

Further on I would like to support the struct.pack and
struct.unpack methods
so that they operate directly on the contents of the
StringIO buffer.

In the end I was wondering what people think of adding some,
or all of these
features to the cStringIO module, or perhaps to a completely
new module.

Comments and suggestions?

--MDM




From wlav at atlas03.cern.ch  Fri Jun  9 13:11:15 2000
From: wlav at atlas03.cern.ch (Wim Lavrijsen)
Date: 9 Jun 2000 17:11:15 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB333@ALTNET>
Message-ID: <8hr8fj$fu$1@sunnews.cern.ch>

Mike Steed <MSteed at altiris.com> writes:
>> From: nbecker at fred.net [mailto:nbecker at fred.net]
>> 
>> I rarely need to do this for my classes.  Using STL heavily is a big
>> help.

>This is a surprising thing to say.  STL requires that the types used to
>instantiate its various templates be well-behaved in certain ways, e.g., for
>copying, assigning, and comparing (depending on the "concept" to which the
>template conforms).

All that STL containers require is that your dtor and copy assignment
don't throw. The compiler generated members fullfill that requirement.

>  This means you have to implement your own copy
>constructor, assignment operator, and/or comparison operators for each
>(non-trivial) class.  It may not be difficult, but it is overhead that
>Python does not impose.

Please inform yourself better before making such statements. Blindlessy
drawing conclusions based on the C++ implementation of one particular
vendor from Redmond isn't good for your credability.

Best regards,
      Wim Lavrijsen


From hzhu at mars.localdomain  Sat Jun 17 21:58:46 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Sun, 18 Jun 2000 01:58:46 GMT
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com> <slrn8kljmr.56e.hzhu@rocket.knowledgetrack.com> <uitv9qb72.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8kobd5.1vs.hzhu@mars.localdomain>

On 16 Jun 2000 22:43:29 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>hzhu at knowledgetrack.com (Huaiyu Zhu) writes:
>
>> OK, so I agree that the (setq-default tab-width 4) in my .emacs is evil (it
>> sits there since way back when I copied my first .emacs from somebody else),
>> I'm wondering what's the easiest way to revert this on multiple files,
>> either within or without emacs.  If I'm going to change my default emacs
>> behavior I want to do this on all the files, and safely.
>
>One fairly simple way is to untabify your file, change the tab width,
>and then if you want re-tabify.  Emacs will use a mixture of TABs
>(using the new tab width definition) and spaces to achieve the same
>spacing you already have.
>

This is simple to do for one file, but not for many files.  I'd prefer
running something in batch mode on a list of files.

Huaiyu


From xtianNOxtSPAM at hyperactive.co.nz.invalid  Wed Jun 14 23:51:03 2000
From: xtianNOxtSPAM at hyperactive.co.nz.invalid (xtian)
Date: Wed, 14 Jun 2000 20:51:03 -0700
Subject: boolean true and false values.
References: <3947f922.425016156@nntp.interaccess.com> <CIW15.1255$OZ.357618@news1.rdc2.tx.home.com>
Message-ID: <0043ea86.2c4e5ff3@usw-ex0101-006.remarq.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote:
>Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
>news:3947f922.425016156 at nntp.interaccess.com...
>> Does python have true and false values.
>> I would like to do something like
>>
>> class A:
>>      def _init_(self):
>>            self.processed=false
>>      def process(self):
>> self.processed=true
>>
>> at the moment I am using
>> false=(1==0)
>> true=(1==1)
>> but would like something more elegant, or at least less
sloppy.
>
>false = 0
>true = 1
>

This is a bit risky, depending on that for which you'll be using
it - there are a lot of other expressions that evaluate to false
but not to zero, as Peter was illustrating (including user
defined objects, if the author has set it up that way).

If you've defined false, you might be trying to use checks like

  if (expression) == false:
    do_stuff()

when you should probably be doing

  if not expression:
    do_stuff()


Of course, if you're just using them for parameters or some
such, then I'll be quiet.

Xtian


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



From android at micron.net  Thu Jun 15 08:58:20 2000
From: android at micron.net (Peter Immarco)
Date: Thu, 15 Jun 2000 08:58:20 -0400
Subject: Best link for Python add-ins?
Message-ID: <3948D2EB.424B5E42@micron.net>

Just read about another wonderful library for Python, NumPy.  (Is that
pronounced like "lumpy" or "lum-pie":)?  What's the best link that has a
list of the libraries available for Python, windows 32 version?

Peter.




From kens at sightreader.com  Fri Jun 16 01:54:29 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 22:54:29 -0700
Subject: callbacks in python
References: <3947D6B5.664ABA2B@nih.gov>
Message-ID: <3949C115.7EE4B928@sightreader.com>

Sure, that's one good way to do it.  It's particularly easy to do in python.

>>> class Eggs:
...     def __init__(self, spammer):
...         self.spammer = spammer
...     def OnSpam(self, a,b):
...         self.spammer(a,b)
...
>>> def SpamCallback(a,b):
...     print 'spam and ' + a + ' and spam and spam and ' + b
...
>>> e = Eggs(SpamCallback)
>>> e.OnSpam('green eggs', 'ham')
spam and green eggs and spam and spam and ham
>>>

Jonathan Epstein wrote:

> I'm working on my first Python project, and am encountering the issue of
> callbacks.  In particular, whenever my class undergoes a change which
> modifies its print representation, I would like to notify a callback
> routine which has been specified by the user of the class.  This
> callback routine would usually be some sort of GUI.
>
> Is this the right way to do things in python?  I read a reference in
> _Python Programming on Win32_ which seemed to suggest that this was not
> the right way to do things, but I am not sure.
>
> Thanks in advance for any help, especially including any pointers to
> examples...
>
> - Jonathan

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/a68eef03/attachment.vcf>

From darrell at dorb.com  Mon Jun 12 02:16:08 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 11 Jun 2000 23:16:08 -0700
Subject: Python Productivity over C++
Message-ID: <047e01bfd435$b2f35570$6401a8c0@home>

A co-op working with me tells the story of how he learned C++ in
excruciating syntactic detail. All before he had any idea why anyone would
dream all this up. I'm sure most of it was lost on him.

Looking over my shoulder one afternoon, has him all fired up for Python. We
focused on the ultimate goal instead of how deep the language is.

I only have so much mental bandwidth and want to minimize how much is spent
on language/tools. My goal is to solve customer problems fast and leave them
with something they can understand and use for years to come.

Just happened to be looking for info on g++ exception handling.
Came across this site saying not to use all the fancy C++ features for
portability reasons. http://wxwindows.org/standard.htm

--Darrell Gallion







From kmself at ix.netcom.com  Tue Jun 27 02:36:19 2000
From: kmself at ix.netcom.com (kmself at ix.netcom.com)
Date: Mon, 26 Jun 2000 23:36:19 -0700
Subject: profiler -- numeric overflow?
References: <h579j8.grg.ln@kmself.nntp.ix.netcom.com>
Message-ID: <31i9j8.dbh.ln@kmself.nntp.ix.netcom.com>

kmself at ix.netcom.com wrote:
> I'm new to Python.

> ...yes, I've also considered running a smaller sample of the log, but I
> suspect that runtime may grow nonlinearly, large input samples are good.

Followup to Self:

...well, at least things appear to grow linearly.  Runtimes varying
input lines:

  Input     Runtime (sec)   Lines/s
  -----     -------------   -------
      100        0.210        476
    1,000        3.860        259
   10,000        41.30        242
  100,000      415.670        240

...which is darned close to straight-line growth.  However, performance
is about an order of magnitude less than I'd like to see.

-- 
Karsten M. Self <kmself at ix.netcom.com>         http://www.netcom.com/~kmself
  Evangelist, Opensales, Inc.                       http://www.opensales.org
   What part of "Gestalt" don't you understand?      Debian GNU/Linux rocks!
     http://gestalt-system.sourceforge.net/      K5: http://www.kuro5hin.org
GPG fingerprint: F932 8B25 5FDD 2528 D595  DC61 3847 889F 55F2 B9B0


From jhe at webde-ag.de  Tue Jun 27 09:41:08 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Tue, 27 Jun 2000 15:41:08 +0200
Subject: Improving cStringIO API (Request For Comments)
References: <l03102801b57e21040bf2@[193.78.237.126]>
Message-ID: <8jaaqt$ltb$1@pollux.ip-plus.net>

"Just van Rossum" <just at letterror.com> schrieb im Newsbeitrag
news:l03102801b57e21040bf2@[193.78.237.126]...
> Check out the array module. It should be easy enough to whip up a class
> that does what you want, using arrays underneath.

Isn't the more or less recently introduced buffer() built-in and the buffer
interface exactly what is needed here? Namely, refer to byte chunks of a
larger object.

Ciao, J?rgen

--
J?rgen Hermann (jhe<at>webde-ag.de)
WEB.DE AG, Amalienbadstr.41, D-76227 Karlsruhe
Tel.: 0721/94329-0, Fax: 0721/94329-22




From hinsen at cnrs-orleans.fr  Tue Jun 20 09:33:00 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 20 Jun 2000 15:33:00 +0200
Subject: Computer Algebra System
References: <8inpa4$8iv$1@vishnu.jussieu.fr>
Message-ID: <m366r41nqr.fsf@chinon.cnrs-orleans.fr>

Rodolphe Pollet <pollet at liliput.lct.jussieu.fr> writes:

> Is anyone aware of a Computer Algebra System (CAS) ---
> i.e a mathematical software that is able to derivate, integrate,
> simplify, find limits, ... analytically --- in python ?
> It should compete with Mathematica or Maple.

Pythonica isn't there yet, but it's an interesting start. More at:

       http://www.strout.net/python/pythonica.html

-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From garyl at saeweb.com  Tue Jun  6 13:10:51 2000
From: garyl at saeweb.com (garyl at saeweb.com)
Date: Tue, 06 Jun 2000 17:10:51 GMT
Subject: gtk and GdkColor and set_forground
Message-ID: <8hjbab$98j$1@nnrp1.deja.com>

I am having a problem with set_forground() in a clist.
the error I get back is that I am not sending the color val as a
GdkColor   but I can't figure out how to get a GdkColor val

I am new to python so it is probably something dumb I am doing

in c I would call gdk_parse_color("color name" &varname);
and that would work  but I can't figure out what I am doing wrong


Thanks for any help .


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


From pecora at anvil.nrl.navy.mil  Thu Jun 15 07:10:55 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Thu, 15 Jun 2000 11:10:55 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <20000614155125.A26436@xs4all.nl>
Message-ID: <150620000710550920%pecora@anvil.nrl.navy.mil>

>On Wed, Jun 14, 2000 at 11:16:50AM +0000, Louis M. Pecora wrote:
>
>> > Wow, thanks for looking that code over, Bernhard.  What a horrible
bunch
>> > of holes.
>
>> And who could have seen all that so clearly without an expert on
hand. 
>> Even then I couldn't understand all the reasoning.  In Python we are
all
>> happy, but the ugly side is when we return to C.  It's obvious to me:
>
>> CAN'T SOMEONE UPDATE THE EXTENSIONS PDF MANUAL, ADDING A BETTER
TUTORIAL
>> ON REFERENCE COUNTS AND (!!) LISTING Py_XXXX FUNCTIONS AND THEIR
REFERENCE
>> EFFECTS ??
>
>Sure, anyone can. The address to submit your new sections is
>python-docs at python.org, I believe. Note that you shouldn't send in PDF
>documents. The actual documentation is written in TeX, which is used to
>generate PostScript, PDF, HTML, info and what not. I'm sure someone
will
>volunteer to rewrite your plaintext-version to TeX if you write a good
>manual on reference counts ;-)

The old "give the assignment to the guy who complains the most," eh? 
:-) 

Of course, if I knew all that, I wouldn't complain.  

>> Sorry to yell, but it's obvious that many people like me are
stumbling
>> with this one.  I'd hate to have all that time I save using Python
eaten
>> up by debugging C extensions.
>
>The API interface documentation is pretty clear in which functions
give or
>take a reference. *most* functions do not touch references, they take
the
>object you give to them, do their thing, and return a result. Those
that
>return objects are documented as returning either a new reference or a
>borrowed reference (in which case, if you keep the object around, you
have
>to INCREF yourself) and those that steal a reference (take an Object
>argument without returning it, and without INCREFing it) are
documented to
>do so as well.

Maybe I missed something in the documentation - -and I will go back and
check it out--but I don't remember seeing a list of functions which
said whether INC- or DECREF's were necessary with it.  There were some
examples, but there are a lot of functions in the Python library, it
seems, that are not mentioned.  

>The only trouble I had with refcounting (while writing the
>augmented_assignment patch) was not bothering to see wether
PyDict_SetItem()
>stole a reference or not. It doesn't, but I did INCREF, so I was
leaking ;-P

OK, good example. How _did_ you find out what your problem was and what
was _really_ required by PyDict_SetItem()?  Did you have to dig into
the actual code for  PyDict_SetItem()?


From gmcm at hypernet.com  Wed Jun 28 19:27:26 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 28 Jun 2000 23:27:26 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com>
Message-ID: <8F61C6D29gmcmhypernetcom@199.171.54.194>

Sean Blakey wrote: 

>On Wed, Jun 28, 2000 at 09:56:27PM +0000, Huaiyu Zhu wrote:

[snip]

>OK, I think I see what you mean.  You are objecting to the way that
>python searches first for modules in the local directory before
>searching the normal PYTHONPATH, correct?

>If you don't want python to search the current directory for imports,
>there is a very simple solution: put this snippet above all of your
>other imports: import sys
>sys.path.remove('.')

Not a real solution. If module x is part of a package, and does an "import 
y", then the package is searched first. This has nothing to do with  '.' 
being on sys.path. It's called a "relative import", and it means that if 
you have "package.os", then the real "os" is unavailable anywhere in 
"package". Period.

I think the "relative import" feature should be shot in the head (2 posts 
today), but unfortunately most packages rely heavily on it.

- Gordon


From eric at estinc.com  Tue Jun 27 19:15:03 2000
From: eric at estinc.com (Eric Lee Green)
Date: Tue, 27 Jun 2000 23:15:03 GMT
Subject: more python evangelising...
References: <200006271406.AAA06218@mbuna.arbhome.com.au> <D9465.71358$7o1.1704294@news2.rdc1.on.home.com>
Message-ID: <395935AB.4C91AF73@estinc.com>

Tom wrote:
> 
> But you're falling into the old trap of thinking that one language is best
> for everything.
> 
> Python is excellent for what it was designed for, but this does not include
> large scale app development.

Huh? What is "large scale"?

Just curious. I've written a large app in Python (sorry, NDA), and I can't see
anything in the language that renders it useless for "large scale" app
development. About the only painful thing is Python's speed, and in most cases
that can be solved by simple application of a "C" module for time-critical
pieces. 

In fact, Python is excellent for large-scale apps in particular, because the
Python interpreter and runtime is small enough that you can distribute it as
part of your app, complete with any "C" modules that you needed, and still be
more compact than the typical bloated monstrosity that is the typical modern
C++ program. For more modest Python projects that's not true (the size of the
interpreter becomes a real issue). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From etoffi at softhome.net  Wed Jun 14 02:25:34 2000
From: etoffi at softhome.net (etoffi)
Date: Wed, 14 Jun 2000 02:25:34 -0400
Subject: tuple question
Message-ID: <3947255E.7CF77923@softhome.net>

*** post for free via your newsreader at post.newsfeeds.com ***

admittedly this is a very [insert descriptive word here] question to ask, but
what is a tuple and why is it used (as opposed to an array, etc).  i suppose the
answer could be found in some comp sci book somewhere, but i dont have access to
those books.

ps: i know:

	(x, y) = point.get()

i just want to know what was the reason for creating/using tuples, more from a
language implementor standpoint than a witing-a-program-in-Python standpoint...

--
Hopefully somebody will understand what I'm trying to say

  **** Post for FREE via your newsreader at post.newsfeeds.com ****

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
*** Newsfeeds.com - The #1 Usenet Newsgroup Service on The Planet! ***
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
  http://www.newsfeeds.com???    |         http://www.newsfeeds.com
                                 |
* Anonymous posting server!      | * Totally Uncensored!
* SUPER Servers!                 | * Over 80,000 Newsgroups!
* BINARIES ONLY Servers!         | * 16 seperate Newsgroup Servers!
* SPAM FILTERED Server!          | * Instant access!
* ADULT ONLY Server!             | * Multiple OC 3's and OC 12's!
* MP3 ONLY Server!               | * 99% Article Completion!
* MULTIMEDIA ONLY Server!        | * Months of Retention!
* 7 UNCENSORED Newsgroup Servers | * Lightning FAST downloads!
                                 |
  http://www.newsfeeds.com       |         http://www.newsfeeds.com
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 ?      **** Point your newsreader to post.newsfeeds.com ****
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 


From Ted.Horst at ubsw.com  Thu Jun 22 14:10:35 2000
From: Ted.Horst at ubsw.com (Ted Horst)
Date: Thu, 22 Jun 2000 13:10:35 -0500
Subject: Roman Number Representation
Message-ID: <200006221810.AA05499@ch1d2833nwk>

Ok, here is my take on how to do roman numerals in python.  rn and an are  
simple, basic routines that are only valid for positive numbers.  rn2 and  
an2 expand into negative numbers and use a goofy notation that I made up for  
large numbers.  Their range is limited only by your memory and patience.

------8<------romans.py------------8<--------------------------------
# basic roman numeral conversion by Ted Horst <Ted.Horst at ubsw.com>
# used as part of JulianFormatting
# which is part of Julian, a date, time, timezone package
# comments, fixes, extensions appreciated

'''module to convert between roman and arabic numerals.
rn does a simple conversion from a number to a roman numeral string.
an does a simple conversion from a roman numeral string to an int.
rn2 converts a numer to a roman numeral string
 using a goofy, made up notaion for large numbers.
an2 converts from a roman numeral string into a long integer
 taking in the goofy notation output from rn2.'''

import string
_join = string.join
del string

_romans = ((1000, 'M'), (900, 'CM'), (500, 'D'), (400, 'CD'),
          (100, 'C'), (90, 'XC'), (50, 'L'), (40, 'XL'),
          (10, 'X'), (9, 'IX'), (5, 'V'), (4, 'IV'), (1, 'I'))

_couples = {'CM':900, 'CD':400, 'XC':90, 'XL':40, 'IX':9, 'IV':4}
_singles = {'M':1000, 'D':500, 'C':100, 'L':50, 'X':10, 'V':5, 'I':1}

def rn(num):
    'convert from a number to a roman numeral string'
    res = ''
    for a in _romans:
        res = res + a[1]*int(num/a[0])
        num = num%a[0]
    return res

def an(roman):
    'convert from a roman numeral string to an integer'
    place = 0
    num = 0
    while place < len(roman):
        if roman[place:place + 2] in _couples.keys():
            num = num + _couples[roman[place:place + 2]]
            place = place + 2
        else:
            num = num + _singles[roman[place]]
            place = place + 1
    return num

def rn2(num):
    'convert from a number to a possibly goofy roman numeral string'
    num = long(num)
    if num < 0:
        res = '-'
        num = -num
    else:
        res = ''
    if num < 4000:
        res = res + rn(num)
    else:
        if res:
            res = '%s(%s)' % (res, _rnb(num))
        else:
            res = _rnb(num)
    return res

def an2(roman):
    'convert from a possibly goofy roman numeral string to a long'
    return eval(_subit(roman) or '0')

def _rnb(num):
    exp = (len(repr(num)) - 2)/3
    if exp == 1:
        expr = 'M'
    else:
        if exp < 4000:
            expr = 'M**%s' % rn(exp)
        else:
            expr = 'M**(%s)' % rn2(exp)
    mag = 1000L**exp
    co = num/(mag)
    if co == 1:
        cor = ''
    else:
        cor = '%s*' % rn(co)
    rem = num - co*mag
    if rem:
        remr = '+%s' % rn2(rem)
    else:
        remr = ''
    return '%s%s%s' % (cor, expr, remr)

def _subit(as):
    state = 0
    ret = []
    tn = ''
    for a in as:
        if a in 'MDCLXVI':
            state = 1
            tn = tn + a
        else:
            if state == 1:
                num = long(an(tn))
                ret.append(repr(num))
                tn = ''
                state = 0
            ret.append(a)
    if state == 1:
        num = long(an(tn))
        ret.append(repr(num))
    return _join(ret, '')

def test():
    import traceback
    try:
        for i in range(4000):
            if an(rn(i)) != i:
                raise ValueError
        for i in range(-3999, 4000):
            if an2(rn2(i)) != i:
                raise ValueError
        for j in range(10):
            for i in (10L**j, 10L**j+1, 10L**j-1,
                      -10L**j, -10L**j+1, -10L**j-1):
                if an2(rn2(i)) != i:
                    raise ValueError
        for j in range(4):
            for i in (10L**10**j, 10L**10**j+1, 10L**10**j-1,
                      -10L**10**j, -10L**10**j+1, -10L**10**j-1):
                if an2(rn2(i)) != i:
                    raise ValueError
        for i in (long(1e200), 9L**3000):
            if an2(rn2(i)) != i:
                raise ValueError
    except:
        print 'FAILED TO PRESERVE', i, 'IN TRANSLATION'
        traceback.print_exc()
    else:
        print 'ALL TESTS PASSED'

-----8<-----end romans.py-----8<-------------------
Ted Horst <Ted.Horst at ubsw.com>



From embed at NOSPAM.geocities.com  Tue Jun 27 13:16:13 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 27 Jun 2000 13:16:13 -0400
Subject: How do you copy a Dictionary object from C code?
References: <Ge565.9419$227.220401@nnrp1.uunet.ca>
Message-ID: <0j565.9440$227.220430@nnrp1.uunet.ca>

"Warren Postma" <embed at NOSPAM.geocities.com> wrote in message
news:Ge565.9419$227.220401 at nnrp1.uunet.ca...
> Please Guido, add PyDict_Copy( dict ) to the API in the future!!!!
>
> This is an annoying oversight in the Dictionary's design.

Nope. I was wrong. I'm an idiot. I'm sorry.

I just realized that I shouldn't automatically assume the implementation is
actually that of dictobject.c, in other words another "sort of a dictionary
thing" would also work in place of a real dictobject.c object, but only if I
treating it with respect. It is some kind of Object, and it has an attribute
named "copy" which I can invoke. Much better.  Kudos to the Abstract and
Weird way that Guido's mind works.

Warren




From thomas at xs4all.net  Wed Jun 21 13:57:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 21 Jun 2000 19:57:48 +0200
Subject: mailing list duplicates?
In-Reply-To: <3950B648.29F8FC2B@bam.com>; from scherbi@bam.com on Wed, Jun 21, 2000 at 12:34:17PM +0000
References: <3950B648.29F8FC2B@bam.com>
Message-ID: <20000621195747.I26436@xs4all.nl>

On Wed, Jun 21, 2000 at 12:34:17PM +0000, Bill Scherer wrote:

> Is anyone else seeing a lot of duplicates on the list?

Yes, since last night or so. Barry, the list maintainer (and list-software
maintainer ;) is already tracking down the problem.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From andres at corrada.com  Wed Jun  7 08:13:09 2000
From: andres at corrada.com (andres at corrada.com)
Date: Wed, 7 Jun 2000 08:13:09 -0400
Subject: DEBUGGER question
In-Reply-To: <8hjrrg+tc6s@eGroups.com>; from al@servana.com on Tue, Jun 06, 2000 at 09:52:48PM -0000
References: <8hjrrg+tc6s@eGroups.com>
Message-ID: <20000607081309.A6719@corrada.com>

On Tue, Jun 06, 2000 at 09:52:48PM -0000, al at servana.com wrote:
> 
> When I have a script with a MAIN that is envoked when you 'execute 
> the script' (like: python scriptname.py -a -b -c arg1 arg2)
> I cannot find a way to get it 'in' to PDB.
> 
> If I try pdb.run('scriptname.main()') it does not actually execute 
> the script.
> I tried pdb.runcall also, and also used the __main__ notataion - 
> still no go.
> 
> If you look at the WEBCHECKER scripts which come with python, you can 

Try something like:

python /path/to/pdb.py scriptname.py

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From richard_chamberlain at ntlworld.com  Wed Jun  7 09:13:31 2000
From: richard_chamberlain at ntlworld.com (Richard Chamberlain)
Date: Wed, 7 Jun 2000 14:13:31 +0100
Subject: Tkinter: treeview widget
References: <393E3782.FF7A5DCA@proceryon.at>
Message-ID: <8hlhn9$1t2$1@gxsn.com>

Idle uses a tree widget in 0.5 that you could look at.

Richard

Horst Gassner <horst at proceryon.at> wrote in message
news:393E3782.FF7A5DCA at proceryon.at...
> Hi all!
>
> I am looking for a treeview widget. Does anyone know if such a widget is
> available somewhere?
> At the moment I am using PMW which has a very good collection of meta
> widgets but unfortunately it does not contain a treeview widget.
>
> Thanx
> - Horst




From maxm at normik.dk  Tue Jun 20 07:59:32 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Tue, 20 Jun 2000 13:59:32 +0200
Subject: SV: How to sell Python ( and OpenSoruce )
Message-ID: <7BD10B680501D411B9DF009027E06F32011AD1@EXCHANGE>

Fra: thomas at cintra.no [mailto:thomas at cintra.no]

>I need selling points for OpenSource, the idea and philosophy,
>examples of successful technology and eh .. moral support. I feel like
>I`m in the lions den right now. ( " Think of a happy place. Calm blue
>ocean, calm blue ocean ..." )

OS, Database, webserver, scripting language etc. is only building blocks for
making an applicatiuon server, where all the web functionality is divided
into managable modules that can be reused.

It probably is to expensive to develop your own application server
(appserver) with all the nessecary modules. Besides it will quickly fall
behind other open and commercial appservers in features, as they will have
more resources to develop them. So in the long run you must use either an
open or a commercial appserver.

If on the other hand you develop your own appserver you have the advantage
of owning the source code, so you don't risk devolping a lot of
functionality/modules/components on a platform with a loose foundation.

Ie if the company developing your appserver, stops supporting it, goes
bankrupt, or merges with another company, hostile takeover etc. You will no
longer be able to support your customers software. And that is BAD
business!!!!

I have been in this situation and it has been my worst business experience
ever. You have build a trust relationship with customers, and suddenly you
can no longer support them.

It also did cost us a few customers!

Therefore I believe that the ONLY possible way to develop web apps is with
an open source solution. You get the same scale as a commercial appserver,
and you have full controll/ownership of the code.

Currently I believe that Zope is the best solution when considering
language, features, performance etc. If only it was better documented there
would be more components and it would start to really take off.

The price of SQL server, NT, etc. doesn't really matter a hoot, except in
very small companies.

When somebody (like us) charges aprox. $150/hour+Tax, then the cost of
software is a very small percentage of doing business. Something that saves
time, improves customer relationships, gives better functionality and makes
it easier to estimate project cost, is a far better argument. This is what
an open source app server will do.

Kind regards

	Max M



From shapr at uab.edu  Wed Jun  7 09:48:01 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 07 Jun 2000 13:48:01 GMT
Subject: Ruby -- A better OO Perl than Perl? Python 3000 features available 
 now?
References: <8hl866$bt5$1@news.jump.net>
Message-ID: <393E523E.713D6732@uab.edu>

Conrad Schneiker wrote:

<snip>
I agree with an earlier reply, stop trolling, this is comp.lang.python,
not comp.lang.recruiting. If ruby is a good language, it will get people
interested on its own merits.

I've already looked at ruby, the main things that I didn't like are:
lack of first class functions, 'end' is required to end a function, and
ruby isn't as explicit as Python (it seems to me to cater somewhat to
P*** people).

For me and my programming style, Python is a better fit than Ruby.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From eduard.hiti at t-online.de  Sun Jun 11 07:20:34 2000
From: eduard.hiti at t-online.de (Eduard Hiti)
Date: Sun, 11 Jun 2000 13:20:34 +0200
Subject: Python and ADO 2.5?
References: <39430E84.2679DD45@micron.net>
Message-ID: <8hvsr1$rbs$12$1@news.t-online.com>

"Peter Immarco" <android at micron.net> wrote in message
news:39430E84.2679DD45 at micron.net...
> I am thinking of replacing my VBScript ASP's with Python.  VBScript is
> excellent for integrating with ADO (Active Data Objects) databases.  Is
> Python as facile?
>

I used ADO 2.5 in my last Python project, and had no problems at all. Most
of the time, ADO access in VB translates almost directly to python syntax.
Features of ADO I used with no problem include parameterized commands, batch
updating and client side cursors.

One thing to keep in mind is the way pythoncom handles out-parameters of
method calls: These values are returned as part of the method result, e.g
the Execute method of the Connection object returns a tuple consisting of an
RecordSet object and the RecordsAffected value (see MSDN):

    recordset, affected = db_connect.Execute("select * from ...")

In VB you would only get the recordset returned and RecordsAffected would
get stuffed in a reference parameter of the call.

Another point is handling of date values: They are returned as objects with
type pywintypes.TimeType. The documentation for this type is very terse, and
I had to dig into the pythoncom demos to find their interface: One useful
method is Format(<formatstring>), which will return a string from TimeType:

    import pywintypes
    if type(x) == pywintypes.TimeType:
        return x.Format("%d.%m.%Y")

For more information on COM and Python I recommend Mark Hammond's "Python
Programming on Win32" (which covers almost the entire COM body, albeit
sometimes not too deeply: "Advanced Python Programming..." ?), which has a
chapter on database programming.

Good luck,

    Eduard





From pete at visionart.com  Fri Jun  2 15:45:12 2000
From: pete at visionart.com (Pete Shinners)
Date: Fri, 02 Jun 2000 12:45:12 -0700
Subject: Help, Image Filters with numpy
References: <393803A2.9AE23E9C@visionart.com> <8h9284$rc2$1@info3.fnal.gov>
Message-ID: <39380EC8.3100E9F5@visionart.com>

Charles G Waldman wrote:
> The name for the operation you're looking
> for is convolution.  Any book on digital image
> processing will describe this - the book of
> Russ is an good general reference.
> 
> Numeric Python has a "convolve" method but it
> unfortunately only supports 1D convolution.  However
> the filter coefficients you give describe a separable
> kernel - i.e. the 2D convolution is a composition of
> 2 1D convolutions, in the row and column directions
> (this is not true for all convolution kernels).
> 
> Also note that the Python Imaging Library ("PIL") has
> builtin support for convolution filters.
> 
> Finally, if you are convolving with very large kernels,
> it is more efficient to use Fourier transform techniques.


wow, thanks for this abundance of info, charles.
i'll look into PIL, but i'm doing limited operations
on images, and it seems like more than i need.
(that, and "YAD" - Yet Another Dependency, heh)

i'll be sticking with smaller kernels, no larger than
5x5 likely. nonetheless, i've seen a lot about FFT's
but not much 'learning' or 'beginning' info. is there
somewhere i can go to get a handle on the basics of FFT?


From jvickroy at sec.noaa.gov  Thu Jun 29 10:33:10 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Thu, 29 Jun 2000 08:33:10 -0600
Subject: ODBC
References: <8jf7nm$n48$1@www.univie.ac.at>
Message-ID: <395B5E26.77666E49@sec.noaa.gov>

I have not seen this behavior when repeatedly accessing an SQL Server
database using Python 1.5.2 on a Win NT 4 platform.

What happens if you repeatedly open and close the database **without
executing any other statements** between successive "opens" ?


Matthias Huening wrote:

> Hi,
>
> I use the ODBC module that comes with Marc Hammond's winall.
> Everything works fine; I open a connection, do something, close the
> connection, like this:
>
> import odbc
>
> conn = odbc.odbc(DSN)
> cur = conn.cursor()
> sql = "SELECT col FROM table"
> cur.execute(sql)
> x = cur.fetchall()
> cur.close()
> conn.close()
>
> The problem: when I try to open another connection in the same script
> (after I closed the first one), Python crashes. Any ideas?
>
> Matthias
>
> --
> matthias.huening at univie.ac.at
> http://www.ned.univie.ac.at/dt/



From jmwarfe at darwin.cwru.edu  Tue Jun 20 12:07:15 2000
From: jmwarfe at darwin.cwru.edu (Mike Warfe)
Date: Tue, 20 Jun 2000 12:07:15 -0400
Subject: Changing directory ACLs
Message-ID: <394F96B3.34E31C68@darwin.cwru.edu>

Hello, I have been using Python to do administrative tasks on my NT PDC.

Is there a Win32 module that I can use to set directory privileges, that

has the functionality of cacls.exe using python? I am currently using
cacls to set the privileges by doing a os.system command in my script.
Thanks for the help in advance!


Mike Warfe



From matt at nightrealms.com  Thu Jun 22 06:20:22 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Thu, 22 Jun 2000 02:20:22 -0800
Subject: Module of built-in wrappers
Message-ID: <GXk45.1367$9W1.102607@news-west.usenetserver.com>

The Python FAQ gives an example of how to make a class wrapper around
a builtin type, like a dictionary.  Is there any module that has
class wrapers for every builtin type for which this can be done?

Thanks in advance.





From JamesL at Lugoj.Com  Fri Jun  2 00:14:44 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 01 Jun 2000 21:14:44 -0700
Subject: And now for something completely different. Was: String.join    revisited    (URGENT for 1.6)
References: <28052000.5@sanctum.jae.ddns.org> <sj3ejda85pj169@corp.supernews.com> <hjoY4.6071$wYl.223614464@newsb.telia.net> <39328edc$0$18384@personalnews.de.uu.net> <_pyY4.10102$Za1.160118@newsc.telia.net> <3932e628$0$6681@personalnews.de.uu.net> <39348990.F7E82EC0@Lugoj.Com> <3934fbdd$0$8957@personalnews.de.uu.net>
Message-ID: <393734B4.11E2D780@Lugoj.Com>

Johannes Stezenbach wrote:
> 
> James Logajan <JamesL at Lugoj.Com> wrote:
> >Speaking of interesting syntax, did you know the following is perfectly
> >valid C/C++:
> >
> >var = 2["abc"];
> >
> >Points awarded for determining the type of "var" and the value assigned to
> >it. Extra credit points for explaining why this is valid C/C++. (Hint: this
> >has been valid C for as long as the language has been around.)
> 
> The magic of C pointer arithmetics makes a[i] equivalent to i[a], so
> in the above var is of type char and has a value of 'c'.
> 
> Explanation:
>   After e.g.    char a[] = "abc";
>   the following are true:
>         a == &a[0]
>         *a == a[0]
>         a + i == &a[i]
>         *(a + i) == a[i]
>         *(i + a) == a[i] == i[a]
> 
> The good thing about this is that nobody ever uses it,
> except maybe in obfuscation contests.
> 
> Johannes

You get all available points! Because any combination of expressions e1[e2]
is immediately reduced to *(e1 + e2) by the compiler (by definition) and
because the "+" operator commutes, e1[e2] == e2[e1]. HOWEVER, no equivalent
line of reasoning applies to array declarations. "Type var[size]" can not be
changed to "Type size[var]"; again by definition of the language.


P.S. A friend once pointed out some old code where someone had mapped DOS
drive numbers 0, 1, ... to drive letters 'a', 'b', ... using the simple
expression "abcd"[drive_num].


From confused at americasm01.nt.com  Thu Jun  8 11:59:14 2000
From: confused at americasm01.nt.com (Morris, Brad [WDLN2:2W40:EXCH])
Date: Thu, 08 Jun 2000 11:59:14 -0400
Subject: Nympy for win
References: <Pine.LNX.4.21.0006061109170.30720-100000@chelin.chtf.stuba.sk> <fAdMjhAnqNP5Ew3P@jessikat.demon.co.uk> <m3aegy60g1.fsf@brant.geog.ubc.ca>
Message-ID: <393FC2D2.F751741E@americasm01.nt.com>

Thanks, I've been wondering where Win NumPy could be found. Having
looked through the files I couldn't see an obvious way to install it.

Is there a README or web page that says where to put all the files from
the zip archive into the python directories?

Also, where is the python path information stored under windows to allow
other directories to be searched by python (I'm a Mac user using a
PC...)?

Regards,
BJM

Phil Austin wrote:
> 
> Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> 
> > >I would like to use the Numpy package with my windoze/python1.5.2 system,
> > >but it needs distutils and distutils need m$ visual c++ which I don't own.
> > >Is here anybody who has it compiled and could send me the
> > >binaries? Thanks.
> 
> The binary file at http://download.sourceforge.net/numpy/python-numpy-15.2.zip
> works for me with python 1.5.2.  Unfortunately, the new sourceforge site
> hasn't made it onto the numerical topics webpage at:
> 
> http://www.python.org/topics/scicomp/numpy_download.html
> 
> and the LLNL archive no longer exists.  Is there someone besides webmaster at python.org
> who should get this bug report?
> 
> Regards, Phil


From bwarsaw at python.org  Thu Jun  8 00:05:44 2000
From: bwarsaw at python.org (Barry A. Warsaw)
Date: Thu, 8 Jun 2000 00:05:44 -0400 (EDT)
Subject: "syncmail" in CVS repository
References: <20000606145723586.AAA450.401@hermes.cinetic.de>
Message-ID: <14655.7064.244727.512321@anthem.concentric.net>

>>>>> "JH" == Juergen Hermann <jhe at webde-ag.de> writes:

    JH> I made the following change to the syncmail script (fixes an
    JH> ugly error msg when committing a file for the first time).

Thanks Juergen.  I will apply a slightly modified version of this
patch to the syncmail script on SourceForge.

-Barry




From digitig at cix.co.uk  Wed Jun 14 17:32:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Wed, 14 Jun 2000 22:32 +0100 (BST)
Subject: Where are the math functions?
References: <owen-97F32C.10015712062000@news.u.washington.edu>
Message-ID: <memo.20000614223200.31901D@digitig.compulink.co.uk>

In article <owen-97F32C.10015712062000 at news.u.washington.edu>, 
owen at astro.washington.edu (Russell E. Owen) wrote:

> In article <slrn8k8dfv.tu.hzhu at mars.localdomain>, 
> hzhu at users.sourceforge.net wrote:
> 
> >I'm quite puzzled by not being able to find most of the common special
> >functions, such as (incomplete) gamma, beta and bessel and erf.
> >
> >This arrises from effort to provide statistical functions for the MatPy
> >package.  Many stat functions require incomplete gamma/beta and some
> >variants.
> >...
> 
> The random module (which is available on all platforms) offers beta and 
> gamma functions. I can't comment on incomplete gamma, bessel and erf. I 
> found the information about the random module in "Python Essential 
> Reference" by David Beazley; the index lists beta and gamma, but not 
> the others.

No it doesn't. It includes the beta and gamma /distributions/. These are 
related to, but not the same as, the beta and gamma /functions/.

When I wanted the gamma function I just ported code from Numerical Recipes 
in C, which only took a couple of minutes (longer to test, but I'd have 
spent that time if I'd just typed in the C). Probably still be a copyright 
violation to post it, though :-(


From anders.eriksson at morateknikutveckling.se  Thu Jun 22 04:22:04 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 22 Jun 2000 10:22:04 +0200
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <3gh3lsgg9dtkfgrsq1ml2i9jcfkknrb6ei@4ax.com>

On Wed, 21 Jun 2000 20:25:24 GMT, "William Dandreta"
<wjdandreta at worldnet.att.net> wrote:

>I don't really need to use Dbase but I need something that works with DOS
>and is platform independent.
>

Hmmm... How fancy database action are you going to use? Maybe the
shelve module is enough?

There also is a database called MetaKit, which I'm kind of found of.
it really small and easy (don't have SQL, but neither has dBASE). I'm
not sure if it exist for DOS but the creator mr Jean-Claude Wippler
maybe sees this and replies!

>I found XBase which is an open source C++ database library that uses dbase
>data structures and is platform independent. If Python does not already have
>Dbase access, I guess I will have to figure out how to put a Python wrapper
>around Xbase.
>

I remembered xBASE++ after I had pressed send ;-) This library has
changed name and 'owner'. It's now called xDB and you can find it at:
http://linux.techass.com/projects/xdb/

>What is SWIG?
>

SWIG is a 'program' that creates a wrapper around C libraries. Which
means that you can write a function in C, run it thru SWIG and get a
Python module that works like a middle tier between your Python code
and the C library. 
http://www.swig.org

It's a bit more difficult on C++ (.or so they say.)

I do recommend you to find an another solution than wrapping xDB,
maybe you could use one of the XML packages that exist for Python. the
speed will probably not be as fast as a database, but ... If you tell
us a bit more about what your application will do maybe we can steer
you in the 'right direction'

// Anders



From rob at hooft.net  Wed Jun 14 01:52:49 2000
From: rob at hooft.net (Rob Hooft)
Date: 14 Jun 2000 07:52:49 +0200
Subject: Newbie: Pass Value
References: <8horhs$rfl$1@news-int.gatech.edu>
Message-ID: <m2aegoolla.fsf@hooft.net>

>>>>> "SL" == Shengquan Liang <liang at cad.gatech.edu> writes:

 SL>  in the main progam, i call a method populate (), and pupulate
 SL> call another method assign() .

 SL>  like this: populate(assign())

The example is not what you describe.

 SL>  e.g.  populate(assign(post= form1["InputNumber"]) )

This will work if "assign" is defined with a parameter "post":

def assign(post):
    print post

You can even specify a default value for "post" if you want to be
able to call assign without arguments.

Regards,

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


From fiona at sitegnome.com  Tue Jun 20 08:53:42 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 20 Jun 2000 12:53:42 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 20th, 2000
Message-ID: <20000620125342.29489.qmail@synop.com>

Hi guys,

Below are the latest entries into http://python.faqts.com

cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Is getstatusoutput() the prefered method to run os commands from within a python script or is there something more generic?
http://www.faqts.com/knowledge-base/view.phtml/aid/3850
-------------------------------------------------------------
Fiona Czuczman
Matthew Schinckel

Try using os.system(command)


-------------------------------------------------------------
How can I find out how much memory a remote computer has installed? The computers are on the same NT network and have the same login.
http://www.faqts.com/knowledge-base/view.phtml/aid/3866
-------------------------------------------------------------
Fiona Czuczman
Roger Upole

Try the win32pdh module from the win32 extensions.  The NT performance
monitor can access memory stats on another machine if you have 
sufficient privilege.  The hard part is getting the path string right, 
but there is a path browser to help with that.  I think it's 
win32/lib/win32pdhutil.py.


-------------------------------------------------------------
How do I create a fully stand alone programme to put on systems where python is not installed?
http://www.faqts.com/knowledge-base/view.phtml/aid/3868
-------------------------------------------------------------
Fiona Czuczman
Matthias Huening, Matthew Schinckel

You should try Gordon McMillan's "Installer". It's at:
http://starship.python.net/crew/gmcm/distribute.html

One option is to use mxCGIPython (can't remember the URL, find it on
Parnassus <http://www.vex.net/parnassus>, and just have a compiled
single-file binary and the program.


-------------------------------------------------------------
I'm trying to get a user's username and password but by using raw_input() the password is echoed to the screen.  How can I make it so that their password isn't shown?
http://www.faqts.com/knowledge-base/view.phtml/aid/3869
-------------------------------------------------------------
Fiona Czuczman
Janko Hauser, David Goodger

Look for the module getpass in the library reference. It also supports
windows according to the document string.

    import getpass
    password = getpass.getpass()

If you still see the password echoed to the screen, then you're missing 
some functionality, probably termios. Check out the getpass.py code for 
the decision tree that is followed.


-------------------------------------------------------------
Could someone give me a simple explanation of regular expressions?
http://www.faqts.com/knowledge-base/view.phtml/aid/3870
-------------------------------------------------------------
Fiona Czuczman
Andrew Kuchling

Regular expressions are a mini-language for defining a certain set 
of strings; the re module then takes a regular expression and tells you
if a given string matches the expression.  For example, [abc] means
'a', 'b', or 'c'; \bA.*?\b means a word beginning with 'A' (\b is a
special sequence meaning "match a word boundary").  As you can see,
regular expressions are concise but can easily be obscure.

Regular expressions will practically never be faster than a simple
string.find, so if you're searching for a fixed string, don't use
regular expressions; their extra expressive power costs you in
performance.  But they can do things that aren't possible with the
string module alone, and would require several lines of Python code 
to implement.

Consult the Library Reference, or the Regular Expression HOWTO for
more info.  (http://www.python.org/doc/current/lib/module-re.html,
http://www.python.org/doc/howto/regex/)


-------------------------------------------------------------
Can two threads get in each others' way if they simultaneously try to mutate a list?
Can two threads get in each others' way if they simultaneously try to mutate a list?
http://www.faqts.com/knowledge-base/view.phtml/aid/3831
-------------------------------------------------------------
Fiona Czuczman
Aahz Maruch

It depends how they mutate it.  I believe that single builtin functions
(e.g. [].append()) are thread-safe.  OTOH, [].sort() may not be
thread-safe, particularly if you've got user-defined classes on the list
with __cmp__() defined.


-------------------------------------------------------------
When and why should I use Intvar() etc?
When and how do I use the textvariable option?
http://www.faqts.com/knowledge-base/view.phtml/aid/3867
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

from Tkinter import *
root=Tk()
myIntVar = IntVar()
myStringVar= StringVar()
Label(root,textvariable=myStringVar).pack(side=TOP)
def setLabel():
    myStringVar.set(myIntVar.get())
Radiobutton(root,text="First Radiobutton",variable=myIntVar,value=1,
                    command=setLabel).pack(side=LEFT)
Radiobutton(root,text="Second Radiobutton",variable=myIntVar,value=2,
                    command=setLabel).pack(side=LEFT)
root.mainloop()

I'm using both IntVar and textvariable in this example.

I create new instances of Intvar and StringVar. I then create a label 
and assign the textvariable option to the instance of StringVar. I can 
now set the text of this label by called myStringVar.set('Hello') or get 
the text by myStringVar.get().

Next I've created two Radiobuttons and assigned their variable options 
to the instance of IntVar. I give them two separate values. Any changes 
to the radiobuttons will now be reflected in myIntVar.

I created a function called setLabel which uses the set and get methods 
of the instances. I then assign it to the label by called 
myStringVar.set(myIntVar.get()) which changes the value of the 
textvariable and the change is immediately reflected on the label.


## Edited Entries ##############################################


-------------------------------------------------------------
How can I extract all the keys in a formatting string to a list?
How can I extract all the keys in a formatting string to a list?
http://www.faqts.com/knowledge-base/view.phtml/aid/1505
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Michael Str?der, Johannes Stezenbach

For example, go from:

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

to get:

  ['language','count']

Try something like this:

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

gk = GrabKeys()

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

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


-------------------------------------------------------------
Are colons unnecessary in my Python code?
http://www.faqts.com/knowledge-base/view.phtml/aid/2910
-------------------------------------------------------------
Fiona Czuczman
Courageous, David Porter, Fran?ois Pinard, Moshe Zadka, Dirck Blaskey

Lead in discussion:

Sometimes colons seem syntactically unnecessary. For example:

        if <condition>:
                statement
        else:
                statement

Really, else doesn't need a colon, as far as I can tell (I can see the 
need for the if, supposing you want to have the statement on the same 
line).

Answer(s):

- When using python-mode in Emacs (or jed), the colon facilitates
auto-indention. Also, if you forget the colon, the next line will not be
indented, so you will catch your mistake. 

- Granted, but it is always good for a language to have a bit of 
redundant information.  When properly thought, such redundancy prevents 
various mistakes from programmers (once they are used to it, of course 
:-), and often increase overall legibility.

- Theoretically, a colon is only necessary in things like

if yes: print "yes"

Since otherwise the parser can figure out when to stick a colon. 
However, usability studies show people are more comfortable when the 
beginning of a block is signaled, and I can see why:

if yes
        print "yes"

Seems....naked. Much less readable then 

if yes:
        print "yes"

Guido didn't want 10 ways (or even 2) to spell things, so the colon is
mandated for all.

- > if yes: print "yes"

Oddly enough, the parser doesn't really need the colon here either.
It can manage to figure out where the if expression ends without it.

The colon is almost entirely for readability purposes.
(there are a couple of places where ambiguity occurs without it).

If you're curious about the other thread,
or about Python without colons,
or to test my above assertion,
check out:

http://www.danbala.com/python/colopt


-------------------------------------------------------------
What is a tuple and why is it used (as opposed to an array, etc)?
http://www.faqts.com/knowledge-base/view.phtml/aid/3795
-------------------------------------------------------------
Fiona Czuczman
Stephen Hansen, Aahz Maruch, Warren Postma,Russell E Owen

A Tuple is of a fixed length, the length it was when it was created. A
list is of a dynamic length -- its actual size, I believe, from a 
memory-ish standpoint, that lists are actually larger then they appear 
to be in the Python environment. On the C level, my understanding from 
reading various posts is that the list allocates additinoal memory for 
elements which may be added in later, so that it doesn't have to 
allocate One Piece Of Memory every time an element is added. A Tuple is 
immutable. Once created, with its given values, it can not be changed. a 
= (1,2,3) can not have a[1] = 5 done. A list is mutable. You can change 
the individual elements at any time.

My unknowing guess? The overhead to create the truly 'dynamic' lists is 
completely wasted in enough circumstances that it warrented making an 
array which was fixed, in both size, and content.

----------

>From my POV, the primary purpose of tuples is for dictionaries: they
allow you to do things like

foo = {}
foo[(1,2)] = 3

This essentially permits the use of multi-column primary keys (to use
database terminology).  In addition, as Stephen Hansen pointed out,
tuples are a fair bit more efficient than lists.

---------

I first came across the term tuple in relation to Relational Algebra and 
in particular, in the realm of SQL and Databases.

A row of data in a database is often called a tuple. Really, it's a way 
of creating a Structure like you would do in C, very efficiently, which 
once created, is immutable. This means you can create a hash on it, and 
then use the hash to get the piece of data out of a database table. This 
is exactly how Tuples work, and they are hashed and then placed into an 
associative array (Python calls associative arrays Dictionaries).

So whereas there is one native Array type in C, the List type in Python 
is a mutable dynamically sized array, and a Tuple is a fixed immutable 
grouping of items, and a Dictionary is like an array where the key 
doesn't have to be a number.  So, three kinds of things which have 
something in common with a C array is better than just having a return 
to the Bad Old Days where C arrays were fixed in size, mutable to the 
point that you could easily crash your system, and totally 
non-polymorphic.

-----------

Tuples are immutable only at the highest level (the level of a shallow 
copy). The entries in a tuple cannot be added or removed, but if an 
entry is a mutable object, that object is still mutable and hence the 
tuple is mutable.

I suspect the main reason for the existence of tuples is for use as 
dictionary keys. In this situation it is important to use a truly 
immutable tuple (one containing only immutable entries). I'm not sure 
how much protection Python offers, but if you do succeed in using a 
mutable tuple as a dictionary key, and the tuple changes, the key will 
no longer point to the entry.

tuples appear to be basically a "const" form of lists. If Python had 
"const" (from C++) then it probably would not need tuples.


-------------------------------------------------------------
Is the match() function any different to having search() with a ^ at the beginning of the string?
http://www.faqts.com/knowledge-base/view.phtml/aid/3348
-------------------------------------------------------------
Fiona Czuczman
Tim Peters

Yes, although the difference doesn't become clear until you exploit the
optional slicing arguments of compiled regexps:

>>> import re
>>> p = re.compile("^a")
>>> p.search("cba", 2)
>>> p = re.compile("a")
>>> p.match("cba", 2)
<re.MatchObject instance at ead590>
>>>

That is, while there is an "a" at index 2 of "cba", and "match" finds 
it, a caret does *not* match starting at index 2 (caret means 
start-of-string, and may or may not mean more than that depending on 
"multiline" mode -- using "match" instead cuts all that magical crap out 
of the equation).


-------------------------------------------------------------
How can I choose an element from a list at random?
How can I choose an element from a list at random?
http://www.faqts.com/knowledge-base/view.phtml/aid/1361
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Bruce Wolk

import random
random.choice(list)


-------------------------------------------------------------
Is there a function returning the intersection of two lists?
http://www.faqts.com/knowledge-base/view.phtml/aid/3283
-------------------------------------------------------------
Fiona Czuczman
Penfold

Example:

l1 = [1,2,3,4,5]
l2 = [4,5,6,7,8]

l3 should be [4,5]

Solution:

You could try using the ever useful kjbuckets module (find it on
http://www.vex.net/parnassus).

This implements mathematical sets and graphs which mean your below 
becomes as easy as

kjSet([1,2,3,4,5]) & kjSet([4,5,6,7,8])


-------------------------------------------------------------
I'd like to create a subclass of list that loads its contents from a file when you instantiate it. What do I subclass to do this?
http://www.faqts.com/knowledge-base/view.phtml/aid/3424
-------------------------------------------------------------
Fiona Czuczman
Remco Gerlich

You can't directly subclass types, like lists and integers. This is 
commonly viewed as a flaw in Python, maybe this will be changed once.

However, you can subclass UserList.UserList, a Python wrapper around a 
list (see Lib/UserList.py for its source).


-------------------------------------------------------------
I'm searching information regarding the use of pointers for linked and double linked lists.
http://www.faqts.com/knowledge-base/view.phtml/aid/3770
-------------------------------------------------------------
Fiona Czuczman
Martijn Faassen

Linked lists aren't hard:

class Node:
    def __init__(self, next):
        self.next = next

linked_list = Node(Node(Node()))

Every name in Python's a reference, so no pointers are needed. Just 
think of every name in Python as a pointer (to an object), if you like.

Doubly linked lists are along the same pattern, but have the trouble 
that they introduce circular references, which is bad for Python's 
reference counting based garbage collection scheme. You have to break 
one of the references yourself for it to work:

class Node:
    def __init__(self, prev, next):
        self.prev = prev
        self.next = next

node1 = Node(None, None)
node2 = Node(None, None)
node1.next = node2
node2.prev = node1

# and now to clean up so that refcounting works:
node2.prev = None






From paul at prescod.net  Mon Jun  5 23:00:20 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 05 Jun 2000 22:00:20 -0500
Subject: Python <--> XML data binding.
References: <8hhlr6$eoh$1@slb3.atl.mindspring.net>
Message-ID: <393C6944.2FBCE9B1@prescod.net>

Werner Fouche wrote:
> 
> Does anyone know if a tool exist that will generate Python data binding code
> for an XML DTD similar to what Breeze ( http://www.breezefactor.com ) does
> for Java.

I'm working on such a library. Check back in a couple of weeks.

Nevertheless, the Breeze website vastly exaggerates the difficulty of
doing object creation "by hand" using the output of a parser. I wouldn't
even bother with my data binding technology if the data binding did not
come "for free" as a side effect of other features I need.
-- 
 Paul Prescod
Those who profess to favor freedom and yet depreciate agitation, are men
who want crops without plowing up the ground. They want rain without
thunder and lightning. They want the ocean without the roar of its many
waters. - Fredrick Douglass
 http://www.informamerica.com/Articles/Quotes.htm



From entropiamax at jazzfree.com  Tue Jun  6 09:25:14 2000
From: entropiamax at jazzfree.com (entropia)
Date: Tue, 06 Jun 2000 15:25:14 +0200
Subject: Newbie question.
References: <393CE029.27B87C91@dulcesol.es>
Message-ID: <393CFBB9.F23565D@jazzfree.com>


Pablo Prieto escribi?:

> Hi all of you there!
>
> #This is my question
>
> AYS = ""
>
> def setEnviroment():
>   AYS = "NUMBER 9"
>
> def getEnviroment():
>   return AYS
>
> #This is the end...
>
> The function getEnviroment always return "". I guess AYS is declared
> twice, One is module-scope and the other is local to setEnviroment, so
> I've got two differents variables.
>
> How can I make the variable AYS global to setEnviroment, so I'd be able
> to return "NUMBER 9" in getEnviroment?. (Don't say return "NUMBER9".
> I'll get annoyed :))
>
> Well, Have a nice summer all of you (or Winter if you are in the other
> side of the floor).
>
> Pablo.
>
> --
> http://www.python.org/mailman/listinfo/python-list

That's easy:
    AYS=""
def set():
    global AYS
    AYS="NUMBER 9"

def get():
    return AYS





From 06361993941-0001 at t-online.de  Sat Jun  3 18:35:24 2000
From: 06361993941-0001 at t-online.de (Klaus Baldermann)
Date: Sat, 03 Jun 2000 22:35:24 GMT
Subject: Python to call commands, e.g. "find , perl scripts , nmap , and others" is this possible?
References: <393692FB.8610EBCA@dallas.net> <007901bfcbf2$c3b4e680$1906a8c0@fc.fenx.com> <3936FEFB.BA248CA3@sightreader.com> <003301bfcc30$57851820$1906a8c0@fc.fenx.com>
Message-ID: <r11ch8.cl.ln@qcok.lan>

On Thu, 1 Jun 2000 18:17:27 -0700, "Emile van Sebille"
<emile at fenx.com> wrote:

>see if that makes a difference.  Pipes in winxx are somewhat=20
>flakey as I understand.

>>>> mylist = os.popen('dir').readlines()
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>OSError: (0, 'Error')

On Windows, you usually have 2 Python interpreters:
- Python.exe, which gives you a dos style window, os.popen works
- PythonW.exe, which gives you only the windows opened by your
scripts, but os.popen doesn't work.

So, if you don't mind the (additional) dos style window, you can start
idle via Python.exe instead of PythonW.exe, and have a working
os.popen.

HTH
Klaus


From mikeb at mitre.org  Wed Jun 21 07:13:28 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Wed, 21 Jun 2000 07:13:28 -0400
Subject: Enumeration Types (when required)
References: <20000621095914.AFBD61D045@dinsdale.python.org>
Message-ID: <3950A358.FB6F9E@mitre.org>

Enumerations that hard-code to integers are more useful than
dictionaries in at least two cases. These are hard-realtime
implementation of the states and classes of finite state machines
and colored Petri nets whose transitions vary at run-time. 

(Since the transitions vary, they cannot be hardcoded as gotos. Hash
coded dictionaries would add a few cycles of overhead to lookup the
value of each enumeration. The hard realtime requirement means that
it would be required to know the worst case number of cycles added,
and that there could be no possibility of memory leakages or hanging
pointers. An array lookup of a constant, on the other hand, takes a
predictable number of cycles.)

For non-realtime algorithms, especially involving user interaction,
I cannot think of an algorithm where a dictionary would not do in
place of an enumeration. But inside servers, medical instruments,
chips, orbitting satellites, semantic artificial intelligence
technologies, nuclear safety devices, Jini federations, autonomous
intelligent software agents, FAA airspace deconfliction, weapon
fail-safes, auto-pilots, military battle-shorts, and other
non-client applications, the stability and predictability of code is
more important than speed. 

Admittedly, Python is not commonly used aboard satellites that must
run reliably for 30 years. But Python is quick to code in. It is
therefore being used to prototype algorithms that will be
reimplemented in hard realtime technologies including Ada and, where
permitted by license, realtime Java. Currently various Java and C++
tool licenses forbid the use of those languages/tools for military,
medical instruments, or any life-saving devices, and, until the
hanging pointer and memory leakage problems are solved, it is good
to think about where exactly are hard realtime techinques needed
versus where hangs, wait states, reboots, reformats, and reinstalls
are okay.

Mike Brenner
The MITRE Corporation


Cameron Laird wrote in message ...
>I'll say that more forcefully than Aahz does here:  while
>enums are often a marker of good taste in C coding, Python
>generally achieves the same results with a considerably
>cleaner reformulation in terms of dicts.  If you feel a
>need for enums while writing Python, you're probably missing
>out on a better (and not-so-reminiscent-of-C) coding.
>
>I've become rather dogmatic on this point.  I need the
>practice of working to falsify it.  If you have a specific
>algorithm you can share that you see as ripe for a Python
>enum, please tell us here; maybe we can suggest an
>alternative.




From cut_me_out at hotmail.com  Fri Jun  9 11:04:21 2000
From: cut_me_out at hotmail.com (Alex)
Date: 09 Jun 2000 11:04:21 -0400
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no>
Message-ID: <etdn1kuhp56.fsf@w20-575-109.mit.edu>

> I use Emacs under Linux and it works just fine. Just wish the
> code-completion-thingy in Pythonwin could somehow be implemented in
> Emacs.

The dabbrev package in emacs is a sort of poor-man's code-completion.  I
use it extensively.

Alex.


From glc at well.com  Wed Jun  7 15:16:06 2000
From: glc at well.com (Greg Chapman)
Date: Wed, 7 Jun 2000 12:16:06 -0700
Subject: Small bug in zlib module
Message-ID: <003a01bfd0b4$d485ea60$95730e3f@versa>

On 6 Jun 2000 16:42:06 -0000, Peter Schneider-Kamp wrote:
> Greg Chapman wrote:
> >
> > Using Python 1.5.2 (under Windows 98), this script causes python to
> > enter a loop in which it progressively allocates larger and larger
> > blocks of memory:
> >
> >   import zlib
> >   s = zlib.decompress("")
> > I assume PyZlib_decompress should have a check for an empty input
> > string and (if found) should generate an appropriate exception.
> >
> On the current CVS version it immediately (as far as I can see without
> looping) returns with an error.
>

I just checked ver. 2.32 (which seems to be the latest in the SourceForge
web CVS repository).  If the check you're referring to is this:

  if (r_strlen <= 0)
      r_strlen = 1;

that doesn't catch the condition I described.  Instead, you need to check
the "length" variable, which has the length of the input string (and I don't
see such a check; perhaps I missed it?).  Anyway, as it turns out, simply
checking for an empty input string is not sufficient.  The following causes
Python to enter the same memory allocation loop:

  import zlib
  s = zlib.decompress("\010")

I received a reply from eito which I believe correctly diagnoses the problem
and solution (this reply apparently did not go to the list):

On Tue, 6 Jun 2000 13:07:58 -0700, eito wrote:
> I think because, in zlib module, when it gets Z_BUF_ERROR, it does not
> check for the cause of the error. I don't know if the author of this
> module intentionally left the checking. (the same is true in the
> latest CVS version) Because it will return an error (MemoryError)
> anyway.
>
>
>   [from zlib.h (version 1.13)]
>   "Inflate() returns ... Z_BUF_ERROR if no progress is possible or
>    if there was not enough room in the output buffer when Z_FINISH
>    is used."
>
>
> The zlib's description is not so clear to me, but I modified zlib
> module anyway... The following patch (in the end of this e-mail) for
> zlibmodule.c (of version 1.5.2) should fix your problem. This patch
> checks for the output buffer, if the inflate() gets Z_BUF_ERROR. And
> if there is a room in the output buffer, it return an error.
> Otherwise, it goes to case(Z_OK), which tries to allocate more output
> buffer.
>

Eito's patch (see below) looks OK to me, provided that the test avail_out >
0 is a reliable way to distinguish between the two kinds of Z_BUF_ERROR
(i.e., cases where no progress is possible and cases where progress is
possible, but a bigger output buffer is needed).

By the way, I apologize for posting this to the main list.  I didn't realize
until yesterday that www.python.org has a bug list to handle this sort of
report.

Greg Chapman


-----zlibmodule.patch-----
*** zlibmodule.c.orig   Tue Jun  6 09:42:12 2000
--- zlibmodule.c        Tue Jun  6 09:53:12 2000
***************
*** 222,227 ****
--- 222,240 ----
          case(Z_STREAM_END):
            break;
        case(Z_BUF_ERROR):
+             /*
+              * If there is at least 1 byte of room in zst.next_out
+              * and get this error, I assume that it means zlib cannot
+              * process inflate.
+              */
+             if (0 < zst.avail_out)
+             {
+               PyErr_Format(ZlibError, "Error %i while decompressing data",
+                            err);
+               inflateEnd(&zst);
+               Py_DECREF(result_str);
+               return NULL;
+             }
          case(Z_OK):
            /* need more memory */
            if (_PyString_Resize(&result_str, r_strlen << 1) == -1)
-----zlibmodule.patch-----




From eviltofu at rocketmail.com  Sat Jun 10 14:43:11 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 10 Jun 2000 18:43:11 GMT
Subject: Type problem on Windows
References: <eviltofu-B4ED2B.12472010062000@news.earthlink.net> <02d101bfd318$8091ae10$6401a8c0@home>
Message-ID: <eviltofu-9D1688.14430510062000@news.earthlink.net>

In article <02d101bfd318$8091ae10$6401a8c0 at home>, "Darrell Gallion" 
<darrell at dorb.com> wrote:

> 
> sockets are platform dependent as expected :)
> 

Eeek. So maybe I should wrap the socket object in my own class and then 
test for that class.


From stevena at permanent.cc  Mon Jun  5 11:34:37 2000
From: stevena at permanent.cc (Steven D. Arnold)
Date: Mon, 05 Jun 2000 11:34:37 -0400
Subject: How to prevent a web client from timing out
In-Reply-To: <8hgfn8$3uq$1@nnrp1.deja.com>
Message-ID: <4.3.1.2.20000605113201.019893e0@phear.dementian.com>

At 03:07 PM 6/5/2000 +0000, aiglauer wrote:
>Not a purely python question, but.....
>
>Can I rig it so that the calculation process will still complete and
>write its output to a file so that I can use it later?  (Writing to a
>file is not a prob - preventing the thing getting slapped is).

If you are producing a report that is basically the same every time, you 
might have a cron job do the report every morning at 3 am and then simply 
send a cached copy to the user.  That way, the report should load almost 
instantaneously.

If you need to be able to do a fresh report every time, you might simply 
email the report to the user.  Excel can read HTML, so if you send the file 
with the correct MIME type and extension, the user can double-click it and 
it will open into Excel.  You could also have it open to a web browser, but 
for your app, it sounds as if Excel might be more useful.


steve


--
Steven D. Arnold            Que quiero sera         stevena at permanent.cc
"Fear: The best way out is through."                     -- Helen Keller




From h.stenstrom at ieee.org  Tue Jun  6 15:14:02 2000
From: h.stenstrom at ieee.org (Helge Stenstr�öm no spam please)
Date: Tue, 06 Jun 2000 19:14:02 GMT
Subject: File dialog
Message-ID: <m38zwiwrnq.fsf@ieee.org>

Is there any package available, which has a file dialag available,
similar to that of programs like Acrobat Reader, Netscape etc, i.e.:

a directories field with vertical scrollbar
a Files field with vertical scrollbar
a Selection field
and a Filter field
an Open/OK button
a Cancel button.

This thing would return a string representing the choosen file.


-- 
Helge Stenstr?m


From mwh21 at cam.ac.uk  Sun Jun  4 12:46:58 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 04 Jun 2000 17:46:58 +0100
Subject: Upcoming Starship changes
References: <200006021501.LAA11161@amarok.cnri.reston.va.us> <m2aeh1n1yn.fsf@pchoel.privat.uni-hamburg.de>
Message-ID: <m3g0qtxukd.fsf@atrus.jesus.cam.ac.uk>

bhoel at server.python.net (Berthold=?iso-8859-1?q?_H=F6llmann?=) writes:

> Could it be, that changing the machine broke mail forwarding again?
> Mail send to bhoel at starship.python.net seems to be lost again in
> nowhere. (It neither arrives at the address where it should be
> forwarded to, nor do I get an error message to the sending address.)

It seems the mx records for starship have disappeared, so your mail is
probably sitting in some queue somewhere.  I guess/hope they'll be
back sometime soon.

"starship.python.net" still points to the old machine here.

HTH,
Michael

-- 
6. Symmetry is a complexity-reducing concept (co-routines include
   subroutines); seek it everywhere.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From xxx at yellow13.mimuw.edu.pl  Wed Jun 21 11:34:16 2000
From: xxx at yellow13.mimuw.edu.pl (Robert Szefler)
Date: 21 Jun 2000 15:34:16 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <FvuAy4.51q@cwi.nl> <8hogt2$57t$1@nnrp1.deja.com> <8houlu$j9s$1@nntp.itservices.ubc.ca> <394889C3.1ED5@esatclear.ie> <t7545.239$ZR1.2944@news.uswest.net>
Message-ID: <slrn8l1np4.1jm.xxx@yellow13.mimuw.edu.pl>

In article <t7545.239$ZR1.2944 at news.uswest.net>, Jeff Massung wrote:
>Please, this interests me -- I've had differention equations, but that's as
>far as I go. In theory, pi (and e) are infinite, but has that been proven
>that there is not end to their decimals?
>
>Jeff

oh, come on, pi and e are irrational, so their decimal representations are
infinite and non-periodic. It's quite easy to prove e's irrationality. And
it's a bit harder for pi.
Moreover, they're transcendental numbers (do I spell it right?) ie. not a
root of any integer-coefficient-polynomial. This is substantially harder
to prove than the above ones.

steve


From freytag at szs.ira.uka.de  Mon Jun 19 12:32:02 2000
From: freytag at szs.ira.uka.de (Thomas Freytag)
Date: Mon, 19 Jun 2000 18:32:02 +0200
Subject: Kjbuckets-Module, Python 1.6 and NT
Message-ID: <394E4B02.5F4A0AB4@szs.ira.uka.de>

Dear all,

we have severe trouble getting the kjbuckets-module running with Python
1.6 and Windows NT. We have tried the following so far:

1)Compile the kjbucketsmodule.c as a dll. Fails and produces some
"Initialiser is not a constant" error, whatever...

2)Re-compile the Python interpreter using Visula C++ 6.0 and adding
kjbucketsmodule.c to the files to be considered in the project. Plus we
edited the "setup.in" in the Modules directory.  C++ claims to copile
"kjbucketsmodule.c", however, when we try  "import kjbuckets" with the
freshly compiled interpreter the answer is "No module....".

Sooo...has anyone managed to get kjbuckets running with 1.6 and NT? Any
hints are greatly appreciated.

Thanks a lot.

Thomas.






From fiona at sitegnome.com  Fri Jun  9 05:26:54 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 9 Jun 2000 09:26:54 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 9th, 2000
Message-ID: <20000609092654.30250.qmail@synop.com>

Hi Guys!!

Below are the entries I've entered into http://python.faqts.com

Cheers,

Fiona Czuczman


## New Entries #################################################


-------------------------------------------------------------
Has anyone successfully compiled the Python Interpreter on a Solaris 2.6?  If so, do you know where DIR is declared?
http://www.faqts.com/knowledge-base/view.phtml/aid/3612
-------------------------------------------------------------
Fiona Czuczman
Thomas Wouters,Dieter Maurer

Problem:

I downloaded the source for the Python Interpreter from
http://www.python.org/download/download_source.html.  When trying to
build it on Solaris 2.6 (SunOS 5.6), using gcc for Solaris 2.6, I get 
the following errors:

---------------------------------------------------------------------
./posixmodule.c: In function `posix_listdir':
./posixmodule.c:944: `DIR' undeclared (first use in this function)
./posixmodule.c:944: (Each undeclared identifier is reported only once
./posixmodule.c:944: for each function it appears in.)
./posixmodule.c:944: `dirp' undeclared (first use in this function)
./posixmodule.c:945: parse error before `struct'
./posixmodule.c:958: `ep' undeclared (first use in this function)
make: *** [posixmodule.o] Error 1
----------------------------------------------------------------------

Solution:

Did you run "./configure" before "make"?

We compiled Python on Solaris 2.4/5/6/7 without any problems.

"DIR" usually is defined in "sys/dir.h" or "ndir.h" or "sys/ndir.h".

If you did run "configure", then maybe you "gcc" is not properly 
installed.

DIR is a standard define in one of the standard UNIX header files, 
usually dirent.h (either directly or indirectly.) If you are missing 
dirent.h, you have a very broken system. Are you able to compile 
anything else ? Did you install the proper Solaris 'packages' ? (From 
what I recall, long ago, about SunOS installs and compilers, and gcc and 
the like, you have to be careful about what you install where.)


-------------------------------------------------------------
I am new to python and have tried to install numpy, but the darn thing doesn't recognise it's existence.
http://www.faqts.com/knowledge-base/view.phtml/aid/3617
-------------------------------------------------------------
Fiona Czuczman
David Porter

What happens if you start python from the directory that numpy is in and
then import Numeric? If that works, then you need to put the directory 
on your path. See site.py in the standard library for how this works. [ 
Quick explanation: if you have numpy at 
/usr/lib/python/site-packages/numpy, then create a file named numpy.pth 
in /usr/lib/python/site-packages with the word numpy as its contents. ]


-------------------------------------------------------------
Is there a site that lists what functions and or libraries are available?
http://www.faqts.com/knowledge-base/view.phtml/aid/3613
-------------------------------------------------------------
Fiona Czuczman
Shae Erisson, Justin Sheehy

If you want a listing of freely distributed functions and libraries that
don't come with the base Python install, check out parnassus:

http://www.vex.net/parnassus/

You may also not have been aware of the very nice online documentation 
for Python itself, at:

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

Current library references are found at:

http://www.python.org/doc/current/lib/


-------------------------------------------------------------
How do you get the IP number from a request for a web page?
http://www.faqts.com/knowledge-base/view.phtml/aid/3614
-------------------------------------------------------------
Fiona Czuczman
Michal Wallace, Chuck

Try putting this in your script:

#--start
import os
print os.environ["REMOTE_HOST"]
#--end

By the way, using the weblib wrapper ( http://weblib.sourceforge.net/ ), 
that's the entire script.

----------

Ah, but using the CGI Wrapper from Webware, it would only be:

#--start
print environ["REMOTE_HOST"]
#--end

  ;-)

http://webware.sourceforge.net    -- (forgot the URL)


-------------------------------------------------------------
I am looking for a treeview widget.  Is there one available?
http://www.faqts.com/knowledge-base/view.phtml/aid/3595
-------------------------------------------------------------
Fiona Czuczman
Richard Chamberlain

Idle uses a tree widget in 0.5 that you could look at.


-------------------------------------------------------------
I'm trying to capture an Entry box's value and then print it out. How can I do this?
http://www.faqts.com/knowledge-base/view.phtml/aid/3596
-------------------------------------------------------------
Fiona Czuczman
richard_chamberlain

from Tkinter import *
root=Tk()
entry=Entry(root)
var=StringVar()
entry2=Entry(root,textvariable=var)
entry.pack(side=TOP)
entry2.pack(side=TOP)

def calculate():
    var.set(entry.get())

button=Button(root,text="Calculate",command=calculate)
button.pack(side=RIGHT)
root.mainloop()

Getting and setting an Entry widget is straightforward. In the above 
example I've used entry.get() to get the contents of a widget and I've 
used the textvariable option to set the value.

var=StringVar()

This creates an instance of StringVar that we can assign to our entry.

entry2=Entry(root,textvariable=var)

So every change to var is now reflected in the widget.

var.set(entry.get())

I'm simply changing var to the contents of entry. Obviously here you'll 
need to do some calculations and pass that as a string.

In the Grayson book look at the appendices at the back to check out the
methods and options of Entry.


-------------------------------------------------------------
Is there any drag-and-drop feature now in Tkinter or any other Python public graphical interface?
http://www.faqts.com/knowledge-base/view.phtml/aid/3618
-------------------------------------------------------------
Fiona Czuczman
Richard Chamberlain

Have a look at Tkdnd.py in the standard distribution.


-------------------------------------------------------------
Can I explicitly request garbage collection from C?
http://www.faqts.com/knowledge-base/view.phtml/aid/3615
-------------------------------------------------------------
Fiona Czuczman
Courageous

Things are destroyed instantly when their reference counts reach zero. 
There's a function somewhere that prints out an object's reference 
count; that's probably what you want. Either that, or if you're writing 
a C extension, you can examine the object's reference count directly. In 
virtually all cases, it will be one higher than you expect, because the 
tuple that was sent as an argument to your external function also 
happens to own a reference to your object.


-------------------------------------------------------------
How do I print out the current time?
http://www.faqts.com/knowledge-base/view.phtml/aid/3600
-------------------------------------------------------------
Mike Hostetler
http://www.python.org/doc/current/lib/module-time.html

I didn't think that the official documentation was very clear on this (b 
ut it is there), so I'm adding it here.

If you do

>> import time
>> print time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(time.time()))

You'll see something like:

2000-06-08 15:05:09

To dissect this:

time.time() gets the time in seconds since the epoch

time.localtime(secs) creates a tuple

time.strftime(string,tuple) creates a string of the time, using the 
formatting in "string" (that is what the "%"'s are doing there).
To see the full table of formatting options, see 
http://www.python.org/doc/current/lib/module-time.html


-------------------------------------------------------------
Is there such a thing as a toString() for objects, like java has, that will be automatically called when an object is used "as a string"?
http://www.faqts.com/knowledge-base/view.phtml/aid/3616
-------------------------------------------------------------
Fiona Czuczman
Andres Corrada-Emmanuel, Quinn Dunkan

Problem:

ie. when you create an object, __init__ gets called automatically, so if 
I say:

x = Spam()
print x

Can I put a method in Spam such that it will print out something useful 
based on the class fields rather than the ugly data definition?

Solution:

Yes, it's __str__  See the Language Reference 3.3 "Special method names" 
for all the double-underscore magic you can accomplish.

In your class definition for Spam include a __str__ method, something 
like the following:

class Spam:
  .
  .
  .
  def __str__( self ):
     return self.someAttribute


## Edited Entries ##############################################


-------------------------------------------------------------
How can I call Python scripts from C/C++ program?
Is it possible for C   to call Python routines and vice-versa?
http://www.faqts.com/knowledge-base/view.phtml/aid/2633
-------------------------------------------------------------
Fiona Czuczman
Phil Austin

Here's a solution using CXX (http://CXX.sourceforge.net).  

You'll want the current CVS snapshot, as there have been several
bugs fixed in CXX-4.2

Note: put FibSeries into a file called myfun.py:

callable.cxx:

#include "Python.h"
#include "CXX_Objects.h"
#include <iostream>

using namespace Py;
using namespace std;

extern "C" void Py_Initialize();
extern "C" void Py_Finalize();

int
main(int argc, char* argv[])
{
  Py_Initialize();
  Module myfun("myfun");
  Callable fib = myfun.getAttr("FibSeries");
  Tuple argtuple(3);
  argtuple[0]=Int(2);
  argtuple[1]=Int(2);
  argtuple[2]=Int(30);
  Object output = fib.apply(argtuple);
  cout << List(output) << endl;
  Py_Finalize();
  return 0;
}


Compiling and executing this gives:

<peacock ~> callable
2
4
6
10
16
26
[2, 4, 6, 10, 16, 26]






From eableson at mac.com  Tue Jun 27 15:30:34 2000
From: eableson at mac.com (Erik Ableson)
Date: Tue, 27 Jun 2000 15:30:34 -0400
Subject: more python evangelising...
In-Reply-To: <200006271406.AAA06218@mbuna.arbhome.com.au>
Message-ID: <B57E791A.264E%eableson@mac.com>

Just to confuse things:

We are talking about MS - and they're more interested in making money in
training packages, ms certification programs, etc.  It stands to reason that
they have no interest in an open source product that they don't have the
ability to change arbitrarily to ensure that everyone in the corporate world
is sent on MS Certified retraining courses.

An open source product with a solid user base is of no interest to them
since they can't control and brand the add on services.  As always these
decisions are business decisions and in no way related to the relative
technical merits of a product :)

Cheers,

Erik Ableson

> From: Anthony Baxter <anthony at interlink.com.au>
> Reply-To: anthony at interlink.com.au
> Date: Wed, 28 Jun 2000 00:06:57 +1000
> To: python-list at python.org
> Subject: more python evangelising...
> 
> And an even more cunning route was at hand, as the redoubtable maintainer
> of the Java site Cafe Au Lait, Rusty Harold Elliot points out. Elliot,
> a long time critic of Sun's licensing tactics, notes: "If Microsoft
> wanted to really challenge Java, they should have gone with Python. I
> just don't believe it's possible for any major advances in language
> design to be made while restricting oneself to the mistakes Kernighan
> and Ritchie made 30 years ago."
> 
> Well, quite. The open source language Python has not only proven itself
> macho enough in heavy duty production environment, it's simple enough
> to be taught as a first language to kids. And it's interfaced with COM
> interface for several years now. Maybe someone in Redmond thought it
> was time to unleash some "innovation" at the world, and cometh the hour,
> cometh the new programming language.




From mrfx at etang.com  Mon Jun 12 04:38:01 2000
From: mrfx at etang.com (mrfx)
Date: Mon, 12 Jun 2000 16:38:01 +0800
Subject: IDLE  can be rewrited using wxPython as GUI widget?
Message-ID: <8i27c7$29gs$1@news.cz.js.cn>

The current IDLE uses TCL/TK as GUI widget.
IDLE  can be rewrited using wxPython as GUI widget?








From wtanksle at dolphin.openprojects.net  Tue Jun  6 21:20:23 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 07 Jun 2000 01:20:23 GMT
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu>   <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA>   <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <8h06uc$bnc$1@newshost.accu.uu.nl> <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net> <8hg4q0$6eo$1@newshost.accu.uu.nl> <slrn8josqp.qnj.wtanksle@dolphin.openprojects.net> <8hiq8d$ngm$1@newshost.accu.uu.nl>
Message-ID: <slrn8jr8qm.b4q.wtanksle@dolphin.openprojects.net>

On 6 Jun 2000 12:19:25 GMT, Martijn Faassen wrote:
>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>> On 5 Jun 2000 12:01:04 GMT, Martijn Faassen wrote:
>>>William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
>>>> On 30 May 2000 10:59:24 GMT, Martijn Faassen wrote:

>>>>>That's nice to say, but you don't agree with that yourself if you think
>>>>>Python's use of indentation is a good thing. It can help a lot if the
>>>>>language makes it impossible to be inconsistent. Unless you think the
>>>>>_only_ advantage of Python's indentation strategy is the lesser line 
>>>>>count.

>>>> This is a classical example of false dichotomy ;-).

>>>It's not.

>> It actually is.  You're claiming that I hold one of two opinions; I hold
>> neither one.

>I'm claiming you haven't shown yet that consistency has nothing to do
>with 'code looking good'. Anyway, you should've said something like that
>in your first post -- now you explain what you meant. Now I'll explain
>what I meant. :)

You're claiming something that you've never written before -- I don't even
understand what you're claiming that you claimed.  (Whew.)

What I disagreed with was your false reduction of my argument to two
views, neither of which I held.

>>>> I liked Python's
>>>> indentation not because it enforced consistency, but rather because it
>>>> makes code look good.  In fact, it could be argued that Python does _not_
>>>> enforce consistency; the programmer is free to choose any indentation for
>>>> each block, so long as each block can be parsed (which does require
>>>> internal consistency).

>>>"""
>>>I liked Python's
>>>case sensitivity not because it enforced consistency, but rather because it
>>>makes code look good. In fact, it could be argued that Python does _not_
>>>enforce consistency; the programmer is free to choose any case-spelling for
>>>each variable name, as long as each variable name can be parsed.
>>>"""

>> But now you've converted my paragraph from a true statement to a false
>> one.  The programmer is not in fact able to choose any case-spelling for
>> each variable name; they must use the same one which was originally used,
>> unless they wish to create or access another variable.

>Perhaps that's unclear, but my reading of 'for each variable name' is for
>each unique variable name -- I mean, a name can be referred to multiple
>times, but it's not a different name each time.

But again, this doesn't match with what I said about indentation.
Python's indentation is flexible because you can choose your own
indentation for each block, regardless of how other blocks are indented.
This can be very handy!

Python's case sensitivity is not flexible: once you write a variable name
one case, you're required to write it the same way.

I think I'm just stretching the metaphor, though.  I don't really think
there's enough similarity between variables and indentation to really use
effectively.

>>>Consistency in style happens to one of the most important factors that
>>>makes code look good (on the syntactic level). Your indentation looks
>>>better in Python because Python complains loudly if you don't indent properly.
>>>That's not the most important thing though; you can indent properly in
>>>any language. It's just that everybody indents properly in Python, and
>>>more or less in the same way.

>> Again, Python code looks good not because consistent indentation is
>> enforced, but rather because indentation alone is used to draw a tree
>> structure.

>But indentation is *always* used to draw a tree structure. Python *enforces*
>this tree structure, but it's not as if you can't draw tree structures
>in C.

"indentation alone" I said, and "alone" is critical.

>> Consider the following example of inconsistent indentation:

>> if something:
>>        do_it()
>> else:
>>  dont_do_it()

>> Python will happily accept this.  It'll even happily accept this for long
>> blocks, here it can be inconvenient.  This makes the code a bit harder to
>> read, but it'll still not bad -- because Python expresses a 2-dimensional
>> structure (the syntax tree) in a 2-dimensional form.

>> Most ugly code I've seen isn't ugly because it has poor indentation
>> (although I have seen some); most of it's ugly because of horrible naming,
>> or bad structuring, or ...

>Well, obviously indentation can't help in case of horrible naming and
>bad structuring. But you say indentation makes code look better, and that's
>not because it improves horrible naming or bad structuring but because
>it improves something else. So either you're saying you like Python's
>indentation because it doesn't have braces (and that's the only reason),
>or you're buying into the consistency argument after all. I believe I
>gave you the braces option before, speaking of dichotomies.

You did NOT give me the brace option before, as you can clearly see by
reading the first paragraph.  Now you give me it as though it were the
ONLY reason; clearly, by my own discussion, it's not the ONLY reason;
however, I will agree that it's a strong one.

>>>> More recently I've come to realise that the reason Python indentation
>>>> looks good is that Python, like most modern languages, is a
>>>> tree-structured language, and so is best represented in two dimensions.
>>>> Most modern languages, however, attempt to represent themselves in one
>>>> dimension: an ASCII stream which ignores lines and whitespace.

>>>Well, why shouldn't a programmer enforce such indentation in any
>>>language?

>> Of course they should.  Lisp programmers get really mad if people post
>> non-standard formatted code on c.l.lisp.  EMACS enforces it when writing,
>> and angry programmers enforce it when reading ;-).

>So in your case there's no reason to prefer Python's way of indenting,
>*except* that there are no delimiters. Okay, thanks for making that 
>clear. Or am I still missing something? :)

You got it right.  Not *exactly* right, since you still seem to insist
that I'm only allowed to like Python's indentation for exactly one reason.

>>>Indentation looks good in any language, not just Python, so
>>>that's not really an argument for Python's indentation style.

>> Well, there you go -- my point exactly.  I was saying the same for case
>> (which Python enforces much more consistently than it enforces
>> indentation).

>Yes, you were saying the same for case. Now, if I understand you right,
>you like Python's indentation strategy more than the delimiter strategy.
>The only two reasons I can come up with is:

>  * less redundancy because there are no delimiters
>  * less space taken by the delimiters

>(because the two-dimensional structure argument makes sense for any language
>if only you indent I'm not including that)

You'd best include it anyhow -- because it's the reason.  It does NOT make
sense for languages in which it's not, in fact, used.  There are not many
languages in which indentation is used.

>> In fact, the nice thing isn't that Python enforces indentation consistency
>> (because it doesn't really, any more than the minimum): the nice thing is
>> that Python doesn't throw anything more in your way than you really need
>> to parse the tree.

>Right, so that's the redundancy argument. Of course, if you used Lisp or C
>without any indentation, you'd have that same feature too. Why do you
>prefer the indentation approach to the delimiter approach? According to
>your opinions on case-spelling one would think you'd prefer the former, as
>it offers more freedom while the programmer still has the option just
>fine to enforce any indentation strategy they want, if they want it.

Excellent question, and this will hopefully clarify EXACTLY why you should
include reason #3 in your list of reasons why I like Python's indentation.

Almost all modern languages are tree-structured.  Therefore, they're most
efficiently written and displayed as a tree.  Trying to write them as a
delimited stream may be effective for computers (although it's still
difficult), but it's terrible for humans.

Therefore, we write even languages which use only delimiters as though
they were trees.

Python breaks away from that.

>>>Besides, if you ignore case you ignore that words in Python are spelled
>>>consistently in case, which is *also* a property of lots of modern languages.
>>>Sometimes not enforced, but still a desired property of the programmer.

>> But in case-sensitive languages, words can be spelled *differently* in
>> case: NO is not the same as No or no.  If the language were
>> case-insensitive this would be impossible.

>Hm, I'd say those would be different words, which vary only due to 
>case-spelling. Not that the same word can be spelled differently in 
>case.

Notice how hard it is to communicate that those are different words,
though?  Even in typed communication we're not consistent, and in spoken
communication we can't communicate that info at all.  Is it any wonder
that newbies get confused?

>Ignoring that point, spelling the same word in different case-spellings
>is only possible in case *insensitive* languages!
>So I'm not quite sure what you mean here?

Those words are pronounced and subvocalised the same -- unless, of course,
the programmer was boorish enough to use any two of them in the same
routine, in which case you'll probably have to read the routine again,
this time subvocalising differently.  With case-insensitivity, you won't
have to; your first reading was correct.

You know, for the first time it occurs to me that our learning styles may
differ -- I'm somewhat auditory, and from the way you talk, I'm guessing
you're primarily visual.

The obvious solution is to compromise so that both styles of perceptivity
are accomodated: the visual learners must be able to see things exactly
the same (or they don't recognise them as being the same), and the
auditory learners have to be able to vocalise and subvocalise things.

Both of these are solved not by language fiat, but rather by convention:
always use lower case.

>No, because I as a programmer happen to like the freedom different
>case-spellings happen to offer. It's up to the *programmer* to
>enforce the right conventions here. Throwing your arguments back at you
>again. :)

EXACTLY!  Now you're arguing 100% on my side -- only thing that's puzzling
me is that you imagine you're disagreeing with me.  Evey single thing
you're saying there is an argument ONLY for case-insensitive languages;
because only with them can you choose case-spellings with freedom.

And it's always up to the programmer to enforce conventions; I'm glad you
finally agree on that.

>>>> So I don't really care about languages enforcing consistency; if I need
>>>> it, I'll enforce it myself.

>>>A computer language would be worth less if it brought no consistency at
>>>all.

>> Granted.  But this doesn't seem to justify your slippery-slope treatment --
>> we don't need to enforce your amount of consistency just because no
>> consistency is bad.  We already enforce _some_.

>In the case-sensitive version we have now, we already enforce some, but
>not to the point the language becomes unwritable. It's also a simple
>rule, which helps fit it in people's minds. The case-insensitive rule
>would also be simple, but the code becomes less readable and not 
>significantly more writable (in my opinion). The case-enforcing rule
>makes code even more readable (possibly -- there are arguments about
>this by the people who like different case-spellings), but it becomes
>far less writable, and the rule has a bit of extra complexity.

Code with case-insensitivity becomes vastly more writable, and offers a
danger (but not a certainty) of becoming less readable.  The writer must
be instructed to prefer lower case -- but this is the same instruction you
have to give to writers in a case-sensitive language.

>> And I like polymorphism (which implies typechecking).

>I'm not quite sure what you mean by 'polymorphism' in this context. We
>seem to have polymorphism as far as I understand it already?

Sorry.  I meant parametric polymorphism, which does imply static
typechecking.

Of course, I also like to be able to do things the current way, with no
static typechecking.

>> And a lot of other
>> things, all of which I hope to see.  I'm curious to see whether other
>> people like them, and I'll defend them -- but unlike most of the people
>> arguing about case sensitivity, an ENORMOUSLY minor issue,

>Note that you're arguing about it as well. It's a minor issue with a lot
>of effects on readability. And it's easy to argue about. So that's why
>we do. :)

Oh, of course.  I'm not annoyed by arguing about anything; I'm annoyed by
the overstatement of the importance of case-sensitivity.

>> I won't
>> consider them prerequisites to using the new Python -- I'll use the new
>> Python if it feels like an improvement.

>Oh, I'll likely use the new Python if it's case-insensitive. But I'd
>prefer it would not be so, and I'm just trying to spread a meme here;
>the link with indentation. Your own meme-complex seems to reject it,
>so I'll stop trying with you. :)

It's not a terrible link, really; it just works the opposite of the way
you'd like it to.

>>>According to your argument, you will just enforce this consistency
>>>yourself. The problem is that, *even* if everybody enforces self
>>>consistency, people will enforce *different* consistency and their
>>>coding style will vary over time. From a maintenance and communication
>>>point of view, this is bad. 

>> And for case sensitivity it _just doesn't matter_.  The only people who
>> are strongly helped or hurt by case sensitivity appear to be newbies, and
>> they're being hurt.

>I'd like some more evidence about how the newbies are being hurt,

Guido cites some studies in CP4E, and the Alice people claim the same
(although I've not read the studies).

>but
>if we grant that, I still disagree that case-sensitivity only matters
>for newbies. How does it not matter for case sensitivity? If Foo writes
>a module and you use it in your module, but you use a different case
>spelling for common variables (but consistently!), and then I get to
>maintain it while I've been using another case spelling when I used
>that module all the time, wouldn't that hurt me?

Honestly, it wouldn't hurt me; I wouldn't even care for more than a couple
of seconds, as I decided which tradition I should follow (I'm good at
adapting my coding style to that of the code I'm maintaining).

>It's easy to say that it won't matter for communication and maintenance,
>but could you try backing that up?

It seems to be easy to say it'll hurt you -- do you mind backing that up?
How much experience do you have with large projects (or small ones) in
case insensitive languages?

>>>> Take a lesson from the people complaining
>>>> about Python's indentation: one of the leading false accusations is that
>>>> it's inflexible, that it won't allow people to choose thier own
>>>> indentation.

>>>I don't see how this maps to the case sensitivity argument at all. 
>>>Python's indentation *is* less flexible in what code-layouts are
>>>allowed, and it *is* a silly accusation.

>> It may or may not be silly -- but it's _definitely_ false.  Python is
>> _not_ inflexible; you can lay your code out in any way you want, so long
>> as it's in the same tree shape as its semantic structure implies.

>It's more inflexible. I can't do:
>  start()
>    do_a()
>    do_b()
>  commit()
>And occasionally that would be nice.

Granted.  But as I said, Python is still not "inflexible"; it's at worst
less flexible (as you accurately said).  Case sensitivity _is_ accused of
inflexibility, and in fact it IS inflexible.  Furthermore, there's not
even any linguistic reason for its existance; its only effect is to
increase the number of possible symbols, and no programmer wants to
maintain a program which actually uses case sensitivity to its full power.

And for that matter, this example is easy enough to "solve" using a blank
try-block:

try:
   start()
   do_a()
   do_b()
finally: commit()

>There are other examples, which I'm
>sure the complainers can come up with. Where the complainers are wrong
>(though it's their personal taste, but they should actually *try* it
>before they complain) is that they don't see the significant advantages
>which outweigh these minor disadvantages.

I agree here.

>> I don't think it's silly, either; I've complained enough about languages
>> which don't do this right, such as Make.

>Right -- Make is rather unwritable. There are no advantages there.
>I still don't see how these people relate to our present case;
>these people are complaining (without having actually tried it; anyone else
>complaining is of course in his or her right to complain) about decreased
>freedom you get with Python's indentation scheme.

They're not complaining, by and large, about decreased freedom; if they
were it would be a lot easier to ignore them (because they'd be correct).
They're complaining about total lack of freedom.  They're wrong because
Python has considerable freedom.

>Now you say we should cater to people like that, offering them more
>freedom (in case spelling), while nobody of *those* people appears to be
>complaining about that, and while the reason many people (not you) like
>Python's indentation *is* increased consistency and therefore increased
>readability.

Actually, I like the enforced consistency as well -- but I think it's far,
drastically, more important that there be consistency between how the
languages behaves and how it looks.  I really don't care if the programmer
is required to dot his i's, keep case consistent, or use the same
indentation for all 'if' clauses.  Those consistencies are _minor_
compared to the ones we really need -- things like good variable names,
clear syntactic structure, and so on.

>> There are two obviously terrible errors: inconsistency and inflexibility.
>> There's no reason to assume that avoiding one will force everyone into the
>> other, and I'm not making that mistake about you.

>I'm not quite sure I understand the language. Right now we avoid
>inconsistency and we also avoid inflexibility in the case-spelling
>domain.

Right now we are almost as inflexible as we could get.  We're not at a
level I would call "bad" (I would imagine recieving errors when using
alternate casings of words to be "bad").  On the other hand, loosening up
a bit isn't going to bring us anywhere near to the realm of anarchy some
Of yOu SeEm to PiCtuRe.

>There are minor disadvantages which apparently confuse newbies,
>so we ought to think about how to improve matters.

Good way of stating things.  Do you have any alternate ideas on this
issue?  I don't -- but that's because I'm reasonably satisfied with this
solution.

>But case-insensitivity gives up consistency and case-enforcing gives up
>flexibility; so it seems we shouldn't go anywhere. :)

A false consistency is the hobgoblin of little minds.

>Martijn

-- 
-William "Billy" Tanksley


From bvdpoel at uniserve.com  Fri Jun 16 23:26:31 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Fri, 16 Jun 2000 20:26:31 -0700
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com>
Message-ID: <394AEFE7.6F3B718C@uniserve.com>

Grant Edwards wrote:
> 
> I can't figure out how to create a Tk bitmap and then use it in
> a button.  According to the docs I can find, something like this
> ought to work:
> 
>     myBitmap = BitmapImage(name='myBitmap', data='[...]')
> 
>     Button(root,bitmap='myBitmap')
> 
> But, Tk complains that there is no bitmap named 'myBitmap'.  If
> I leave out the "name=" when I create the bitmap and then use
> the name retrieved with str(myBitmap), I get the same error.
> 
> I have managed to get it to work with bitmap data from a file:
> 
>     Button(root,bitmap='@filename')
> 
> But I don't want to have to write the data to a file just so Tk
> can read it back in.  Anybody have any examples of defining new
> bitmaps?
>

The data is an exact copy of the file. For example:

	bobpic = BitmapImage (data = """
	#define bob_width 107 
        #define bob_height 154
	static char bob_bits[] = {

0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,  
          0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff
< blessedly snipped >
	0xff,0xff,0xff}; """ )  

BTW, this applies to tcl/tk as well. Nice, isn't it, to always have
those clear docs to fall back on :).

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


From bvdpoel at uniserve.com  Mon Jun 19 20:18:11 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Mon, 19 Jun 2000 17:18:11 -0700
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net> <j8y35.1071$ul.67488@news1.rdc2.tx.home.com>
Message-ID: <394EB843.EEB3B68F@uniserve.com>

Yes, but I'm using __name to keep the variable private/static to a
particular instance of the Goption(). 

Rainer Deyke wrote:
> 
> Jeff Massung <jmassung at magpiesystems.com> wrote in message
> news:riu35.36$JM5.4708 at news.uswest.net...
> > I'm still learning, but lemme take a crack and let others correct me ;)
> >
> > __name isn't a member of MyDialog class, so you can't use self with it,
> > because then it looks for MyDialog.__name, which doesn't exist. But
> perhaps
> > you can use Goption.__name instead of self.__name?
> 
> This would indeed work if it wasn't for name mangling.  Don't call it __name
> if you want to access it from a nested class.
> 
> --
> Rainer Deyke (root at rainerdeyke.com)
> Shareware action/role-playing games      -      http://rainerdeyke.com
> "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor

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



From fgbrown at ionet.net  Wed Jun  7 19:56:19 2000
From: fgbrown at ionet.net (F. G. Brown)
Date: Wed, 7 Jun 2000 18:56:19 -0500
Subject: I need a lesson.
Message-ID: <8hmn80$4vu$1@ionews.ionet.net>

Greetings,
I just recently started learning to program using Python, and to be honest
when it comes to programming I'm almost dumb as a rock.  I've done some
Qbasic, but I consider it a dead language as far as the future is concerned
and wish to concentrate my efforts with Python.
Here's the rub.  The following is a module called howfast.py that I can't
get to repeat.  In Qbasic I would have used a goto at the end and it would
have worked fine.  Of course there is no goto in Python and that where I can
use anyone's help.  See if you can understand what I'm trying to accomplish
below, it's a little scrambled towards the end, and tell me the error of my
ways.  I have the book "Learning Python" but I'm still stuck.  If there are
any questions I'll be glad to reply.
Thanks,
Fred
?????????????????????????????????????????????

howfast.py

a = input("What is the distance you traveled in miles?  ")
b = input("How many seconds did it require?  ")
c = a*3600.00/b
print "You were going %d mph!" %c
if c > 75:
 print "Hey you better slow down!"
d = raw_input ("Would you like to do another? (y)es or (n)o")
if d == 'y':
 print 'Here we go!'
else:
 print 'See ya!'



From: sansonj72 at aol.com (Sansonj72)
Newsgroups: comp.lang.python
Subject: Help
Lines: 5
NNTP-Posting-Host: ladder07.news.aol.com
X-Admin: news at aol.com
Date: 07 Jun 2000 23:17:32 GMT
Organization: AOL http://www.aol.com
Message-ID: <20000607191732.11126.00000004 at ng-ci1.aol.com>
Path: news!uunet!ffx.uu.net!newsfeed.mathworks.com!portc01.blue.aol.com!audrey04.news.aol.com!not-for-mail
Xref: news comp.lang.python:98542
To: python-list at python.org
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 2.0beta3
Precedence: bulk
List-Id: General discussion list for the Python programming language <python-list.python.org>

I download python from http://www.python.org and I openned the file, and then
moved both files to c:\ and I run my learning fil called helloworld.py and I
get an error. It acts like the program is not there. Please give me a few tips.

james at lemode.com


From tdixon.no at spam.fwi.com  Fri Jun 16 13:47:22 2000
From: tdixon.no at spam.fwi.com (Tim Dixon)
Date: Fri, 16 Jun 2000 17:47:22 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <394a5109.80394991@10.1.1.28> <394A5795.F4DE98FB@san.rr.com> <394a5ed7.83928743@10.1.1.28> <8idoj0$lea$1@nntp.itservices.ubc.ca>
Message-ID: <394a680a.86283939@10.1.1.28>

On 16 Jun 2000 17:36:32 GMT, israel at math.ubc.ca (Robert Israel) wrote:

>In article <394a5ed7.83928743 at 10.1.1.28>,
>Tim Dixon <tdixon.no at spam.fwi.com> wrote:
>>On Fri, 16 Jun 2000 16:33:59 GMT, Courageous <jkraska1 at san.rr.com>
>>wrote:
>
>
>>>> Suppose, however, that you could get digit 'n' of pi directly (or e,
>>>> or any other inifinte, nonrepeating sequence).  All I have to do is
>>>> communicate which digit of pi the message key starts at, and I can
>>>> generate the rest of the key.
>>>
>>>This is a collosally bad idea, IMO.
>
>>Just for curiosity, why?
>
>Because the real key is the number of the digit that you start at, and
>that's pitifully small by cryptographic standards.  The time to generate
>the n'th digit is not much less than the time to generate all the digits
>up to the n'th, given that you have lots of memory available.  So the
>cryptanalyst with a computer somewhat bigger than yours could try all
>the keys you could be using.  
>

Right, but in the hypothetical example where I can get *any* digit of
pi, that could be any of an infinite number of starting points.
>Robert Israel                                israel at math.ubc.ca
>Department of Mathematics        http://www.math.ubc.ca/~israel 
>University of British Columbia            
>Vancouver, BC, Canada V6T 1Z2



From jens.arnfelt at wanadoo.dk  Thu Jun 29 09:17:10 2000
From: jens.arnfelt at wanadoo.dk (Jens Arnfelt)
Date: Thu, 29 Jun 2000 14:17:10 +0100
Subject: How to convert a string to hex?
References: <3952EC0D.D5CF9066@esec.com.au> <wdD45.7813$_55.169378@news2-win.server.ntlworld.com>
Message-ID: <395B4C56.19EFB075@wanadoo.dk>

Is there any easy way to go the other way, like convert hex number
"0x26100511FFFF" to a string with the the following value: "26100511FFFF"
?

richard_chamberlain wrote:

> Hi Sam,
>
> import string
> b = "6b584"
> i=string.atoi(b,16)
> print hex(i)
>
> 0x6b584
>
> Richard
>
> Sam Wun <swun at esec.com.au> wrote in message
> news:3952EC0D.D5CF9066 at esec.com.au...
> > Suppose here is:
> >
> > b = "6b584"
> >
> > I would like to b to hex. Does anyone know how to do that in python?
> >
> >
> > Thanks
> > Sam.
> >
> >



From petersc at stud.ntnu.no  Wed Jun 14 07:10:21 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 14 Jun 2000 13:10:21 +0200
Subject: tuple creation in C extensions
References: <l03102801b569016fbd23@[193.78.237.174]> <l03102800b56cf504d1d6@[193.78.237.168]>
Message-ID: <3947681D.54583FD5@stud.ntnu.no>

Just van Rossum wrote:
> 
> >Just van Rossum <just at letterror.com> writes:
> >
> >This is a shorter-to-write, slower-to-run version.  If you use
> >Py_BuildValue, the code has to parse the format string in order to
> >build the object.  If you are coding in C because you're trying to
> >speed up a speed-critical part of your program, the code you posted is
> >the better way.
> 
> On the other hand, creating tuples is probably not often something one does
> inside an inner loop...

On yet another hand it was for an IndexError patch for the
main Python distribution. Not that it really matters as the
patch was (for good reasons) not accepted. Just that speed 
was the main problem. ;)

thnx-for-the-help-though-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From dalke at acm.org  Thu Jun  8 00:26:20 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 7 Jun 2000 22:26:20 -0600
Subject: Nth digit of PI
References: <393C053A.58DD8220@home.com> <8hk9iq$502$1@news.wrc.xerox.com>
Message-ID: <8hn770$o6c$1@slb2.atl.mindspring.net>

Mark Jackson wrote:
>Well, if you're captured by an alien who asserts that s/he has
>calculated pi to a gazillion decimal places, and your survival depends
>on determining whether the claim is true or not, you can check a few
>digits at random and either be certain it's false or 99.99. . .% sure
>it's true.  Never know when *that's* going to come in handy; in fact
>IIRC this very plot was once used on McGyver.


I heard a related comment once.  If you are taken by aliens and want
to be able to prove the event, ask them for a list of the first 100
Mersenne primes.  This assumes they are willing and technologically
able, but if it happens, its pretty good proof.

Hmm, or you could be the inventor of a marvelous new primality
test method and want to play a grand hoax on the world - which is
more likely? :)

                       Andrew





From moshez at math.huji.ac.il  Fri Jun  9 03:57:07 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 10:57:07 +0300 (IDT)
Subject: wishlist Python3K: subclass ints
In-Reply-To: <8hpl5f$ila$1@slb2.atl.mindspring.net>
Message-ID: <Pine.GSO.4.10.10006091052200.21387-100000@sundial>

On 9 Jun 2000, Aahz Maruch wrote:

> Oh, I see.  Hmmmm....  That makes a lot of sense, yes.  I'm still having
> a little trouble wrapping my head around how this works with
> user-defined numeric types, but I'll let the people who are into
> metaclass programming deal with it.  ;-)

Oh, it doesn't really. I mean, the best you can do (in a type/class
dichotomy solved world), is inherit from "number", and change any of the
methods: but that wouldn't help you at all to implement constructive
reals the "right way".

on-the-other-hand-you-can-probably-use-this-to-implement-quaternions-ly
y'rs, Z.

(It's easy: simply define a quaternion as a pair (a, b) where a and b are
numbers, have coercion of quaternions defined by a --> (a, 0), and just
ask issubclass(number) about things (which quaternions are). OTOH, you'd
need interfaces and a better coercion architecture then what is in Python
right now to support this correctly)
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From donn at oz.net  Wed Jun  7 01:01:18 2000
From: donn at oz.net (Donn Cave)
Date: 7 Jun 2000 05:01:18 GMT
Subject: popen in real time?
References: <8hjdn4$6kf$1@bmerhc5e.ca.nortel.com>
Message-ID: <8hkkuu$n0p$1@216.39.151.169>

Quoth msoulier at nortelnetworks.com (Michael P. Soulier):
| 	Hey guys. I realize that you can loop on the output of popen if you want
| to read from a pipe, but it would seem from experience that the output of the
| child process is not available until the process exits. I've tried this with
| the output of long-running processes like find. Is there a way to get the
| output of a child in real time, as it's printed?

If it's like everything else, find's output is C library stdio,
like fputs() and so forth.  This is a buffered system, where the
real device I/O (like write()) happen when the buffer fills up.
That's where your output is, waiting in find's process memory,
and there isn't a heck of a lot you can do about it.  (Assuming
you can't modify find.)

There is a significant exception, though, inasmuch as when the
output device is a tty, stdio doesn't buffer output (or doesn't
block buffer it, anyway.)  UNIX provides a pipe-like device with
tty driver interface, called a pseudo-tty or pty, and if you're
really determined to get this to work, that's what you have to
work with.  There's a terrific utility for this called "Expect",
a Tcl extension, that knows how to deal with the pty devices of
many different platforms.  Python has a pty module in its standard
distribution, albeit able to deal with a much narrower range of
platforms, and there are some contributed modules that support
expect-like functions but I don't know much about them.

	Donn Cave, donn at oz.net


From millja at my-deja.com  Mon Jun 26 15:57:42 2000
From: millja at my-deja.com (millja at my-deja.com)
Date: Mon, 26 Jun 2000 19:57:42 GMT
Subject: Python COM does not seem to like me.
References: <8j0ddv$fqb$1@nnrp1.deja.com> <39541965.45DCE978@san.rr.com> <8j813b$if7$1@nnrp1.deja.com>
Message-ID: <8j8cja$ruj$1@nnrp1.deja.com>

Okay, Im sorry.  Its all working now.  I am not at all used to these
stupid Windows computers where you have to restart after you do
anything.  Linux is so much nicer, too bad I dont have a say in what we
use.  Thanks again.
jason

In article <8j813b$if7$1 at nnrp1.deja.com>,
  millja at wwc.edu wrote:
> Thanks!  I still cant get the import issue taken care of though.  I
> have changed the PYTHONPATH key in the registry (WinNT server 4.0) to
> include the dir in which the modual is contained in
> c:\InetPub\BRC\case13.py.  Then I include it using from case13 import
> *.  Is there something Im doing incorrectly?  Any ideas?  Thanks
again.
> jason
>
> In article <39541965.45DCE978 at san.rr.com>,
>   Courageous <jkraska1 at san.rr.com> wrote:
> >
> > > First off, is there something I dont know about the import
> statement?
> >
> > The python module must either be in your PATH or your PYTHONPATH
> > for python to be able to import the module, IIRC.
> >
> > > TypeError: cant multiply sequence with non-int
> > > The statment is:
> > > CM1 = roll[1] * roll[2]
> > > I am pretty much positive roll[1] and roll[2] are ints.
> >
> > You'd know for sure if on the line before you wrote:
> >
> > print `roll[1]` + " " + `roll[2]`
> >
> > I suspect you're going to be surprised by the results.
> >
> > C/
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From jkraska1 at san.rr.com  Thu Jun 15 02:18:14 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 15 Jun 2000 06:18:14 GMT
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie>
Message-ID: <394875B7.CAAA3C1F@san.rr.com>

 
> As someone who uses and likes both C++ and Python, I think estimates
> like that are excessively simplistic.

Me, too. And yet I will claim that I've become *more* productive
in Python after a brief 4-6 weeks of use than I am in C++ after
15 years of use, and that's no exaggeration. If Python executed
as fast as other languages, I'd be tempted to *require* that our
staff use it. And please don't take this wrong; I'm no language
groupie -- I use C, C++, Java, Python, and Lisp -- all of them both
lately and frequently.

> I use C++ for commercial programming, because it gives me the power and
> control I need.  Productivity isn't the best when measured in time to
> get the first prototype running, but I've found nothing better when
> measured in time to develop, deploy, maintain and support something a
> customer will pay for.

Shrug. If it weren't for execution speed, your C++ environment would
be toast. Admit it. :)-



C/


From dmoebius at gmx.net  Thu Jun  1 12:54:57 2000
From: dmoebius at gmx.net (Dirk Moebius)
Date: Thu, 01 Jun 2000 18:54:57 +0200
Subject: check for a running subprocess
Message-ID: <39369561.2633A7A9@gmx.net>

Hello!

I don't know whether that's a python or unix question, but:

How do I check whether a subprocess of the current process is still
running? Let's say, the current process is called 'dial' and the
subprocess is called 'fetchmail'. The subprocess is not necessarily a
_direct_ successor of the current process.


-Dirk.


From Jonathan_Epstein at nih.gov  Mon Jun 26 10:10:17 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Mon, 26 Jun 2000 10:10:17 -0400
Subject: determining an assigned IP address
References: <8j6lnu$jhd$1@nnrp1.deja.com>
Message-ID: <39576449.CFC3E4EC@nih.gov>

The following Perl code works:

  use Sys::Hostname;
  use Socket;

  $ip=gethostbyname(hostname());
  $theip=inet_ntoa($ip);


It should be relatively straightforward to convert this to Python.

- Jonathan


Bill Tolbert wrote:
> 
> I've just set up Linux with my cable modem and I'd like to use Python to
> help me launch remote X apps. I need to know the ip address assigned by
> my isp, but so far I'm striking out. gethostbyname_ex in the socket
> module returns the address I supplied when I thought it was static.
> Right now the only way I know to get the right address is to log into
> another machine and use who.
> 
> All help is appreciated!
> 
> Bill
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From yonik at my-deja.com  Tue Jun 27 17:38:59 2000
From: yonik at my-deja.com (yonik at my-deja.com)
Date: Tue, 27 Jun 2000 21:38:59 GMT
Subject: Microsoft's C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <8jb6th$j4$1@nnrp1.deja.com>

I skimmed over the C# (C sharp) spec... it addresses many of the
things that I didn't like about java (still leaves some though).
 - preprocessor!!! (I always want this for debugging code)
 - no stupid restrictions of 1 public class per file (bad for
   fast prototyping)
 - unified type system.. everyting is an object, even primitive types
 - addition of goto and unsafe... good for sparing use in high perf
   code.
 - foreach loop iterator
 - enums
 - structs (paired down class that can be stored on stack, etc)
   this one is *very* important.  It allows greater abstraction without
   a runtime cost.
 - properties&indexers (can make code execute on get/set of members)
 - versioning of classes
 - out parameters to methods
 - verbatim string literals (also alows multiline)
 - control of overflow checking
 - attributes (can attach to objects, classes, interfaces, etc)
What is missing that I would like to see:
 - templates (parameterized types)
 - operator overloading (so I can make classes on par with builtin ones)
 - Multiple inheritance for classes (they currently only do interfaces).
   - I don't like the separation of class/interface...  One may have to
     use classes by someone else that did not have the foresight to
     base the hierarchy on interfaces.  Without MI for classes, you are
     now stuck unless you can modify their source code.  Even Sun didn't
     have enough foresight with many of their classes and had to rewrite
     using interfaces later.
 - built in vector and map primitives (like python)

Overall, it is a better language than Java... I just don't know if it
will ever get the support (and good JIT compilers) that Java has now.
Everytime I use Java, I run up against artificial barriers that Sun put
up to "protect" users that they though was bad in C++... makes me
crazy.

I hope MS will standardize the runtime env also and do a better job on
the VM than Sun did.

Of course, if speed isn't critical, I just use Python :-)

-Yonik



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


From pecora at anvil.nrl.navy.mil  Thu Jun 15 07:13:10 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Thu, 15 Jun 2000 11:13:10 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil> <etdya48z0m4.fsf@w20-575-109.mit.edu>
Message-ID: <150620000713119061%pecora@anvil.nrl.navy.mil>

In article <etdya48z0m4.fsf at w20-575-109.mit.edu>, Alex
<cut_me_out at hotmail.com> wrote:

> I was thinking it ought to be possible to write a script that would pull
> the signatures of the API functions from the docs, and write some C code
> to determine their refence count effects.  You might be able to get a
> long way to the sort of table we need that way.

Alex,  I think I sense what you're suggesting here (checking argument
and return types, I think), but I don't know how to do that with a
script and even then, how could one tell what is being done inside the
function in terms of memory allocation or just pointers assigned to
existing memory, i.e. no ref. count increase ?


From liang at cad.gatech.edu  Thu Jun  8 17:41:21 2000
From: liang at cad.gatech.edu (Shengquan Liang)
Date: 8 Jun 2000 21:41:21 GMT
Subject: Bypass Sytem Output?
Message-ID: <8hp3u1$5dg$1@news-int.gatech.edu>

i 'm running a Python script that generates
an HTML file under Solaris 7.  The script also
generates a PDF file and puts a link to it. 

on thing i don't like is that Solaris gives 
a message like "saved /usr/local/apache/htdocs/PDF/1234602.pdf "
and printed on the generated HTML file. 

Q: how do i filter out this message?

one thing about generting the HTML file:

 i didn't use the HTML library from Python, instead,
 kinda cheated by printing all the tags "manually"
in the program.

Thanks .


From thomas at xs4all.net  Fri Jun  9 19:01:03 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Jun 2000 01:01:03 +0200
Subject: PATCH: Augmented assignment
In-Reply-To: <Bme05.1164$vp3.76577@news1.rdc2.tx.home.com>; from root@rainerdeyke.com on Fri, Jun 09, 2000 at 10:39:29PM +0000
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net> <8hrdkn$3uf@stochastic.eng.umd.edu> <Bme05.1164$vp3.76577@news1.rdc2.tx.home.com>
Message-ID: <20000610010103.G20700@xs4all.nl>

On Fri, Jun 09, 2000 at 10:39:29PM +0000, Rainer Deyke wrote:
> Warren B. Focke <moron at Glue.umd.edu> wrote in message
> news:8hrdkn$3uf at stochastic.eng.umd.edu...
> > But it does.  Accessing an attribute of an object or an item in a
> > container type can invoke arbitrary code, with arbitrarily freakish
> > side effects.  Evaluating x[17] might suck up hours of CPU time and
> > gibibytes of RAM, redefine builtin functions, launch the missiles, or
> > annoy your mother.  Having all of these things happen once, as in
> > x[17] += 1, is much preferable to, and potentially significantly
> > different from, having them happen twice, as in x[17] = x[17] + 1.
> > It's not just syntactic sugar that

> What if x is a database that loads an object from file on __getitem__ and
> writes it on __setitem__?  Then x[17] += 1 would work correctly only if
> x[17] has no += operation defined and the interpreter falls back on
> executing x[17] = x[17] + 1.

Well, I can't speak for other patches<0.7wink> but in the patch I wrote, and
in any patch that would make sense, 'x[17] += 1' still triggers a
__getitem__ *and* a __setitem__. It's probably easiest to see the
augmented-assignment syntax as a macro rewrite (which it isn't, because it's
a single opcode, which makes it threadsafe.) Looking at it like that,

x[spam(y, 10)-1] += 10 + eggs(10)

would not turn into:

x.__magical_getset_item__(spam(y, 10) - 1, 10 + eggs(10))

which you seem to think (and I thought, before Tim straightened me out), but
in the much simpeler:

index = spam(y, 10) - 1 # The exact order of evaluation of index and rhs
addval = 10 + eggs(10)  # elude me at this moment -- it's late.

x[index] = x[index].__add_ad__(addval)

The __add_ad__ method (or the C-API-variant of it) can decide for itself, on
a case to case basis, wether to return 'self' or a new object.

This way of looking at it makes it very simple, and to me, very natural. Of
course, it may take people a little getting used to, but I dont really see a
reason for it to remain as incomprehensible as, say, regular expressions or
the difference between __getattr__ and __setattr__. (I'll purposely ommit
''.join() in that list ;-)

It's-all-very-logical-in-bytecode-ly yr's,

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From hniksic at iskon.hr  Wed Jun 28 05:47:48 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 28 Jun 2000 11:47:48 +0200
Subject: Multithreading Embedded C/Python : Giving each thread its own sys.stdout
References: <w6265.9118$227.218565@nnrp1.uunet.ca>
Message-ID: <dpog4mp263.fsf@mraz.iskon.hr>

"Warren Postma" <embed at NOSPAM.geocities.com> writes:

> I am trying to set up an embedded Python as a multithreaded
> environment where each thread gets it's own namespace, and in each
> namespace, sys.stdout has been separately redirected.  I am very
> close. The problem is, it appears that the command "print" actually
> checks __main__.sys.stdout, not whatever namespace's sys.stdout you
> are currently in.
...
> It's like there can only be one sys.stdout instance, not one sys
> module per namespace, each with it's own sys.stdout object.  What
> can I do?

You can take a different strategy: instead of giving threads separate
namespaces with multiple hacked `sys.stdout' objects, you could have
only *one* hacked sys.stdout object that checks which thread it's in
and calls its `write' method.  For instance:

    class HackedStdout:
      def write(arg):
        thread_stdout_objs[threading.currentThread].write(arg)

    sys.stdout = HackedStdout

Now all you need to do at thread-creation time is set up the thread's
"stdout" object:

    thread_stdout_objs[newthread] = <code>

...where <code> is the incantation that you now use to create the
thread-specific sys.stdout object.

Hope this helps.


From thomas at xs4all.net  Fri Jun 16 08:58:37 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 16 Jun 2000 14:58:37 +0200
Subject: Python segfault
In-Reply-To: <20000616141903.A23985@foxboro.com>; from hjwidmai@foxboro.com on Fri, Jun 16, 2000 at 02:19:03PM +0200
References: <20000616141903.A23985@foxboro.com>
Message-ID: <20000616145836.O26436@xs4all.nl>

On Fri, Jun 16, 2000 at 02:19:03PM +0200, Hans-Joachim Widmaier wrote:
> Thomas Wouters wrote:

> > The problem is not the 'height' of this graph, or the total number of
> > 'nodes' (which is limited only by memory), but the 'width' of the first
> > tree. The 'node' struct is defined like this:
> > 
> > typedef struct _node { 
> > short n_type; 
> > char *n_str; 
> > short n_lineno; 
> > short n_nchildren; 
> > struct _node *n_child; 
> > } node; 
> > 
> > and the 'n_nchildren' member is the number of immediate subexpressions
> > (not grandchildren, only children) -- and it is limited to a short,
> > which is 16 bits on most systems. And the short is signed, which means
> > the maximum number of children is 32767. This includes the operators
> > itself, so you can operate on just 16384 items in a single expression.
> > If you try to operate on more, the short wraps, and the parser tries to
> > index node->n_child[-32768], which usually segfaults.

> Why not simply make 'n_nchildren' an (32 bit) int? And, while at it,
> 'n_lineno' also? It would cost some memory, though.

Why bother ? It would do nothing to solve the real problem, which is that
there is a finite number of direct children to a node. The current number is
practically unreachable -- the example case of ('+'.join(['2+2'] * 8193)) is
hardly reminiscent of real code.

Fixing the parser so it doesn't crash, and instead says 'add a few
parentheses' would be the better solution, IMHO ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From MarkH at ActiveState.com  Thu Jun 22 19:20:30 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Thu, 22 Jun 2000 23:20:30 GMT
Subject: Win32 Serial Port Setup
References: <01bfdc3f$b19a6800$6642c3d1@Cyberus>
Message-ID: <2bx45.1835$Sn2.5162@news-server.bigpond.net.au>

125 is definately missing some stuff you need.  Unfortunately, even the
latest builds are missing some of the stuff you need for setting up the
ports and getting asynch notifications.

The good news is that Active State agreed to fund this work, so the CVS
version of win32file (and a few new test files) provide everything you
need for complete, native, asynch port access on Windows.  The bad-news is
that a new build wont be out for a while.

Mark.

--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!
"Gordon Williams" <g_will at cyberus.ca> wrote in message
news:01bfdc3f$b19a6800$6642c3d1 at Cyberus...
> Hi,
>
> I would like to know how to set up the serial port using the win32
package
> by Hammond.  I see some stuff in the MS SDK on it but dont see the
> equivalent in the win32 package or docs.  I think that I am using build
125
> if that makes a difference.
>
> I know there is a Sio module but I would like to do it directly.
>
> A code snip would be useful as I am new to this.
>
> Thanks
>
> Gordon Williams




From mogul at primus.ca  Tue Jun  6 21:08:45 2000
From: mogul at primus.ca (Robin Porter)
Date: Tue, 6 Jun 2000 21:08:45 -0400
Subject: Example of Python code (C code to Python code)
Message-ID: <Ych%4.758$0l2.6362@news2.tor.primus.ca>

Could someone please give me an example of some Python code.  I am just new
to this language and would appreciate if you could re-write this simple
piece of C code in Python.

Thank you,

Robin


#include "stdafx.h"
#include <stdio.h>
#include <math.h>

float
HoursWorked,PayRate,Overtime,IncomeTaxes,GrossPay,NetPay,TaxPayable,Overtime
Hours;

const float TaxRate=.23;
const MinPay=300;
const RegHours=40;

main()
{

TaxPayable=0;
Overtime=0;
OvertimeHours=0;

 printf("\nPlease enter the hours worked per week? ");
 scanf("%f",&HoursWorked);

 printf("\nPlease enter the rate of pay? ");
 scanf("%f",&PayRate);

 if (HoursWorked>RegHours)
 {
  OvertimeHours =HoursWorked-RegHours;
  Overtime = (OvertimeHours)*(PayRate*1.5);
 }

 GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime;

 if (GrossPay>MinPay)
 {
  TaxPayable = (GrossPay-MinPay)*TaxRate;
 }

 NetPay = (GrossPay-TaxPayable);

 printf("\nThe gross pay is %f",GrossPay);
 printf("\nThe net pay is %f",NetPay);


 return 0;
}





From millja at my-deja.com  Tue Jun 13 23:27:36 2000
From: millja at my-deja.com (millja at my-deja.com)
Date: Wed, 14 Jun 2000 03:27:36 GMT
Subject: ASP, Python, VBScript and sharing like friends
References: <0a9f5150.911771a5@usw-ex0109-068.remarq.com>
Message-ID: <8i6u39$re3$1@nnrp2.deja.com>

Note: I'm reposting because those messed up
commas are ugly.

I am having a bit of an issue.  My current
situation is that I am able to use Python with
ASP and I am able to use VBScript with ASP.
However, I am still trying to figure out how to
use the two together.

What is occurring is that I am working on a
website, most of which is already written with
VBScript.  I have a rather large function that I
would like to integrate.  Is there some way to
call this function from within the VBScript?
 Or, it would work if I can get numbers,
generated by the VBScript, into the Python
function and the results of this function back to
the VBScript.

Thanks for you time.

jason milliron



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


From mwh21 at cam.ac.uk  Thu Jun  1 04:42:21 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 01 Jun 2000 09:42:21 +0100
Subject: starship status?
References: <m33dnfklsl.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <m3g0qxyeqa.fsf@atrus.jesus.cam.ac.uk>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Is starship now more or less functional?  I tried to post to
> crew at starship, but didn't get it back...
> 
> If it is mostly working, perhaps it's time to replace the current
> rather forbidding front page with something less likely to scare off
> the passer-by; some links to the rest of the starship would be a nice
> start!

Hello?

-- 
34. The string  is a stark data  structure and  everywhere it is 
    passed there is much duplication of process. It is a perfect 
    vehicle for hiding information.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html


From MarkH at ActiveState.com  Fri Jun 30 01:06:28 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 30 Jun 2000 05:06:28 GMT
Subject: Python COM Server Type Library
References: <8jfvvc$iql$1@nnrp1.deja.com>
Message-ID: <oVV65.1838$Tb7.13079@news-server.bigpond.net.au>

The answer, as you guessed, is "no".

However, an effort to support this keeps popping up - it even looks like
it may have resurfaced this very day.

This is usually discussed on the "Python COM" developers list -
http://mailman.pythonpros.com/mailman/listinfo/pycom-dev

Mark.
--
markh at activestate.com - but if you think I speak for them, you dont know
Dick!

"Charles Medcoff" <cmedcoff at pyrasol.com> wrote in message
news:8jfvvc$iql$1 at nnrp1.deja.com...
> I think I already know the answer to this but I'll ask it anyway...
>
> Is there anyway to generate a type library for a COM server object
> written in Python?  I know this has implications as far as what type of
> binding is used.  That being the case I suspect the answer is no.
>
> The reason that I ask is that this capability may make it easier to
> utilize Python COM objects from C++.  This is especially the case in
> Visual C++ where one can use #import.
>
> Any other sources of Python COM information would be helpful.
>
> --
> Charles Medcoff, MCP
> Senior Systems Engineer
> Pyramid Solutions
> 1850 Research Drive, Suite 300
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From neilh at scintilla.org  Wed Jun 21 01:31:31 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 21 Jun 2000 05:31:31 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <39504E63.691C86FB@san.rr.com>
Message-ID: <TqY35.4054$qz4.31321@news-server.bigpond.net.au>

> It would be nice if an editor had a vertical guide, much as some
> of the drawing programs do. You can click and drag the guide around,
> leaving a vertical line drawn across your editor, allowing you to
> move up and down in the file to see what matches that indentation
> level.

   I've been adding a similar feature to Scintilla/SciTE, motivated by this
thread. Rather than a generic draggable guide, there is one 'indentation
guide' displayed (as a light coloured, dotted line) for each indentation
level. When brace highlighting highlights a pair of braces on separate lines
then the indentation guide associated with those braces is highlighted. For
Python the ':' matches to the end of the statement. Well, most of the time
it does in the current implementation :(

   Neil




From pehansen at mail.danbbs.dk  Sat Jun  3 10:59:33 2000
From: pehansen at mail.danbbs.dk (Poul-Erik Hansen)
Date: Sat, 03 Jun 2000 16:59:33 +0200
Subject: Mathematica and python 1.5
Message-ID: <39391D55.2A8F1ED7@mail.danbbs.dk>

Hi

I need help in interfacing Mathematica 4 and python 1.5 via pyml. When I
try to compile the makefile for pyml I get the following error message
"arrayobject.h: No such file or directory". I have tried to play round
with the makefile but without any succes.

PEH


From jcm at bigskytel.com  Wed Jun 21 02:51:45 2000
From: jcm at bigskytel.com (David Porter)
Date: Wed, 21 Jun 2000 00:51:45 -0600
Subject: Roman Number Representation
In-Reply-To: <fHQ8yKAgg6T5EwnH@jessikat.demon.co.uk>; from robin@jessikat.demon.co.uk on Tue, Jun 20, 2000 at 06:21:36PM +0100
References: <S9Q1YMA7B5T5Ew1U@jessikat.demon.co.uk> <TyN35.7946$Q7.118988@news-east.usenetserver.com> <fHQ8yKAgg6T5EwnH@jessikat.demon.co.uk>
Message-ID: <20000621005145.A23166@bigskytel.com>

* Robin Becker <robin at jessikat.demon.co.uk>:

> >[ A UUEncoded file (roman.py) was included here. ]
> >
> Thanks. Any idea if it's a timbot production?

Nope: http://starship.python.net/crew/donp/
(Yes, Google knows more than I do)

I also just saw something similar here: http://hem.passagen.se/eff/bot.htm
 
  david



From loredo at spacenet.tn.cornell.edu  Tue Jun 27 16:02:07 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Tue, 27 Jun 2000 16:02:07 -0400
Subject: profiler -- numeric overflow?
References: <h579j8.grg.ln@kmself.nntp.ix.netcom.com>
Message-ID: <3959083F.C30D25BF@spacenet.tn.cornell.edu>

Karsten-

This doesn't really address your question, but if you're tired of
seeing all those 0s in your profiler output, I've copied below
a slightly modified version of pstats.py (loaded by the profiler)
that modifies the output to give times in seconds (1.111), 
milliseconds (1.111m) or microseconds (1.111u) as appropriate.
Note that if you capture the profiler output for some reason, this
might screw you up (e.g. if you try to read the strings in as numbers
to sort on).

The mods are the f9() function added at the end, and changes
in print_title() and print_line().

-Tom Loredo

~~~~~~~~~~ slightly hacked pstats.py ~~~~~~~~~~~~~
#
# Class for printing reports on profiled python code. rev 1.0  4/1/94
#
# Based on prior profile module by Sjoerd Mullender...
#   which was hacked somewhat by: Guido van Rossum
#
# see jprofile.doc and jprofile.py for more info.

# Copyright 1994, by InfoSeek Corporation, all rights reserved.
# Written by James Roskind
# 
# Permission to use, copy, modify, and distribute this Python software
# and its associated documentation for any purpose (subject to the
# restriction in the following sentence) without fee is hereby granted,
# provided that the above copyright notice appears in all copies, and
# that both that copyright notice and this permission notice appear in
# supporting documentation, and that the name of InfoSeek not be used in
# advertising or publicity pertaining to distribution of the software
# without specific, written prior permission.  This permission is
# explicitly restricted to the copying and modification of the software
# to remain in Python, compiled Python, or other languages (such as C)
# wherein the modified or derived code is exclusively imported into a
# Python module.
# 
# INFOSEEK CORPORATION DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS
# SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
# FITNESS. IN NO EVENT SHALL INFOSEEK CORPORATION BE LIABLE FOR ANY
# SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER
# RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF
# CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
# CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.


import os
import time
import string
import marshal
import re

#**************************************************************************
# Class Stats documentation
#**************************************************************************
# This class is used for creating reports from data generated by the
# Profile class.  It is a "friend" of that class, and imports data either
# by direct access to members of Profile class, or by reading in a dictionary
# that was emitted (via marshal) from the Profile class.
#
# The big change from the previous Profiler (in terms of raw functionality)
# is that an "add()" method has been provided to combine Stats from
# several distinct profile runs.  Both the constructor and the add()
# method now take arbitrarilly many file names as arguments.
#
# All the print methods now take an argument that indicats how many lines
# to print.  If the arg is a floating point number between 0 and 1.0, then
# it is taken as a decimal percentage of the availabel lines to be printed
# (e.g., .1 means print 10% of all available lines).  If it is an integer,
# it is taken to mean the number of lines of data that you wish to have
# printed.
#
# The sort_stats() method now processes some additionaly options (i.e., in
# addition to the old -1, 0, 1, or 2).  It takes an arbitrary number of quoted
# strings to select the sort order.  For example sort_stats('time', 'name')
# sorts on the major key of "internal function time", and on the minor
# key of 'the name of the function'.  Look at the two tables in sort_stats()
# and get_sort_arg_defs(self) for more examples.
#
# All methods now return "self",  so you can string together commands like:
#    Stats('foo', 'goo').strip_dirs().sort_stats('calls').\
#                               print_stats(5).print_callers(5)
# 
#**************************************************************************
import fpformat

class Stats:
	def __init__(self, *args):
		if not len(args):
			arg = None
		else:
			arg = args[0]
			args = args[1:]
		self.init(arg)
		apply(self.add, args).ignore()
			
	def init(self, arg):
		self.all_callees = None  # calc only if needed
		self.files = []
		self.fcn_list = None
		self.total_tt = 0
		self.total_calls = 0
		self.prim_calls = 0
		self.max_name_len = 0
		self.top_level = {}
		self.stats = {}
		self.sort_arg_dict = {}
		self.load_stats(arg)
		trouble = 1
		try:
			self.get_top_level_stats()
			trouble = 0
		finally:
			if trouble:
				print "Invalid timing data",
				if self.files: print self.files[-1],
				print


	def load_stats(self, arg):
		if not arg:  self.stats = {}
		elif type(arg) == type(""):
			f = open(arg, 'rb')
			self.stats = marshal.load(f)
			f.close()
			try:
				file_stats = os.stat(arg)
				arg = time.ctime(file_stats[8]) + "    " + arg
			except:  # in case this is not unix
				pass
			self.files = [ arg ]
		elif hasattr(arg, 'create_stats'):
			arg.create_stats()
			self.stats = arg.stats
			arg.stats = {}
		if not self.stats:
			raise TypeError,  "Cannot create or construct a " \
				  + `self.__class__` \
				  + " object from '" + `arg` + "'"
		return

	def get_top_level_stats(self):
		for func in self.stats.keys():
			cc, nc, tt, ct, callers = self.stats[func]
			self.total_calls = self.total_calls + nc
			self.prim_calls  = self.prim_calls  + cc
			self.total_tt    = self.total_tt    + tt
			if callers.has_key(("jprofile", 0, "profiler")):
				self.top_level[func] = None
			if len(func_std_string(func)) > self.max_name_len:
				self.max_name_len = len(func_std_string(func))
					
	def add(self, *arg_list):
		if not arg_list: return self
		if len(arg_list) > 1: apply(self.add, arg_list[1:])
		other = arg_list[0]
		if type(self) != type(other) or \
			  self.__class__ != other.__class__:
			other = Stats(other)
		self.files = self.files + other.files
		self.total_calls = self.total_calls + other.total_calls
		self.prim_calls = self.prim_calls + other.prim_calls
		self.total_tt = self.total_tt + other.total_tt
		for func in other.top_level.keys():
			self.top_level[func] = None

		if self.max_name_len < other.max_name_len:
			self.max_name_len = other.max_name_len

		self.fcn_list = None

		for func in other.stats.keys():
			if self.stats.has_key(func):
				old_func_stat = self.stats[func]
			else:
				old_func_stat = (0, 0, 0, 0, {},)
			self.stats[func] = add_func_stats(old_func_stat, \
				  other.stats[func])
		return self
			


	# list the tuple indicies and directions for sorting,
	# along with some printable description
	sort_arg_dict_default = {\
		  "calls"     : (((1,-1),              ), "call count"),\
		  "cumulative": (((3,-1),              ), "cumulative time"),\
		  "file"      : (((4, 1),              ), "file name"),\
		  "line"      : (((5, 1),              ), "line number"),\
		  "module"    : (((4, 1),              ), "file name"),\
		  "name"      : (((6, 1),              ), "function name"),\
		  "nfl"       : (((6, 1),(4, 1),(5, 1),), "name/file/line"), \
		  "pcalls"    : (((0,-1),              ), "call count"),\
		  "stdname"   : (((7, 1),              ), "standard name"),\
		  "time"      : (((2,-1),              ), "internal time"),\
		  }

	# Expand all abbreviations that are unique
	def get_sort_arg_defs(self):
		if not self.sort_arg_dict:
			self.sort_arg_dict = dict = {}
			std_list = dict.keys()
			bad_list = {}
			for word in self.sort_arg_dict_default.keys():
				fragment = word
				while fragment:
					if not fragment:
						break
					if dict.has_key(fragment):
						bad_list[fragment] = 0
						break
					dict[fragment] = self. \
						  sort_arg_dict_default[word]
					fragment = fragment[:-1]
			for word in bad_list.keys():
				del dict[word]
		return self.sort_arg_dict
			

	def sort_stats(self, *field):
		if not field:
			self.fcn_list = 0
			return self
		if len(field) == 1 and type(field[0]) == type(1):
			# Be compatible with old profiler
			field = [ {-1: "stdname", \
				  0:"calls", \
				  1:"time", \
				  2: "cumulative" }  [ field[0] ] ]

		sort_arg_defs = self.get_sort_arg_defs()
		sort_tuple = ()
		self.sort_type = ""
		connector = ""
		for word in field:
			sort_tuple = sort_tuple + sort_arg_defs[word][0]
			self.sort_type = self.sort_type + connector + \
				  sort_arg_defs[word][1]
			connector = ", "
					
		stats_list = []
		for func in self.stats.keys():
			cc, nc, tt, ct, callers = self.stats[func]
			stats_list.append((cc, nc, tt, ct) + func_split(func) \
				           + (func_std_string(func), func,)  )

		stats_list.sort(TupleComp(sort_tuple).compare)

		self.fcn_list = fcn_list = []
		for tuple in stats_list:
			fcn_list.append(tuple[-1])
		return self


	def reverse_order(self):
		if self.fcn_list: self.fcn_list.reverse()
		return self

	def strip_dirs(self):
		oldstats = self.stats
		self.stats = newstats = {}
		max_name_len = 0
		for func in oldstats.keys():
			cc, nc, tt, ct, callers = oldstats[func]
			newfunc = func_strip_path(func)
			if len(func_std_string(newfunc)) > max_name_len:
				max_name_len = len(func_std_string(newfunc))
			newcallers = {}
			for func2 in callers.keys():
				newcallers[func_strip_path(func2)] = \
					  callers[func2]

			if newstats.has_key(newfunc):
				newstats[newfunc] = add_func_stats( \
					  newstats[newfunc],\
					  (cc, nc, tt, ct, newcallers))
			else:
				newstats[newfunc] = (cc, nc, tt, ct, newcallers)
		old_top = self.top_level
		self.top_level = new_top = {}
		for func in old_top.keys():
			new_top[func_strip_path(func)] = None

		self.max_name_len = max_name_len

		self.fcn_list = None
		self.all_callees = None
		return self



	def calc_callees(self):
		if self.all_callees: return
		self.all_callees = all_callees = {}
		for func in self.stats.keys():
			if not all_callees.has_key(func):
				all_callees[func] = {}
			cc, nc, tt, ct, callers = self.stats[func]
			for func2 in callers.keys():
				if not all_callees.has_key(func2):
					all_callees[func2] = {}
				all_callees[func2][func]  = callers[func2]
		return

        #******************************************************************
	# The following functions support actual printing of reports
        #******************************************************************

	# Optional "amount" is either a line count, or a percentage of lines.

	def eval_print_amount(self, sel, list, msg):
		new_list = list
		if type(sel) == type(""):
			new_list = []
			for func in list:
				if re.search(sel, func_std_string(func)):
					new_list.append(func)
		else:
			count = len(list)
			if type(sel) == type(1.0) and 0.0 <= sel < 1.0:
				count = int (count * sel + .5)
				new_list = list[:count]
			elif type(sel) == type(1) and 0 <= sel < count:
				count = sel
				new_list = list[:count]
		if len(list) != len(new_list):
			msg = msg + "   List reduced from " + `len(list)` \
				  + " to " + `len(new_list)` + \
				  " due to restriction <" + `sel` + ">\n"
			
		return new_list, msg



	def get_print_list(self, sel_list):
		width = self.max_name_len
		if self.fcn_list:
			list = self.fcn_list[:]
			msg = "   Ordered by: " + self.sort_type + '\n'
		else:
			list = self.stats.keys()
			msg = "   Random listing order was used\n"

		for selection in sel_list:
			list,msg = self.eval_print_amount(selection, list, msg)

		count = len(list)

		if not list:
			return 0, list
		print msg
		if count < len(self.stats):
			width = 0
			for func in list:
				if  len(func_std_string(func)) > width:
					width = len(func_std_string(func))
		return width+2, list
		
	def print_stats(self, *amount):
		for filename in self.files:
			print filename
		if self.files: print
		indent = "        "
		for func in self.top_level.keys():
			print indent, func_get_function_name(func)
		
		print  indent, self.total_calls, "function calls",
		if self.total_calls != self.prim_calls:
			print "(" + `self.prim_calls`, "primitive calls)", 
		print "in", fpformat.fix(self.total_tt, 3), "CPU seconds"
		print
		width, list = self.get_print_list(amount)
		if list:
			self.print_title()
			for func in list:
				self.print_line(func)
			print 
			print
		return self

			
	def print_callees(self, *amount):
		width, list = self.get_print_list(amount)
		if list:
			self.calc_callees()

			self.print_call_heading(width, "called...")
			for func in list:
				if self.all_callees.has_key(func):
					self.print_call_line(width, \
						  func, self.all_callees[func])
				else:
					self.print_call_line(width, func, {})
			print
			print
		return self

	def print_callers(self, *amount):
		width, list = self.get_print_list(amount)
		if list:
			self.print_call_heading(width, "was called by...")
			for func in list:
				cc, nc, tt, ct, callers = self.stats[func]
				self.print_call_line(width, func, callers)
			print
			print
		return self

	def print_call_heading(self, name_size, column_title):
		print string.ljust("Function ", name_size) + column_title


	def print_call_line(self, name_size, source, call_dict):
		print string.ljust(func_std_string(source), name_size),
		if not call_dict:
			print "--"
			return
		clist = call_dict.keys()
		clist.sort()
		name_size = name_size + 1
		indent = ""
		for func in clist:
			name = func_std_string(func)
			print indent*name_size + name + '(' \
				  + `call_dict[func]`+')', \
				  f8(self.stats[func][3])
			indent = " "


# Modified to note use of f9().
	def print_title(self):
		print '   ***** m = milliseconds, u = microseconds *****'
		print string.rjust('ncalls', 9),
		print string.rjust('tottime', 9),
		print string.rjust('percall', 9),
		print string.rjust('cumtime', 9),
		print string.rjust('percall', 9),
		print 'filename:lineno(function)'


# Modified to use f9().
	def print_line(self, func):  # hack : should print percentages
		cc, nc, tt, ct, callers = self.stats[func]
		c = `nc`
		if nc != cc:
			c = c + '/' + `cc`
		print string.rjust(c, 9),
		print f9(tt),
		if nc == 0:
			print ' '*9,
		else:
			print f9(tt/nc),
		print f9(ct),
		if cc == 0:
			print ' '*9,
		else:
			print f9(ct/cc),
		print func_std_string(func)


	def ignore(self):
		pass # has no return value, so use at end of line :-)


#**************************************************************************
# class TupleComp Documentation
#**************************************************************************
# This class provides a generic function for comparing any two tuples.
# Each instance records a list of tuple-indicies (from most significant
# to least significant), and sort direction (ascending or decending) for
# each tuple-index.  The compare functions can then be used as the function
# argument to the system sort() function when a list of tuples need to be
# sorted in the instances order.
#**************************************************************************
class TupleComp:
	def __init__(self, comp_select_list):
		self.comp_select_list = comp_select_list

	def compare (self, left, right):
		for index, direction in self.comp_select_list:
			l = left[index]
			r = right[index]
			if l < r:
				return -direction
			if l > r:
				return direction
		return 0

		

#**************************************************************************

def func_strip_path(func_name):
	file, line, name = func_name
	return os.path.basename(file), line, name

def func_get_function_name(func):
	return func[2]

def func_std_string(func_name): # match what old profile produced
	file, line, name = func_name
	return file + ":" + `line` + "(" + name + ")"

def func_split(func_name):
	return func_name

#**************************************************************************
# The following functions combine statists for pairs functions.
# The bulk of the processing involves correctly handling "call" lists,
# such as callers and callees. 
#**************************************************************************

    # Add together all the stats for two profile entries
def add_func_stats(target, source):				
	cc, nc, tt, ct, callers = source
	t_cc, t_nc, t_tt, t_ct, t_callers = target
	return (cc+t_cc, nc+t_nc, tt+t_tt, ct+t_ct, \
		  add_callers(t_callers, callers))


    # Combine two caller lists in a single list.
def add_callers(target, source):
	new_callers = {}
	for func in target.keys():
		new_callers[func] = target[func]
	for func in source.keys():
		if new_callers.has_key(func):
			new_callers[func] = source[func] + new_callers[func]
		else:
			new_callers[func] = source[func]
	return new_callers

     # Sum the caller statistics to get total number of calls recieved
def count_calls(callers):
	nc = 0
	for func in callers.keys():
		nc = nc + callers[func]
	return nc

#**************************************************************************
# The following functions support printing of reports
#**************************************************************************

def f8(x):
	return string.rjust(fpformat.fix(x, 3), 8)

# Added to handle short times.
def f9(x):
	if (x >= 0.1):
		return string.rjust(fpformat.fix(x, 3), 8) + ' '
	elif (x >= 0.001):
		return string.rjust(fpformat.fix(1.e3*x, 3), 8) + 'm'
	else:
		return string.rjust(fpformat.fix(1.e6*x, 3), 8) + 'u'


From danunder at columbus.rr.com  Sat Jun 17 09:42:36 2000
From: danunder at columbus.rr.com (Dan Underwood)
Date: Sat, 17 Jun 2000 13:42:36 GMT
Subject: Pythonwin: editor nits
References: <39439B04.76F15F39@columbus.rr.com> <bWB25.700$qz4.6064@news-server.bigpond.net.au>
Message-ID: <394B8131.87D3573D@columbus.rr.com>

Mark,
Thanks, turning off Smart Tabs makes it do what I want... 
rather obvious now that I know it!
I find Pythonwin to be an excellent piece of work!
Dan.

Mark Hammond wrote:
> 
> Check out the documentation in the Pythonwin help file about the tabs (in
> the overviews section).  If you can't find it, then update your build!
> 
> Basically, "Smart Tabs" is designed to work the way you are seeing it
> work - it uses the first indent in the source file to set the options for
> the file.  If you want your options to be in effect _always_, turn
> smart-tabs off.
> 
> This is identical to the way IDLE works - we share the same auto-indent
> code (courtesy of the timbot)
> 
> Mark.
> 
> "Dan Underwood" <danunder at columbus.rr.com> wrote in message
> news:39439B04.76F15F39 at columbus.rr.com...
> > I'm using build 129 on Win98SE. I find that the editor does not
> > (usually) remember my tab and indent settings. Nor does it remember that
> > I set "Use Tabs" on and "Show Whitespace" on. Actually, the options
> > dialog shows these settings as remembered but the editor seems to ignore
> > them for existing files. I do not want to deal with questions about why
> > I change tab and indent ;-) but I would like to know if I am just not
> > setting something right or are these just some bugs to be eventually
> > fixed. If a later build has already fixed them, "never mind".
> > Thanks.
> > Dan.


From thomas at cintra.no  Tue Jun 20 09:34:02 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 20 Jun 2000 13:34:02 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <8inq3u$p0i$1@newshost.accu.uu.nl>
Message-ID: <39586f3d.366407626@news.online.no>

>> Linux with Windows  2000/ NT 4.0.
>
>Why switch from Linux to Windows NT/2000, though? ;)

Well, first off it`s my OS of choice, but mostly it *was* because it
costs money. I`ve realized later that the cost of an os is scraps in a
huge project so that`s not a very important issue anymore. But I would
still go for Linux. It has remote-administration out-of-the-box with
telnet/ssh, ready-and-running web-server ( Apache ) , dns, ftp - well,
do I need to go on. AND PYTHON by default.

Ok, a little off topic, but Linux is probably THE development platform
at the moment, and have been for some time. If you look at
distributions like Mandrake Linux 7.1 you`ll see that developers get
alot of free and great stuff. When I had installed for about a week
ago I had MySQL, PostgreSQL, PHP, Apache, Python, Perl, C/C++ and so
on up and running in less than 2 hours. It at the cost of a blank cd.
And that`s including documentation. 

Oh, the big-ol` Raving-Linux-Fanatic took over, sorry.

Thanks for your input anyway.

Thomas



From jacobs at darwin.epbi.cwru.edu  Sun Jun 11 19:31:19 2000
From: jacobs at darwin.epbi.cwru.edu (Kevin Jacobs)
Date: 11 Jun 2000 23:31:19 GMT
Subject: Mega-dir() or my cool snippet of the day
Message-ID: <8i17g7$392$1@eeyore.INS.CWRU.Edu>

Hi all,

  I'm usually pretty quiet around here, but I feel like making a little
noise.  Here is my cool snippet of the day, to enhance the behavior of the
somewhat anemic dir builtin.  Very useful when used in conjunction with
rlcompleter.py.  e.g.:

  import readline, rlcompleter, megadir
  rlcompleter.dir = megadir.megadir
  readline.parse_and_bind("tab: complete")

megadir.py:
-------------------------------------------------------
# Not the most efficient implementation, but not bad for
# a 10 minute hack.

import types

def unique(seq):
  out = []
  if len(seq):
    out.append(seq[0])
  for i in range(1,len(seq)):
    if seq[i] != seq[i-1]:
      out.append(seq[i])
  return out

def megadir(obj = globals()):
  ldir = []
  for b in getattr(obj, '__bases__', []):
    ldir.extend(megadir(b))
  if type(obj) == types.InstanceType:
    ldir.extend( megadir(obj.__class__) )
  ldir.extend(dir(obj))
  ldir.sort()
  return unique(ldir)

def test():
  class A:
    def a(): pass
  class B:
    def b(): pass
  class C(A):
    def c(): pass
  class D(B,C):
    def d(): pass
    def a(): pass

  d = D()
  print 'dir =',dir(d)
  print 'megadir =',megadir(d)

if __name__ == '__main__':
  test()


-- 
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8487  <--------
S.A.G.E. Project Technical Coordinator  | Department of Epidemiology
  & System Administrator                |   & Biostatistics
Internet E-mail: jacobs at darwin.cwru.edu | Case Western Reserve University
----------------------------------------------------------------------------


From dworkin at ccs.neu.edu  Wed Jun 28 18:03:46 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 28 Jun 2000 18:03:46 -0400
Subject: (Serious?) package namespace problem (and a proposal)
In-Reply-To: hzhu@knowledgetrack.com's message of "Wed, 28 Jun 2000 20:25:23 GMT"
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>
Message-ID: <vndwvj9bgzh.fsf@camelot-new.ccs.neu.edu>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> from AAA import *
> from BBB.AAA import * 

[various convoluted issues]

"Doctor, it hurts when I do this."

"Don't do that, then."

I'd be quite happy if the 'from' version of import was just not
available at all, except perhaps in interactive mode.

Import your modules directly and you don't have to deal with this kind 
of problem.  As a side effect, your code is usually clearer.

This is a good general rule.  Just don't use 'from' in programs or
modules unless you have a really good reason to.

-Justin

 



From olczyk at interaccess.com  Wed Jun  7 09:44:12 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 07 Jun 2000 13:44:12 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com>
Message-ID: <393f5114.72916406@nntp.interaccess.com>

On Mon, 05 Jun 2000 15:55:08 GMT, "Tom" <tom-main at REMOVEME.home.com>
wrote:

>
>I see that Mark Hammond has gone to ActiveState.
>
>I'm interested to hear that they will be integrating Python into Visual
>Studio 7, but what does this all mean for the future of the PythonWin?
>
>Will PythonWin (the IDE and the MFC wrapper library) continue to be actively
>and publicly developed?
>
>Tom Malcolmson.
>

Ms has said this. Very interesting. Didn't MS pay ActiveState to
develop ActivePerl? Why aren't they integrating Perl too?

Could be good are bad news. Technologies tend to get perverted when
Microsoft starts integrating them into their products. Isn't
there some sort of lawsuit about that ;)

OTOH maybe Microsoft is learning somethings.


From thiele at muc.das-werk.de  Tue Jun 13 08:14:27 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 13 Jun 2000 14:14:27 +0200
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE>
Message-ID: <394625A3.4489C696@muc.das-werk.de>

>

NO!

I think the more complex the program is the more C++ is the better choice.

It is very easy and fast to write short effective programs with python.
But in larger programs a lot of time is spend for developing and testing the
algorithm.

And I think using C++ is faster because the compilation time is relativly small
compared with the
whole time for program developing. And the the additional time to program the
nice
pythonfeatures (like dictionaries, lists etc.) in C++ is also very small. In
large programs
 the disadvatages of python (like rumtime errors instead of compiler errors,
typemismatch,
functions defined twice...) dominates.



From trentm at activestate.com  Mon Jun 26 22:45:37 2000
From: trentm at activestate.com (Trent Mick)
Date: Mon, 26 Jun 2000 19:45:37 -0700
Subject: learning-python question
In-Reply-To: <20000626170753.C2492@activestate.com>
References: <8j8ep3$tkn$1@nnrp1.deja.com> <20000626170753.C2492@activestate.com>
Message-ID: <20000626194537.A3557@activestate.com>

On Mon, Jun 26, 2000 at 05:07:53PM -0700, Trent Mick wrote:
> This is because the global
> namespace is not searched when in a Python function.


Whoa. I was just plain wrong here. Sorry about that. Just pay attention to
Aahz Maruch and Walter Hofmann, who responded much more intelligently to your
question.

Using the 'global' keyword *is* the correct answer though. :)


shamedly yours,
Trent

-- 
Trent Mick
trentm at activestate.com



From ullrich at math.okstate.edu  Sat Jun 10 10:52:42 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: 10 Jun 2000 14:52:42 GMT
Subject: spaces in filenames
References: <000001bfd253$ce6f9180$5b2d153f@tim> <slrn8k2o30.3r4.nascheme@cranky.arctrix.com>
Message-ID: <01bfd2ed$42b67880$29cf8ad1@daves-dell>


Neil Schemenauer <nascheme at enme.ucalgary.ca> wrote in article
<slrn8k2o30.3r4.nascheme at cranky.arctrix.com>...
> Tim Peters <tim_one at email.msn.com> wrote:
> >I strongly recommend one deviation from the defaults, though:
> >when it asks for the Python installation directory, force it to
> >"c:\python15" (where "c:\" is the root of his boot drive, almost
> >certainly exactly "c:\"). The default is some monster with
> >embedded spaces in its name, and that just makes life harder.
> 
> This is something that pissed me off about Windows 95.  Microsoft
> just had to show off their new feature allowing spaces in
> filenames by putting a space in one of the most important
> directory names on the system.  Unix has allowed weird characters
> in filenames for ages.  Just because you can do it doesn't mean
> its a good idea.  I wonder how much time has been wasted
> worldwide because Microsoft did not call "Program Files"
> "Programs".

          My opinion is they did it on purpose. A program is required to
be able to handle funny filenames to qualify as Windows 95 
compatible blah blah, so this is like a test.

          Pisses me off too. Long filenames great, but allowing
spaces is just silly, when spaces have been command-line
delimiters all along. (Wasted a lot of time once finding out
why a certain TeX file didn't work - turns out there's a reason
MikTeX doesn't install itself under Program Files...)

>     Neil
> 
> -- 
> "Reliability means never having to say you're sorry." -- D. J. Bernstein
> 


From erno at iki.fi  Wed Jun 28 16:45:04 2000
From: erno at iki.fi (Erno Kuusela)
Date: 28 Jun 2000 23:45:04 +0300
Subject: postgreSQL successes
References: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com> <395A5033.DF103F0C@estinc.com> <ku66qtv945.fsf@lasipalatsi.fi>
Message-ID: <kuzoo5tu0f.fsf@lasipalatsi.fi>

>>>>> "Erno" == Erno Kuusela <erno at iki.fi> writes:

    Erno> this calls the flush() system call.

doh, fsync(2) that is.

  -- erno


From root at rainerdeyke.com  Wed Jun  7 14:44:47 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 07 Jun 2000 18:44:47 GMT
Subject: Embedding + extending without DLLs
References: <wkbt1da7w2.fsf@turangalila.harmonixmusic.com>
Message-ID: <zKw%4.8444$r7.1747658@news1.rdc2.tx.home.com>

Dan Schmidt <dfan at harmonixmusic.com> wrote in message
news:wkbt1da7w2.fsf at turangalila.harmonixmusic.com...
> I would like to simultaneously embed and extend Python in a Win32
> project.

See: Demo\embed\demo.c.


--
Rainer Deyke (root at rainerdeyke.com)
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From arta at NOSPAMx-stream.nl  Tue Jun 13 11:57:14 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Tue, 13 Jun 2000 17:57:14 +0200
Subject: Client with select.select()
References: <#F2#YuI1$GA.106@net025s> <8i36rt$42mq$1@nntp6.u.washington.edu>
Message-ID: <emt7cXW1$GA.180@net025s>

> I think you also may have the wrong idea about the write-set parameter
> to select().  The test with select() is ``will I block?''  That is,
> if read-set returns a file descriptor, you may read from that descriptor
> without blocking - either there's data, or end of file or whatever may
> cause a non-blocking state on a blocking descriptor.  The same is true
> of the write set, mutatis mutandi.  The reader on the other end will
> probably ignore what you're writing, you just won't block.  Sockets
> are usually in this state.  I guess it's a good thing to check anyway,
> but by no means is there any implication that the other end of the socket
> has requested data.  ("block" means "wait".)

But if you want to write a telnet-client. The server requests for data,
isn't he?
Then you can check with the write-set and send some data.
Or am I missing something?


A[r]TA

--
-Life can kick your ass, but all you got to do is stand up and move on...






From rhays at interaccess.com  Sat Jun 17 09:12:03 2000
From: rhays at interaccess.com (Bob Hays, Computer Geek)
Date: Sat, 17 Jun 2000 08:12:03 -0500
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com>

On Sat, 17 Jun 2000 07:49:43 GMT, olczyk at interaccess.com (Thaddeus L.
Olczyk) wrote:

The following are just my opinions to take or leave (probably the
latter) as you wish.

>1) The first is rather stylistic and may not appeal to most, but I
>have often found it useful. When debugging code I often put in
>code to help me figure out whats happening. I also avoid indenting
>such code so that it easy to remember where it is later when I want to
>take it out. An example ( assuming I use begin and end to denote
>blocks ):

I've gotten in the habit of putting the comment DEBUG on such lines in
all programming languages; depending upon indention is very dangerous
for debug lines since someone else may reindent the line properly and
then it gets left in (since they didn't know it was a debug line).

>2) You can easily insert a tab introducing a bug.  Consider the
>following code ( similar to a bug I recently fixed ):
{Example skipped}
>Initially the line in error was indented correctly, but someplace
>along the line the extra tab was introduced. A lot of effort then had
>to go in to finding this bug. With begin...end or braces this is less
>likely to happen.

Yep, this happens.  Its a tradeoff - speed vs. occasional extra work.

In addition, I had an example, in C++, where someone had commented out
a block of code, but then had an embedded inline comment inside that
block, and, to top it all off, the less-than-desired commented out
block would still compile - the guy worked on that bug for a day
before coming to me.  I loaded it into emacs and the color coding told
him immediately what had happened; he was very ashamed and angry, but
I told him sh*t happens:-).

>3) Finally. Python is a language where it is almost easy to write
>extensible programs. To illistrate what I mean consider this example
>of a python program I recently wrote:

Here's were we will agree to disagree, I think.  I think its way
faster and easier to extend Python stuff than C++ (and Java) due to
the lack of compile/link cycle and due to inheritance inside of
Python.  Yep, you have to be careful how you indent, but if you have a
good editor it will cut/copy/paste as-is without reindenting.  I see
more problems due to bad cut/copy/paste reuse in all programming
languages than I do from mis-indenting so far.

My interesting tale is a consultant I have right now working on some
production Python code.  He started making changes using vi and
realized that we had carefully used spaces for indention (I had that
holy war in a previous life [at a shrink wrap vendor that had to run
on almost anything], and we finally agreed that tabs were evil in a
cross-platform environment).  He started using spaces and now isn't
having problems.

>Anyway I hope this causes TPTB to think about how blocks should be
>handled in Pyhton 3000.

My personal moral - no programming language will save you from
yourself - use good tools and do your best.

Have fun! - Bob

--
Bob Hays                           "Discipline is never an end in
181 W. Madison                      itself, only a means to an end."
MS 4000                            -- King Crimson
Chicago, IL  60602                 "Whoever flees history will be
(312) 904-4668 (V)                  pursued by history."
(312) 904-5774 (F)                 -- Janusz Korczak
Bob.Hays at abnamro.com               "Faster! Faster! Until the thrill
rhays at interaccess.com               of speed overcomes the fear of
                                    death."
                                   -- Hunter S. Thompson
                                   "Security is very frequently the
                                    commencement of calamity." 
                                   -- Paterculus
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS d+(-)@ s+:+ a C++(+++)$ U$ P+$ L@ E+++$ W+++$ N++ !o-- K--? w$ !O 
M$ V@ PS+ PE !Y+ PGP t !5-- !X !R tv+ b++() DI+ !D- G-- e++ h--- r+++ 
y+++** 
------END GEEK CODE BLOCK------


From PClaerhout at CREO.BE  Thu Jun 22 08:27:42 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 14:27:42 +0200
Subject: Versionstamp script included with PythonWin
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301215@msgeuro1.creo.be>

> Hello all,
> 
> I found a script called verstamp.py under python\win32\scripts\VersionStamp,
> which looks like a part of the win32 package from Mark Hammond. This script
> makes it possible to give an exe or a dll a version stamp. Does anyone knows
> if this works for reading the version stamp or did someone already made a
> module for that?
> 
> Kind regards,
> 
> 
> Pieter



From al at servana.com  Tue Jun 13 11:21:17 2000
From: al at servana.com (al at servana.com)
Date: Tue, 13 Jun 2000 15:21:17 -0000
Subject: How do you FTP ASCII if ftplib strips CR?
Message-ID: <8i5jhd+h2lu@eGroups.com>

If you look at the FTPLIB docs, it states that RETRLINES will strip 
CR when passing the text to the RETRLINES callback.

So, how do you put the CR back in if you want them?

When you FTP ascii from a non-similar platform, the loss of the CR 
causes the text to FTP as 'one line'. (like UNIX-to-UNIX and NT-to-NT 
work, UNIX-to-NT does not)

Yet ALL FTP tools do not do this, so only the Python scripts tend to 
munge ASCII files.

I was using the following notation, but cannot come up with another 
where I can control the append of a CR:

ourFTP.retrlines('RETR ' + filename, open(os.path.join(localdir, 
filename), 'w').write)

Any help would be appreciated.
We have two people on this for days and cannot come to a resolution 
to this apparently simple problem.

Thanks.




From djc at itga.com.au  Tue Jun  6 20:08:11 2000
From: djc at itga.com.au (Dave Cole)
Date: 07 Jun 2000 10:08:11 +1000
Subject: SV: How to execute python scripts under Apache?
References: <7BD10B680501D411B9DF009027E06F32011AAB@EXCHANGE>
Message-ID: <87itvms68k.fsf@heresy.itga.com.au>

>>>>> "Max" == Max M?ller Rasmussen <maxm at normik.dk> writes:

Max> Does anybody has some good technical reasons to go with Python
Max> instead of PHP?

I have been using Python and HTMLgen here for about a year.  I am very
happy with the results.  The excellent object model in Python really
shines for this type of work.

I recently bought the "Professional PHP Programming" book to see what
PHP was all about.  I have read all of the book and I am currently
browsing through the case studies.  PHP has quite a bit of implicit
functionality which makes it quite powerful.  I get the impression
that for small projects, you would get off the ground faster with PHP.
The book does not really talk much about the PHP object model, so it
is hard to see if there are any equivalents for the various Python
__getattr__ (for instance) tricks.  There was also no discussion of
extension modules.

It is not clear from the book how well PHP would scale for larger
projects.  My gut feeling is that it would not scale as well as
Python.

On a non-technical level, the language is no where near as pretty as
Python.  The fast template system is as ugly as a hat full of
arseholes (as a friend of mine would say).

- Dave


From emile at fenx.com  Thu Jun  1 21:17:27 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 1 Jun 2000 18:17:27 -0700
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and others" is this possible?
References: <393692FB.8610EBCA@dallas.net> <007901bfcbf2$c3b4e680$1906a8c0@fc.fenx.com> <3936FEFB.BA248CA3@sightreader.com>
Message-ID: <003301bfcc30$57851820$1906a8c0@fc.fenx.com>

Ken,

Are you trying this in IDLE?  I get the same error there.  
If so, try starting python from within a dos window and
see if that makes a difference.  Pipes in winxx are somewhat 
flakey as I understand.

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message ----- 
From: Ken Seehof 
To: Emile van Sebille 
Cc: python-list at python.org 
Sent: Thursday, June 01, 2000 5:25 PM
Subject: Re: Python to call commands, e.g. "find , perl scripts , nmap , and others" is this possible?


On NT I get: 
>>> mylist = os.popen('dir').readlines() 
Traceback (innermost last): 
  File "<interactive input>", line 1, in ? 
OSError: (0, 'Error') 

(same thing with any command) 

popen (command[, mode[, bufsize]]) 
    Open a pipe to or from command. The return value is an open 
    file object connected to the pipe, which can be read or written 
    depending on whether mode is 'r' (default) or 'w'. The bufsize 
    argument has the same meaning as the corresponding argument 
    to the built-in open() function. The exit status of the command 
    (encoded in the format specified for wait()) is available as the 
    return value of the close() method of the file object, except that 
    when the exit status is zero (termination without errors), None is 
    returned. Availability: Unix, Windows. 

- Ken Seehof 

Emile van Sebille wrote: 

  Jim, Take a look at os.popen, as in: import osmylist = os.popen('locate myname').readlines()for eachfile in mylist: do something with eachfile HTH,  
  Emile van Sebille 
  emile at fenx.com 
  -------------------   
  ----- Original Message ----- 
  From: Jim
  To: python-list at python.orgSent: Thursday, June 01, 2000 9:44 AMSubject: Python to call commands, e.g. "find , perl scripts , nmap , and others" is this possible?
    
  Dear list, 

  I'm new to Python, and I'm sorry if I'm asking about anything that has been 
  already discussed. 

  I would like to use Python to call commands, e.g. "find , perl scripts  , nmap , and others" is this possible? 

  I do not need GUI. All I need is to link the out put and work with it. Is perl or shell scripts my only 
  answer? 

  jim 

-- 
Proprietary & Confidential.  The information transmitted is intended only for the person or entity to whom it is addressed and may
contain confidential and/or privileged material.  Any review, retransmission, dissemination, or other use of, or taking of any
action in reliance upon, this information by persons or entities other than the intended recipient is prohibited.  If you received this in error, please
contact the sender and delete the material from any computer.
   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000601/b2d8d4e6/attachment.html>

From jerome.quelin at insalien.org  Tue Jun 13 08:01:20 2000
From: jerome.quelin at insalien.org (Jerome Quelin)
Date: Tue, 13 Jun 2000 14:01:20 +0200
Subject: How many times inner functions are compiled ?
Message-ID: <960897680.1890969865@news.libertysurf.fr>

Hi all,

def outer():
    def inner():
        print "spam"
    inner()
outer()
outer()

In the snippet code above, I call outer() two times. I would like to
know how many times python will compile the inner function ? One time
when compiling the script/module, or two times, ie at every invocation of the
outer function ?

Thanks,
Jerome
--
jerome.quelin at insalien.org



From MSteed at altiris.com  Fri Jun  9 13:39:38 2000
From: MSteed at altiris.com (Mike Steed)
Date: Fri, 9 Jun 2000 11:39:38 -0600 
Subject: Python Productivity over C++
Message-ID: <65118AEEFF5AD3118E8300508B1248774CB334@ALTNET>

> From: wlav at atlas03.cern.ch [mailto:wlav at atlas03.cern.ch]
> Sent: Friday, June 09, 2000 11:11 AM
> To: python-list at python.org
> Subject: Re: Python Productivity over C++
> 
> 
> Mike Steed <MSteed at altiris.com> writes:
> >> From: nbecker at fred.net [mailto:nbecker at fred.net]
> >> 
> >> I rarely need to do this for my classes.  Using STL heavily is a big
> >> help.
> 
> >This is a surprising thing to say.  STL requires that the types used to
> >instantiate its various templates be well-behaved in certain ways, e.g.,
for
> >copying, assigning, and comparing (depending on the "concept" to which
the
> >template conforms).
> 
> All that STL containers require is that your dtor and copy assignment
> don't throw. The compiler generated members fullfill that requirement.

True.  And although it's not required, it *is* a good idea to make sure that
each class's copy constructor and assignment operator behave properly.  The
default copy ctor often does not do what you want, e.g., if your class
contains pointers to other objects.
 
> >  This means you have to implement your own copy
> >constructor, assignment operator, and/or comparison operators for each
> >(non-trivial) class.  It may not be difficult, but it is overhead that
> >Python does not impose.
> 
> Please inform yourself better before making such statements. Blindlessy
> drawing conclusions based on the C++ implementation of one particular
> vendor from Redmond isn't good for your credability.

And assuming that I'm blindlessly drawing conclusions based on Microsoft's
compiler is not good for your credibility.  The default copy ctor
(regardless of the compiler) simply does a member-by-member copy, which is
often the wrong thing to do.  The point is that, even if the default is
acceptable, the C++ programmer has to (or should) think about it.

This discussion illustrates the original point: these are details that the
Python programmer does not need to deal with.

> Best regards,
>       Wim Lavrijsen

--
M.



From david_ullrich at my-deja.com  Wed Jun 28 13:03:59 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 28 Jun 2000 17:03:59 GMT
Subject: long lines, long string
References: <j9tcj8.i0b.ln@iiab.soluma.de>
Message-ID: <8jdb5g$ikj$1@nnrp1.deja.com>

In article <j9tcj8.i0b.ln at iiab.soluma.de>,
  "Holger M. F??ler" <fuessler at solutec.de> wrote:
> Hallo!
>
> Is there a possibility to put long lines of python-source-code in
multiple
> lines (for better clarity)?
> How can I accomplish the same with strings?

    Triple quotes allow you to include line breaks
in a quoted string. Putting a back-slash at the end
of a line tells Python to ignore the line break and
read the next line as a continuation of the current
one. This backslash thing works inside triple-quoted
strings:

s = """A long string\
 not containing line breaks"""

print s

s = """A string
containing line breaks"""

print s

> Greetings
>
> Holger
>
>


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


From petersc at stud.ntnu.no  Sat Jun  3 18:15:12 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 00:15:12 +0200
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <Pine.GSO.4.10.10006030844560.14749-100000@sundial>
Message-ID: <39398370.2D597F02@stud.ntnu.no>

Moshe Zadka wrote:
> 
> Martijn, do you happen to know how much time do I have to learn dutch
> before....ummmm....I can sleep more easily?

If you know, can you slip me a note? That'd sure be nice to know.

> or-is-haskell-enough-ly y'rs, Z.

answer :: Integer -> Boolean
answer 42 = "goed"
answer x  = "verkeerd"

so-haskell-does-only-work-in-special-cases-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From db3l at fitlinxx.com  Sun Jun 18 19:29:31 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Jun 2000 19:29:31 -0400
Subject: TABs (was: Waffling between Python and Ruby)
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com> <l03102800b570d8042c72@[193.78.237.143]>
Message-ID: <u1z1ua7qc.fsf@ctwd0143.fitlinxx.com>

Just van Rossum <just at letterror.com> writes:

> In general (ie. in the rest of the world, outside of emacs), a TAB has
> several perfectly good definitions. One of them is that it is always equal
> to 8 spaces. Another is that it is a variable thing.

Hmm, I've found a fixed space definition to be far less common than
anything else - in particular fixed column locations, which grew out
of the first TTYs with fixed tab locations.

> The fact that the "tab == 8 spaces" rule is a) not generally accepted and
> b) not generally enforcable across editors makes it IMHO a pretty useless
> definition. Tools that break when using anything but 8 are badly designed:
> it's like using two digits to store a year...

In this particular case, it's not so much that anything is actually
breaking (since the use in question was solely using TABs for
indentation), but that the visual indentation being enforced by
setting the editor to use 4 instead of the more common 8 just makes it
harder to get the same visual representation in other tools - without
somehow also adjusting each of those tools.

But I do agree with you that the fact that this sort of discussion
takes place over what a TAB really should be defined to do makes it
largely unsuitable for use in code that is meant to be distributed and
shared.

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


From hzhu at knowledgetrack.com  Fri Jun 16 12:51:54 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 16 Jun 2000 16:51:54 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com>

Oops!  You're right, as several people also pointed out to me in email.
Actually I had this in .emacs and didn't realize it for years:

(setq-default tab-width 4)

I am not aware of serious adverse effect, though.  The good thing is that
changing indentation is really easy.  You can use one keystroke backspace or
C-d, for example.  Presumably the evil will manifest itself when dealing
with other programs, but I also set 'less -x4' so I don't even see it.

Wondering what other people will see when reading my programs ...

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


On 14 Jun 2000 21:57:29 -0400, David Bolen <db3l at fitlinxx.com> wrote:
>(dropped comp.lang.ruby since the whitespace is Python only)
>
>hzhu at knowledgetrack.com (Huaiyu Zhu) writes:
>
>> If you use emacs:
>> 
>> C-x h M-x tabify
>> 
>> will tabify the whole buffer.  I always do this when doing copy and paste
>> from news groups, and do an untabify the other way round.  This is because
>> the default tab space is 4 in emacs (which is sane) but 8 everywhere else.
>
>The default tab stop for a physical TAB character in Emacs is 8, which
>is the most sane value.  A TAB character in an Emacs buffer is, by
>default, treated as moving to the next boundary that is a multiple of
>8 by default.  The tab stop width can in theory be overridden (see
>variables default-tab-width and buffer-local tab-width), but doing so
>is just plain evil.
>
>Instead of changing tab stops, it is far preferable to set the
>indentation for coding in various modes to anything you want (and
>Python mode will default to 4).  When inserting indentation, Emacs
>will use a mixture of TABs and spaces to get you where you need to be
>in the most efficient manner, assuming the 8-character TAB multiple.
>
>Or to put it more succinctly - the default Emacs interpretation of the
>TAB physical character and Python's interpretation are the same, and
>you should be perfectly safe letting Emacs mix TABs and spaces for
>Python code, unless you've reset the default tab stops, which as I
>mentioned above would be evil :-)
>
>That said, the Python mode keeps everything with spaces anyway.
>
>--
>-- David



From gchiaramonte at mail.com  Wed Jun 28 15:46:57 2000
From: gchiaramonte at mail.com (Gene)
Date: Wed, 28 Jun 2000 19:46:57 GMT
Subject: Why won't exec accept this?
Message-ID: <RCs65.11832$_b3.285418@newsread1.prod.itd.earthlink.net>

I am trying to store scripts in a dictionary, then exec the scripts as
necessary. I have stored the following text in a dictionary:

# test ...
def FooBar(x):
    return x + x

# here is the dictionary...
{'script': '# test ...\015\012def FooBar(x):\015\012    return x + x',
'version': 1.0, 'type': 'script'}

# and here is what happens...
>>> print d['script']
# test ...
def FooBar(x):
    return x + x
>>> exec d['script']
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "<string>", line 2
    def FooBar(x):
                  ^
SyntaxError: invalid syntax

If anyone can tell what is wrong, I would appreciate any help you could give
me.

Thanks,

Gene




From claird at starbase.neosoft.com  Fri Jun 30 23:17:34 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 30 Jun 2000 22:17:34 -0500
Subject: Web server executing Python scripts
References: <2B1262E83448D211AE4B00A0C9D61B030130125E@msgeuro1.creo.be>
Message-ID: <80EB8C0C818F320D.2A60B0748B6BCCB3.6A1C28BCC752E3D8@lp.airnews.net>

In article <2B1262E83448D211AE4B00A0C9D61B030130125E at msgeuro1.creo.be>,
Pieter Claerhout  <PClaerhout at CREO.BE> wrote:
>Hello all,
>
>I'm working on a web based system, which consist of a whole
>bunch of Python scripts, which connect to a bunch of machines
>on a network (either using ethernet or dailup networking) and
>execute some to code to for example get things from the
>windows registry of those machines.
>
>What do you think would be the best idea?
>- Use whatever webbrowser with CGI-scripting
>- Use MS IIS with ASP (and Python)
>- Use Zope to build the whole system?
>
>It should also connect to a database running on the host.
>
>
>Pieter
>

Others have already demonstrated that there's a large
... situational, if not subjective, component to the
decision about which Web server to use.

Your first paragraph interests me.  Do you realize
that Python can retrieve registry information directly
through your LAN, without a need for installation of
"some to code" on the reporting hosts?
-- 

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


From fiona at sitegnome.com  Tue Jun 27 09:38:03 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Tue, 27 Jun 2000 23:38:03 +1000
Subject: Books
References: <8jarl9$9mi1@news.qualitynet.net>
Message-ID: <3958AE3B.A19827B5@sitegnome.com>

Hi,

Have a look here ->

I'm new to Python and would appreciate advice on what book(s) I should
be looking
at?
http://www.faqts.com/knowledge-base/view.phtml/aid/2795/fid/199/lang/en

HTH,

Fiona


From gmcm at hypernet.com  Mon Jun  5 17:31:16 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 05 Jun 2000 21:31:16 GMT
Subject: threading with time limit
References: <Pine.LNX.4.10.10006051515580.30679-100000@Gloria.CAM.ORG>
Message-ID: <8F4AB9E3Bgmcmhypernetcom@199.171.54.155>

intmktg at Gloria.CAM.ORG (Marc Tardif) wrote in 
<Pine.LNX.4.10.10006051515580.30679-100000 at Gloria.CAM.ORG>:

>How can x threads be started at once and stopped y seconds later,
>processing only the successfully returned threads?

Threads have to stop themselves. You can't "kill" a thread without 
jeopardizing the integrity of the process.

>For example, consider a situation where all the links from multiple URL's
>must be retrieved at the same time. By setting a time limit, slow sites
>don't slow down the whole process and appropriate error messages are
>returned stating which sites wheren't completed in y seconds.

Then you need to use a method of retrieving that recognizes time limits. In 
your case, you have 3 choices:

- a version of urllib where urlopen uses select with a timeout and aborts 
if the timeout expires (Aahz may have done this?).
- a version of urllib that uses non-blocking sockets (then no threads 
needed).
- use separate processes instead of separate threads (because you *can* 
kill a process and properly release resources).

If none of those options are available to you, then you're doing roughly 
the right thing - just ignore the thread if it's too slow.

- Gordon


From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Fri Jun 23 06:10:13 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Fri, 23 Jun 2000 03:10:13 -0700
Subject: python & excel
References: <39533572.AB90FD1F@tcomhp43.epfl.ch>
Message-ID: <086ff63a.3e6f872e@usw-ex0105-035.remarq.com>

Hi Cedric,

It's possible to generate an excel file using win32com
extensions on a windows box. Get the extensions and a bit of
documentation at:

http://starship.python.net/crew/mhammond/

You might want to pick up a copy of Python Programming on Win 32
By Mark Hammond, Andy Robinson which has a chapter on this very
subject.

Richard



Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From moshez at math.huji.ac.il  Tue Jun 13 08:12:40 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 15:12:40 +0300 (IDT)
Subject: No Boolean type in "types"
In-Reply-To: <394612a2.0@katana.legend.co.uk>
Message-ID: <Pine.GSO.4.10.10006131512230.20303-100000@sundial>

On Tue, 13 Jun 2000, Thomas Mangin wrote:

> Hello,
> 
> Can someone explain me why the module "types" have no boolean type defined ?
> 
> true = (0==0)
> false = (not true)
> 
> BooleanType = type(0==0)

BooleanType == IntType
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From maxm at normik.dk  Thu Jun 22 08:24:43 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Thu, 22 Jun 2000 14:24:43 +0200
Subject: Am I the only one to receive the messages several times????
Message-ID: <7BD10B680501D411B9DF009027E06F32011AD2@EXCHANGE>

This starting to be a drag.

Furthermore I received a message from BeOpen about one of my messages that
was to long for their listserver.

Perhaps a BeOpen list is mailing to the python list, and the Python list is
mailing back to them???? This should cause a loop and make messages appear
several times.

Regards
   Max M



From millja at wwc.edu  Fri Jun 23 15:22:58 2000
From: millja at wwc.edu (millja at wwc.edu)
Date: Fri, 23 Jun 2000 19:22:58 GMT
Subject: Python COM does not seem to like me.
Message-ID: <8j0ddv$fqb$1@nnrp1.deja.com>

Im using Python with ASP.  Am using the WIN32 package is on the same
d/l page as python 1.5.2.

First off, is there something I dont know about the import statement?
Using it, it says that it cant find a module thats is the same
directory as the .asp file.  I checked the file name and everything.

Second, after getting rid of the import statement I get errors in my
python code that werent showing up when I was using the interpreter.  I
had just cut and pasted it straight across.

The error is:
TypeError: cant multiply sequence with non-int

The statment is:
CM1 = roll[1] * roll[2]

I am pretty much positive roll[1] and roll[2] are ints.  As I said,
this statement worked great in interpreter.

Thank!
jason milliron

BTW, this is being called using <SCRIPT RUNAT = SERVER LANGUAGE=Python>
& </SCRIPT> because there is also VBS code on the page.
Im using Python with ASP.  Am using the WIN32 package is on the same
d/l page as python 1.5.2.


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


From jam at quark.emich.edu  Mon Jun 19 19:58:09 2000
From: jam at quark.emich.edu (Jeff)
Date: Mon, 19 Jun 2000 19:58:09 -0400
Subject: SOCK_RAW with python socket library?
In-Reply-To: <slrn8ksreq.i2.tim.lavoie@goibniu.beyondtv.net>; from tim.lavoie@mts.net on Mon, Jun 19, 2000 at 07:08:03PM +0000
References: <slrn8ksreq.i2.tim.lavoie@goibniu.beyondtv.net>
Message-ID: <20000619195809.B13763@quark.emich.edu>

On Mon, Jun 19, 2000 at 07:08:03PM +0000, Tim Lavoie wrote:
> 
> Hello,
> 
> I've been tinkering with some socket programming lately, basically trying
> out ideas I get from browsing W. Richard Stevens' "UNIX Network
> Programming". Anyway, I'd like to play in Python too, so I'm using the
> socket library functions. When I try to use SOCK_RAW though, I get:
> 
>    socket.error: (93, 'Protocol not supported')
> 
> I understand the "operation not permitted" if I try it as a non-root user,
> but why this? Is SOCK_RAW not supported yet? The docs mention that it isn't
> generally useful, but it is quite specifically useful for say, ping.
> 
> 
>   Thanks,
>     Tim Lavoie
> 

greetings,

if all you really want is to implement 'ping' with python, this page:
http://the-tech.mit.edu/~jeremy/python/

has an implementation which I have used and I know works if run as root. in
fact, since it's kind of old (pre-1.5), it would be a good excercise to
reimplement using the 'struct' module (as the page suggests).

hope that helps...

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 



From ejfc at my-deja.com  Tue Jun 20 06:15:08 2000
From: ejfc at my-deja.com (ejfc at my-deja.com)
Date: Tue, 20 Jun 2000 10:15:08 GMT
Subject: converting hex number contained in a string to an integer
Message-ID: <8ing7d$o6r$1@nnrp2.deja.com>

Hello all,

I have a string containing a number. This is an
hex number. How can I convert this to an integer?

For instance:

>>> hexnumber
'32'
>>> 0x32
50
>>> int(0x32)
50  # this is the result I want to get
>>> hexnumber = '0x' + hexnumber
>>> hexnumber
'0x32'
>>> int(hexnumber)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
ValueError: invalid literal for int(): 0x32
>>>

How can I do this? I am sure it is really easy,
but I am just beginnig with
Python.

Best Regards.
--
Eduardo Fernandez


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


From Linus_Chang1 at bigpond.com  Sun Jun 11 07:38:16 2000
From: Linus_Chang1 at bigpond.com (Linus Chang)
Date: Sun, 11 Jun 2000 21:38:16 +1000
Subject: win32gui question - is my window the foreground window?
Message-ID: <39437A27.1020403@bigpond.com>

Hi all,

I'd like to detect if the window of my wxPython application is currently 
the foreground window.

The way I'm thinking of doing this is to use the GetForegroundWindow 
function to grab the handle of the current foreground window and compare 
it with my own handle (got by using the GetActiveWindow function)

(eg. http://msdn.microsoft.com/library/psdk/winui/windows_08tv.htm)

However, win32gui only gives me SetForegroundWindow, not Get...

Is there a better way of doing this? A wxPython method would be 
preferred but failing that a win32 specific way would be still be 
appreciated :-)

Assuming there's no better way of doing this, how would I go about 
calling the GetForegroundWindow and GetActiveWindow functions? Do I have 
to write c extensions for python?

Thanks in advance,

Linus Chang





From phd at phd.russ.ru  Wed Jun 21 08:38:55 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 21 Jun 2000 12:38:55 +0000 (GMT)
Subject: [python] Comparison feature in Python
Message-ID: <Pine.LNX.4.21.0006211237320.5939-100000@fep132.fep.ru>

---------- Forwarded message ----------
Date: Wed, 21 Jun 2000 15:33:56 +0400 (MSD)
From: Denis S. Otkidach <den at analyt.chem.msu.ru>
Subject: Re: [python] Comparison feature in Python


Python documentation states that comparison in Python is consistent, but it's
not true. The comparision is performed first by operands types and then by
their ids. The problem is that this statement is true only when both operands
are bult-in. The code listed below demonstrates bug in the builtin comparision
routine. The bug can be easily fixed (at least for instances without __cmp__
method) by changing comparison rules or turned into a feature by documenting
it.


-------------------------------------------------------------------------
#!/usr/bin/env python
"""
This sample code demonstrate that comparison in Python is not consistent
(as documentation say) even when buggy __cmp__ method is not used.
"""
import sys

class B:
    def __repr__(self):
        return "<Object B>"

objects = ['C'*2, B(), 'A'*2]

trash = []

while id(objects[1]) < id(objects[0]):
    trash.append(objects[1])
    objects[1] = B()
while id(objects[2]) < id(objects[1]):
    trash.append(objects[2])
    objects[2] = 'A'*2

print "The source list:", objects # => ['CC', <Object B>, 'AA']

objects.sort()
print objects # => ['CC', <Object B>, 'AA']

objects.reverse()
print objects # => ['AA', <Object B>, 'CC']

objects.sort()
print objects # => [<Object B>, 'AA', 'CC']

# => [<Object B>, 'AA', 'CC'] but one expects ['CC', <Object B>, 'AA']
# as it is after first call of sort method
-------------------------------------------------------------------------

Many thanks to everybody in Python Russian (python-rus) mailing list for
useful discussion.




From gmcm at hypernet.com  Wed Jun 28 20:37:04 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 00:37:04 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <8F61BBD91gmcmhypernetcom@199.171.54.194> <slrn8ll4ak.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <8F61DA247gmcmhypernetcom@199.171.54.194>

Huaiyu Zhu wrote: 

>On 28 Jun 2000 22:51:18 GMT, Gordon McMillan <gmcm at hypernet.com> wrote:

>>Besides avoiding using the same name, you could make the top level a 
>>package, then python.BBB.AAA could import python.AAA and it would be 
>>completely unambiguous. Which is a good reason to avoid relative
>>imports - they are by nature ambiguous.
>
>I don't follow you here.  Both AAA and BBB are packages.   In fact they
>are 
>
>python/
>   AAA/
>      __init__.py
>   BBB/
>      __init__.py
>      aaa.py

I meant
 python/
   __init__.py
   AAA/
   ...

In other words, put a package around the whole thing.

But your real problem looks like this:

Gnuplot/
  __init__.py
  Gnuplot.py

MatPy/
  __init__.py
  gnuplot.py

You are being bitten by (1) case insensitivity of Windows and (2) relative 
imports.

[Incidentally. "from Gnuplot import Gnuplot" will *not* try to import 
itself - it could fail because it finds the module Gnuplot before it finds 
the package Gnuplot (this is the relative import problem), or because it 
finds (on Windows) the gnuplot module instead (both problems).]

Quite simply, the authors of Gnuplot are asking for trouble by having a 
package and a module within it with the same name. You are asking for 
trouble by having a module with the same (according to Windows) name as a 
package you want access to.

But the situation is not *quite* as bad as you say. That is:

AA/
 __init__.py
 snarf.py
BB/
 __init__.py
 snarf.py

presents no problem at all, (unless someone uses "from ... import *" which 
can cause name clashes no matter what you do). So we're not talking about a 
"flat" namespace. Just a tricky one <wink>.

>My question is exactly how to make imports absolute instead of relative.

You can't avoid the fact that Python will first try a relative import.

>>>Although it is not advisable to put '.' in path, especially at front,
>>>a lot of people actually do. And the functionality of a package should
>>>not depend on this detail.  Otherwise what happens if one imports
>>>several packages each expecting a different ordering of sys.path?
>>
>>Actually it is very useful, since it allows you to have a script have
>>some private support modules that don't clutter up sys.path for other
>>scripts. 
>
>Hmm, what is useful?  Having '.' in path or allowing packages to demand
>different content of path?  Yes I'd agree to the former (but it's biting
>me now) but I don't see the latter.

No, it's not biting you. That would bite you when you were in Gnuplot/ and 
wrote "from Gnuplot import Gnuplot". You are being bitten by the fact that 
Python *always* looks first in the current package before looking at 
sys.path, (in combination with Windows...).


Lets say you have a (gag) 2,000 line module: foo.py. To speed things up 
(avoid the compile), you write a 2nd script:
------
import foo
foo.main()
------

You shouldn't have to add the directory this is in to PYTHONPATH for this 
to work. (This being why "." *should* be on the path.)

>>You could use Greg Stein's imputils, but in this case just packagizing
>>the whole thing is a more straightforward solution.
>>
>
>Where can I find it?  

http://www.lyra.org/greg/python/


From wjdandreta at worldnet.att.net  Sun Jun 18 19:32:55 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 18 Jun 2000 23:32:55 GMT
Subject: Why is Python so slow ?- revisited.
Message-ID: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net>

Let me preface my remarks with:

--I am using the Python 1.2 binary for DOS with the Python 1.5.2 library
which causes some problems. I avioded some problems by importing functions
directly from strop instead of string. Replace is not in strop so I used the
one in string but I had to comment out (put an octothorpe at the start of
each line) the function's descriptive string to get it to work.--

A while ago I posted a message called Why is Python so slow? At that time
this Python script took 90 minutes to run. By making some relatively minor
changes, I was able to reduced the time to 6 minutes.

The biggest improvement came (about a factor of 6) when I changed the
replace(x,y) function in string with
joinfields(splitfields(x,y),''). Considering that this is exactly what the
replace function in string does, I was quite surprised at the results.
Essentially I reduced 3 nested function calls to 2.

The only thing that makes sense to me is that Python is spending 85% of the
time processing the comment lines and 15% of the time doing useful work.
There is a string.pyc so I would have assumed that it would not contain any
comments or descriptive strings but apparently that's not true.

I am not very familiar with interpretive languages, Python is the first one
I have used, but it seems that comments can cause a serious perfomance hit
because they have to be processed along with the code.

Bill




From ge at nowhere.none  Wed Jun 28 10:53:03 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 28 Jun 2000 14:53:03 GMT
Subject: more python evangelising...
References: <B57E791A.264E%eableson@mac.com> <Uj865.1905$iN5.408850@ptah.visi.com> <Rn965.73297$7o1.1735998@news2.rdc1.on.home.com> <3959376D.DA8DC41F@estinc.com>
Message-ID: <jjo65.1999$iN5.438087@ptah.visi.com>

In article <3959376D.DA8DC41F at estinc.com>, Eric Lee Green wrote:
>Tom wrote:
>> "Grant Edwards" <ge at nowhere.none> wrote in message
>> news:Uj865.1905$iN5.408850 at ptah.visi.com...
>> ...
>> > Since MS is a publicly held company, the people who run it
>> > have a fiduciary responsibility to the stockholders.  They are
>> > obligated to make them as much money as they can (legally, that
>> > is).  If they didn't, they could be held liable in court.
>> 
>> This is an important point.  This is the heart of capitalism.
>
>Gosh, I thought the heart of capitalism was making a better
>product at a better price. Maybe I ought to re-think that.

The heart of capitalism is to make the minimum that you can
sell and charge what the market will bear.

>Let's see:
>
>Make a shoddy product and market the hell out of it while
>charging double the price. That's the heart of capitalism,
>yeah.

If it works, and it makes more money than the "better product"
approach, (and you know this) then you have a fiduciary
responsibility to the owners of the company (John and Jane
Doe's 401K plan) to do that.

If customers didn't put up with such crappy products, then the
people running MS would be obligated to do make better ones.
Customers seem to love the crap MS dishes out, so they keep
dishing it out.

> it works for Microsoft.

So far.

-- 
Grant Edwards                   grante             Yow!  Gee, I feel kind of
                                  at               LIGHT in the head now,
                               visi.com            knowing I can't make my
                                                   satellite dish PAYMENTS!


From famvandenberg at home.com  Wed Jun 21 17:05:30 2000
From: famvandenberg at home.com (bo Vandenberg)
Date: Wed, 21 Jun 2000 21:05:30 GMT
Subject: wxpython open source c++
Message-ID: <39512DFD.6F70D16E@home.com>

I was wondering if anyone had used the BLOODSHED C++ package to make
widjets for WXwindows and WXpython?

bo



From robin at alldunn.com  Wed Jun 14 00:05:32 2000
From: robin at alldunn.com (Robin Dunn)
Date: Tue, 13 Jun 2000 21:05:32 -0700
Subject: ANNOUNCE: wxPython 2.1.16
Message-ID: <003c01bfd5b5$c9cac810$0b01a8c0@ARES>

wxPython 2.1.16 is available for download at http://wxpython.org/Download.
This is hopefully the last release before 2.2.

>From the CHANGES.txt file:

New in 2.1.16
-------------

Added an attribute named labelDelta to the generic buttons that
specifies how far to offset the label when the button is in the
depressed state.

Added wxTipProvider and friends.  See the demo for an example.

wxGrid can now change the cell highlight colour.

Added wxDragImage.

Fixed printing on wxGTK.

Added wxDateTime, wxTimeSpan, and wxDateSpan to wxPython.utils.

Added wxCalendarCtrl.

WARNING: A while back I asked what should be done about the Magic
Method Names.  (Methods that are automatically turned into event
handlers by virtue of their name.)  The consensus was that it is more
confusing to have them than to try and expand them to have greater
coverage.  I am finally getting around to removing the code that
generates the event binding.  This means that if you are using any of
the following method names without a EVT_* call that you need to
modify your code to add the EVT_* to hook the event to the method.

    OnChar
    OnSize
    OnEraseBackground
    OnSysColourChanged
    OnInitDialog
    OnPaint
    OnIdle
    OnActivate
    OnMenuHighlight
    OnCloseWindow
    OnScroll

Added wxSpinCtrl.

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





From echuck at mindspring.com  Tue Jun  6 13:03:19 2000
From: echuck at mindspring.com (echuck at mindspring.com)
Date: Tue, 06 Jun 2000 17:03:19 GMT
Subject: Memory leak: assigning function object to instance variable
References: <8hh0pf$i7$1@phys-ma.sol.co.uk>
Message-ID: <8hjasb$8pc$1@nnrp1.deja.com>

In article <8hh0pf$i7$1 at phys-ma.sol.co.uk>,
  "John Parkey" <jparkey at nospamrapidbs.com> wrote:
> I have registered this bug, with code fragment below, but wondered if
> anybody else was falling foul of memory leaks of this type.  We use
xmllib,
> which relies on assigning methods to instance variables for later
callbacks.
> Our recent volume testing has shown serious leakage relating to this
problem
> (I use it in other places too, so it's pretty serious for us).
Environment
> is 1.5.2 on NT, build 132.  To see the problem, run the example below
and
> watch memory usage in task manager.
>
> (Quick plug for Mark Hammond - thanks for loads of help, and
finding/fixing
> COM related leaks). I hope my week of leak hell will soon be over!
>
> class fred:
>  def __init__(self):
>   self.indirectFunc = self.theFunc
>
>  def theFunc(self):
>   return "blah"
>
> def test():
>  f = fred()
>  del f
>
> if __name__ == "__main__":
>  for x in xrange(1000):
>   test()
>
>

John,

This is not an actual bug. When you say:

        self.foo = self.someMethod

What you get is a _bound method_. e.g., bound to 'self'. I bet when you
invoke your method with apply() you don't pass self. That's because
self is already stored in the bound method. This causes a circular
reference, and as you probably already know, Python doesn't collect
those (by design).

We had this problem in Webware for Python. Our solution:

        self.foo = self.__class__.someMethod

And the apply() has to pass self. Now the circular refs are gone.

I did a performance benchmark and the second method seems to be ever so
slightly slower, but not much. In the context of everything else you
do, you won't see it.

Happy coding.

-Chuck
http://webware.sourceforge.net


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


From thomas at xs4all.net  Tue Jun 13 05:30:18 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 11:30:18 +0200
Subject: Newbie questions
In-Reply-To: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>; from jar@mminternet.com on Mon, Jun 12, 2000 at 09:10:35PM -0700
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
Message-ID: <20000613113018.S20700@xs4all.nl>

On Mon, Jun 12, 2000 at 09:10:35PM -0700, jim roush wrote:

> I'm just learning Python.  The following code does not work.  It's
> supposed to calculate the circumference after the user enters the
> radius.  The program displays the prompt, but when I enter the radius
> and the press enter, nothing happens.

> FYI, I'm using Python 1.5.2 on WinNT.

Your program should work fine, and it does, here. However, if you are using
IDLE, the problem is a known bug in IDLE ;) You have to type the number and
press enter in a different window than the one that shows the 'Enter is the
radius:' text. I dont have the exact text of the bug at hand, sorry, but if
you fiddle around, you'll find the right window.

It's fixed in newer versions of IDLE, by the way, as well as a number of
other things. If you use IDLE a lot, you might want to grab the newer
version: http://www.python.org/idle

> pi = 3.141592654
> radius = input("Enter is the radius: ")
> circ = 2 * pi * radius
> print circ

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From dag at orion.no  Fri Jun 23 04:12:29 2000
From: dag at orion.no (Dag Sunde)
Date: Fri, 23 Jun 2000 08:12:29 GMT
Subject: "TypeError: read-only buffer, None" returned from file.write...
References: <MIS35.664$PQ4.13634@news1.online.no> <8ip2q3$2jrk$1@news.aros.net> <mu145.13877$R7.2335108@juliett.dax.net> <6E1AD32AF87E515E.2A23DF4E6211A01A.00440780EBC1EDDA@lp.airnews.net>
Message-ID: <NZE45.14713$R7.2481010@juliett.dax.net>

I do that now...

It was a typo in the function that filled the buffer
in the first place. I tested it in the interactive window,
and when I felt that it did what I wanted, I moved it into
my code, introducing a bug in the process... :-(

Dag.

"Jeremy Hylton" <jeremy at beopen.com> wrote in message
news:6E1AD32AF87E515E.2A23DF4E6211A01A.00440780EBC1EDDA at lp.airnews.net..
.
> It would be better, though, to test explicitly for None and not write
> the result to a file when it is None.
>
> sBuffer = m_XMLDoc.transformNode(xslDoc)
>
> if sBudder is None:
> print "Error: transformNode returned None"
> else:
> f = open( outFile, 'w')
> f.write(sBuffer)
> f.close()
>
>
> Jeremy




From johngrayson at home.com  Thu Jun 15 06:25:09 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 15 Jun 2000 10:25:09 GMT
Subject: Threads and GUI
References: <3948A32A.25025A54@t-online.de>
Message-ID: <8iaatp$htj$1@nnrp1.deja.com>

In article <3948A32A.25025A54 at t-online.de>,
  Heinz Folker <folker at t-online.de> wrote:
> I want to create window like this:
>
> tk=Tkinter.Tk()
> widget=Test(tk)
> thread.start_new_thread(widget.mainloop(), ())
>
> I need a thread because I want to use the interpreter while the
mainloop
> is running. I want to interact with this window.
> But this code doesn't work.
>
> Thanks for help
>

If you search this newsgroup, you'll find numerous postings on
this topic.

The basic rule is: Leave the mainloop in the main thread then
thread off subordinate tasks. Also, wherever possible,
arrange for the main thread to update the gui.

>


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


From jsantaniello at my-deja.com  Mon Jun  5 21:23:02 2000
From: jsantaniello at my-deja.com (jsantaniello at my-deja.com)
Date: Tue, 06 Jun 2000 01:23:02 GMT
Subject: SSL and POST with httplib or urllib
Message-ID: <8hhjpn$c7$1@nnrp2.deja.com>

Hi,

Sorry about the dejanews, but my newsfeed isn't
working just now...

Thanks to the good folks here on
comp.lang.python, I've gotten 1.6 going with SSL
support and it works. But now I'm having problems
using POST with a cgi. I've tried using httplib's
HTTPS and urllib's openurl, but I don't know what
I'm doing wrong. GET seems to work fine, but the
cgi I'm trying to access requires POST. Are there
any particular strangenesses to using POST that
anyone knows about? Does someone have a
suggestion on how I should be doing this? I'm
basically just following the examples from the
Python web page's docs. All help greatly
appreciated.

Thanks,

Joseph Santaniello


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


From dalke at acm.org  Wed Jun 28 07:01:48 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 28 Jun 2000 05:01:48 -0600
Subject: Requiring a certain version of interpreter?
References: <Lqj65.4362$h8.209230@news-west.usenetserver.com> <s2k65.1080$MZ3.2911@newsfeed.slurp.net>
Message-ID: <8jclp8$t6a$1@slb7.atl.mindspring.net>

Matthew Cline <matt at nightrealms.com> asked:
> Is there any builtin feature, or common module, that handles requiring
> at least a certain version of the Python interpreter to be used for a
> script?  Perl has 'require' for this purpose.


Chris Armstrong replied:
>>>> import sys
>>>> print sys.version
>1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian GNU/Linux)]
>
>There may a way to just get the version number only, but if nothing else
this
>will work.

Python's Makefile.pre uses sys.version[:3] to get the version.  At the
very least, this implies there is no other way to get that data - I've
never seen another way.

The other answer is, why do you need this?  You should write your code
to check for capability, rather than version.  There are two major versions
of Python, one in C and the other in Java.  There's no guarantee that the
version numbers will be the same on both platforms and mean the same thing.

For example, suppose your code wants to use list extend, which was added
in (I think) 1.5.1.  You could force a check like:

if "extend" not in dir([]):
  raise ImportError, "This version of Python does not support list.extend()"

Perl doesn't have this problem because only one person has been, umm..,
as adventerous as Larry Wall to write a Perl implementation from scratch.

                    Andrew
                    dalke at acm.org






From Eckhard at pfluegel.fsnet.co.uk  Thu Jun 15 17:20:38 2000
From: Eckhard at pfluegel.fsnet.co.uk (Eckhard Pfluegel)
Date: Thu, 15 Jun 2000 22:20:38 +0100
Subject: Reloading modules dynamically
Message-ID: <394948A6.8B9B5FF0@pfluegel.fsnet.co.uk>

I am writing a Python application which consists of two parts: a server
and a client. The client sends requests to the server. The requests
contain names of modules which the server is supposed to load and to
use.
The problem is that these modules might change at runtime, and I want to
send refresh requests to the server so that it reloads the modules. The
modules contain class definitions, I know that reloading such modules is
not recommended since previously instanciated objects will continue to
exist with the old class definition. However I make sure that all these
old objects won't be used after the reload.
In order to deal with a module name stored in "name", the code on the
server side goes

m = sys.__dict__["modules"]
if not m.has_key(name):
    # this means the module is not yet loaded
    exec("import " + name)
else:
    # reload the module
    reload(m[name])

This look like a bad hack to me, but it works. But now I am running into
problems because one of the modules "m" itselfs imports another module
say "n". This other module "n" as well changes during runtime, and this
is not detected even after a refresh request because the import
statement in module "m" does not reload "n".
I wonder how I could solve that problem or, even better, whether one of
you guys out there have another idea of realising the whole matter...

Thanks very much in advance,

Eckhard





From brent.fulgham at xpsystems.com  Wed Jun  7 20:13:30 2000
From: brent.fulgham at xpsystems.com (Brent Fulgham)
Date: Wed, 7 Jun 2000 17:13:30 -0700 
Subject: I need a lesson.
Message-ID: <EDFD2A95EE7DD31187350090279C6767958677@THRESHER>

> The following is a module called howfast.py that I can't
> get to repeat.  In Qbasic I would have used a goto at the end 
> and it would have worked fine.  Of course there is no goto in 
> Python and that where I can use anyone's help.  See if you can 
> understand what I'm trying to accomplish below, it's a little 
> scrambled towards the end, and tell me the error of my
> ways.  I have the book "Learning Python" but I'm still stuck. 
> If there are any questions I'll be glad to reply.

Turn to the chapter on "Flow Control" constructs (for/while loops).

You could solve this using any of the following:

1) a for loop
2) a while loop:

e.g.,
while 1:
  # text of your program
  # will loop for ever

Have fun,

-Brent



From fplancke at my-deja.com  Wed Jun 28 15:23:46 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Wed, 28 Jun 2000 19:23:46 GMT
Subject: list.pop([i]) and list.remove(x) w/ for loops
References: <8jdfb6$ugh$1@netnews.upenn.edu>
Message-ID: <8jdjbc$pmf$1@nnrp1.deja.com>

In article <8jdfb6$ugh$1 at netnews.upenn.edu>,
  dwhite2 at blue.seas.upenn.edu (David White) wrote:

> I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].
> I wan't to delete all the a objects out of the list while
> preserving the order of the b objects.

You can also use the 'filter' builtin function:

  def f(o):
    return o.type != a

  olist = filter(f, olist)
  # filter(f, olist) returns a copy of olist with only the
  # items o such that f(o) returns a true value

or shorter:

  olist = filter(lambda o: o.type != a, olist)


--
Frederic van der Plancke
e-mail address: reverse and add dots etc. to:
      <com writeme vdplancke frederic>


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


From samwun at yahoo.com  Fri Jun 30 23:24:33 2000
From: samwun at yahoo.com (Sam Wun)
Date: Sat, 01 Jul 2000 13:24:33 +1000
Subject: how to make popen secure?
Message-ID: <395D6471.79CF0908@yahoo.com>

I want to use popen execute an external C program. The C program is
written to read kmem network traffic form the kernel memory. Therefore
the C program is given a special permission to read the kmem device.
How do I make popen secure? or how to make kmem secure even thought it
is read by the C program?

thanks
Sam.



From tanzer at swing.co.at  Sun Jun 11 02:23:51 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Sun, 11 Jun 2000 08:23:51 +0200
Subject: PATCH: Augmented assignment 
In-Reply-To: Your message of "Sat, 10 Jun 2000 23:28:06 +0300."
             <Pine.GSO.4.10.10006102326430.9354-100000@sundial> 
Message-ID: <m13119z-000wcEC@swing.co.at>

Moshe Zadka <moshez at math.huji.ac.il> wrote:

> On Sat, 10 Jun 2000, Christian Tanzer wrote:
> 
> > It doesn't obfuscate code looking like:
> > 
> >    nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1
> > 
> > And it saves more than a few keystrokes.
> 
> Anyone doing *that* isn't  familiar with the pydiom:
> 
> s, i = nested_strucre[some_index], another_index
> s[i] = s[i] + 1

I'm familiar with the pydiom and I consider it one of Python's strong
points.

But I want to keep complex algorithms as short as possible. Adding
another line to unpack doesn't always increase readability.

Use-the-right-tool-for-the-job-ly y'rs, 

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From krussell4 at videon.home.com  Thu Jun 15 00:30:58 2000
From: krussell4 at videon.home.com (Kevin Russell)
Date: Thu, 15 Jun 2000 04:30:58 GMT
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my>
Message-ID: <39485C09.306BF3DC@videon.home.com>

root wrote:

>    the matrix is fixed, which is not good. my solution use dictionary (i
>    know this is not the right way). the key is a tuple of the matrix
>    indices. the value is '0'. then base on game of life rule, update the
>    dictionary and print for next generation.
>
>            m={}
>            m[(0,0)] = 0, m[(0,1)] = 1,....

If it works, then it IS the right way.  After all, there's more than one
way to... oops, sorry, wrong newsgroup.

As Trent Mick has pointed out, there's NumPy for doing serious
matrix manipulation.  But the demands of the Game of Life and the
8-Queens problem are minimal enough that a list-of-lists solution
or your dictionary solution are perfectly adequate.



>
> 2. 8 non-attacking queen.
>
>    use backtracking. the function call itself back. how to do recursive call
>    in python? roughly, the C program I study look like this;

It's pretty straightforward to translate a standard backtracking
algorithm from practically any AI textbook into Python.  I tried it
out as an exercise when I was first learning Python, but the result
died in a disk crash.

You might want to look at Mark Lutz's Holmes -- a proof-of-
concept expert system shell in Python.  It illustrates a few different
backward chaining and forward chaining inference algorithms.

[Aside to peanut gallery:  Did the only web-accessible copy of
Holmes die in the Starship disk crash?  It seems to me to be
something worth having in the Vaults of Parnassus.  If Mark is
reading this, is willing to put it there, but is too busy, I'd be happy
to send them my own copy.]

And, with the advent of Stackless Python, the possibilities for doing
backtracking directly with continuations are just beginning.

Good luck.

-- Kevin




From alexandre.ferrieux at rd.francetelecom.fr  Wed Jun 28 10:41:13 2000
From: alexandre.ferrieux at rd.francetelecom.fr (Alexandre Ferrieux)
Date: Wed, 28 Jun 2000 16:41:13 +0200
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <3958C648.1D76@rd.francetelecom.fr> <3959DDA0.5414CF@hursley.ibm.com>
Message-ID: <395A0E89.1A94@rd.francetelecom.fr>

Paul Duffin wrote:
> 
> Apparently there is a port of Expect for Python so why not use that
> rather than try and reimplement it all over again. 
> Expect has over 7 years of development and centuries of testing on many
> different platforms, do you really want to reproduce that. If so why,
> what exactly is wrong with Expect.

The same as with TkInter: it needs a full Tcl inside. Again I've got
next to nothing against Tcl, it's just that the separation of orthogonal
things is such a natural thing for me.

-Alex


From erikd at zip.com.au  Sat Jun 10 01:05:30 2000
From: erikd at zip.com.au (Erik de Castro Lopo)
Date: Sat, 10 Jun 2000 05:05:30 +0000
Subject: Distutils: Problem with install_path?
Message-ID: <3941CC9A.7E88788B@zip.com.au>

Hi all,

I've got Python 1.5.2 installed in /usr/local and I'm trying
to install NumPy. 

I first installed DistUtils and then tried to install NumPy
as per the documentation and got the following error:

Traceback (innermost last):
  File "setup.py", line 29, in ?
    ext_modules = [('_numpy',
  File "/usr/local/lib/python1.5/site-packages/distutils/core.py", line 72, in setup
    dist = klass (attrs)
  File "/usr/local/lib/python1.5/site-packages/distutils/dist.py", line 196, in __init__
    raise DistutilsSetupError, \
distutils.errors.DistutilsSetupError: invalid distribution option 'install_path'

Anybody have any ideas on whats going wrong and/or how to fix it? I've 
had a look at the files mentioned in the error report, but I'm rather 
new to Python and I'm stumped.

Thanks in advance,
Erik
-- 
+-------------------------------------------------+
     Erik de Castro Lopo     erikd at zip.com.au
+-------------------------------------------------+
"No Silicon Heaven?  Preposterous!  Where would
all the calculators go?" -- Kryten, Red Dwarf


From aahz at netcom.com  Thu Jun  8 16:43:30 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jun 2000 20:43:30 GMT
Subject: wishlist Python3K: subclass ints
References: <Pine.GSO.4.10.10006082253450.17560-100000@sundial>
Message-ID: <8hp0hi$oh4$1@slb3.atl.mindspring.net>

In article <Pine.GSO.4.10.10006082253450.17560-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>On 8 Jun 2000, Aahz Maruch wrote:
>>
>> Close, but no cigar.  Start with the base class "number".  Then
>> everything else follows: int, float, rational, complex.
>
>Actually, in the number case specifically, I'd go for the
>no-hierarchy-have-a-good-model-optimize-behind-the-covers strategy:

You have a point, but I do still think there should be some sort of
virtual base class behind all numbers.  That way, it's at least clear
what operations should be available in any number class.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From senn at maya.com  Fri Jun  9 15:10:52 2000
From: senn at maya.com (Jeff Senn)
Date: 09 Jun 2000 15:10:52 -0400
Subject: Stackless Python for the Mac?
In-Reply-To: Ian Maurer's message of "Fri, 09 Jun 2000 18:30:32 GMT"
Message-ID: <zoouk6v7.fsf_-_@SNIPE.maya.com>

Has anyone contemplated (or started...or finished) a port of stackless
to the Mac?  Other than syncronizing the sources, there shouldn't be
any complications, right?

[Chris, Jack?]

-- 
-Jas




From clarence at netlojix.com  Thu Jun 15 10:34:46 2000
From: clarence at netlojix.com (Clarence Gardner)
Date: Thu, 15 Jun 2000 07:34:46 -0700
Subject: Can I build a single file containing all the contents of LIB?
References: <Vg525.3907$227.93950@nnrp1.uunet.ca>
Message-ID: <961079960.717336290@news.silcom.com>

On Thu, 15 Jun 2000, Warren Postma wrote:
>I have an embedded system which has no support for long filenames. It's also
>a pain to upload the whole LIB subdirectory to this system.  The embedded
>system has a FAT filesystem and only supports 8 character filenames with a 3
>character extension. The first problem then is with the "exceptions.py" and
>from there on.    Since my system recompiles for both native NT/Win32 and
>embedded mode, what I would ideally like to do is have a python script
>"bundle up" the standard Python library into a file, and then just upload
>that bundle. Then I would like it if python would search the \LIB directory
>FIRST for xxx.py and if not found, then search in the compressed (or not)
>file for it.   I would like the file to be separate from the executable, not
>linked in, so I can change the python library without recompiling the
>executable.
>

You could probably write your own importing function which would
use __import__ rather than import, and in your embedded system would
map the module name to your short-named copy of the normally long-named
file before applying __import__ to it.  Of course, you want to bind the module
thus imported to the actual module name rather than the short file name.
For example (totally untested code)

def myimport(ModuleName):
    ModuleName = ModuleMap.get(ModuleName, ModuleName)
    return __import__(ModuleName)

exceptions = myimport('exceptions')


From ge at nowhere.none  Thu Jun 29 11:49:00 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 29 Jun 2000 15:49:00 GMT
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com>
Message-ID: <MdK65.2186$iN5.485797@ptah.visi.com>

In article <395B6960.26AB52C1 at estinc.com>, Eric Lee Green wrote:

>> >Mostly because Python doesn't "really" have classes. What Python
>> >"really" has are hash tables whose entries are objects (either methods
>> >or other data objects), and some syntactic sugar [...]
>> 
>> By this logic, all C++ has is pointers and some syntactic sugar.
>
>Exactly :-).
>
>C++ is syntactic sugar for "C". As should be obvious from the fact that most
>early C++ compilers generated "C" as their output :-). 

And C is just semantic sugar for assembly...

-- 
Grant Edwards                   grante             Yow!  HELLO, little boys!
                                  at               Gimme a MINT TULIP!! Let's
                               visi.com            do the BOSSA NOVA!!


From jajs at NOSPAMretemail.es  Tue Jun 13 20:09:29 2000
From: jajs at NOSPAMretemail.es (Juanjo =?iso-8859-1?Q?=C1lvarez?=)
Date: Wed, 14 Jun 2000 00:09:29 GMT
Subject: "import foo" and "from foo import bar"
References: <slrn8kaa4u.1r0.jajs@Pascual.es> <vnd3dmiwfci.fsf@camelot-new.ccs.neu.edu>
Message-ID: <slrn8kb8f9.7of.jajs@Pascual.es>

Justin Sheehy dijo sobre Re: "import foo" and "from foo import bar",
en fecha: 12 Jun 2000 21:15:57 -0400

|jajs at NOSPAMretemail.es (Juanjo ?lvarez) writes:
|
|> It's better to write the "from string import join" form or it's exactly
|> the same (in memory use and performance) that "import string"?
|
|There is a small performance benefit to the 'from' approach, but I
|usually find that it is outweighed by the fact that it is much easier
|to read and understand the code later if you don't do that.

True.

|The only time that I would recommend doing that sort of thing (other
|than in modules that explicitly suggest that you do so) is if you are
|going to reference that name _many_ times, such as inside a huge for-loop.

I have followed your advice to make aliases to module symbols in a huge loop 
and now my program runs 15% faster!

|In that case, it may be worth it to get it into a local variable first.

Thanks Justin and Thomas. This newsgroup amazes me increasingly, I post
the message and in 20 minutes, Wham! two good answers.



From jvickroy at my-deja.com  Fri Jun  9 14:42:02 2000
From: jvickroy at my-deja.com (jvickroy at my-deja.com)
Date: Fri, 09 Jun 2000 18:42:02 GMT
Subject: OpenDX Python Wrapper -- any interest?
References: <20000515112219.C1304991@vislab.epa.gov>
Message-ID: <8hrdpr$ahj$1@nnrp2.deja.com>

In article
<20000515112219.C1304991 at vislab.epa.gov>,
  Randall Hopper <aa8vb at yahoo.com> wrote:
>      A while back I wrote a Python wrapper for
OpenDX, complete with
> callbacks to Python and exceptions thrown
across the DXL layer.  Before I
> devote any time toward preping, packaging, and
posting it, I thought I'd
> first see if any others are interested in this
tool (e.g. to build Python
> apps which use DX for graphics/visualization
under-the-hood).
>
>      For those that don't know, OpenDX
(formerly IBM Data Explorer) is an
> industrial-strength scientific/data
visualization package, now open-source:
>
>      http://www.research.ibm.com/dx/
>      http://www.opendx.org/
>
> --
> Randall Hopper
> aa8vb at yahoo.com
>
>

I am interested in this.


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


From aahz at netcom.com  Fri Jun 23 10:06:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jun 2000 14:06:51 GMT
Subject: SAP R/3 and Python ?
References: <3953212C.84E3127F@lemburg.com>
Message-ID: <8ivqtr$oo8$1@nntp9.atl.mindspring.net>

In article <3953212C.84E3127F at lemburg.com>,
M.-A. Lemburg <mal at lemburg.com> wrote:
>
>does anyone have an idea of how to interface to SAP R/3 from Python ?
>
>If there are no such Python interfaces, how much effort would it take
>to add orders from a web system into an R/3 system ?
>
>[I know, my ignorance shines through here: I've never used R/3, but a
>customer has confronted me with that need.]

Disclaimer: I know little about R/3, too.

IIRC, R/3 uses a database backend.  At worst, it should be possible for
you to dump data directly into R/3 as long as you're careful to follow
the virtual schema.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Why don't they make poppasmarts or didjeridon'ts?  --Aahz


From tchur at bigpond.com  Sun Jun 11 19:26:11 2000
From: tchur at bigpond.com (Tim Churches)
Date: Mon, 12 Jun 2000 09:26:11 +1000
Subject: Mozilla, XUL and...Python?
Message-ID: <39442012.692F02C2@bigpond.com>

The whole architecture of the Mozilla browser is very impressive,
especially the ability to create applications using an XML-based
specification (XUL). The only wrinkle is that the only client-side
scripting language mentioned is JavaScript. Clearly support for
JavaScript is essential, but is it possible for other scripting
langauges, in particular Python, to be embedded in Mozilla? I know that
David Ascher at ActiveState is working on a Mozilla/XUL-based Python
(and Perl?) development environment, but its not clear whether as a
result of this project Python will end up embedded in Mozilla as an
alternative scripting language, with full support for the DOM2 object
model used by Mozilla. Is that being worked on? Mozilla has been
described as an "Application Virtual Machine" which will allow
application deployment on any platform on which Mozilla runs (just like
the promise of the Java Virtual Machine...but Mozilla might succeed
where java has partially failed). Surely this is a boat which Python
can't afford to miss.

Tim C



From thiele at muc.das-werk.de  Wed Jun 14 14:09:37 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 14 Jun 2000 20:09:37 +0200
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i88b0$hlc$1@nntp9.atl.mindspring.net>
Message-ID: <3947CA61.D5A6CF8F@muc.das-werk.de>

Aahz Maruch wrote:

> Suppose you needed a file object to connect to some chunk of memory.  In
> other languages, you would need to inherit from the Python file object
> in order to ensure that random functions that expect file objects would
> still work.  In Python, that's completely unnecessary.  It's probably a
> lot simpler to create your own class, complete with open(), close(),
> read(), and write() methods -- and because you've maintained *interface*
> consistency, you can use your new class *ANYWHERE* a file object would
> go.
> --

But there is no mechanism to ensure that the behaviour of your memfileobjectsis
like a fileobject. For instance if you don't note that you forgot to write the
write()-function.
In a worst case scenario it works for many months until a user calls a function
in which
the write() -function is needed.



From e.smith at cs.bham.ac.uk  Tue Jun 20 11:02:12 2000
From: e.smith at cs.bham.ac.uk (Elliot Smith)
Date: Tue, 20 Jun 2000 16:02:12 +0100
Subject: Widget resizing
Message-ID: <394F8774.138B@cs.bham.ac.uk>

Hello all,

Can anyone tell me how to get Tkinter to resize frames etc. when their
parents (e.g. a Toplevel instance) are resized?

Thanks,
Elliot
-- 
Elliot Smith
School of Computer Science, University of Birmingham
email: e.smith at cs.bham.ac.uk
homepage: http://www.cs.bham.ac.uk/~ezs/


From andy.elvey at paradise.net.nz  Sun Jun  4 19:13:01 2000
From: andy.elvey at paradise.net.nz (Andy Elvey)
Date: Sun, 4 Jun 2000 23:13:01 GMT
Subject: Query about using "DataHandlers" module
References: <393A0796.4E5254D0@paradise.net.nz>
Message-ID: <393AE27D.A7B8123@PARADISE.NET.NZ>

Oops ...... the "example_07b" that I posted didn't quite show what I meant
with my question .....

 The code attached here *does* show what I mean..... it has the SQL code in it
with the WHERE statement that doesn't work .....

Andy Elvey wrote:

> Hi all.  I have a query about the use of the DataHandlers module
> (available at "Vaults of Parnassus").
>
>  I have attached a simple  example which uses this module (this example
> works fine).  But .... when I amend the SQL code to say  'select * from
> test where team == "Yankees" '  ,  the code still retrieves all teams -
> the where statement is ignored.
>
>   My question is - "does anyone have an example (or two!) of some Python
> code similar to my example , but which has a select statement that has a
> "where" clause in it?
>
>  ( and -  are there any examples which use the DataHandlers module and
> use * 2 tables*   with a join ... ? )
>
>  Many thanks in advance for any help offered .... :-)
>
>  ( and by the way .. this DataHandlers module is *great*  ..... :-)
>
>   ------------------------------------------------------------------------
> #!/usr/bin/env python
>
> # Illustrate the parsing a data file and creating SQL insert statements
> # The table definition comes from an XML file
>
> from DataHandlers.SQLTable import SQLTable
> from DataHandlers.dbdefXMLParser import dbdefXMLParser
> from DataHandlers.FileConnection import Connection
>
> # create a parser
> xml = dbdefXMLParser()
> # return a dictionary of table definitions
> tables = xml.parseFile( 'test.xml' )
> # get the baseball table
> tbl = tables['baseball']
> # the data file is delimited by commas
> tbl.setReader( 'Delimited' )
> tbl.setDelimiter( ',' )
> # open a file connection
> fc = Connection()
> fc.connect( 'teams.txt' )
> # we need to do this fake query so we can get the cursor
> # it actually means something when we are doing an SQL thing
> cursor = fc.query('')
> if not cursor:
>     print 'Unable to execute query'
>     sys.exit(1)
> # now walk thru the lines of data
> while cursor.next():
>     tbl.parse( cursor.getData())
>     # print out the insert statements
>     print tbl.getInsert()
>
> print "Query data returned..."
> cursor = fc.query( 'select * from test' )
> while cursor.next():
>     print cursor.getData( )
>
> cursor.close()
>
>
>   ------------------------------------------------------------------------
> <dbdef>
>   <sqltable name="baseball">
>     <field name="team" size="20" type="char" />
>     <field name="city" size="20" type="char" />
>     <field name="stadium" size="30" type="char" />
>     <field name="league" size="1" type="char" />
>     <field name="division" size="1" type="char" />
>   </sqltable>
> </dbdef>
>
>   ------------------------------------------------------------------------
> Orioles,Baltimore,Camden Yards,A,E
> Red Sox,Boston,Fenway Park,A,E
> Angels,Anaheim,Anaheim Stadium,A,W
> White Sox,Chicago,Comiskey Park,A,W
> Indians,Cleveland,Jacobs Field,A,E
> Tigers,Detroit,Tiger Stadium,A,E
> Royals,Kansas City,Royals Stadium,A,W
> Brewers,Milwaukee,County Stadium,A,E
> Twins,Minnesota,Hubert H. Humphrey Stadium,A,W
> Yankees,New York,Yankee Stadium,A,E
> Athletics,Oakland,Oakland Coliseum,A,W
> Mariners,Seattle,Kingdome,A,W
> Rangers,Texas,Arlington Stadium,A,W
> Blue Jays,Toronto,Toronto Stadium,A,E
> Braves,Atlanta,Fulton County Stadium,N,W
> Cubs,Chicago,Wrigley Field,N,E
> Reds,Cincinnati,Three Rivers Stadium,N,W
> Astros,Houston,Astrodome,N,W
> Dodgers,Los Angeles,Dodger Stadium,N,W
> Expos,Montreal,Exposition Stadium,N,E
> Mets,New York,Shea Stadium,N,E
> Phillies,Philadelphia,Veterans Stadium,N,E
> Pirates,Pittsburgh,Pirate Stadium,N,E
> Cardinals,St. Louis,Busch Stadium,N,E
> Padres,San Diego,Jack Murphy Stadium,N,W
> Giants,San Francisco,Candlestick Park,N,W
-------------- next part --------------
#!/usr/bin/env python

# Illustrate the parsing a data file and creating SQL insert statements
# The table definition comes from an XML file

from DataHandlers.SQLTable import SQLTable
from DataHandlers.dbdefXMLParser import dbdefXMLParser
from DataHandlers.FileConnection import Connection

# create a parser
xml = dbdefXMLParser()
# return a dictionary of table definitions
tables = xml.parseFile( 'test.xml' )
# get the baseball table
tbl = tables['baseball']
# the data file is delimited by commas
tbl.setReader( 'Delimited' )
tbl.setDelimiter( ',' )
# open a file connection
fc = Connection()
fc.connect( 'teams.txt' )
# we need to do this fake query so we can get the cursor
# it actually means something when we are doing an SQL thing

print "Query data returned..."
cursor = fc.query( 'select * from test where team eq "Yankees"' )
if not cursor:
    print 'Unable to execute query'
    sys.exit(1)
while cursor.next():
    tbl.parse( cursor.getData())
    print cursor.getData( )

cursor.close()


From teyc at bigfoot.com  Fri Jun 30 08:01:04 2000
From: teyc at bigfoot.com (Chui Tey)
Date: Fri, 30 Jun 2000 22:01:04 +1000
Subject: command-line history in win
References: <395c304e.1224185@news.student.lu.se>
Message-ID: <8ji1tk$1ui$1@bunyip.cc.uq.edu.au>

http://www.is.lg.ua/~paul/devel/

There is a port of the GNU readline library for Win32 at the link above.

Chui

"Alexander K" <invalid at dev.null> wrote in message
news:395c304e.1224185 at news.student.lu.se...
> aloha!
>
> i did a backsearch for this on deja. found a question but no answers.
> and the question is: isnt there a historyfunction in the python
> command-line interpreter for win?.
>
> you know, the bash-kinda thing.
>
>
>   thanks in adv. / alex k




From thiele at muc.das-werk.de  Wed Jun 28 15:59:36 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Wed, 28 Jun 2000 21:59:36 +0200
Subject: "+" does not match?
References: <395A577A.C83AE15D@muc.das-werk.de>
Message-ID: <395A5927.80448DD5@muc.das-werk.de>

Ok, I know why!



From aahz at netcom.com  Wed Jun 14 12:19:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 14 Jun 2000 16:19:57 GMT
Subject: Python Productivity over C++
References: <65118AEEFF5AD3118E8300508B1248774CB334@ALTNET> <8i890f$4fj$1@slb7.atl.mindspring.net> <3947ABA0.4987D728@home.com>
Message-ID: <8i8bbd$679$1@slb1.atl.mindspring.net>

In article <3947ABA0.4987D728 at home.com>,
Ivan Van Laningham  <ivanlan at home.com> wrote:
>Aahz Maruch wrote:
>> In article <65118AEEFF5AD3118E8300508B1248774CB334 at ALTNET>,
>> Mike Steed  <MSteed at altiris.com> wrote:
>>>
>>>ctor
>> 
>> Okay, does anyone else think this is a misspelling of "Chtorr"?
>
>Yes.  When the hell is Gerrold going to get back to the series?

http://www.chtorr.com/

(No, I'm not joking.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From pete-python-5-17 at kazmier.com  Thu Jun 15 17:08:02 2000
From: pete-python-5-17 at kazmier.com (Peter Kazmier)
Date: Thu, 15 Jun 2000 21:08:02 GMT
Subject: cmd, readline, and rlcompleter problems
Message-ID: <slrn8kihdh.jok.pete-python-5-17@view.kazmier.com>

I need help debugging a very strange problem.  I'm not a python expert
so it could be something very simple, on the other hand, I did use the
debugger already, but the problem disappears miraculously when I use
the debugger.  It makes no sense to me.

So, I'm basically trying to write my own cmd and rlcompleter.  I've
attached both the module and a small example of its use.  Its almost
identical to the standard cmd however I am trying to offer command
line completion.  

For example, if I have the following commands defined: do_quit and
do_show_version.  You should be able to type just 'qu' and be able to
hit TAB and have it complete to 'quit'.  That works fine of course.
Here is the problem, if a user types 'show ver' it SHOULD complete to
'show version'; however, the readline.completer function only seems to
pass in the current word that needs completing, in this case, 'ver'.
I want to complete based on 'show ver'.  It shouldn't be a problem
because readline provides a function called get_line_buffer() which
returns the contents of the line buffer.  So here's my completer
function: 

    def complete(self, text, state):

        if state == 0:
	    # Without this line, it would only complete on the 
	    # current word, in the above example, that would have
	    # been 'ver'.  But adding this, I should be able to 
	    # complete on the full line, or 'show ver'.
	    text = readline.get_line_buffer()
            self.matches = self.global_matches(text)
        try:
            return self.matches[state]
        except IndexError:
            return None

    # Find the matches, basically, look through this instance's class
    # for any attribute that begins with 'do_', for each attribute
    # that matches, convert underscores to spaces and ditch the 'do'
    # component.  This provides a list of possible completions.
    def global_matches(self, text):
        matches = []
        n = len(text)

        print "DEBUG text =", text
        for word in self.__class__.__dict__.keys():
            if word[0:3] == 'do_':
                word = string.join(string.split(word, '_')[1:], ' ')
                print "DEBUG", word
                if word[:n] == text:
                    matches.append(word)

        return matches

Everything works fine IF I comment out the get_line_buffer() line.
Unfortunately, that would only do completion on the current word.
When I uncomment the get_line_buffer() line, everything seems to work,
but during the 'for word in self.__class__.__dict__.keys()' loop, half
of the keys mysteriously disappear!  The DEBUG output, shows only some
of the keys, not all of them.  Comment out the get_line_buffer() line
and all of the keys() are back!  When I stepped through this in the
debugger, all of the keys() were also there!!  It appears that the
get_line_buffer() call is somehow munging my environment in a weird
way.  Oddly enough, whwen debugging, his munging does not appear.

Could someone PLEASE help me debug this??  I have no idea what I am
doing wrong and the symptoms are like none I've ever seen.  I've
attached two files, cli.py which is my cmd.py replacement, and
shell.py which is a simple test shell.  Run shell.py, and type in 'sh'
and hit TAB, it won't work.  But comment out that get_line_buffer()
line and then do the same.  Viola, it works.  But how come and how do
I get it to work the other way??

Thanks,
Pete

============
shell.py
------------
#!/usr/bin/python

import cli

class CLI(cli.cli):
    def do_quit(self, args):
        """Exit the command interpreter"""
        return 1

    def do_show(self, args):
        """Display information on various components"""
        print "Show function was called:", args
        return None

    def do_show_version(self, args):
        """Display the version of the shell"""
        print "Version is ..."

    def do_show_version_cli(self, args):
        """Display the version of the CLI"""
        print "CLI version is ..."

CLI().cmdloop()

============
cli.py
------------
import string
import readline

# Modified version of cmd.py

class cli:
    prompt = '# '
    
    def __init__(self):
        readline.parse_and_bind("tab: complete")
        readline.set_completer(self.complete)

    def cmdloop(self):
        stop = None
        while not stop:
            try:
                line = raw_input(self.prompt)
            except EOFError:
                print 
                break

            line = self.precmd(line)
            stop = self.onecmd(line)
            stop = self.postcmd(stop, line)

        self.postloop()

    def postloop(self):
        pass

    def precmd(self, line):
        return line

    def postcmd(self, stop, line):
        return stop

    def onecmd(self, line):
        line = string.strip(line)
        if not line:
            return None
        tokens = string.split(line)

        help = 0
        if tokens[len(tokens)-1] == 'help':
            help = 1
            tokens.pop()

        func = None
        for i in range(len(tokens), 0, -1):
            try:
                func = getattr(self, 'do_' + string.join(tokens[:i], '_'))
                args = tokens[i:]
            except AttributeError:
                func = None
            else:
                break

        if help:
            self.do_help(func)
        elif func:
            return func(args)

        return None
        
    def do_help(self, func):
        """Welcome to the help system"""
        if func:
            if func.__doc__:
                print func.__doc__
            else:
                print "No help available"
        else:
            print "No help available"

    def complete(self, text, state):
        
        if state == 0:
            text = readline.get_line_buffer()
            self.matches = self.global_matches(text)
        try:
            return self.matches[state]
        except IndexError:
            return None

    def global_matches(self, text):

        matches = []
        n = len(text)

        print "DEBUG text =", text
        for word in self.__class__.__dict__.keys():
            if word[0:3] == 'do_':
                word = string.join(string.split(word, '_')[1:], ' ')
                print "DEBUG", word
                if word[:n] == text:
                    matches.append(word)

        return matches

    






From bill_tolbert at bigfoot.com  Mon Jun 26 00:21:29 2000
From: bill_tolbert at bigfoot.com (Bill Tolbert)
Date: Mon, 26 Jun 2000 04:21:29 GMT
Subject: determining an assigned IP address
Message-ID: <8j6lnu$jhd$1@nnrp1.deja.com>

I've just set up Linux with my cable modem and I'd like to use Python to
help me launch remote X apps. I need to know the ip address assigned by
my isp, but so far I'm striking out. gethostbyname_ex in the socket
module returns the address I supplied when I thought it was static.
Right now the only way I know to get the right address is to log into
another machine and use who.

All help is appreciated!

Bill


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


From sholden at bellatlantic.net  Mon Jun 26 23:47:35 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Tue, 27 Jun 2000 03:47:35 GMT
Subject: smtplib
References: <000801bfdfca$61227820$b6cb45c6@timaratz>
Message-ID: <395823C5.63388650@bellatlantic.net>

Peter Timaratz wrote:
> 
> As a kindness to all the idiots like me perhaps this intro can be added to
> the smtplib module:

You aren't an idiot.  You are just a little disappointed that SMTP is
only a protocol, and you still have to prepare RFC822-structured messages.

However, there are the likes of the "mimecntl" module (available from
the starship last time I wanted it) which allows you to construct messages
you can edit the headers in, attach files, and anything else the aspirant
Pythin mail hacker might want.  Quite powerful enough to, for example,
take four years' e-mail and index it all in a SQL Server 7 relational
database.

> 
> "This module implements SMTP and does not send e-mail. If you want to send
> e-mail with Python then learn SMTP. If you don't want to learn SMTP then you
> can use one of the many Perl modules available that don't require you to
> know SMTP."
> 
Well, I might argue that since smtplib implements the protocol you don't
need to learn it, only the API offered by smtplib (one of many possible
APIs to the RFC 821 SMTP protocol).  But your remarks about Perl might be
considered inflammatory in a less well-mannered newsgroup.  Fortunately
the average Pythonista is a gregarious type, and rarely attacks even
trolls :-)

> > -----Original Message-----
> > From: python-list-admin at python.org
> > [mailto:python-list-admin at python.org]On Behalf Of Grant Edwards
> > Sent: Monday, June 26, 2000 10:09 AM
> > To: python-list at python.org
> > Subject: Re: smtplib
> >
> >
> > In article <001501bfdf74$6dff1fe0$14b745c6 at timaratz>, Peter
> > Timaratz wrote:
> >
> > >Yes, I figured out that smtplib requires the subject and
> > >headers to be imbedded in the message. I'm new to Python and I
> > >love the language. But I think that the design of this module
> > >reflects poorly upon Python.
> >
> > The module impliments SMTP, hence the name.  SMTP does not
> > concern itself with message contents.  Writing smtplib to make
> > certain assumptions or force certain conventions regarding
> > message content would be just plain incorrect.
> >
Which isn't to say this is the only possible approach.

> > >Almost every CGI programmer who looks at Python will want to
> > >use this module. It would only take a few lines of code to make
> > >it much more usable.
> >
> > Maybe you should write a higher level e-mail module that does
> > what you are requesting.  It does not belong in smtplib.
> >
> > >I'm saying this not to flame Python, but to encourage those
> > >responsible for it's distribution to look more critically at
> > >what they are putting out.
> >
A noble cause.  I would say that "those responsible" are pretty
open to reasonable suggestions, so don't let this stop you from
saying what els eneeds fixing.  But expect them to argue back
when they donlt think anything is broken!

> > Perhaps you should look more critically at what you expect from
> > a library that claims only to impliment SMTP.  If you want a
> > layer of functionality on _top_ of SMTP, then that's something
> > another module can impliment.
> >
> > >I hope that this module is atypical of the quality of this
> > >product. I'd like to see Python become more popular. There are
> > >many factors involved in that, but the quality of the modules
> > >is certainly one of them.
> >
> > There is nothing wrong with smtplib.  You seem to be expecting
> > it to do things other than SMTP.
> >
> > --
> > Grant Edwards                   grante             Yow!  I'm an East Side
> >                                   at               TYPE...
> >                                visi.com
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
regards
Steve Holden
-- 
Helping people meet their information needs with training and technology.
703 967 0887   sholden at bellatlantic.net   http://home.cox.rr.com/sholden/


From gregor at hoffleit.de  Tue Jun  6 05:45:51 2000
From: gregor at hoffleit.de (Gregor Hoffleit)
Date: Tue, 6 Jun 2000 11:45:51 +0200
Subject: [Zope-dev] [ANN] Zope at Linuxtag 2000 in Stuttgart
In-Reply-To: <4.3.1.0.20000602131515.00aae710@198.78.130.6>; from srichter@cbu.edu on Fri, Jun 02, 2000 at 01:51:33PM -0700
References: <4.3.1.0.20000602131515.00aae710@198.78.130.6>
Message-ID: <20000606114551.A1907@53b.hoffleit.de>

I'll try very hard to be able to show up at least one day at the LinuxTag;
I'm going to try to help out at the Debian booth as well.

How about I18N as a topic for one of the discussions ? I guess many European
sites and developers struggle with problems of I18N.

Another topic I'm very interested in is packaging and distribution of Zope
extensions: keywords are Distutils, packages, zexp's, a package/dependency
tool like Debian's APT, distribution etc. pp., cf. Simon Michael's recent
rant about ZWiki/ZWikiWeb.

Looking forward to meet you,

    Gregor

    

On Fri, Jun 02, 2000 at 01:51:33PM -0700, Stephan Richter wrote:
> The Linuxtag will be from June 29th (Thursday) to July 2nd (Sunday). It 
> will be in the Exhibition Center (Messegelaende) in Stuttgart, Germany.
> 
> I know we have little time to organize the conference (26 days). We have 
> one day to fill with speeches and discussions, so please let me know if 
> there are people willing to give some talks or want to have "Special 
> Interest Group" discussions, like E-commerce, XML or Portals in Zope. I 
> would also like to know, who would be able to come, so I can give the 
> Linuxtag staff some feedback.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/2b1e1fdb/attachment.sig>

From ben at co.and.co  Sat Jun 24 07:16:53 2000
From: ben at co.and.co (ben at co.and.co)
Date: Sat, 24 Jun 2000 11:16:53 GMT
Subject: Now for the flames...
References: <39549529.F5FFBD89@mindspring.com>
Message-ID: <FM055.14873$6t3.20011@afrodite.telenet-ops.be>

David Broadwell <dbroadwell at mindspring.com> wrote:
> Ok, now, what else am i doing that really isn't necessary?
>    (ie: point out my c hangups, please.)

Some minor optimisations:

On a Unix system:

$ find . -name '*.m3u' -exec cat {} \; > outfile
 
In python:

import os.path

def all_m3u(outputfile, dirname, names):
    ext = ''
    root = ''
    f = None
    for name in names:
        root, ext = os.path.splitext(name)
        if ext == '.m3u':
            f = open(os.path.join(dirname, name))
            outputfile.write(f.read())
            f.close()

def doit(dir, outputfile):
    f = open(outputfile, 'w')
    os.path.walk(dir, all_m3u, f)
    f.close()

if __name__ == '__main__':
    doit('.', 'outputfile')

Greetings,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From thomas at xs4all.net  Mon Jun 19 10:24:06 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 19 Jun 2000 16:24:06 +0200
Subject: Python vs Perl
In-Reply-To: <slrn8krpbf.gc.andyl@azaal.dircon.co.uk>; from andyl@azaal.dircon.co.uk on Mon, Jun 19, 2000 at 09:18:39AM +0000
References: <8ikai5$v3d$1@nnrp1.deja.com> <394ebe95.255648282@news.online.no> <slrn8krpbf.gc.andyl@azaal.dircon.co.uk>
Message-ID: <20000619162406.Z26436@xs4all.nl>

On Mon, Jun 19, 2000 at 09:18:39AM +0000, Andy Leighton wrote:

> >* Python is more object-oriented than Perl

> In what way?  Admittedly Python programmers tend to use OO code more than
> Perl programmers.  But I believe that is mainly due to culture.

That may be so, but then it's interesting to see that even the Perl culture
agree with that statement ;) Many a perl hacker has commented on the fact
that Perl OO concepts are 'bolted on'. Also, extending Python is a pleasant,
very object-oriented experience (heck, *altering* [C]Python is a pleasant
experience), which can't be said of Perl.

(Except perhaps in relation to, ooh, C or assembler ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From thomas at xs4all.nl  Mon Jun 26 03:41:02 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 07:41:02 GMT
Subject: Control cmp used with 'in'?
References: <52D55.2660$Gh6.168003@news-west.usenetserver.com> <ehD55.95136$WS3.791305@typhoon.we.rr.com>
Message-ID: <slrn8le281.2oh.thomas@centurion.xs4all.nl>

On Mon, 26 Jun 2000 07:05:46 GMT, Stephen Hansen
<stephen at cerebralmaelstrom.com> wrote:
>Matthew Cline <matt at nightrealms.com> wrote in message
>news:52D55.2660$Gh6.168003 at news-west.usenetserver.com...
>> Is there any way to control the function that is used during
>> an 'in' search?

>I believe its the __contains__ method.

In 1.6, yes. In 1.5.2, overloading 'in' is not possible, except by
overloading __getitem__ -- the 'in' implementation for user classes and most
builtin types simply __getsitems__ until it finds a match or reaches
IndexError. In 1.6, it's possible to override it in both and provide a more
efficient implementation. (See 'Objects/rangeobject.c' for an example ;-)

plug-plug-ly y'rs,
Thomas.


From morten at src.no  Thu Jun 22 06:23:50 2000
From: morten at src.no (Morten W. Petersen)
Date: Thu, 22 Jun 2000 12:23:50 +0200
Subject: Deleting mails with imapib
Message-ID: <Pine.LNX.4.21.0006221223060.22204-100000@lade.src.no>

The imaplib.IMAP connection doesn't seem to have a method that lets you
delete mails? only expunge etc..

-Morten

-------------------------------------------------------------
How to reply to email:
  http://home.sol.no/~vidaandr/news/OBSquoting.html (norsk)
  http://home.sol.no/~vidaandr/news/FAQquoting.html (english)

Who is the bigger fool? The fool or the fool who follows him?



From stephen at cerebralmaelstrom.com  Fri Jun  9 21:46:44 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 10 Jun 2000 01:46:44 GMT
Subject: RFC822 compliance in modern mail clients?
Message-ID: <86h05.51947$WS3.494817@typhoon.we.rr.com>

    This isn't entirely on-topic, but in the process of working on a lexical
analyzer for rfc822 (Yes, i know that the module rfc822 already has one),
i've been slowly reading over the standard and translating it into actually
Readable-By-Me reference notes, and I was wondering..

    Do any clients pay attention to headers  such as Resent-etc? Anytime
anyone's ever forwarded me something, its always seemed to come from them :)

    Just curious how much attention I should pay to it :)

--S




From helmit_world at hotmail.com  Tue Jun 27 10:09:09 2000
From: helmit_world at hotmail.com (Helmit)
Date: Tue, 27 Jun 2000 17:09:09 +0300
Subject: Books
Message-ID: <8jb1ec$9mh2@news.qualitynet.net>

Does anyone know any good python books for teaching beginers?




From Vladimir.Marangozov at inrialpes.fr  Thu Jun 29 17:38:03 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Thu, 29 Jun 2000 23:38:03 +0200
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <395A66C5.E9AC4F3E@roguewave.com> <slrn8lkta6.1v9.hzhu@rocket.knowledgetrack.com> <20000628151611.A2972@freei.com> <8F61C6D29gmcmhypernetcom@199.171.54.194> <ffcmlsc6p4lqohdd6tequdu918l0temns4@4ax.com> <395B6027.B47094CA@letterror.com> <8F6281530gmcmhypernetcom@199.171.54.194>
Message-ID: <395BC1BB.89F6ED06@inrialpes.fr>

Gordon McMillan wrote:
> 
> Their main feature is "ease of use", which can be pejoratively cast as
> "programmer laziness".

:-)

Gordon, why aren't you pointing all these people to the reference
reading about naming and binding of objects so that they refrain
submitting countless proposals on the subject without giving it
some more thought?

http://web.mit.edu/Saltzer/www/publications/nbo/nbo.html

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


From gregholmes at my-deja.com  Thu Jun 29 11:46:33 2000
From: gregholmes at my-deja.com (gregholmes at my-deja.com)
Date: Thu, 29 Jun 2000 15:46:33 GMT
Subject: [Tkinter] Centering a window in the screen ?
References: <962215194.1421627556@news.libertysurf.fr> <00dbb77f.cec66729@usw-ex0104-032.remarq.com>
Message-ID: <8jfr0i$ei7$1@nnrp1.deja.com>

  Richard Chamberlain
<richard_chamberlainNOriSPAM at ntlworld.com.invalid> wrote:
...
>
> I'm considering starting a Tkinter web site (including a more
> comprehensive FAQ) if anyone shows any interest.
>
You've got my vote - I'd sure be interested in such a website.

-Greg Holmes
gregholmes at SPAMMENOTmy-deja.com


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


From mkx at excite.com  Mon Jun 12 02:29:26 2000
From: mkx at excite.com (mkx at excite.com)
Date: Mon, 12 Jun 2000 02:29:26 -0400
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i1uv0$hki$0@216.39.141.21>
Message-ID: <1d09kssdshm96qgilpm7rabmibt2blmv5j@4ax.com>

On 12 Jun 2000 06:11:44 GMT, dana_booth <dana at oz.net> wrote:

>mec: So my question is this: Is Python a more "Windows-friendly" language?
>
>Okay, I couldn't leave this alone. :)
>
>I think "Windows-friendly" is relative. I know Windows only programmers who
>can out-Perl just about anyone. (FWIW, I'm in the UNIX crowd. :)

I should have replaced "Windows-friendly" with "less painful to
non-Unix users"...

I come from a DOS->NetWare->OS/2->Windows path. Due to the nature of
my local marketplace, you are either Windows, Unix, or unemployed. So
by default, I do Windows. It is not always fun, but it pays the bills
well.

At this point, Python represents a bit of sanity...I can use it for
many of my Windows projects, without having to be totally Windows
bound (VB, etc.) While most commercial projects have defined tools
where I must grin and bear it, some just allow me to "get the job
done", and I can choose the best tool for the job, and for my self.
So, to a degree, some of the attraction to Python is that can me "more
fun." And after a while, that can be important too.




From dfan at harmonixmusic.com  Fri Jun 16 13:59:25 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 16 Jun 2000 13:59:25 -0400
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com> <ud7ljln92.fsf@ctwd0143.fitlinxx.com> <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com>
Message-ID: <wkn1klmrr6.fsf@turangalila.harmonixmusic.com>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

| Oops!  You're right, as several people also pointed out to me in email.
| Actually I had this in .emacs and didn't realize it for years:
| 
| (setq-default tab-width 4)
| 
| I am not aware of serious adverse effect, though.
|
| [snip]
| 
| Wondering what other people will see when reading my programs ...

That is, in fact, the main problem with tabs.

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


From frey at weasel.bme.unc.edu  Thu Jun 15 08:14:50 2000
From: frey at weasel.bme.unc.edu (Eric Frey)
Date: 15 Jun 2000 12:14:50 GMT
Subject: Problem displaying Numeric Array as image
Message-ID: <8iahbq$5q2$1@news2.isis.unc.edu>

I a trying to display a floating point Numeric (NumPy) array as a
grayscale image via Tkinter.  The way I am trying to do it is via a
PhotoImage created via PIL. The problem is that the image I get looks like
it is a binary (black or white) image. Here is sample code

from Numeric import *
from Tkinter import *
import Image, ImageTk
#...
# im is a floating point array computed using numeric
bits=im.tostring()
pilim=Image.fromstring('F',(width, height), bits, "raw", 'F')
print pilim.format, pilim.size, pilim.mode

root=Tk()
pilphoto=ImageTk.PhotoImage(pilim)
l=Label(root,image=pilphoto,bd=0)
l.pack()
b=Button(root,text="Quit",command=root.quit)
b.pack()
root.mainloop()
---
I have tried lots of things including using astype to convert the im
to various other types (Int16, Int8, Int32) (and modifying the mode
strings in the Image.fromstring call. I have also tried setting the
maximum pixel value to various values using
im = (im/(max(im.flat))*NNN
where NNN is the value I want the maximum to be mapped to. This which
pixels are displayed as if there is some kind of thresholding going on.
What I really want is a way to display floating point pixels as a
grayscale/pseudocolor image in some kind of gui. Is Tkinter+PIL the way
to go? Would I be better of with PyGtk or some other toolkit? Thanks
for any advice.



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


From thomas at xs4all.net  Tue Jun 13 10:06:06 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 13 Jun 2000 16:06:06 +0200
Subject: importing module through variable
In-Reply-To: <8i5e9p$nm0$1@zingo.tninet.se>; from andre@beta.telenordia.se on Tue, Jun 13, 2000 at 03:51:25PM +0100
References: <8i5e9p$nm0$1@zingo.tninet.se>
Message-ID: <20000613160606.W20700@xs4all.nl>

On Tue, Jun 13, 2000 at 03:51:25PM +0100, Andr? Dahlqvist wrote:

> I have a variable which is assigned the name of a module. Is it possible
> to use this variable in an import statement? Someone told me I could use
> "exec("import %s" %var) and while this works I am not sure how I can
> access the module by it's name after that. Using something like 
> var.modfunc() won't do it since 'var' is just a string containing the 
> modules name.

You want the __import__ function:

module = __import__(var)

>>> print __import__.__doc__
__import__(name, globals, locals, fromlist) -> module

Import a module.  The globals are only used to determine the context;
they are not modified.  The locals are currently unused.  The fromlist
should be a list of names to emulate ``from name import ...'', or an
empty list to emulate ``import name''.
When importing a module from a package, note that __import__('A.B', ...)
returns package A when fromlist is empty, but its submodule B when


-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From r2d2 at mao.acc.umu.se  Thu Jun 22 04:10:31 2000
From: r2d2 at mao.acc.umu.se (Niklas Frykholm)
Date: 22 Jun 2000 08:10:31 GMT
Subject: Linux/NT python startup speed
Message-ID: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se>

I've been noticing a significant difference in the start-up speed of
Python on my NT machine and on a Linux system where I have an account.

On the Linux system

	python -c "print 'hello'"

times to about 0.1s.

However, on the NT system, the very same command takes 4.0 s.

Can someone explain this difference. (The processor on the Linux system is
certainly not 40 times faster.)

The startup speed on the NT system is a real problem. If you write a
simple system utility (such as wc) you don't want to wait 4 s for it
to start each time you run it.

// Niklas


From andre at UltraShell.Net  Wed Jun 28 06:52:18 2000
From: andre at UltraShell.Net (Andre van Straaten)
Date: Wed, 28 Jun 2000 10:52:18 GMT
Subject: adding mail attachments
References: <8ivdl4$90k$1@supernews.com>
Message-ID: <CNk65.16132$l_1.350676@news-east.usenetserver.com>

geoff munday <geoffm at nojunkmailqsiuk.com> wrote:
> OK, I  can send mail via the smtplib but am at a loss as how to attach files
> to the Email.
> Can anyone advise where to look for instructions or examples??

> Geoff

An example of the use of MimeWriter is on my Web site. Go "scripting".

I'll just add the line with the MIME version as suggested by Juergen
Herrmann in this thread.
As all my mail problems never complained about it, I wasn't aware that it
was missing. Thanks.

-- avs

Andre van Straaten
http://www.vanstraatensoft.com
______________________________________________
flames please to /dev/null at vanstraatensoft.com



From panse at muk.tu-darmstadt.de  Fri Jun  9 05:33:41 2000
From: panse at muk.tu-darmstadt.de (Frank Panse)
Date: Fri, 09 Jun 2000 11:33:41 +0200
Subject: pyc -> py can i transform it?
Message-ID: <3940B9F5.639FD84A@muk.tu-darmstadt.de>

Can I transform pyc files to py files?


From michael at stroeder.com  Wed Jun 14 04:36:49 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 14 Jun 2000 10:36:49 +0200
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com>
Message-ID: <39474421.E63F098B@stroeder.com>

Grant Edwards wrote:
> 
> Is there a
> handy list somewhere of which Linux distributions can be
> expected to have Tkinter installed?

S.u.S.E. comes with packages for Python, Tkinter and other handy
Python modules. Well, the person who installs it has to choose it in
the install application called YaST (series d).

The new Debian potato has also packages for Python and several
modules.

It might help if you provide some ready-to-install-packages for your
own software or find somebody who creates e.g. RPM- and/or Debian
packages because you can set dependencies for needed software in the
packages.

Ciao, Michael.


From moshez at math.huji.ac.il  Tue Jun 13 14:27:42 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 21:27:42 +0300 (IDT)
Subject: Two newbie questions
In-Reply-To: <39465D3B.3356DCC4@bucodi.com>
Message-ID: <Pine.GSO.4.10.10006132126390.21681-100000@sundial>

On Tue, 13 Jun 2000, BuCoDI-Rony Steelandt wrote:

> Don't know exactly what you mean, could it be that??
> st = "asdfasfasdf"
> l = []
> for c in st:
>     l.append(c)
> #process here
> #....
> #now get string back
> s=''
> for c in l:
>     s= s+ c

Don't do that! That's O(n^2) rather then O(n)....
See my other reply for a more efficient way.




From akuchlin at mems-exchange.org  Thu Jun  8 17:28:58 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Jun 2000 17:28:58 -0400
Subject: Obscure Exception message
References: <75F7304BB41CD411B06600A0C98414FC0A418B@ORSMSX54>
Message-ID: <3dr9a7yi91.fsf@amarok.cnri.reston.va.us>

"Daley, MarkX" <markx.daley at intel.com> writes:
> TypeError: string member test needs char left operand
> I have no idea what this is telling me, although the 'char' comment makes me
> wish I had a flamethrower!

Good point.  Hmm... how about:
TypeError: in <string> requires character as left operand

In the above message, <string> isn't the contents of the string
involved, but the actual literal '<string>'.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
If I tended toward frivolity as a boy, I am incorrigibly settled in it now.
  -- Robertson Davies, _The Diary of Samuel Marchbanks_


From aa8vb at yahoo.com  Wed Jun 28 08:10:28 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 28 Jun 2000 08:10:28 -0400
Subject: more python evangelising...
In-Reply-To: <Uj865.1905$iN5.408850@ptah.visi.com>; from ge@nowhere.none on
 Tue, Jun 27, 2000 at 08:41:24PM +0000
References: <B57E791A.264E%eableson@mac.com>
 <Uj865.1905$iN5.408850@ptah.visi.com>
Message-ID: <20000628081028.B228150@vislab.epa.gov>

Grant Edwards:
 |...then they could take lower profit, higher "social good" directions.
 |But that's not what stockholders want.  Stockholders (at least in the US)
 |only care about short-term profits. 

Not quite.  Be careful about including everyone in the US with your
statements.  Despite the media rage, serious folks invest in companies for
the long term.  CAGR is a big deal on the minds of many, and you don't get
that without compounding.  Of course, selling short-term kills this in a
taxable account.

And you might have heard of "ethical investing".  For now, I won't invest
in MSFT no matter what CAGR I can get from them.  If MSFT changes their
arrogant business attitude shafting customers and dealing out frequent
misinformation just to confuse consumers, then I might.

 |If there was a large enough group interested in investing in
 |socially-conscious software companies perhaps things would be
 |different.  Perhaps.

Well, consider.  MSFT was great when they could strong-arm the industry.
Not doing so good anymore; so maybe there is a correlation between
maximizing profits and treating your customers and suppliers well...

-- 
Randall Hopper
aa8vb at yahoo.com



From mvanier at endor.bbb.caltech.edu  Tue Jun 13 20:59:47 2000
From: mvanier at endor.bbb.caltech.edu (Michael Vanier)
Date: Tue, 13 Jun 2000 20:59:47 -0400
Subject: stupid thread question
References: <jl7lbu4jdf.fsf@endor.bbb.caltech.edu> <8F51D9DB3gmcmhypernetcom@199.171.54.154>
Message-ID: <642A954DD517D411B20C00508BCF23B00129698B@mail.sauder.com>

gmcm at hypernet.com (Gordon McMillan) writes:

> Michael Vanier: 
> 
> >I wrote a little python script using the thread module the other day.  I
> >wanted to be able to set a flag in the parent thread that would be read
> >in the child thread.  So I set a flag variable (an integer in this case)
> >in the parent, created the child thread, and then checked the value of
> >the flag every time through a loop in the child thread to see if it had
> >changed.  It turns out that the child thread apparently made a copy of
> >the flag, because when it changed in the parent it didn't change in the
> >child. 
> 
> >To fix this, I changed the flag to be a list with one integer element. 
> >Now, when I change the list element in the parent, the child sees it
> >too.  
> 
> 
> Let me guess. You did something like make "flag" a module level global, and 
> then did "from ... import *" in the other.

Even dumber than this; it was all in the same file.  I set the global in the
parent thread, and expected that the new thread would see the new value when
it was changed.

> 
> If you had referenced "module.flag", all would have worked. But that 
> stinks, too. Your thread should have a flag member, and the parent thread 
> should set it (through a setter, if you really want to be virtuous <wink>). 
> That way the flag becomes part of the thread's API, instead of a bass-
> ackwards dependency.
> 

Well, the thread wasn't an object; I just called a function and expected that
it would see the same data (see below).

> 
> > I gather that what this means is that there is a fundamental
> >distinction between atomic types like integers and reference types like
> >lists or dictionaries.  However, since python treats everything as a
> >reference this seems inconsistent to me.  So what I want to know is:
> >what is the cause of this behavior, and what is the justification for
> >it?  Or could it be a bug? 
> 
> Aahz is correct (as usual). It has to do with assignment and immutability. 
> To see assignment to any immutable (not in your global or local namespace), 
> you have to see the immutable through some larger context. Say, a module 
> namespace, or (as you did) through a container object.
> 
> - Gordon

So what you're saying is that a thread's namespace is not the same as the
global namespace of the parent.  This is puzzling, since that's what I
thought a thread was: effectively a new process which shares the parent's
namespace.  Is this in a FAQ somewhere?  Or am I just not getting it?

To further complicate things, here is a stripped-down version of the code
that actually works!

#! /usr/bin/env python

import os, thread

def thread_func():
    while 1:
        print "running child thread..."
        if stop:  
            print "child thread is done"
            break


stop = 0  
thread.start_new_thread(thread_func, ())
os.system("sleep 2")
stop = 1
print "parent thread is done"
os.system("sleep 2") # Wait for child thread to finish.
# end of test program

The original program has a Tk interface, which may account for the
difference.  Nevertheless, given what you've said, why does this work?

Mike


--------------------------------------------------------------
Mike Vanier	mvanier at bbb.caltech.edu
Department of Computation and Neural Systems, Caltech 216-76
GNU/Linux: We can't lose; we're on a mission from God.



From jsantaniello at my-deja.com  Tue Jun  6 00:16:10 2000
From: jsantaniello at my-deja.com (jsantaniello at my-deja.com)
Date: Tue, 06 Jun 2000 04:16:10 GMT
Subject: SSL and POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com> <etdog5fcx7a.fsf@w20-575-109.mit.edu>
Message-ID: <8hhttv$7r6$1@nnrp1.deja.com>

The cgi (which I have no control over) returns a nice html page that
says it can't process my request because of a "format or context error".

A simple test html page I made:
<html>
<body>
<form method=POST action=https://www.experian.com/cgi-bin/ccredit.cgi>
<input type=hidden name=0 value=1>
<input type=SUBMIT>
</form>
</body>
</html>

works from a browser but:

import httplib
h = httplib.HTTPS('www.experian.com')
h.putrequest('POST', '/cgi-bin/ccredit.cgi?0=1')
h.putheader('Accept', 'text/html')
h.putheader('Accept', 'text/plain')
h.endheaders()
errcode, errmsg, headers = h.getreply()
print errcode # Should be 200
f = h.getfile()
print f.read()
f.close()

from a module returns a different page with the "context" error
business. And it's not a cookie issue.

Am I doing something wrong with specifiing POST? Is there a better way
to do that?

And if I try with urllib:

>>> import urllib
>>> params = urllib.urlencode({'0': 1})
>>> f = urllib.urlopen("https://www.experian.com/cgi-bin/ccredit.cgi",
params)

I get a traceback:

Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.6/urllib.py", line 62, in urlopen
    return _urlopener.open(url, data)
  File "/usr/local/lib/python1.6/urllib.py", line 163, in open
    return getattr(self, name)(url, data)
TypeError: too many arguments; expected 2, got 3


Any ideas?

Thanks,

Joe



In article <etdog5fcx7a.fsf at w20-575-109.mit.edu>,
  Alex <cut_me_out at hotmail.com> wrote:
>
> What happens when you try to do a POST?
>
> Alex.
>


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


From matthias.huening at univie.ac.at  Mon Jun 26 12:16:07 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Mon, 26 Jun 2000 18:16:07 +0200
Subject: SMTP with Python 1.6
References: <EVH55.7839$pW4.206854@news1.rdc1.on.wave.home.com> <8j7n8d$4u64$1@www.univie.ac.at> <VqK55.1715$iN5.345609@ptah.visi.com>
Message-ID: <8j7vnb$1kue$1@www.univie.ac.at>

> Gosh, on all my computers, "localhost" is the address of a valid
> SMTP-server.  Are your computers all broken?
>

Because of this line in Colin's script:

> #!c:\python\python.exe

I assumed that he is on a Windows-computer. And that makes it unlikely that
localhost is a SMTP-server. (No, I am not going to discuss the question
whether Windows computers are broken by definition...)

Matthias




From hgg9140 at seanet.com  Mon Jun 12 18:44:16 2000
From: hgg9140 at seanet.com (Harry George)
Date: 12 Jun 2000 15:44:16 -0700
Subject: Testing for a file in python?
References: <39450724.231994953@nntp.interaccess.com>
Message-ID: <m3wvjusenz.fsf@wilma.george.org>

If you use pyperl, you get:

  import pyperl
  p=pyperl.PyPerl()
  ...
  if p.f(filename):
    ...

Similarly the vaious other file operators.

See: http://www.seanet.com/~hgg9140/comp/index.html

olczyk at interaccess.com (Thaddeus L. Olczyk) writes:

> How would you go about testing if a certain file exists in python?
> In most scripting languages you have the -f test:
> 
> if -f filename do
>    whatever_you_were_going_to_do
>    fi
> else do
>    print "Error" file not found
>    fi
> 
> stat doesn't work, it aborts if the file does not exist.

-- 
Harry George
hgg9140 at seanet.com


From jfr at genyinfo.com  Wed Jun 28 14:33:27 2000
From: jfr at genyinfo.com (Jean-Francois Richard)
Date: Wed, 28 Jun 2000 14:33:27 -0400
Subject: list.pop([i]) and list.remove(x) w/ for loops
References: <8jdfb6$ugh$1@netnews.upenn.edu>
Message-ID: <fyr65.5792$jT5.147787@wagner.videotron.net>

You can't modify a list while iterating over it since the result is
unexpected.
Try to do something this way instead:

   i = 0
   while i < len(olist):
      if olist[i].type == a:
         olist.remove(olist[i])
      else:
         i = i + 1

"David White" <dwhite2 at blue.seas.upenn.edu> wrote in message
news:8jdfb6$ugh$1 at netnews.upenn.edu...
> I was getting some unexpected behavior when using list.remove(x),
wondering if
> someone can tell me what's going on.
>
> I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].  I wan't
to
> delete all the a objects out of the list while preserving the order of the
b
> objects.  My code looks like this:
>
> for o in olist:
>     if (o.type == a):
>        olist.remove(o)
>
> Now for some reason I get a left over a object that wasn't removed.
Anyone
> know why?  Using python 1.5.2.
>
> thanks,
> david




From Jonathan_Epstein at nih.gov  Fri Jun 23 12:45:31 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Fri, 23 Jun 2000 12:45:31 -0400
Subject: win32pdutil.FindPerformanceAttributesByName slowness
Message-ID: <3953942B.303A5AE9@nih.gov>

Hi,

I'm running the sample killProcName.py which is provided with the
Windows' Python 1.52 (using Pythonwin build 129 and patches, from Mark
Hammond's web site).

There are 33 processes as reported by the Windows Task Manager.  I'm
running W2K on a 650 Mhz laptop with 256MB of RAM.

I have profiled killProcName.py, and it is consistently taking its call
to FindPerformanceAttributesByName 28 seconds (!) to run.

I saw a reference to a related problem, but no posted solution:
 
http://x59.deja.com/getdoc.xp?AN=589911330&CONTEXT=961778125.2116812828&hitnum=1

Any suggestions as to why this is occurring, or how I can track it down?

TIA,

-Jonathan


From stevena at neosynapse.net  Fri Jun 30 15:19:02 2000
From: stevena at neosynapse.net (Steven D. Arnold)
Date: Fri, 30 Jun 2000 15:19:02 -0400
Subject: Web server executing Python scripts
In-Reply-To: <2B1262E83448D211AE4B00A0C9D61B030130125E@msgeuro1.creo.be>
Message-ID: <4.3.1.2.20000630151218.021a7858@phear.dementian.com>

At 08:24 PM 6/30/2000 +0200, Pieter Claerhout wrote:

>What do you think would be the best idea?
>- Use whatever webbrowser with CGI-scripting

I would use apache and either CGI or mod-python, more likely the latter.


>- Use MS IIS with ASP (and Python)

IIS is quite probably the most insecure server on the Internet.  In 
addition, I don't like the closed nature of the system, I don't trust MS to 
introduce the features I want/need (where I can introduce them myself in 
Apache), and I don't want to depend on MS's support services to solve 
problems that only they can fix because only they have the source.


>- Use Zope to build the whole system?

I don't know enough about Zope to give an answer from personal experience 
here.  I've heard many good things about Zope; I've also heard that the 
docs are poor and the learning curve steep.  I have faith in the speed and 
flexibility of Apache, where (to me) Zope is an unknown in these regards; 
are there any benchmarks?


steve


--
Steven D. Arnold           Que quiero sera         stevena at neosynapse.net
"If you have  built castles in the air, your work  need not be lost; that
is where they should be.  Now put the foundations under them!" -- Thoreau




From wjdandreta at worldnet.att.net  Wed Jun 28 11:32:51 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 28 Jun 2000 15:32:51 GMT
Subject: float, '%.2f' help... please!
References: <39596682.108805343@news1.on.sympatico.ca>
Message-ID: <DUo65.13046$kK.348843@bgtnsc05-news.ops.worldnet.att.net>

Hi chiba,

Here is a method I sometimes use to do that.

roundedPriceTimes100 = int((price+.005)*100))
roundedPrice = roundedPriceTimes100[:-2]+'.'+roundedPriceTimes100[-2:]

chibaA at TinterlogD.Tcom wrote in message
<39596682.108805343 at news1.on.sympatico.ca>...
>Hi,
>
>I'm just kinda stuck for answers...  I need to display a price -
>rounded up to the proper cent.  I was using this line:
>
>print '%.2f' % (price+0.005)
>
>Thinking that it would just chop off the float at the second decimal.
>But when I had a value of 0.00 for price, it seemed to round up.  To
>be sure, I wanted to find a method of printing that would just chop
>off any decimals after 2 - and NOT round up or down.  Is there such a
>method?
>
>Thanks!




From cut_me_out at hotmail.com  Tue Jun 27 17:42:36 2000
From: cut_me_out at hotmail.com (Alex)
Date: 27 Jun 2000 17:42:36 -0400
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com>
Message-ID: <etdd7l2vm0j.fsf@w20-575-116.mit.edu>

It seems like you could get what you want with this sort of code:

while (whatever):
    new_forward = whatever
    l = forward[domain] = forward.get (domain, [])
    l.append (new_forward)

Alex.


From aahz at netcom.com  Thu Jun 15 13:06:14 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jun 2000 17:06:14 GMT
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com>
Message-ID: <8ib2e6$ens$1@slb1.atl.mindspring.net>

In article <8ib1bs$1pv at autodesk.autodesk.com>,
Akira Kiyomiya <akira.kiyomiya at autodesk.com> wrote:
>
>I wonder if Python runs faster than Perl.

Python almost never runs "faster" than Perl.  And there are some things
for which Perl is indisputably faster than Python (such as regular
expressions on a text file).  Where you gain in Python, though, is that
it's a much cleaner language, so you stand a greater chance of writing
an efficient algorithm in the first place -- and faster algorithms are
what get you faster programs.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From MarkH at ActiveState.com  Fri Jun 30 01:07:56 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 30 Jun 2000 05:07:56 GMT
Subject: ExitWindowsEx()
References: <8jg1mk$e1s$1@nnrp2.deja.com>
Message-ID: <MWV65.1839$Tb7.13008@news-server.bigpond.net.au>

I doubt you want this:
> win32api.ExitWindowsEx(4 Or 2,0)

ie:
>>> 4 or 2
4

You probably mean this:

>>> 4 | 2
6

I'm afraid I have no idea if that change will force the reboot!

Mark.





From amk1 at erols.com  Wed Jun  7 21:35:01 2000
From: amk1 at erols.com (A.M. Kuchling)
Date: Wed, 7 Jun 2000 21:35:01 -0400
Subject: ANNOUNCE: early alpha Quixote Web development tool
Message-ID: <200006080135.VAA04781@mira.erols.com>

The MEMS Exchange gang has been slowly working on a new framework for
writing Web apps.  The Web page for it is
http://www.mems-exchange.org/software/python/quixote/ .

The basic idea is that a Web application is written as a Python
package, and URLs are then treated as paths walking down into the
package.  Objects, functions, and methods are accessible through URLs
if they're listed in an __exports__ attribute of the object.
(Obviously this is inspired by Zope's object publishing.)  

Templating is done in Python Templating Language (PTL), which tries to
stick close to Python's statements and semantics:

<? template form(request, response) ?>
  <? block ?>
import string
from mems.ui import standard
<?/block?>

  <? standard.header(request, response, "Send Feedback") ?>

  <!-- Loop -->
  <? for item in compute_list_value() ?>
     <p> <? item.attribute ?>
  <? /for ?>

<? /template ?>

Through an import hook, if you put the above in a file called page.ptl
you can then do "import page ; string = page.form(...)", just as if
the template was a regular Python function.

Your Web application's package might look like this:
webapp/                         # Root of package
        __init__.py
        module.py (defining functions f1, f2, ...)
        pages.ptl (defining templates t1, t2, ...)

A URL like http://wherever/cgi-bin/quixote.cgi/module/f1 or
.../pages/would call function f1()

Quixote is still only partially finished, and we need to experiment
with it and figure out if it's a useful approach.  For example, not
every statement in PTL has been implemented yet, it needs to write
.ptlc files analogous to .pyc files, &c.

Anyway, we'd welcome opinions and suggestions about the design
(posting them to c.l.python is fine), and if you want to poke around
in the code, be our guest.  However, if you want to start actually
using Quixote for real work at this point, you're extremely brave (and
rather foolhardy); it might well get changed in incompatible ways in
the future.

--amk




From rupole at compaq.net  Tue Jun 13 17:59:27 2000
From: rupole at compaq.net (Roger Upole)
Date: Tue, 13 Jun 2000 21:59:27 GMT
Subject: Python and Windows Scripting Host
References: <9Ov15.1411$vM.56343@typhoon1.ba-dsg.net>
Message-ID: <39y15.839$Zh1.63369@news.easynews.com>

The Wscript object is not referenced directly as a Python variable.
You create it via win32com.client.Dispatch:
    import win32com.client
    wsh=win32com.client.Dispatch('Wscript.Shell')
    wnt =win32com.client.Dispatch('Wscript.Network')
    ...
Then you can use the their functions as
    wsh.somefunction()
    wnt.somefunction()
    ...
>From Pythonwin, if you first run the makepy utility from the tools menu
and select Windows Scripting Host, autocomplete will magically show
you what functions are available.
       HTH
                Roger Upole

"Jim Bower" <jbower at lims.com> wrote in message
news:9Ov15.1411$vM.56343 at typhoon1.ba-dsg.net...
> Just getting started with Python. Downloaded the basic distribution and
the
> Win32All distribution. Python appears to register OK (ran pyscript.py
> separately just to verify).
>
> However, I am getting "NameError: wscript" anywhere the wscript object is
> mentioned. Have I missed something in the install?
>
> Jim Bower jbower at lims.com
>
>
>




From daniel at visionart.com  Wed Jun 14 13:20:47 2000
From: daniel at visionart.com (Daniel Kramer)
Date: Wed, 14 Jun 2000 17:20:47 GMT
Subject: tcl/tk vs. python/Tkinter - Example attached
References: <3946B170.17A2630@visionart.com> <DCO15.563$fb5.9835@news2-win.server.ntlworld.com>
Message-ID: <P9P15.540$3E6.17373@news1.alsv1.occa.home.com>

"richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote in message
news:DCO15.563$fb5.9835 at news2-win.server.ntlworld.com...
> Hi Daniel,
>
> What platform are you working on?
>
> On my win98 machine I don't see any difference in speed (except python
> starts up very slightly slower).
>
> Other than that the readability of the python is the real clincher.
>
> Richard




Hmm.. interesting.  I also got a second response to my email stating
basically the same thing.  I am running on Win2k and I also tried it at work
on an SGI Octane.  I notice it less on the Octane but it still feels
different.

For me when I move a node around really fast it feels slightly choppy
compared to the tcl ver.. it's still responsive and it is a subtle
difference.  In the tcl ver the node is very fluid.. right there with you..
in tkinter there are these little laggy bumps in the motion... it stutters
slightly.

There is no question that the python code is much clearer.. your always
jumping though alot of hoops to get things formatted correctly in tcl I
find.  I was surprised that I liked coding TK more in python then tcl.

I'd like to do a benchmark but that would rely on doing a loop on the move
method and timing it rather then just "feeling" what it's like moving with
the mouse.. I just don't know if that would be an accurate benchmark since
it may only test the speed of the loop between the 2 languages and not the
tk move command itself.  I guess I could do the loop in both lang w/out the
move and with the move to isolate the move cmd timing.. hmm.

=Daniel



> Daniel Kramer <daniel at visionart.com> wrote in message
> news:3946B170.17A2630 at visionart.com...
> >
> > I have just started to use python this last week.. I wanted to port a
> > tcl/tk app to it and see how I like working with it.  As far as
> > programming with it I like it more.. but just as I got started on my app
> > I noticed some speed differences in TK.  The tcl/tk app is much more
> > responsive then the python/Tkinter app.
> >
> > I have put together a simple example done both ways.. it's attached to
> > this mail.
> >
> > Within the tar you will find:
> >
> > canvas.tcl - This is just a simple tcl/tk prog.. 2 rectangles are added
> > to a canvas and you can move them around with the mouse.
> >
> > canvas.py - This is the same thing written in python/Tkinter.
> > nodeUI.py - This contains most of the classes/functions to drive the
> > canvas.py.. it should be in the same dir as canvas.py
> >
> >
> > Since I am new to this lang I was just wondering if there was a way to
> > get better python/TK performance... if anyone could try out the examples
> > and let me know.. for me the tcl/tk one feels much more responsive but I
> > don't have any hard numbers to prove that.
> >
> > I have been reading that wxWindows is the much faster solution.. but I
> > am very familiar with TK and it's plenty fast for what I need.. but I
> > worry that the Tkinter ver won't be fast enough once I add alot more to
> > it.
> >
> >
> > -Daniel
> >
>




From 12klat at sightreader.com  Tue Jun  6 18:49:53 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 15:49:53 -0700
Subject: operator overloading
References: <f3trf8.08.ln@kazan> <0bbb01bfcf2a$bc5c39c0$1906a8c0@fc.fenx.com>
		<393C2386.307A160D@sightreader.com> <20000605235010.C469@xs4all.nl>
		<8hj2f9$1rn$1@nnrp1.deja.com> <oqwvk2df1q.fsf@titan.progiciels-bpi.ca>
		<8hjb8c$96f$1@nnrp1.deja.com> <oqog5eboig.fsf@titan.progiciels-bpi.ca>
Message-ID: <393D8011.93CAA037@sightreader.com>

Comparisons are evaluated from left to right:
   "a>b>c" is the same as "(a>b) and (b>c)"

Operator functions are searched from left to right.
  "a>b" looks for a.__cmp__
   If not found, (or a type exception is thrown), it looks for b.__cmp__.

>>> class A:
...  def __cmp__(self,b):
...   print 'A.__cmp__'
...   return 0
...
>>> class B:
...  def __cmp__(self,b):
...   print 'B.__cmp__'
...   return 0
...
>>> a = A()
>>> b = B()
>>> a>b
A.__cmp__
0
>>> b<a
B.__cmp__
0
>>> a < b < "vorpal bunny"
A.__cmp__
0
>>> a < "vorpal bunny" < b
A.__cmp__
0
>>> "vorpal bunny" < b < a
B.__cmp__
0

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

Fran?ois Pinard wrote:

> david_ullrich at my-deja.com ?crit:
>
> > > If one wants tools like `sort()' to work properly, `__cmp__' should
> > > establish a proper order relation, yes.  Otherwise, rather cryptic
> > > bugs might pop out.
>
> > If so this should probably be mentioned in the docs.
>
> _If_ the documentation does not make it clear, it should indeed.  Could
> someone check and, if the documentation is improper, report it appropriately?
>
> --
> Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/d3a1abcd/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/d3a1abcd/attachment.vcf>

From bjorn at roguewave.com  Thu Jun 29 13:06:34 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 29 Jun 2000 11:06:34 -0600
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com>
Message-ID: <395B821A.9C835D32@roguewave.com>

Eric Lee Green wrote:
> 
> Gordon McMillan wrote:
> >
> > Eric Lee Green wrote:
> >
> > >Steve Juranich wrote:
> > >> I am *brand* new to python (as of Sunday), and I just got to the
> > >> section in the tutorial about classes.  I was wondering why there
> > >> really isn't such an idea as a "private" member of classes?
> > >
> > >Mostly because Python doesn't "really" have classes. What Python
> > >"really" has are hash tables whose entries are objects (either methods
> > >or other data objects), and some syntactic sugar [...]
> >
> > By this logic, all C++ has is pointers and some syntactic sugar.
> 
> Exactly :-).
> 
> C++ is syntactic sugar for "C". As should be obvious from the fact that most
> early C++ compilers generated "C" as their output :-).

Non Seqitur.  Many compilers for a variety of languages (including
Python) produce C as their output.  That doesn't mean they are syntactic
sugar for C, just that C is close enough to assembly that you're not
going to lose much performance by using it as a target language.

-- bjorn



From corporate at cpsgroupinc.net  Tue Jun 20 11:51:18 2000
From: corporate at cpsgroupinc.net (CPS Group, Inc)
Date: Tue, 20 Jun 2000 11:51:18 -0400 (EDT)
Subject: Profitable opportunity - Join our affiliate program.....
Message-ID: <20000620155118.30FF71CF52@dinsdale.python.org>

Hello

Recently, I visited your web site at http://www.python.org:80/doc/FAQ.html & wanted to compliment you on it.
I would like to inform you, of what can turn out to be an extremely profitable 
venture for both of our companies.  "The CPS Group Inc. Affiliate Program".  
Our Affiliate Program is FREE to join, can instantly generate a new and ongoing 
stream of income without any cost or obligation on your part.  
Before I get into the details of the program, let me give you a brief backround 
of our company.

CPS Group Inc is a leading developer of total Internet Payment Solutions.  
Our services provide new and existing businesses with their own online merchant 
processing system, which allows them to accept all major credit cards, personal 
& business checks online right from their own web site.  We provide both the 
online interface (real time and virtual terminal) as well as the merchant account, 
so that your clients have everything they need under one roof to begin accepting 
credit cards & checks from their web site.  To get better details on the services 
we provide, you can visit us at www.cpsgroupinc.net/ecommerce.htm.  

So here is how the program works.  You simply recommend our services to your web 
site visitors by placing a banner on your web site that we provide you with or 
simply by giving them our toll free number. If they sign up for an e commerce 
merchant account from us and gets "approved", you will get paid a referral commission.  
It's that simple! 

We handle all customer questions, requests and application processing. 

I invite you to visit our Affiliate site www.cpsgroupinc.net/affiliates.htm to get 
all the details on the program. There is absolutely nothing to lose, this is 
truly a WIN WIN situation.  Working together, we can both earn and enjoy the 
substantial profits that this program has to offer.  I look forward to a long, 
mutually profitable relationship with you. 

Sincerely,
Anthony Santiago
Affiliate Program Coordinator
asantiago at cpsgroupinc.net
(800) 933-0064






From kens at sightreader.com  Thu Jun 15 23:07:22 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:07:22 -0700
Subject: how does python invoke an external C function??
References: <3949843F.37B5C997@esec.com.au>
Message-ID: <394999EA.9045ED56@sightreader.com>

That's the easy part.  Same as calling a python function.  If your
function is called "spam" in an extension module called "eggs", you
import eggs and call eggs.spam.

>>> import eggs
>>> eggs.spam()

Probably the question you should be asking is: "How do I write an
external C function?"

I refer you to the "extending and embedding" section of your favorite
python book.  You are interested in "extending" which means "writing a
python module in C".  Warning, this is a somewhat advanced topic, but
easy once you get the hang of it.

Sam Wun wrote: [how does python invoke an external C function?]

> thansk
> Sam.
>
> --
> Sam Wun                            Firewalls / Security
> OneGuard Senior Software Engineer  Electronic Commerce
> eSec Limited                       Phone: +61 3 83715376
> mailto:swun at eSec.com.au            C++/JAVA/UNIX/OOP/OOD
>
--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/0c191dc2/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/0c191dc2/attachment.vcf>

From loredo at spacenet.tn.cornell.edu  Thu Jun 15 15:54:16 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 15 Jun 2000 15:54:16 -0400
Subject: Where are the math functions?
References: <owen-97F32C.10015712062000@news.u.washington.edu> <memo.20000614223200.31901D@digitig.compulink.co.uk> <20000615134357.A1087@better.net>
Message-ID: <39493468.80FB5FF4@spacenet.tn.cornell.edu>

Howdy-

Below are some pure Python special functions, some adapted from
Numerical Recipes, some from formulas in standard references.  They are
mostly related to the Gamma function.  This
is old stuff, practically my first Python, so don't expect much!  8-)
I didn't know about cephes at the time, and mostly use cephes now.
But these have the virtue of being pure python, so they may be helpful
to you if you can't compile cephes on your platform.

I've done some other Python translations of NR routines, soon
to appear on a web page.  Others are already on the web:

http://www.python.org/topics/scicomp/recipes_in_python.html

Peace,
Tom loredo

PS: One of the Numerical Recipes coauthors is in the office next
door, and knows I've been up to this stuff.  When I first told him
I'd been translating some NR routines into Python, his response
was, "Why?!"  8-)

##### sp_funcs.py --- Not too thoroughly tested as yet!
from math import *
import Numeric
N = Numeric

# Exceptions:

max_iters = 'Too many iterations: '

#============= Globals ===============

rt2 = sqrt(2.)
gammln_cof = N.array([76.18009173, -86.50532033, 24.01409822,
	-1.231739516e0, 0.120858003e-2, -0.536382e-5])
gammln_stp = 2.50662827465

#============= Gamma, Incomplete Gamma ===========

def gammln(xx):
	"""Logarithm of the gamma function."""
	global gammln_cof, gammln_stp
	x = xx - 1.
	tmp = x + 5.5
	tmp = (x + 0.5)*log(tmp) - tmp
	ser = 1.
	for j in range(6):
		x = x + 1.
		ser = ser + gammln_cof[j]/x
	return tmp + log(gammln_stp*ser)
	
def gser(a, x, itmax=700, eps=3.e-7):
	"""Series approx'n to the incomplete gamma function."""
	gln = gammln(a)
	if (x < 0.):
		raise bad_arg, x
	if (x == 0.):
		return(0.)
	ap = a
	sum = 1. / a
	delta = sum
	n = 1
	while n <= itmax:
		ap = ap + 1.
		delta = delta * x / ap
		sum = sum + delta
		if (abs(delta) < abs(sum)*eps):
			return (sum * exp(-x + a*log(x) - gln), gln)
		n = n + 1
	raise max_iters, str((abs(delta), abs(sum)*eps))


def gcf(a, x, itmax=200, eps=3.e-7):
	"""Continued fraction approx'n of the incomplete gamma function."""
	gln = gammln(a)
	gold = 0.
	a0 = 1.
	a1 = x
	b0 = 0.
	b1 = 1.
	fac = 1.
	n = 1
	while n <= itmax:
		an = n
		ana = an - a
		a0 = (a1 + a0*ana)*fac
		b0 = (b1 + b0*ana)*fac
		anf = an*fac
		a1 = x*a0 + anf*a1
		b1 = x*b0 + anf*b1
		if (a1 != 0.):
			fac = 1. / a1
			g = b1*fac
			if (abs((g-gold)/g) < eps):
				return (g*exp(-x+a*log(x)-gln), gln)
			gold = g
		n = n + 1
	raise max_iters, str(abs((g-gold)/g))


def gammp(a, x):
	"""Incomplete gamma function."""
	if (x < 0. or a <= 0.):
		raise ValueError, (a, x)
	if (x < a+1.):
		return gser(a,x)[0]
	else:
		return 1.-gcf(a,x)[0]

def gammq(a, x):
	"""Incomplete gamma function."""
	if (x < 0. or a <= 0.):
		raise ValueError, repr((a, x))
	if (x < a+1.):
		return 1.-gser(a,x)[0]
	else:
		return gcf(a,x)[0]

#======== Error function, normal CDF and inverse ================

def ncdf_inv(p):
	"""Inverse of the normal CDF."""
	c0 = 2.515517
	c1 = 0.802853
	c2 = 0.010328
	d1 = 1.432788
	d2 = 0.189269
	d3 = 0.001308
	
	sign = -1.
	if (p > 0.5):
		sign = 1.
		p = 1. - p
	arg = -2.*log(p)
	t = sqrt(arg)
	g = t - (c0 + t*(c1 + t*c2)) / (1. + t*(d1 + t*(d2 + t*d3)))
	return sign*g


def erfcc(x):
	"""Complementary error function."""
	z = abs(x)
	t = 1. / (1. + 0.5*z)
	r = t * exp(-z*z-1.26551223+t*(1.00002368+t*(.37409196+
		t*(.09678418+t*(-.18628806+t*(.27886807+
		t*(-1.13520398+t*(1.48851587+t*(-.82215223+
		t*.17087277)))))))))
	if (x >= 0.):
		return r
	else:
		return 2. - r


def ncdf(x):
	"""Cumulative normal dist'n."""
	global rt2
	return 1. - 0.5*erfcc(x/rt2)


def ncdf_sig (nsig):
	"""Cummulative normal dist'n inside nsig sigmas.
	ncdf_sig = 1 - 2 * (upper tail) = 1 - erfc(sigfac/rt(2))"""
	global rt2
	return 1. - erfcc(nsig/rt2)

#=============== Chi squared dist'n ==============

def pchisq(chisq, nu):
	"""Lower tail area of the chi**2 dist'n with nu dof.
	Note that chisq is *not* the reduced chis**2!"""
	hnu = 0.5 * nu
	hchi = 0.5 * chisq
	return gammp(hnu, hchi)

def qchisq(chisq, nu):
	"""Upper tail area of the chi**2 dist'n with nu dof.
	Note that chisq is *not* the reduced chis**2!"""
	hnu = 0.5 * nu
	hchi = 0.5 * chisq
	return gammq(hnu, hchi)


def chisq_crit(nu, p, tol=1.e-5):
	"""Critical chi**2 with lower tail area of p for nu dof."""

	#  For the first guess, use the assyptotic normal limit of the
	# chi**2 distribution:  chi**2 ~ N(nu,sqrt(2*nu)).
	chi = nu + ncdf_inv(p)*sqrt(2.*nu)
	pcur = pchisq(chi,nu)
	
	# Now do a Newton-Raphson loop...
	while 1:
		dfdc = (pchisq(1.001*chi,nu) - pcur) / (0.001*chi)
		chi = chi - (pcur - p)/dfdc
		pcur = pchisq(chi,nu)
		if (abs(pcur-p) <= tol):
			return chi

# Allow it to run as a script	
if __name__ == "__main__":
	print 'gser: ', gser(5., 5.)
	print 'gcf:  ', gcf(5., 7.)
	print 'gammp, gammq: ', gammp(5.,5.), gammq(5.,5.)
	print 'erfcc: ', erfcc(.5)
	print 'ncdf_inv: ', ncdf_inv(0.977), ncdf_inv(0.5)
	print 'ncdf: ', ncdf(1.)
	print 'ncdf_sig: ', ncdf_sig(2.)
	print 'q & p chisq: ', qchisq(4.,1), pchisq(4.,1)
	print 'chisq_crit: ', chisq_crit(1.,0.954)


From thomas at xs4all.nl  Wed Jun 28 06:41:43 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 28 Jun 2000 10:41:43 GMT
Subject: Requiring a certain version of interpreter?
References: <Lqj65.4362$h8.209230@news-west.usenetserver.com> <s2k65.1080$MZ3.2911@newsfeed.slurp.net>
Message-ID: <slrn8ljlin.l8b.thomas@centurion.xs4all.nl>

On Wed, 28 Jun 2000 06:08:36 -0500, Chris Armstrong <punck at PenguinMints.cx>
wrote:
>In article <Lqj65.4362$h8.209230 at news-west.usenetserver.com>, Matthew
>Cline <matt at nightrealms.com> wrote:
>> Is there any builtin feature, or common module, that handles requiring
>> at least a certain version of the Python interpreter to be used for a
>> script?  Perl has 'require' for this purpose.

>>>> import sys 
>>>> print sys.version
>1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian GNU/Linux)]

>There may a way to just get the version number only, but if nothing else this 
>will work.

Python 1.6 has sys.version_info, which is a tuple with version info:

>>> sys.version_info
(1, 6, 0, 'alpha', 2)

But no, there is no builtin way to 'require' a version. Most new features
break cleanly with old versions of Python (like keyword arguments) so the
usual remedy is to say 'Python 1.5 required!' near the top of the docs, and
point people there when they complain it doesn't work ;-)

And for the bug-prevention scheme, you can use 'if sys.version_info[1] < 6:'
in Python 1.6 (which will, of course, be bugfree.)

Thomas.


From jschmitt at vmlabs.com  Wed Jun 21 14:15:14 2000
From: jschmitt at vmlabs.com (John Schmitt)
Date: Wed, 21 Jun 2000 11:15:14 -0700
Subject: Python complements C++ for productivity
In-Reply-To: <rok8fkm5jc.fsf@gage.i-did-not-set--mail-host-address--so-shoot-me>
Message-ID: <NEBBIOGBDMDPEANLFKGHKEGJCEAA.jschmitt@vmlabs.com>

When I first opened Pythonwin, I used the Python/Perl phrasebook.  Very
useful to me.  I don't know the STL very well (I don't even know what the
SWIG is) so I couldn't write one, but how hard would be to write a Python
<-> C++/STL phrasebook?  I imagine it would be useful for C++ programmers
who want to move to Python and useful to Python programmers who have to move
some stuff to C++.

John

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Don Bashford
> Sent: Tuesday, June 20, 2000 2:01 PM
> To: python-list at python.org
> Subject: Python complements C++ for productivity
>
>
> I've been using C++ for about 10 years, and Python less than a year.
> I recently had the experience of implementing an algorithm for
> Delaunay triangulation in both Python and C++.  I was working from my
> reading of someone's Ph.D. thesis that had parts of the algorithm in
> pseudo-code, and part in English.
>
> I started out in Python.  The work was pleasant and went fairly
> quickly.  It was pleasing that a lot of the Python I wrote ended up
> looking very close to the pseudo-code in the thesis.  The interpreted
> nature of Python was very helpful, since it made it easier to inspect
> results as I went along.  I also benefited from a Python graphics
> package written by a colleague that hepled me visualize what the
> algorithm was doing.  I think in the early phases, development in
> Python was faster than it would have been in C++.
>
> But the resulting code was slow---about 100 times slower than the
> thesis author's C code.  I started re-implementing in C++.  The
> similarity of the basic "class" ideas in the two languages made the
> re-write very easy and fast.  In some cases, I just dropped the Python
> code into the C++ file and edited it.   In the translation process,
> the stricter type rules of C++ helped me find some bugs that I might
> never have found in the Python version, and it also resulted in
> improved design.
>
> The only real unpleasentness in the whole process was when I had my
> implementation partly in Python (higher level) and partly in C++
> (lower, more number crunching levels).  Then I was using SWIG to
> "wrap" the C++ parts for use from Python, but getting proper handling
> of translations like Python lists to C++ containers was quite a pain
> (although I learned a lot about SWIG typemaps in the process).
>
> In short, I found Python to be a great prototyping language, with
> pretty easy movement to C++, which is a great language when you want
> strict type checking, data hiding and performance.
> --
> http://www.python.org/mailman/listinfo/python-list




From embed at NOSPAM.geocities.com  Fri Jun  2 14:56:05 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Fri, 2 Jun 2000 14:56:05 -0400
Subject: historical python releases?
Message-ID: <wvTZ4.42229$HG1.716082@nnrp1.uunet.ca>

Is there a Python 1.0 or Pre-1.0 source code anywhere publically available?
I'm interested to learn the evolution of Python from the earliest source
code.

Warren Postma




From hzhu at rocket.knowledgetrack.com  Fri Jun  9 15:16:28 2000
From: hzhu at rocket.knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 09 Jun 2000 19:16:28 GMT
Subject: PATCH: Augmented assignment
References: <20000609163937.D20700@xs4all.nl> <39442c45.54923673@news.telus.net>
Message-ID: <slrn8k2gq6.pi.hzhu@rocket.knowledgetrack.com>

>>As promised, here is my patch to add augmented assignment (+=, -= and
>>family) to Python. It's still Work In Progress, and probably requires a
>>rewrite of some magnitude before Guido will accept it.
>>

Greate!  I'll use them in the MatPy package as well.

BTW, maybe __add_eq__ would be a better name than __add_ad__, as it
describes the symbols instead of their meaning.


>Why would the Python community want this?  It obfuscates code, in as much
>that newbies aren't likely to intuite what the symbols mean.
> 

Which of the following is clearer to a newbie?  

dictionary[key][index] = dictionry[key][index] + 1
dictionary[key][index] += 1

The first happens to have a typo as well. Besides, the second version is
always faster. For many classes the speed gain can be enormous, as obj=obj+1
is likely to involve a "partitially deep" copy.

The new patch gives ability to define __add_ad__ independently of __init__
and __add__ so that obj+=1 takes advantage of potential efficiency gain.


Huaiyu


From jah at mail.ru  Sun Jun 18 05:59:52 2000
From: jah at mail.ru (Igor E. Poteryaev)
Date: Sun, 18 Jun 2000 15:59:52 +0600
Subject: Advice on a Learning strategy
References: <39490B9E.950C2576@home.com>
Message-ID: <394C9D98.2E5B9FB@mail.ru>

Do you aware of Python for Delphi components ?
If no, look at
http://www.multimania.com/marat/ 



bo Vandenberg wrote:
> 
> I'm trying to put together a basic toolkit for general consultant
> programming in a win32 environment. (I know this is vague but I have to
> be flexible at this point.
> 
> My thoughts are to work in Python and Delphi.
> 
> Python is going to be for all the client side customization and Delphi
> will be my engine to do what doesn't need to be interpreted. I'm
> considerring C++ but I don't know if I want the learning curve.
> 
> VB vs Delphi    Delphi programs seem more self contained than VB. I also
> like the possibility of going to Kylix in the future. If I write it in
> Delphi I think the client is less likely to accidentally break it (no
> promises for Microsoft though).
> 
> I would like to use Tkinter but I think speed concerns will have me move
> to wxwindows and pythonwin.
> 
> Has anyone gone this route, or a compatible route, before who would be
> willing to share some advice???
> 
> These are my first programming languages.



From Michaell.Taylor at reis.com  Thu Jun 15 11:24:05 2000
From: Michaell.Taylor at reis.com (Michaell Taylor)
Date: Thu, 15 Jun 2000 11:24:05 -0400
Subject: Testing for file open
Message-ID: <NDBBJMANPNIHDHDJJBHMEEJBCMAA.Michaell.Taylor@reis.com>


Sorry, another newbie question.

I am writing a script (ScriptA) which writes to a file which could possibly
already be opened (either in read or write mode) by another python script
(ScriptB).

I would like for ScriptA to test the opened status of the file and if
already opened, delay writing for a second or two (at which point it should
be closed again).

Now for the complication - ScriptA and ScriptB will be running on different
machines and different OSs (NT and Linux) both accessing a file on a Samba
share.

Any ideas?

Thanks in advance.

Michaell




From moshez at math.huji.ac.il  Sun Jun 11 16:03:07 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sun, 11 Jun 2000 23:03:07 +0300 (IDT)
Subject: HELP with HTTPLIB/URLLIB
In-Reply-To: <sk7kt5ajh51169@corp.supernews.com>
Message-ID: <Pine.GSO.4.10.10006112300280.13750-100000@sundial>

On Sun, 11 Jun 2000, Alan Alper wrote:

> I need to duplicate the action of a browser POSTing form data to a server.
> GET is easy, but I don't know how to do POST. Any and all help, preferably
> with examples would be welcome.  Essentially, I'm trying to write a small
> agent to fetch classified ad data from several newspaper websites for me so
> that I don't need to visit the sites manually.  I have no clue how to do a
> POST.

urllib.urlopen has a second argument, which is data for a form request.
You can turn dictionaries into data by encoding them: urllib.urlencode.
So, here's a post example:

urllib.urlopen("http://foo.bar.com", urllib.urlencode({'name': 'moshe'})

With a little work you can wrap up this function to be much easier to use.
(Think keyword parameters)
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From msoulier at nortelnetworks.com  Tue Jun 27 11:36:31 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 27 Jun 2000 15:36:31 GMT
Subject: compilation
References: <8j84cf$6bj$1@bmerhc5e.ca.nortel.com> <slrn8lf95c.fu7.thomas@centurion.xs4all.nl>
Message-ID: <8jahlv$rui$4@bmerhc5e.ca.nortel.com>

In article <slrn8lf95c.fu7.thomas at centurion.xs4all.nl>, Thomas Wouters wrote:
>>
>No. (other than 'python -c "import <myscript>"')

	I was thinking of that too. ;-)

>(Then again, contrary to my C and Perl code, both a fair bit, my Python code
>almost never has syntax/programming errors, and most of my programs are
>simple enough to do the unit testing 'by hand' ;)

	Fair enough. While I've had to pour over the extensive API to find what I
need, once I found it, it was easy enough to use. So far my only problems have
been trying to do too much in one line of code, which I do instinctively from
my C/Perl coding. ;-)

	I actually hate finding errors at runtime. It's my biggest complaint about
Perl, in that I can make a function call that doesn't exist, and I won't find
out about it until I hit that piece of code. I suppose python has the same
flaw? Dynamic binding is fine, but a check for typos like that would be good. 

	Mike

-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From steffen.ries at sympatico.ca  Thu Jun 22 20:28:15 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Fri, 23 Jun 2000 00:28:15 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand> <8it320$g9$1@nnrp1.deja.com>
Message-ID: <m3g0q5p7fp.fsf@gondolin.beleriand>

bragib at my-deja.com writes:

> This goes using re:
> 
> >>> p = '\,'
> >>> pat = re.compile(p, re.IGNORECASE)
> >>> t = 'uid=joe, ou=orgunit, o=some\\org'
> >>> pat.split(t)
> ['uid=joe', ' ou=orgunit', ' o=some\\org']

Unfortunately, this does not do what I need:
>>> t='uid=joe, ou=orgunit, o=some\\, org'
>>> pat.split(t)
['uid=joe', ' ou=orgunit', ' o=some\\', ' org']

the last two elements should be one (i.e. 'o=some\\, org')

(string.split has the same problem)

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


From kstar18 at hotmail.com  Mon Jun  5 14:29:25 2000
From: kstar18 at hotmail.com (kstar18 at hotmail.com)
Date: Mon, 05 Jun 2000 18:29:25 GMT
Subject: Grace
Message-ID: <8hgrhd$dmd$1@nnrp1.deja.com>

Hi,

I'm relatively new to using Python, and am currently working on a
program to plot massive amounts of data using Grace.

I am working under Windows NT, and unfortunately was unable to find a
program or module that would allow me to pipe commands to the Grace
program.  I was able to find such a module for Unix, grace_np by
M. Haggerty, but unfortunately, that was not helpful.

Does it have a counterpart in windows?  If not, is there another way to
do this?

Any help would be greatly appreceated.

Thanks!


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


From paul.magwene at yale.edu  Thu Jun 22 15:31:18 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Thu, 22 Jun 2000 15:31:18 -0400
Subject: Python website
References: <3951fe9c_2@nnrp1.news.uk.psi.net>
Message-ID: <39526986.F8D33806@yale.edu>

Simon Brunning wrote:
> 
> Has <http://www.python.org> disappeared, or is it just me?
> 


It's just you. ;)  I was able to bring it up without any problems as of
3:30 pm EDT.

--Paul


From pbleyer at embedded.cl  Thu Jun  8 14:23:07 2000
From: pbleyer at embedded.cl (Pablo Bleyer Kocik)
Date: Thu, 08 Jun 2000 14:23:07 -0400
Subject: My wishlist for Python3K
References: <slrn8jtuum.169.hzhu@rocket.knowledgetrack.com>
Message-ID: <393FE48B.217BB87A@embedded.cl>

Huaiyu Zhu wrote:

> My wishlist for Python3K (This is not to start a flame war):
>
> 1. Everything is an object of a class that can be subclassed.

    ... And everything is painfully slow and error prone. Strange and unexpected
things will begin happening as a result for hidden behavior. Developers will go
mad checking polymorphism when implementing their functions. Moral: Use fixed
objects for what they are meant to be and keep them that way -- that's their
raison d'?tre.

> 2. Augmentation operators: +=, *=, etc.  Dot operators .*, ./, etc (for
>    numerical computation to distinguish matrix and elementwise operations).

    Nice.

> 6. Have a clean mechanism to work with other programming languages other
> than C.  (I have matlab/octave in mind.)

    The core is there. I think it's enough clean already. Perhaps there's a need
for better interfaces/object passing/object marshalling & demarshalling between
Python and other applications and we should rely on a standard (CORBA?). Or
better, make Python objects the standard ;^) ("glue it all with Python").

> 7. NOT case insensitive.  (No followup on this please.  Enough said already.)

    You are not very fond of computers, are you?

> When you see how much computation is done by the matlab/octave crowd you can
> appreciate why convenience to scientific programmers (this term should be
> clear) matters.  Want CP4E?  See http://www.mathworks.com/products/demos/

    I use Python mostly for scientific work (Numpy et al.). I use Matlab, Octave
and Scilab also. Instead of changing radically the core language I think it's
better to concentrate efforts in improving Numpy, for example. What would be nice
is to reinforce the language in it's "dynamic mode/dynamic interface"
capabilities via operator overloading, etc. Then you could run something that
"looks like Matlab" over Python, efficiently.

> Huaiyu
>
> PS. (Shameless advertising):  Want a Matlab clone in Python?  See
> https://sourceforge.net/project/?group_id=6490

    Then use Matlab, and not attempt to change Python from it's bases.

    Cheers!

--
Pablo Bleyer Kocik |
pbleyer            |"Rintrah roars & shakes his fires in the burdend air;
      @embedded.cl | Hungry clouds swag on the deep" ? William Blake




From andy.elvey at paradise.net.nz  Sun Jun  4 19:13:01 2000
From: andy.elvey at paradise.net.nz (Andy Elvey)
Date: Mon, 05 Jun 2000 11:13:01 +1200
Subject: Query about using "DataHandlers" module
References: <393A0796.4E5254D0@paradise.net.nz>
Message-ID: <393AE27D.A7B8123@paradise.net.nz>

Oops ...... the "example_07b" that I posted didn't quite show what I meant
with my question .....

 The code attached here *does* show what I mean..... it has the SQL code in it
with the WHERE statement that doesn't work .....

Andy Elvey wrote:

> Hi all.  I have a query about the use of the DataHandlers module
> (available at "Vaults of Parnassus").
>
>  I have attached a simple  example which uses this module (this example
> works fine).  But .... when I amend the SQL code to say  'select * from
> test where team == "Yankees" '  ,  the code still retrieves all teams -
> the where statement is ignored.
>
>   My question is - "does anyone have an example (or two!) of some Python
> code similar to my example , but which has a select statement that has a
> "where" clause in it?
>
>  ( and -  are there any examples which use the DataHandlers module and
> use * 2 tables*   with a join ... ? )
>
>  Many thanks in advance for any help offered .... :-)
>
>  ( and by the way .. this DataHandlers module is *great*  ..... :-)
>
>   ------------------------------------------------------------------------
> #!/usr/bin/env python
>
> # Illustrate the parsing a data file and creating SQL insert statements
> # The table definition comes from an XML file
>
> from DataHandlers.SQLTable import SQLTable
> from DataHandlers.dbdefXMLParser import dbdefXMLParser
> from DataHandlers.FileConnection import Connection
>
> # create a parser
> xml = dbdefXMLParser()
> # return a dictionary of table definitions
> tables = xml.parseFile( 'test.xml' )
> # get the baseball table
> tbl = tables['baseball']
> # the data file is delimited by commas
> tbl.setReader( 'Delimited' )
> tbl.setDelimiter( ',' )
> # open a file connection
> fc = Connection()
> fc.connect( 'teams.txt' )
> # we need to do this fake query so we can get the cursor
> # it actually means something when we are doing an SQL thing
> cursor = fc.query('')
> if not cursor:
>     print 'Unable to execute query'
>     sys.exit(1)
> # now walk thru the lines of data
> while cursor.next():
>     tbl.parse( cursor.getData())
>     # print out the insert statements
>     print tbl.getInsert()
>
> print "Query data returned..."
> cursor = fc.query( 'select * from test' )
> while cursor.next():
>     print cursor.getData( )
>
> cursor.close()
>
>
>   ------------------------------------------------------------------------
> <dbdef>
>   <sqltable name="baseball">
>     <field name="team" size="20" type="char" />
>     <field name="city" size="20" type="char" />
>     <field name="stadium" size="30" type="char" />
>     <field name="league" size="1" type="char" />
>     <field name="division" size="1" type="char" />
>   </sqltable>
> </dbdef>
>
>   ------------------------------------------------------------------------
> Orioles,Baltimore,Camden Yards,A,E
> Red Sox,Boston,Fenway Park,A,E
> Angels,Anaheim,Anaheim Stadium,A,W
> White Sox,Chicago,Comiskey Park,A,W
> Indians,Cleveland,Jacobs Field,A,E
> Tigers,Detroit,Tiger Stadium,A,E
> Royals,Kansas City,Royals Stadium,A,W
> Brewers,Milwaukee,County Stadium,A,E
> Twins,Minnesota,Hubert H. Humphrey Stadium,A,W
> Yankees,New York,Yankee Stadium,A,E
> Athletics,Oakland,Oakland Coliseum,A,W
> Mariners,Seattle,Kingdome,A,W
> Rangers,Texas,Arlington Stadium,A,W
> Blue Jays,Toronto,Toronto Stadium,A,E
> Braves,Atlanta,Fulton County Stadium,N,W
> Cubs,Chicago,Wrigley Field,N,E
> Reds,Cincinnati,Three Rivers Stadium,N,W
> Astros,Houston,Astrodome,N,W
> Dodgers,Los Angeles,Dodger Stadium,N,W
> Expos,Montreal,Exposition Stadium,N,E
> Mets,New York,Shea Stadium,N,E
> Phillies,Philadelphia,Veterans Stadium,N,E
> Pirates,Pittsburgh,Pirate Stadium,N,E
> Cardinals,St. Louis,Busch Stadium,N,E
> Padres,San Diego,Jack Murphy Stadium,N,W
> Giants,San Francisco,Candlestick Park,N,W
-------------- next part --------------
#!/usr/bin/env python

# Illustrate the parsing a data file and creating SQL insert statements
# The table definition comes from an XML file

from DataHandlers.SQLTable import SQLTable
from DataHandlers.dbdefXMLParser import dbdefXMLParser
from DataHandlers.FileConnection import Connection

# create a parser
xml = dbdefXMLParser()
# return a dictionary of table definitions
tables = xml.parseFile( 'test.xml' )
# get the baseball table
tbl = tables['baseball']
# the data file is delimited by commas
tbl.setReader( 'Delimited' )
tbl.setDelimiter( ',' )
# open a file connection
fc = Connection()
fc.connect( 'teams.txt' )
# we need to do this fake query so we can get the cursor
# it actually means something when we are doing an SQL thing

print "Query data returned..."
cursor = fc.query( 'select * from test where team eq "Yankees"' )
if not cursor:
    print 'Unable to execute query' 
    sys.exit(1)
while cursor.next():
    tbl.parse( cursor.getData())
    print cursor.getData( )

cursor.close()
    

From jkraska1 at san.rr.com  Wed Jun 14 21:32:31 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 15 Jun 2000 01:32:31 GMT
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
Message-ID: <394832DB.B41948AB@san.rr.com>

> 1.  After using Python seriously for 6 months, I must confess I still
> dislike the "whitespace as syntax".  I like the fact that Ruby uses block
> markers (begin, end) instead of tabs. This means cheifly to me that I don't
> need to worry about how my editor handles tabs and spaces.  Having something
> look correct on the screen, but not work because of misuse of tabs versus
> spaces, or incorrect indentation, is something that still continues to bug
> me.

IMO, the solution to this is to tab to be an illegal character
in python. :)-
 
> 2. Python uses ":" to introduce a new scope, something I find extraneous,
> and more irritating than begin/end or Ruby's def and end for functions, for
> example.

Yeah, it does seem to be spurious.


I somewhat wish python had Ruby's/Lisp's :THIS-IS-A-KEYWORD syntax. Would
be nice. Rule: any single run of characters not seperated by whitespace
and beginning with a colon is treated like a Python interned string.



C/


From jfdecd at execpc.com  Sun Jun  4 23:31:52 2000
From: jfdecd at execpc.com (Jerry F. Davis)
Date: Sun, 04 Jun 2000 22:31:52 -0500
Subject: stripping out unwanted characters in string
Message-ID: <393b1f58$0$95987@news.execpc.com>

Howdy:

I have a string that has some unwanted characters in it.
Octal 240 for instance.

I want to actually strip them out of my string.

I have looked at string.replace, but can't seem to figure out how to do
this.

Thanks in advance.

Jerry


From thomas at xs4all.net  Sun Jun  4 19:28:22 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 5 Jun 2000 01:28:22 +0200
Subject: Starship ever gonna fly?
In-Reply-To: <f9oljso6hl7r7blv64an77h8lfj6dhr6f9@4ax.com>; from urner@alumni.princeton.edu on Sun, Jun 04, 2000 at 04:09:58PM -0700
References: <f9oljso6hl7r7blv64an77h8lfj6dhr6f9@4ax.com>
Message-ID: <20000605012821.V469@xs4all.nl>

On Sun, Jun 04, 2000 at 04:09:58PM -0700, Kirby Urner wrote:

> The loss of Starship was a real tragedy.  Glad to
> read it's recovering.  But is it?  The same not-
> so-reassuring text has remained unchanged for some
> time.

> Part of what made the Python community seem so vital
> was all these add-ons (like PyOpenGL).  But with 
> Starship down for the count, so many of the pointers
> in Vaults of Parnassus are just dead, stone dead.

If you search the list archive or browse the newsgroup for the last few days,
you'll see that Starship is moving, or has been moved, this weekend. It'll
take a few days for the DNS changes to settle, though. After the move, it
should be more actively maintained. (By BeOpen, if I'm not mistaken.)

> Whoever killed Starship did real and lasting damage
> it seems.  Did anyone call the FBI?

I was under the impression that the Big Crash was hardware-related, not a
malicious attack.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From moshez at math.huji.ac.il  Thu Jun  8 21:46:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 04:46:05 +0300 (IDT)
Subject: String.join revisited (URGENT for 1.6)
In-Reply-To: <394003D6.366D0B0@concentric.net>
Message-ID: <Pine.GSO.4.10.10006090442460.19646-100000@sundial>

On Thu, 8 Jun 2000, Manus Hand wrote:

> But as the devil's advocate, tt just seems to me
> that [].join() could be implemented such that each member of the list is 
> str()'ed or repr()'ed 

Surely you don't mean "repr": that would simply give the wrong result.
But "str" is just as wrong: consider

u", ".join(u"Zadka", u"Moshe")

(Where "Zadka" and "Moshe" are written in Hebrew, using Unicode).

"str"ing *that* would simply be wrong.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From just at letterror.com  Sat Jun 17 03:49:03 2000
From: just at letterror.com (Just van Rossum)
Date: Sat, 17 Jun 2000 08:49:03 +0100
Subject: TABs (was: Waffling between Python and Ruby)
In-Reply-To: <upuphqqbh.fsf_-_@ctwd0143.fitlinxx.com>
References: <FVN15.3523$227.83704@nnrp1.uunet.ca>
 <8i8bi2$1s8$1@news.wrc.xerox.com>
 <slrn8kg937.328.hzhu@rocket.knowledgetrack.com>
 <ud7ljln92.fsf@ctwd0143.fitlinxx.com>
 <slrn8kkmul.q1.hzhu@rocket.knowledgetrack.com>
Message-ID: <l03102800b570d8042c72@[193.78.237.143]>

At 5:16 PM -0400 16-06-2000, David Bolen wrote:
>A TAB character has a perfectly good definition, and artifically
>mucking with that definition just within the environment of your
>editor is going to pretty much ensure that any other application that
>expects the normal behavior will present or parse your files
>differently.  Besides, Emacs makes it so easy to maintain indentation
>at any level without changing the basic tab size.

In general (ie. in the rest of the world, outside of emacs), a TAB has
several perfectly good definitions. One of them is that it is always equal
to 8 spaces. Another is that it is a variable thing. If you stick to the
first you'll have trouble with people who don't follow the rule and mix
tabs with spaces. If you follow the second, you'll "only" have trouble with
people mixing tabs with spaces. So the only thing we can say for sure is
"don't mix tabs with spaces".

The fact that the "tab == 8 spaces" rule is a) not generally accepted and
b) not generally enforcable across editors makes it IMHO a pretty useless
definition. Tools that break when using anything but 8 are badly designed:
it's like using two digits to store a year...

Just





From johngrayson at home.com  Thu Jun  8 13:10:36 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 08 Jun 2000 17:10:36 GMT
Subject: PLEASE HELP ON SOLARIS!!!
References: <393FBD20.E4664BD4@usa.alcatel.com> <Pine.LNX.4.21.0006081611570.796-100000@fep132.fep.ru>
Message-ID: <8hok1t$7o2$1@nnrp1.deja.com>

In article <Pine.LNX.4.21.0006081611570.796-100000 at fep132.fep.ru>,
  phd at phd.russ.ru wrote:
> On Thu, 8 Jun 2000, Sara Fowler wrote:
> > Is ANYBODY out there trying to run Python on Solaris?  If so, how
did
> > you get it to build?
>
>    I run it on 2.5.1. No problems - just ./configure and make...
>
> Oleg.            (All opinions are mine and not of my employer)
> ----
>     Oleg Broytmann      Foundation for Effective Policies
phd at phd.russ.ru
>            Programmers don't die, they just GOSUB without RETURN.
>
>

My experience is similar. No problems ever with Solaris 2.4.x, 2.5.1
or 7. It is immaterial if you use gcc or Sun compilers. Similarly,
adjust Modules/Setup as indicated in the comments and Tkinter is
fine, too.

    John


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


From karl-sve at dsv.su.se  Wed Jun  7 06:29:25 2000
From: karl-sve at dsv.su.se (Kalle Svensson)
Date: Wed, 7 Jun 2000 12:29:25 +0200
Subject: Newbie Question...
In-Reply-To: <ZTo%4.854$0l2.6963@news2.tor.primus.ca>
References: <ZTo%4.854$0l2.6963@news2.tor.primus.ca>
Message-ID: <Pine.GSO.4.20.0006071204020.8823-100000@luna.dsv.su.se>

On Wed, 7 Jun 2000, Robin Porter wrote:

> I wrote before asking for an example of C code changed to Python code (i.e.
> Payroll example).  Could someone please explain how I go about saving and
> then running the code once I have inputed it in.  I tried entering it into
> the Python GUI and everything looked good until I tried to save it, recall
> it, then run it.

Save it in a plain text file with the extension .py. This can be done in
your favorite text editor, IDLE or PythonWin.
In IDLE, select "New window" from the "File" menu, enter your Python
program in the new window and then choose "Save As" from the "File" menu.
Then you can run it in the preferred way of your platform, i.e. double
click on it or invoke it from the command line.

> My apoligies for such a basic question.  This language looks great...I am
> anxious to learn how to use it.

No problem. There's no such thing as a stupid question, you know... :)
And it is a great language!

> Could someone please recommend a GOOD book on PYTHON for the beginner.

This has been discussed quite well here, try a newsgroup search on
http://www.python.org/search/.

The Quick Python Book was given a quite good review in last month's Linux
Journal, but I haven't read it (the book) myself, so I can't say...
The review is available on
http://www2.linuxjournal.com/lj-issues/issue73/3851.html

Hope this helps,
  Kalle Svensson

--
 * krftkndl at hem.passagen.se * http://www.unix-shells.com/~kalle/ *



From petersc at stud.ntnu.no  Wed Jun 21 12:32:48 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 21 Jun 2000 18:32:48 +0200
Subject: mailing list duplicates?
References: <3950B648.29F8FC2B@bam.com>
Message-ID: <3950EE30.D0953BE3@stud.ntnu.no>

Bill Scherer wrote:
> 
> Is anyone else seeing a lot of duplicates on the list?

Yes, but they seem not to be in the archives. Wonder if
it's something about the news/list gateway.

have-someone-who-deletes-them-for-me-though-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From sabren at manifestation.com  Thu Jun  8 00:51:57 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 8 Jun 2000 00:51:57 -0400 (EDT)
Subject: Newbie Q: CGI
In-Reply-To: <8hn0c1$1a22$1@news1.wire.net.au>
Message-ID: <Pine.BSF.4.10.10006080045100.23617-100000@chi.pair.com>

On Thu, 8 Jun 2000, Steven Adams wrote:

> Hi,
> 
> How do you get the IP number from a request for a web page? gethostbyname()
> and gethostbyaddress() weren't mentioned anywhere in the cgi module docs
> (that I could see, did I miss it?)
> 
> simple problem, but I just can't find the answer...

Hey Steven,

If I get what you're asking, you don't need the cgi module at all.
Try putting this in your script:

#--start
import os
print os.environ["REMOTE_HOST"]
#--end

By the way, using the weblib wrapper ( http://weblib.sourceforge.net/ ), 
that's the entire script.

Cheers,

- Michal
------------------------------------------------------------------------
Zike Interactive    http://www.zike.net/    http://zike.sourceforge.net/
------------------------------------------------------------------------





From digitig at cix.co.uk  Sat Jun  3 19:49:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 4 Jun 2000 00:49 +0100 (BST)
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <slrn8jikci.ctd.wtanksle@dolphin.openprojects.net>
Message-ID: <memo.20000604004935.14417B@digitig.compulink.co.uk>

In article <slrn8jikci.ctd.wtanksle at dolphin.openprojects.net>, 
wtanksle at dolphin.openprojects.net (William Tanksley) wrote:

> This is a classical example of false dichotomy ;-).  I liked Python's
> indentation not because it enforced consistency, but rather because it
> makes code look good.  

But, of course, the programmer could make C look just as good by applying 
consistent layout. The nice thing about Python is that the programmer 
/will/ apply (locally) consistent layout.

> In fact, it could be argued that Python does 
> _not_
> enforce consistency; the programmer is free to choose any indentation 
> for
> each block, so long as each block can be parsed (which does require
> internal consistency).

And, of course, one is free to choose any legal identifier, so the case 
sensitivity doesn't enforce consistency either.

> So I don't really care about languages enforcing consistency; if I need
> it, I'll enforce it myself.

Great. Ever maintain anybody else's code?

> Take a lesson from the people complaining
> about Python's indentation: one of the leading false accusations is that
> it's inflexible, that it won't allow people to choose thier own
> indentation.

Lesson taken. I won't give any credence to those who complain about case 
sensitivity on the same grounds.


From aa8vb at yahoo.com  Mon Jun 12 10:12:39 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 12 Jun 2000 10:12:39 -0400
Subject: Swig: typemap for string out parameter
In-Reply-To: <dn1g0qnm5s3.fsf@verden.pvv.ntnu.no>; from olafb@pvv.org on Fri,
 Jun 09, 2000 at 01:51:24PM +0200
References: <dn1g0qnm5s3.fsf@verden.pvv.ntnu.no>
Message-ID: <20000612101239.B742284@vislab.epa.gov>

Olaf Trygve Berglihn:
 |
 |I guess some of you swig'ers can answer this easily:
 |
 |I want to create a typemap for out parameter strings similar to what
 |swig does for int with the "int *OUTPUT". I know the length of the
 |string.
 |
 |The C prototype would look like this:
 |
 |int getNewString( const char *pInParameter,
 |                        char *pOutString,
 |	                int  iOutStringLen );

If no reply, you might want to search the SWIG mailing list archives and
post if you don't find the answer.  See http://www.swig.org/ for details.

-- 
Randall Hopper
aa8vb at yahoo.com



From thomas at xs4all.net  Thu Jun  8 19:05:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 9 Jun 2000 01:05:19 +0200
Subject: Obscure Exception message
In-Reply-To: <3dr9a7yi91.fsf@amarok.cnri.reston.va.us>; from akuchlin@mems-exchange.org on Thu, Jun 08, 2000 at 05:28:58PM -0400
References: <75F7304BB41CD411B06600A0C98414FC0A418B@ORSMSX54> <3dr9a7yi91.fsf@amarok.cnri.reston.va.us>
Message-ID: <20000609010518.C20700@xs4all.nl>

On Thu, Jun 08, 2000 at 05:28:58PM -0400, Andrew M. Kuchling wrote:

> "Daley, MarkX" <markx.daley at intel.com> writes:
> > TypeError: string member test needs char left operand
> > I have no idea what this is telling me, although the 'char' comment makes me
> > wish I had a flamethrower!

> Good point.  Hmm... how about:
> TypeError: in <string> requires character as left operand

Put quotes around 'in <string>':

TypeError: 'in <string>' requires character as left operand


-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From mdm at corp.eCircles.com  Thu Jun 22 19:03:50 2000
From: mdm at corp.eCircles.com (Michael D. Marchionna)
Date: Thr, 22 Jun 2000 23:03:50 GMT
Subject: Improving cStringIO API (Request For Comments)
Message-ID: <39529B56.D54681E0@CORP.ECIRCLES.COM>

I was messing around trying to convert my Java app to
Python, and
ran into some difficulty with cStringIO module.  In my app I
transmit
a segmented byte stream over a socket connection.  I read
the data
section by section while laying the message out in a single
StringIO
buffer.  I then create a series of section objects that
reference the
segments in the StringIO buffer.  I want the section object
to contain
no data except an offset from the start of the buffer.   So
that when I
query the contents of a section the data is actually pulled
from the
StringIO object.

This led me to a bunch of code like the following:

class MsgSect:
    <<< code deleted >>>
    def getAttrText(self):
        value = None
        if self.type == ATTR and self.size > 9 and
self.atype == 'C':
            oldpos = self.buffer.tell()
            self.buffer.seek(self.offset+9)
            value = self.buffer.read(self.size-9)
            self.buffer.seek(oldpos)
        return value

The point is not to disturb the current read/write pointer.
You might ask
why not just extract the string and add it as a member of
the MsgSect
object.  Well I have some reasons OK.  Anyway this led me to
making some
modifications to the cStringIO module.

At first I just added a peek method so I could read sections
without
modifying the read/write pointer.  For example:

    >>> from cStringIO import  StringIO
    >>> so = StringIO()
    >>> so.write('0123456789ABCDEF')
    >>> so.tell()
    16
    >>> so.peek(4,3)
    '456'
    >>> so.tell()
    16

I also added a poke method to do what you might think:

    >>> so.poke(4,'***')
    >>> so.getvalue()
    '0123***789ABCDEF'
    >>> so.tell()
    16

Then I got to thinking these are still strings why don't
they behave a little
more like them.  So I added a set of sequence methods to the
StringI and
StringO object types.  So now you can do

    >>> si = StringIO('0123456789ABCDEF')
    >>> len(si)
    16
    >>> si[4]
    '4'
    >>> si[-4]
    'C'
    >>> si[4:7]
    '456'
    >>> si[-6:-2]
    'ABCD'

Item assignment works like you would expect for StringO
objects:

    >>> so = StringIO()
    >>> so.write('ABCD')
    >>> so[3] = 'c'
    >>> so[:]
    'ABcD'

Repeat operations work as well:

    >>> soso = so * 2
    >>> soso[:]
    'ABcDABcD'

Things get a little more squirely with the remaining concat
and setslice
methods.  A concat of StringI objects works fine, but the a
concat of
StringO objects leaves the problem of what to do with the
softspace
attribute.  Currently I just copy the value from the self
object.

    >>> so1 = StringIO()
    >>> so1.write('ABC')
    >>> so2 = StringIO()
    >>> so2.write('123')
    >>> so1.softspace
    0
    >>> so2.softspace = 1
    >>> so3 = so1 + so2
    >>> so3.getvalue()
    'ABC123'
    >>> so3.softspace
    0
    >>> so4 = so2 + so1
    >>> so4.getvalue()
    '123ABC'
    >>> so4.softspace
    1

More variations can be dealt with the concat methods.  For
example it is
possible to mix StringI and StringO types, but what should
the resultant
type be.  Also it seems reasonable to allow string objects
in the concat
as well, so5 = so1 + 'XYZ'.   Unfortunately so5 = 'XYZ' +
so1 won't work.

Slice assignments obviously only work on StringO objects.
In some
sence the setslice should overwrite the contents of the
StringO buffer rather
than produce a new StringO object.  But the poke method
handles most of
the setslice operation except it currently doesn't permit
extending the buffer.
It could do that though.  Perhaps optionally via a third
argument:

    def poke(offset, value, extend=0):

The other areas of concern about the cStringIO module
concern its fixed
reallocation scheme.   Currenty a StringO object starts out
at a capacity
of 128 and will double its size each time a write goes
beyond the current
buffer space.  It would be nice to allow this scheme
modified.  Perhaps via
optional keyword arguments to the StringIO factory
function.  For example:

        so = StringIO(capacity=4096, increment=8192)

Would create a StringO object that would start out with a
buffer space of
4096 bytes, and add 8192 each time a write went off the
edge.  Or possibly
something even likethis:

        so = StringIO(capacity=1024*1024, upscale=0.1)

Would create a 1MB buffer and increase the size by 10
percent for each
overwite.  Options could also be added for decreasing the
size of the buffer
when it is truncated, e.g. decrement=65536, downscale=0.5.

Another feature that would be nice is to allow shifting of
the contents of a
StringO buffer.  Similar to what memmove does.

Further on I would like to support the struct.pack and
struct.unpack methods
so that they operate directly on the contents of the
StringIO buffer.

In the end I was wondering what people think of adding some,
or all of these
features to the cStringIO module, or perhaps to a completely
new module.

Comments and suggestions?

--MDM




From eviltofu at rocketmail.com  Thu Jun 15 07:39:37 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Thu, 15 Jun 2000 11:39:37 GMT
Subject: Modulator.py in Pythonwin?
Message-ID: <eviltofu-B92575.07393815062000@news.earthlink.net>

I noticed that modulator.py is missing in the tools folder of Pythonwin. 
Is there a way I can get it without downloading the entire source code 
for Python?


From jkraska1 at san.rr.com  Sun Jun  4 16:19:10 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sun, 04 Jun 2000 20:19:10 GMT
Subject: Questions regarding design decisions in listobject.c
Message-ID: <393ABA63.18E2F46@san.rr.com>


1. For slice operations, ilow/ihigh indices which undeflow
or overflow the list in question are reset to the edges
of the list. Why was it that PyExc_IndexError wasn't
thrown instead?

2. In slice expressions of the form s[i:j], what was the
design decision that lead to s[i:i] always returning an
empty list and s[i:i+1] selecting one item. At first
glance, it would appear that s[i:i] is the obvious and
intuitive correct expression to return just one item on
a slice, where s[i:i+1] ought to select two items and so
forth. Was there some particular situation which arose
which required this slightly counterintuitive idiom?



C/


From DavidA at ActiveState.com  Tue Jun 13 14:00:41 2000
From: DavidA at ActiveState.com (David Ascher)
Date: Tue, 13 Jun 2000 11:00:41 -0700
Subject: O'Reilly Python Conference Early Registration Deadline Looming
In-Reply-To: <PLEJJNOHDIGGLDPOGPJJMEMMCHAA.DavidA@ActiveState.com>
Message-ID: <PLEJJNOHDIGGLDPOGPJJMEMOCHAA.DavidA@ActiveState.com>

> It's that time of year again, the O'Reilly Open Source 
> Convention, including the Python track, is fast approaching.
> You can save up to $280 if you register by Friday July 16. 

Make that June 19.  Sorry about that.

--david ascher



From collins at seal.aero.org  Mon Jun  5 16:00:05 2000
From: collins at seal.aero.org (Jeff Collins)
Date: Mon, 5 Jun 2000 13:00:05 -0700
Subject: A Simple question about a slice of a string.
In-Reply-To: <8hgefr$sok$1@news2.acs.oakland.edu>
References: <8hgefr$sok$1@news2.acs.oakland.edu>
Message-ID: <14652.1351.470741.751258@malibu.aero.org>



Jay R. Wren writes:
 > I'm attempting to accomplish grabbing the last 6 characters of sys.argv[0]
 > so that I can compare it to another string.
 > If 'myprog' is called as /usr/local/bin/myprog.  It could be called from any
 > path, as long as the 'myprog' portion is there, certain things will
 > execute.(this is used as a symlink)
 > 
 > myprogname = sys.argv[0][ len(sys.argv[0])-10:len(sys.argv[0])

The os.path module provides some useful services:

>>> import os
>>> os.path.split("/usr/local/bin/myprog")
('/usr/local/bin', 'myprog')
>>> head, tail = os.path.split("/usr/local/bin/myprog")
>>> tail
'myprog'
>>> head
'/usr/local/bin'


Jeff



From teyc at bigfoot.com  Fri Jun  9 04:50:58 2000
From: teyc at bigfoot.com (Chui Tey)
Date: Fri, 9 Jun 2000 18:50:58 +1000
Subject: win32pipe
Message-ID: <8hqald$d9p$1@bunyip.cc.uq.edu.au>

Hi,

Still trying to grapple with pipes in windows

Why does the following fragment hang on me?

#
# testwinpipe.py
#
def setuptest():
    f=open('C:\\test.txt','w')
    for i in range(50):
        f.write( str(i) + '\n')
    f.close()

def testpopen2():
    from win32pipe import popen2
    from win32pipe import popen
    cmd = 'more C:\\test.txt'
    w,r = popen2(cmd)
    magic=23
    for i in range(30):
        line = r.readline()[:-1]
        print 'i=%d'%i, line
        if i == magic:
            # send a space when more
            # waits for kb input
            w.write(' ')
    w.close()
    r.close()

if __name__ == '__main__':
    setuptest()
    testpopen2()

Thanks.

Chui




From petersc at stud.ntnu.no  Sun Jun 18 17:14:11 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 18 Jun 2000 23:14:11 +0200
Subject: Why is Python so slow?
References: <gC935.12021$Xx5.555547@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <394D3BA3.A5B150D2@stud.ntnu.no>

William Dandreta wrote:
> 
> I posted a message a while ago about a Python script I wrote that took 90
> mins to run when a similar job in C++ took only 2 mins.
> 
> With help from some experts in this news group, most notably, Peter
> Schneider-Kamp, I was able to reduce the time to 37 mins

Thanks, but I don't deserve that title ;)

> Today I was looking at the script and I noticed that the only string
> function I was using that was not in strop was replace. I took a look at it
> and found:
> 
> replace(x,y) = joinfields(splitfields(x,y),'')
> 
> When I replaced, replace with the joinfields(splitfields()) in strop, much
> to my surprise, the time for the job dropped to 6 minutes.

In my Python version (post 1.6a2 CVS) the joinfields/splitfields
solution
caused double the number of function calls (naturally...) and ran almost
half as fast the replace version.

BTW: You probably meant:

replace(x,y,z) = joinfields(splitfields(x,y),z)

Bye
Peter 
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From thomas at xs4all.net  Mon Jun 19 07:07:04 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 19 Jun 2000 13:07:04 +0200
Subject: Why is Python so slow ?- revisited.
In-Reply-To: <m3u2eq55ha.fsf@atrus.jesus.cam.ac.uk>; from mwh21@cam.ac.uk on Mon, Jun 19, 2000 at 11:29:21AM +0100
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <394D6579.8B2FF840@appliedbiometrics.com> <ylg35.9511$C44.540156@bgtnsc05-news.ops.worldnet.att.net> <20000619101816.X26436@xs4all.nl> <m3u2eq55ha.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <20000619130704.Y26436@xs4all.nl>

On Mon, Jun 19, 2000 at 11:29:21AM +0100, Michael Hudson wrote:
> Thomas Wouters <thomas at xs4all.net> writes:

> > It won't matter a hoot in function calls, as all that happens string
> > literals in a void context is that they get added to the functions'
> > constants tuple (function.func_code.co_consts.)  And that happens at
> > compiletime (which can be at runtime, of course, if you use exec()
> > or import an uncompiled module, but if you do that often you
> > shouldn't whine about performance ;-)

> PyTrivia: isolated string constants are *not* added to co_consts,
> they're just dropped.

Indeed ! I specifically tested this, but must've botched it. I'd already
started wondering why string literals were added to co_consts, as it seemed
a waste ;)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From phd at phd.russ.ru  Wed Jun 21 11:25:34 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 21 Jun 2000 15:25:34 +0000 (GMT)
Subject: mailing list duplicates?
In-Reply-To: <3950B648.29F8FC2B@bam.com>
Message-ID: <Pine.LNX.4.21.0006211524390.6990-100000@fep132.fep.ru>

On Wed, 21 Jun 2000, Bill Scherer wrote:
> Is anyone else seeing a lot of duplicates on the list?

   I saw millions of them and already reported to postmaster at python.org.
After that it almost stopped.

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From moshez at math.huji.ac.il  Fri Jun  9 16:53:10 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 23:53:10 +0300 (IDT)
Subject: Reloading on the fly: was: Re: Python Productivity over C++
In-Reply-To: <8hrjg3$dn2$2@news1.xs4all.nl>
Message-ID: <Pine.GSO.4.10.10006092346400.29560-100000@sundial>

On 9 Jun 2000, Boudewijn Rempt wrote:

> Could you post an example of designing the main program to reload
> submodules? I'd like to get that working, too, and coming from
> languages where that's impossible, I've a bit of trouble coming up
> with a solution...

You were addressing a different "you", but I would still like to explain how 
I did that in a program that didn't see yet the light of day. I had a 
main module containing "hooks": lists of functions to execute in different
circumstances. For example:

# File: Hooks.py
on_load_hooks = []

Now, when I started up the program, I looked in a specified directory for
extensions: each of them got read into a string, and then I did

exec s in {} 

Now, an extension usually looked like:

#
import Hooks
def foo(params):
	do something
Hoos.on_load_hooks.append(foo)

Additionally, the user was able to add extensions at run-time. There was
no provision for de-registering extensions, except for places which had
place for exactly one hook. In those, registering a hook meant de-registering
the previous one. This convinced me that it is possible, but it wasn't needed
for that application. One caveat: when you de-register a function, first
delete it from the globals, otherwise you might get some nasty leaks. Python
1.6 might solve this if the gc patch gets in.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From neilh at scintilla.org  Sun Jun 18 21:05:13 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 19 Jun 2000 01:05:13 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <v8N25.63502$WS3.627109@typhoon.we.rr.com> <02b001bfd885$fbefa5e0$1906a8c0@fc.fenx.com> <xg735.63947$WS3.647315@typhoon.we.rr.com>
Message-ID: <dle35.2194$qz4.17968@news-server.bigpond.net.au>

>     I'm currently experimenting with PythonWin's Show-Tabs thing. I need
to
> get used to it, though...because as is it looks very distracting. :)

   You can change the colour to silver to make it less heavy. Or hack the
code to any colour you want. Originally tabs where displayed as round
cornered rectangles but Mark thought this was too heavy. Perhaps vertical
lines at the start of tabs instead would be lighter and make it easy to
match up code.

   Neil




From spamfranke at bigfoot.de  Mon Jun 19 12:36:37 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Mon, 19 Jun 2000 16:36:37 GMT
Subject: Function that knows the names of its actual parameters
References: <394a6395.7217524@news.btx.dtag.de> <8ijgdl$oh8$1@slb6.atl.mindspring.net>
Message-ID: <394e45c3.513888@news.btx.dtag.de>

On 18 Jun 2000 21:53:57 GMT, aahz at netcom.com (Aahz Maruch) wrote:

>Why not simply write a function that takes a dict as its parameter?
That's what I do now.

>Maybe if you gave a better description of an occasion you think you need
>this feature we could give you a better answer.
Sure - see snippet below. Being able to call SQL_update without doubling
the arguments would be nice. 
If we could pass more than one of these dicts to SQL_update(), it would even 
be possible to specify the WHERE clause with current parameters as well.
I'm enclosing the magic arguments here into $...$ for provocative reasons <wink>:

>>> index=1
>>> member_name='you'
>>> is_current=1
>>> key=23
>>> SQL_update(table, $member_index, member_name$, $key, is_current$)
UPDATE my_table SET index=1, member_name='you' WHERE key=23 AND is_current=1

Of course no one really *needs* this feature, it's just syntactic sugar that doesn't
add to the expressive power of the language. In fact, writing this in a function
call notation possibly obfuscates things. 
So I'm not proposing anything here. I just feel a certain preference for musing
about those parts of Python were the border between source code identifiers 
and their string value is crossed (**kwargs, __getattr__, import, to name a few).

Hesitating-to-follow-the-longstanding-tradidtion-to-demand-a-Python-syntax-change-for-any-singular-special-case-you-stumble-upon-ly
y'rs,
Stefan


---------------------------------------
import string

def SQL_update(table, **set):
  return "UPDATE %s SET %s WHERE [...]" %(table, SQL_value_dict(set))

def SQL_value_dict(dict):
  args = []
  for k,v in dict.items():
    if v is not None:
      args.append("%s=%s" %(k, repr(v)))  # simplified: should be SQL_repr here
  return string.join(args, ", ")
 
member_index=1
member_name_before="ok"
is_current=None
print SQL_update("my_table", member_index=member_index,
                 member_name_before=member_name_before, is_current=is_current)



From _delete_this_mlerutte at square.nl  Mon Jun 26 08:07:20 2000
From: _delete_this_mlerutte at square.nl (Maurice le Rutte)
Date: Mon, 26 Jun 2000 14:07:20 +0200
Subject: JPython and Java integration
Message-ID: <8j7h3m$5v18j$1@fu-berlin.de>

I'm thinking of using JPython as our 'scripting'/plug-in language for our
product written in Java. With the risk of being flamed beyond recognition:
like MS's 'VBA', but then better.

My first tests of the JPython interpreter are ok. I got the thing running
and calling it from within a Java program was easy. Finding a way to pass
and retrieve Java objects was slightly more difficult, mostly because there
is almost no documentation, but once I found out through reflection it was
very easy.

Are there people on this list that are using JPython and have some
experience or otherwise useful advise when using JPython as a runtime
language within the Java platform?

I'd greatly appreciate any snakepits and areas to avoid. Success stories are
also very welcome!

Maurice le Rutte.




From cut_me_out at hotmail.com  Wed Jun  7 07:54:11 2000
From: cut_me_out at hotmail.com (Alex)
Date: 07 Jun 2000 07:54:11 -0400
Subject: Another Newbie question....
References: <Xhq%4.864$0l2.6976@news2.tor.primus.ca>
Message-ID: <etdhfb53dwc.fsf@w20-575-109.mit.edu>

Could you post the code?

Alex.


From jamarijr at my-deja.com  Mon Jun 12 09:37:56 2000
From: jamarijr at my-deja.com (Arinté)
Date: Mon, 12 Jun 2000 13:37:56 GMT
Subject: Printing exceptions
Message-ID: <8i2p3i$3pd$1@nnrp1.deja.com>

My embedded python app sets an exception string, python can easily
print this error using
print sys.exc_value
but, when I try to post it in my MessageBox
pos.alert(sys.exc_value)
the application crashes, I even tried putting the failing part in a
try..catch block and it still crashed (What the...)
Of course
pos.alert("testing this messagebox") works fine

Any suggestions?
thanx


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


From aahz at netcom.com  Mon Jun 12 00:20:46 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jun 2000 04:20:46 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <E0C8C2A9B34C2EF9.8F6C18C45B6980ED.AD2ECD05DCCE3B64@lp.airnews.net> <8F50ADB75gmcmhypernetcom@199.171.54.155> <85snuj346q.fsf@one.net.au>
Message-ID: <8i1oeu$vbu$1@slb2.atl.mindspring.net>

In article <85snuj346q.fsf at one.net.au>,
Neurocrat  <neurocrat at one.net.au> wrote:
>
>What is it about Python that encourages the "component" style you
>described?

I think it's that Python modules with even a modicum of "proper" design
truly are components.  That makes unit testing of each module
ridiculously simple.  Back in January, I came up with a quote that a lot
of people seem to think captures the essence of Python design:

  The way to build large Python applications is to componentize and
  loosely-couple the hell out of everything.

Python is all about interfaces, given that there's no static typing.
What's a component?  It's a defined interface.  Python is about
simplicity, to the point where experienced programmers have to fight
their tendency to look for the complexities.

Unless you actually have a complicated algorithm (e.g. income tax
calculations), if your Python code looks complicated, you're probably
doing it wrong.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From tom-main at REMOVEME.home.com  Thu Jun  8 17:03:58 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Thu, 08 Jun 2000 21:03:58 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com> <8hluop$mlo$1@news1.xs4all.nl>
Message-ID: <2TT%4.178184$55.3794691@news2.rdc1.on.home.com>

"Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
news:8hluop$mlo$1 at news1.xs4all.nl...
...
> There's another option, of course. You could use Qt (and PyQt), which
> works on Windows and Unix/X11, offers a great GUI and a good design.
> And there are tons of programs using it you can look at & learn from.
> (Not to mention the great documentation.)

Actually, Qt sounds pretty good, but about six months ago I looked into
trying it out, only to find that I would have to pay a big load of dough to
do so (for C++ on Windows).

Tom.




From scarblac-spamtrap at pino.selwerd.nl  Wed Jun 28 04:14:31 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 28 Jun 2000 08:14:31 GMT
Subject: Complex datastructures in Python
References: <8jb6jr$2uov$1@news.enteract.com>
Message-ID: <slrn8ljcp5.36m.scarblac-spamtrap@flits104-37.flits.rug.nl>

Greg Gallagher wrote in comp.lang.python: 
> I'm a little disheartened here... I know how to do this in Perl, but I'm lost in Python and 
> searching for 'complex data structures' is landing me nowhere.  Basically,  I want
> a dictionary whosevalues are lists of dictionaries.  Simple.
> I want to basically loop through some data and so something like this:
> 
> while loop:
>    new_forward = {'ip' : ip_address,
>                   'hostname' : hostname,
>                   'mac_address' : mac_address}
> 
     if not forward.has_key(domain):
	    forward[domain] = []
>    forward[domain].append(new_forward)

The only problem is that you can't append to it if it's not a list yet,
so you have to initialize forward[domain] to the empty list first.

(btw, this is not a hash of hashes, but a hash of lists of hashes, right?)

> Any pointers?  Also, any good documentation for 'complex datastructures',
> ala Perl's Cookbook ?

I've never had any problem with complex structures in Python, really. They
tend to just work. If they become too complex wrapping them in some OO
usually helps. 

A Python version of Perl's Cookbook may or not be useful (check Deja - it's
been discussed before - someone also mentioned that O'Reilly is actually
working on it). "Python Annotated Archives" is the book that comes
closest I think, but it's had mixed reviews - check Deja again, I suppose.

But just use Python for a while. It's consistent, there's not that much to
learn.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Gblrsh smrzl, Mom 'n' Dad":
   In Runequest III (Avalon Hill), two people from the same village,
   speaking the same language, have a 1 in 3 chance of totally
   misunderstanding each other.


From mjhand at concentric.net  Tue Jun 13 15:28:21 2000
From: mjhand at concentric.net (Manus Hand)
Date: Tue, 13 Jun 2000 13:28:21 -0600
Subject: Two newbie questions
References: <20000613165659.A7828@basajaun> <39468424.BED211C2@concentric.net>
Message-ID: <39468B55.FE9488D5@concentric.net>

Following my own post.  How gauche.

Anyway, a slightly more legible alternative to get a random number
between -sys.maxint and sys.maxint:

import random
yourNumber = random.randrange(sys.maxint) * random.choice([1, -1])

Manus




From crzylzydaze at icqmail.com  Mon Jun 12 08:01:50 2000
From: crzylzydaze at icqmail.com (Joe Wilson)
Date: Mon, 12 Jun 2000 12:01:50 -0000
Subject: Email
Message-ID: <8i2jfe+clld@eGroups.com>


   Why havent I been obtaining any mail from this base?




From olczyk at interaccess.com  Thu Jun  8 18:54:39 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Thu, 08 Jun 2000 22:54:39 GMT
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com>
Message-ID: <3942239b.23771750@nntp.interaccess.com>

On Thu, 8 Jun 2000 14:19:23 -0700, "Jeff" <no_spam at umich.edu> wrote:

>Okay, 2 questions.  Seem like basic enough functionality to be in pythonwin,
>but I dont see it.
>1. How do you kill an infinite loop program without killing the pythonwin
>environment?
>2. How can you 'unload' a module so that you can overwrite the .pyd without
>leaving the pythonwin environment?
>
>If there are no answers to these questions, someone needs to work on
>pythonwin's "development" capabilities.
>
>
>
Have you tried using Emacs?
As a newbie that's the first thing I'm lookingb into.


From mrfx at etang.com  Sun Jun 11 16:04:54 2000
From: mrfx at etang.com (mrfx)
Date: Mon, 12 Jun 2000 04:04:54 +0800
Subject: IDLE  can be rewrited using wxPython?
Message-ID: <8i25jd$1u8g$1@news.cz.js.cn>

The current IDLE uses TCL/TK as GUI widget.
IDLE  can be rewrited using wxPython as GUI widget?




From wlfraed at ix.netcom.com  Wed Jun 14 23:42:43 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 14 Jun 2000 20:42:43 -0700
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au> <8i39ll$93f$1@slb3.atl.mindspring.net> <3947A59F.5016D9F0@sage.att.com> <85em60nozb.fsf@one.net.au> <8i8o8b$83i$0@216.39.141.36>
Message-ID: <crigkss2m7ick2imqdctubg5d4lqjpvtd1@4ax.com>

On 14 Jun 2000 20:00:11 GMT, Dana Booth <dana at mmi.oz.net> declaimed the
following in comp.lang.python:

> Neurocrat <neurocrat at one.net.au> wrote:
> 
> >> i find i'm more enamored of elegance these days.
> 
> N: Me too.
> N: Now if only I could find it in an OS ...
> 
> TRS-DOS or NEW-DOS? :)

	The LS TRSDOS 6, or the Tandy original?

	I'd toss the Amiga OS into the hat too...

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From thomasm at legend.co.uk  Mon Jun 19 11:34:49 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Mon, 19 Jun 2000 16:34:49 +0100
Subject: os.path.isdir strange behaviouring
Message-ID: <394e3d29.0@katana.legend.co.uk>

Hi,

I notice that if a directory is root owned and chmoded 700 (let say /test)
and if it contain a directory called test2
os.path.isdir("/test/test2") return 0 if called by a "normal user"

I was expecting an exception.
Can so explain me why this "strange" behaviour ?

Thomas




From ray at commsecure.com.au  Tue Jun 13 06:45:48 2000
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Tue, 13 Jun 2000 10:45:48 GMT
Subject: can python do this?
References: <3945c87a$0$505$7f31c96c@news01.syd.optusnet.com.au> <3945D022.63951249@san.rr.com>
Message-ID: <3945B926.9AB7762F@commsecure.com.au>

Courageous wrote:
> 
> Q: "Can python do this?"
> A: "Python was born for it, baybee!"
> 
> import string
> file = open ("file")
> lines = file.readlines()
> for line in lines:
>     pair = string.split(line,":")
>     print pair[0] +": "+ `string.split(string.strip(pair[1]),",")`
> file.close()

I think his specification was a little more complex, he wanted to choose a
subset of
lines (those that he indicated by the use of the ** prefix). In addition the
"CMLISTPTRACKS"
field can span multiple lines, since it is a list of the tracks on the album.
And I also believe that he is expecting many of these per file, i.e. that the
input file has a list of the information associated with all of the new albums
that have been added to the catalogue.

The following might do what he wants, it uses the CMTYPE label to
indicate the end of the current run of album info.
Edit TITLE to suit your needs for the first line title.
It reads data from stdin, and writes to stdout.

#!/usr/local/bin/python
import sys, string

TITLE = 'Title\tArtist\tCat no\tAPN\tPrice Code\tMedia Code\tDist
Code\tRelease\tTracks\tPrice\tType'
keep_list = ['CPTITLE', 'CPARTIST', 'CMCATNUMBER' 'CMCATNUMBER', 'CMAPN',
                'CMARIAPRICECODE', 'CMARIAMEDIACODE', 'CMARIADISTRIBUTORCODE',
                'DMRELEASE', 'CMLISTPTRACKS', 'CURMPRICE', 'CMTYPE']
LAST_ATTRIBUTE = 'CMTYPE'

print TITLE
row_strs = []
tracks = []
in_track = 0
for line in sys.stdin.readlines():
        fields = string.split(line[:-1], ':')
        if in_track or fields[0] in keep_list:
                if len(fields) == 2:
			# multi-line record, store in tracks
                        if fields[0] == 'CMLISTPTRACKS':
                                tracks.append(fields[1])
                                in_track = 1
                        else:
                                if in_track:
                                        in_track = 0
                                        row_strs.append(string.join(tracks, '
'))
                                else:
                                        row_strs.append(string.strip(fields[1]))
                else:
                        if in_track:
                                tracks.append(line[:-1])
		# end of record, print result
                if fields[0] == LAST_ATTRIBUTE:
                        print string.join(row_strs, '\t')
                        row_strs = []
                        tracks = []


From rwallace at esatclear.ie  Thu Jun 15 02:14:51 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Thu, 15 Jun 2000 07:14:51 +0100
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at>
Message-ID: <3948745B.941@esatclear.ie>

> stevemul at ozemail.com.au (Steve Mullarkey) wrote:
> 
> > I have read in several places productivity claims of 5 to 10 times over
> > 'C' and "C++". I would like to ask for some feedback from "C++"
> > programmers who have moved to Python as to whether these estimates are
> > realistic.

As someone who uses and likes both C++ and Python, I think estimates
like that are excessively simplistic.

I use C++ for commercial programming, because it gives me the power and
control I need.  Productivity isn't the best when measured in time to
get the first prototype running, but I've found nothing better when
measured in time to develop, deploy, maintain and support something a
customer will pay for.

I use Python for scripting/prototyping/AI research stuff.  (Used to use
Scheme for some of it, but never liked the syntax.)  I find it gives a
substantial productivity advantage over C++ - nowhere near an order of
magnitude, but enough to be well worth having - because it does a lot of
things for you, or at least makes them more convenient.

The-right-tool-for-the-job-ly yrs,

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From root at sirius.ftsm.ukm.my  Wed Jun 14 22:44:03 2000
From: root at sirius.ftsm.ukm.my (root)
Date: Thu, 15 Jun 2000 10:44:03 +0800
Subject: 8 queen/game of life
Message-ID: <20000615104403.A1240@sirius.ftsm.ukm.my>

newbie question.

has anyone wrote the 8 non-attacking queen problem and conway's game of life
in python? i try but face a few newbie problems.

1. conway's game of life:
   how to have a matrix in python? 
   the following is not good;
   
           r0=[0,0,0]
	   r1=[0,0,0]
	   ........
	   
	   m=[r0,r1,....]

   the matrix is fixed, which is not good. my solution use dictionary (i
   know this is not the right way). the key is a tuple of the matrix
   indices. the value is '0'. then base on game of life rule, update the
   dictionary and print for next generation.

	   m={}
	   m[(0,0)] = 0, m[(0,1)] = 1,....
	   
2. 8 non-attacking queen.

   use backtracking. the function call itself back. how to do recursive call
   in python? roughly, the C program I study look like this;
   
   queen_count = -1
   
   main()
   {
     addqueen();
   }
   
   addqueen()
   {
   queen_count++
   if queen_count != 8
        addqueen()
   else
        print result
   }
	
   how to pass a global variable (queen_count) to a function?
   
many thanks in advance.

zamri
   



From akuchlin at mems-exchange.org  Sat Jun 10 19:16:55 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 10 Jun 2000 19:16:55 -0400
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no>
Message-ID: <3dya4d2kk8.fsf@newcnri.cnri.reston.va.us>

Peter Schneider-Kamp <petersc at stud.ntnu.no> writes:
> t = PyTuple_New(2);
> PyTuple_SET_ITEM(t,0,PyInt_FromLong(a));
> PyTuple_SET_ITEM(t,1,PyInt_FromLong(b));
> PyErr_SetObject(PyExc_IndexError,t);

Py_BuildValue would simplify the above code:

t = Py_BuildValue("(ll)", a,b);
if (t==NULL) {DECREF other variables, and return NULL}
PyErr_SetObject(PyExc_IndexError, t);

--amk


From jeremy at beopen.com  Thu Jun 29 23:44:31 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 29 Jun 2000 23:44:31 EDT
Subject: Problem with execv
References: <8jg6j3$o84$1@nnrp1.deja.com>
Message-ID: <blbt0jzv8h.fsf@bitdiddle.concentric.net>

sebv at netscape.net writes:

> I'm trying to execute a simple 'ls' command from python, using
> execv but I have a TypeError. I couldn't find any similar
> problem reported somewhere else. Please give me a advice !
> thanks,
> 
> Python 1.5.2 (#2, Jun 28 2000, 16:52:22)  [GCC 2.8.1] on sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import os
> >>> os.execv('ls','')
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: illegal argument type for built-in operation
> 

The error message will be much clearer in the next Python release!

Python 2.0b1 (#33, Jun 29 2000, 23:26:05)  [GCC egcs-2.91.66
    19990314/Linux (egcs-1.1.2 release)] on linux2 
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> import os
>>> os.execv('ls', '')
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: argv must be tuple or list

The correct calling convention is to supply a tuple or list as the
second argument to execv.  Example:

>>> os.execv('/bin/ls', ('.',))
[lists contents of current directory]

Jeremy

-- Jeremy Hylton, PythonLabs
http://www.pythonlabs.com/


From hgg9140 at seanet.com  Fri Jun  2 18:13:37 2000
From: hgg9140 at seanet.com (Harry George)
Date: 02 Jun 2000 15:13:37 -0700
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee>
Message-ID: <m3g0qv91f2.fsf@wilma.george.org>

Install PyXML, then try this.  Read the xml.dom.core for element names
and the basic functions.  If you are not familiar with xml itself, see
the XML Bible.  Try this routine on some known-good .xml file (I can
supply one if need be.)

I could post one I did to read/write Abiword files.  My heaviest use
of python+xml is at work, gluing together a bunch of engineering apps.


Kaimar Karu <kaimar at ut.ee> writes:

> Hi,
> 
> Where can I find some kind of helping material (tutorials etc) about how
> to use xml in python?
> I need some examples, just pure specification is not enough.
> 
> tia,
> --
> Kaimar Karu | kaimar at ut.ee | 251 13 468
>        = What a wonderful world =
> 
> 
#!/usr/bin/env python
#/* helloxml.py 

"""
Usage: helloxml.py [options] filename
E.g.,  helloxml.py myfile
Options
-h,--help        this message
-v,--version     version

Revisions:
2000-02-23  Harry George   initial version
"""

import sys,os,getopt,time,re,string,exceptions

modname="helloxml"
__version__="0.1"

def debug(ftn,txt):
	sys.stdout.write(modname+'.'+ftn+':'+txt+'\n')
	sys.stdout.flush()

def fatal(ftn,txt):
	sys.stdout.write(modname+'.'+ftn+':FATAL:'+txt+'\n')
	sys.stdout.flush()
	sys.exit(1)

def usage():
	print __doc__

#====================================
from xml.dom import core, utils

def main():
	ftn = "main"
	pargs=[]    #positional args, default is empty
	opts,pargs=getopt.getopt(sys.argv[1:],'hv',['help','version'])
	for opt in opts:
		if opt[0]=='-h' or opt[0]=='--help':
			usage()
			sys.exit(0)
		elif opt[0]=='-v' or opt[0]=='--version':
			print modname+": version="+__version__
			sys.exit(0)

	#---get the DOM "doc"---
	if len(pargs)==0:
			usage()
			sys.exit(0)		
	filename = pargs[0]
	doc = utils.FileReader( filename ).document

	#---dump it out---
	doc.dump();

	#---convert DOM tree back to XML, and dump that
	xml = doc.toxml()
	print xml

	#---dump the doc manually---
	root=doc.get_documentElement()	
	mydump(root)

def mydump(node):
	ftn = "mydump"
	debug(ftn,"name=" +node.get_nodeName())
	#---dump attributes---
	attrs=node.get_attributes()
	keys=attrs.keys()
	for attr in keys:
		debug(ftn,"attr="+attr+" value="+attrs[attr].get_nodeValue())
	#---dump children---
	children=node.get_childNodes()
	for child in children:
        #--report only elements
		if child.get_nodeType()==core.ELEMENT: 
			mydump(child)
			 
if __name__ == '__main__': main()

-- 
Harry George
hgg9140 at seanet.com




From jilanik at tin.it  Mon Jun  5 11:24:16 2000
From: jilanik at tin.it (Jilani Khaldi)
Date: Mon, 05 Jun 2000 17:24:16 +0200
Subject: How to execute python scripts under Apache?
Message-ID: <393BC620.D44A2C26@tin.it>

Hi,
I have Apache compiled with PHP module (I use normally for web
programming), but now I am thinking to give python a try for my future
web applications. How do I modify the file httpd.conf to permit to
Apache to execute python scripts?

Thanks.

--
<jk>
___________________________________________________
http://space.tin.it/scuola/jkhaldi
FreeBSD ~ The Power to Serve





From darrell at dorb.com  Sat Jun  3 02:18:25 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 2 Jun 2000 23:18:25 -0700
Subject: Interprocess comunication
References: <8F456BAC0lminutieidit@192.6.1.10>
Message-ID: <03a801bfcd23$86fe5eb0$6401a8c0@home>

Pyro comes up little effort.
http://www.xs4all.nl/~irmen/python.html

--Darrell Gallion
----- Original Message -----
From: "Luca" <lminuti at eid.it>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, May 31, 2000 1:04 AM
Subject: Interprocess comunication


> I have a question: when occur a specific kind of event I need to notify it
> to some other instance of the same application. But I don't know how to
> realize a comuniction between different process.
>
> thanks.
> --
> http://www.python.org/mailman/listinfo/python-list




From set at ucsee.EECS.Berkeley.EDU  Wed Jun 21 20:43:10 2000
From: set at ucsee.EECS.Berkeley.EDU (Sam E. Trenholme)
Date: 21 Jun 2000 17:43:10 -0700
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHAECICEAA.jschmitt@vmlabs.com>
Message-ID: <8irneu$3j9$1@ucsee.EECS.Berkeley.EDU>

[Discussion of Makefile bugs and why a Python replacment would be nice]

I don't know if this has been brought up in this thread, but people are
working on a Python replacment for make: 

	http://software-carpentry.codesourcery.com/

- Sam
-- 
Please post, and not email, questions you have about my answers
Go to http://samiam.org/cgi-bin/mailme to get my email address


From akuchlin at mems-exchange.org  Wed Jun 21 15:44:33 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 21 Jun 2000 12:44:33 -0700
Subject: Regex engines
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org> <8igpab$7tf$1@slb2.atl.mindspring.net> <3950B2BE.AF756B9F@hursley.ibm.com>
Message-ID: <3dd7laes4x.fsf_-_@mems-exchange.org>

[Posted only to c.l.tcl and comp.lang.python; comp.lang.perl doesn't exist.]

Paul Duffin <pduffin at hursley.ibm.com> writes:
> front end to a C++ program. Note that is Tcl, not Tk. And someone posted
> don't use Tcl, use Python because it has a Tk binding and "Tcl is crap".

Ummm... If you mean the thread "C++ and Tcl", that person was writing
about Perl, not Python.

> I would love to see (and this is not my idea) all of our communities
> working together to produce really simple building blocks which can be
> used by all of us. e.g.
> 	Regular expression engine.

Unfortunately all the regex engines seem inextricably tied in to their
respective languages.  

* Python 1.5 uses a hacked version of PCRE, which is a very
good library, but it doesn't handle Unicode, though its author has said
he's thinking about adding UTF-8 support.  I think PHP and some Scheme
interpreters also have bindings for PCRE.  

* Python 1.6 will use a new regex engine called SRE to handle both
8-bit and UTF-16 strings; the compiler is still written in Python at
this point, though the engine itself is in C.  

* Perl 5's regex engine is unreadable code, and disentangling it from
the rest of Perl seems impossible.

* Tcl uses a Unicode-aware engine written by Henry Spencer, but it's
not available as a standalone library.  At the Ottawa Linux conference
last July I asked about releasing it, but have seen no sign of it
almost a year later.

* Ruby seems to use a modified version of one of the GNU regex
libraries; I can't tell if it's GNU rx or regex.

Unifying this situation seems impossible; maybe when SRE's compiler is
translated to C, or if Spencer's library is released in a standalone
version...

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Thank you for letting me borrow your objects.
  -- Ute Lemper in concert, March 13, 1997


From Wolfgang.Grafen at marconicomms.com  Thu Jun  8 01:07:06 2000
From: Wolfgang.Grafen at marconicomms.com (Wolfgang Grafen)
Date: Thu, 08 Jun 2000 07:07:06 +0200
Subject: Please help!!!!
References: <7WD%4.1099$0l2.8049@news2.tor.primus.ca>
Message-ID: <393F29FA.9E979E5B@marconicomms.com>

Robin Porter wrote:

> I have been trying to run some basic Python code in PYTHON version 1.5.2 in
> a Win9x OS and have been having one problem consistantly when trying to run
> it.  These are the steps I take each time.
>
> 1.)Open PYTHON idle (GUI)
> 2.)Open a new window
> 3.)Open the file I want to run (PAYpython.py)
> 4.)Click EDIT
> 5.)Click RUN
> 6.)A new screen appears and prompts me to enter the number of hours worked.
>
> (Up until this point all seems fine)
>
> 7.)I input the number 50
> 8.)Press enter and expect to be asked the rate of pay.
>
> **I GET NOTHING**
>
> Absolutely nothing happens after this point.....
>
> Could someone please tell me what is wrong.  I have re-installed Python
> 1.5.2.  I have tried other code and all seems to lead me to the same
> problem.  Please help!
>
> Thank you,
>
> Robin

It works fine with idle 0.6 and Solaris2.5.1 (Unix) on a Sun workstation ;)



From denalione at my-deja.com  Thu Jun 29 08:04:40 2000
From: denalione at my-deja.com (Dale Burnett)
Date: Thu, 29 Jun 2000 12:04:40 GMT
Subject: pythonservice on NT
Message-ID: <8jfe0g$3pl$1@nnrp1.deja.com>

Do all services run under one pythonservice process?  If not then is
there a way to change the name pythonservice in NT TaskManager to
something more descriptive?


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


From perlfreak at my-deja.com  Wed Jun 14 11:36:50 2000
From: perlfreak at my-deja.com (Perlfreak)
Date: Wed, 14 Jun 2000 15:36:50 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <8i88qf$t0$1@nnrp1.deja.com>

I have to say, in response to everyone that programming is like a
muscle.  The more you use it the better you get.  If you start with a
language like C++ you are using a large flabby muscle that it is
impossible to do anything with until you have worked and worked it.

However, if you transfer those programming skills (as I did) to Perl or
Python or another language (except Java/VB), suddenly the world seems a
better place because they are smaller muscles (Note: I do not mean less
powerful or important), and therefore easier and quicker to work up get
strong.

I would never advocate learning C++ for use in any other environment
other than high powered applications.  For the Web, Python is an
excellent language, as is Perl (please no hate mail!), because speed is
not so much of an issue, but the powerful constructs are still there.
For other development, I would probably advocate the use of these other
languages purely for speed.

The best way to learn is to find someone who knows the language and get
them to take you through the simple constructs and ideas of the
language.  After that, using this list is a brilliant way of asking
questions.  However, I still think hacking is such a fun way to program
(even though it doesn't produce wonderfully perfect programs all the
time!).

Personally, just learn python and stick with it.  You will learn the
transferrable skills (ie data structures, objects) for something like
C++ as you go along (although they may not seem to some to be as
powerful or well implemented as C++) and then something like python
becomes invaluable.

Currently I know C++, Perl, Python (less than Perl, but I like it),
ASP, Cold Fusion.  I started on C++, hacked it for a year, and then
started web programming with Perl.  I found I could write what it took
me a year to learn and write with C++ in about two weeks with Perl,
including learning it. I am now learning Python because of the simpler
syntax than Perl, and the way the language is implemented. I wish I had
started with Perl or Python, just because they're quicker to learn and
develop with.

I hope this is helpful.

Paul


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


From db3l at fitlinxx.com  Wed Jun 14 21:57:29 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Jun 2000 21:57:29 -0400
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <slrn8kg937.328.hzhu@rocket.knowledgetrack.com>
Message-ID: <ud7ljln92.fsf@ctwd0143.fitlinxx.com>

(dropped comp.lang.ruby since the whitespace is Python only)

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> If you use emacs:
> 
> C-x h M-x tabify
> 
> will tabify the whole buffer.  I always do this when doing copy and paste
> from news groups, and do an untabify the other way round.  This is because
> the default tab space is 4 in emacs (which is sane) but 8 everywhere else.

The default tab stop for a physical TAB character in Emacs is 8, which
is the most sane value.  A TAB character in an Emacs buffer is, by
default, treated as moving to the next boundary that is a multiple of
8 by default.  The tab stop width can in theory be overridden (see
variables default-tab-width and buffer-local tab-width), but doing so
is just plain evil.

Instead of changing tab stops, it is far preferable to set the
indentation for coding in various modes to anything you want (and
Python mode will default to 4).  When inserting indentation, Emacs
will use a mixture of TABs and spaces to get you where you need to be
in the most efficient manner, assuming the 8-character TAB multiple.

Or to put it more succinctly - the default Emacs interpretation of the
TAB physical character and Python's interpretation are the same, and
you should be perfectly safe letting Emacs mix TABs and spaces for
Python code, unless you've reset the default tab stops, which as I
mentioned above would be evil :-)

That said, the Python mode keeps everything with spaces anyway.

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


From cgiard at magerealm.com  Thu Jun 29 01:13:51 2000
From: cgiard at magerealm.com (Chris Giard)
Date: Wed, 28 Jun 2000 22:13:51 -0700
Subject: dl module on redhat
Message-ID: <MPG.13c47aba190946b1989682@news.linkline.com>

Hello.
  Does anyone know where (or if) the dl module is included in the Redhat RPMS 
for python?  I have been trying for the better part of today to find the dl 
module for python so I can call some C functions.  I wouldn't otherwise use 
the feature, but I need to be able to manipulate the password and shadow 
password files, and Python doesn't seem to have a good module for this (the 
pwd module only provides query support, not write support).  Thanks for your 
time.

======================================================================
// Chris Giard (a.k.a. Slyglif Cain) | Life is like a box of dynamite,
// EMAIL: cgiard at magerealm.com       | you never know when it will go.


From darrell at dorb.com  Sun Jun  4 02:03:11 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 3 Jun 2000 23:03:11 -0700
Subject: User Input?
References: <cnj_4.3811$r7.919327@news1.rdc2.tx.home.com>
Message-ID: <008301bfcdea$910ec990$6401a8c0@home>

name=raw_input("Name:")
addr=raw_input("Address:")
print name
print addr

--Darrell Gallion

----- Original Message ----- 
From: "Michael Morrison" <borlak at home.com>
> Is there an easy way to do user input in python?  I found a non-standard
> module, but it looked too complicated (not something I'd expect from the
> ease-of-use I've had so far).
> 
> Example: python blah.py
> Name? (type in name)
> Address? (type in address)
> etc....
> 





From see at below  Fri Jun 30 03:47:38 2000
From: see at below (Paul Foley)
Date: 30 Jun 2000 19:47:38 +1200
Subject: Python design philosophy
References: <Pine.SOL.3.96.1000628105812.1967C-100000@condor.ee.washington.edu> <395A521F.4374B144@estinc.com> <8F61B28C8gmcmhypernetcom@199.171.54.155> <395B6960.26AB52C1@estinc.com> <MdK65.2186$iN5.485797@ptah.visi.com> <395BBC97.F8939100@estinc.com> <geT65.2262$iN5.511399@ptah.visi.com>
Message-ID: <m2lmznzk2t.fsf@mycroft.actrix.gen.nz>

On Fri, 30 Jun 2000 02:03:56 GMT, Grant Edwards wrote:

> C predates the PDP-11 so we'll chalk it up to convergent
> evolution.

Oh no it doesn't.  C was written for the Unix port to PDP-11.
(Perhaps you're thinking of Unix!?)

-- 
Nomina stultorum in parietibus et portis semper videmus.      -- Cicero

(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From bvdpoel at uniserve.com  Sat Jun 17 12:25:10 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sat, 17 Jun 2000 09:25:10 -0700
Subject: How to create a Tk bitmap?
References: <82C25.1322$i06.292933@ptah.visi.com> <394AEFE7.6F3B718C@uniserve.com> <slrn8kn881.sdl.nobody@isis.visi.com>
Message-ID: <394BA666.E5F3B19B@uniserve.com>

Grant Edwards wrote:
> 
> In article <394AEFE7.6F3B718C at uniserve.com>, Bob van der Poel wrote:
> 
> >> I can't figure out how to create a Tk bitmap and then use it in
> >> a button.  According to the docs I can find, something like this
> >> ought to work:
> >>
> >>     myBitmap = BitmapImage(name='myBitmap', data='[...]')
> >>
> >>     Button(root,bitmap='myBitmap')
> >>
> >> But, Tk complains that there is no bitmap named 'myBitmap'.  If
> >> I leave out the "name=" when I create the bitmap and then use
> >> the name retrieved with str(myBitmap), I get the same error.
> 
> >The data is an exact copy of the file. For example:
> >
> >       bobpic = BitmapImage (data = """
> >       #define bob_width 107
> >        #define bob_height 154
> >       static char bob_bits[] = {
> >
> >0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,
> >          0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff
> >< blessedly snipped >
> >       0xff,0xff,0xff}; """ )
> 
> Yes, I know.  The call to BitmapImage() works fine, the error occurs when I
> try to use the bitmap in a button like:
> 
>    Button(root, bitmap='queen')
> 
> This is where the error occurs -- it doesn't matter if I use the name I
> specified in the call to BitmapImage or let Tk generate one and retrieve it
> with str().  What I can't figure out is how to use bobpic as the bitmap for
> a Button.
>

More from my snippet:

	l = Label(w, image=bobpic, borderwidth=4, relief='sunken')

Hope this helps.

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



From akuchlin at mems-exchange.org  Fri Jun  9 10:18:28 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 09 Jun 2000 10:18:28 -0400
Subject: Obscure Exception message
References: <75F7304BB41CD411B06600A0C98414FC0A418B@ORSMSX54> <3dr9a7yi91.fsf@amarok.cnri.reston.va.us> <20000609010518.C20700@xs4all.nl> <m3wvjzujgd.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <3dzoovlyyz.fsf@amarok.cnri.reston.va.us>

Michael Hudson <mwh21 at cam.ac.uk> writes:
> The Rapid-Fire-Python-Error-Message-Improver (me) says:

<K9>Checked in, Master.  Repairs complete.</K9>

The PyUnicode_Contains function needs a bit more work, though, since
it tries to coerce the left-hand operator to a Unicode string first.
This means that None in u'abc' gets "TypeError: coercing to Unicode:
need string or charbuffer".  Here's a proposed patch:

Index: unicodeobject.c
===================================================================
RCS file: /cvsroot/python/python/dist/src/Objects/unicodeobject.c,v
retrieving revision 2.24
diff -u -r2.24 unicodeobject.c
--- unicodeobject.c	2000/06/09 14:04:53	2.24
+++ unicodeobject.c	2000/06/09 14:14:48
@@ -2985,8 +2985,12 @@
 
     /* Coerce the two arguments */
     v = (PyUnicodeObject *)PyUnicode_FromObject(element);
-    if (v == NULL)
+    if (v == NULL) {
+	PyErr_Clear();
+	PyErr_SetString(PyExc_TypeError,
+	    "'in <string>' requires character as left operand");
 	goto onError;
+    }
     u = (PyUnicodeObject *)PyUnicode_FromObject(container);
     if (u == NULL) {
 	Py_DECREF(v);

Two questions are: 1) I suspect the PyErr_Clear() is unnecessary; am I
right?  2) Should it use PyErr_Format and say something like 'requires
character as left operand, not %s' % repr( <thing on LHS> ) ?

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Poor woman, I suppose she led a dog's life, and it made her disagreeable,
which she mistook for being strong.
  -- Robertson Davies, _The Rebel Angels_




From aahz at netcom.com  Fri Jun  2 11:22:43 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jun 2000 15:22:43 GMT
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com>
Message-ID: <8h8jg3$d66$1@slb6.atl.mindspring.net>

In article <8h87u1$sga$1 at news2atm.raleigh.ibm.com>,
Arint? <shouldbe at message.com> wrote:
>
>I am trying to write this escape seq to a device from python
>"H\x1b\x23\x12art\x1b\x23\x01\n"
>The \x1b\x23\x12 is supposed to be a command tell the device to do something
>special with "art", but python is making it look like this \x1b\x23\x12a
>"rt".  Am I missing something on how to work with escape sequences?  Is
>there another way to write hex values in a string.  I rather us \x00 because
>that is what are testers are familiar with.

I'd like to see some of the actual code.  Firing up the Python
interpreter produces correct resulte for me.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From invalid.address at 127.0.0.1  Tue Jun 27 22:05:03 2000
From: invalid.address at 127.0.0.1 (David)
Date: Wed, 28 Jun 2000 02:05:03 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <39595cb0.8198282@news.telus.net>

On Tue, 27 Jun 2000 13:10:46 GMT, "Don Tuttle" <tuttledon at hotmail.com>
wrote:

>In case you haven't heard Microsoft is coming out with a massive new Web
>stratagy that, in part, uses a new language called C# (That sharp, like in
>music).

That'd be a D-flat, eh?



From tanzer at swing.co.at  Fri Jun  9 10:05:48 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 09 Jun 2000 16:05:48 +0200
Subject: Python Productivity over C++ 
In-Reply-To: Your message of "Fri, 09 Jun 2000 07:14:26 EDT."
             <3940D192.97E4ACC3@colconsulting.com> 
Message-ID: <m130PPw-000wcEC@swing.co.at>

"Frank V. Castellucci" <frankc at colconsulting.com> wrote:

> > There are lots of reasons. The ones first springing to my mind are:
> > 
> > - The tools you mention notwithstanding, C++ still forces you to write
> >   lots of low-level code by hand (e.g., memory management, the
> >   `orthodox canonical class form').
> > 
> > - C++ doesn't offer any means of meta-programming. If you want to add
> >   some mechanism (like serialization, introspection, ...) to a class
> >   hierarchy, you'll have to add member functions to every class in the
> >   hierarchy. In Python you can implement the complete mechanism in the
> >   root class of the hierarchy.
> 
> This is crap:
> 
> 1. There are gc, SmartPointer, heap management, etc. class libraries out
> there. 

Guess what. Something like SmartPointer was one of the first classes I
implemented. In C++, that doesn't relieve you of thinking about memory
management, though.
 
> 2. There is the type information in standard conforming compilers.

Nowadays, there is. 

<rant>
If you happen to use a standard conforming compiler. In 1997, there
were still several compilers that didn't even fully conform to the
1990 edition of Stroustrup und Ellis's `The Annotated C++ Reference
Manual'. 
</rant>

RTTI doesn't offer any mechanisms remotely approaching Python features
like `__getattr__'/`__setattr__' though. Not to speak of `eval' and
`exec'. 
 
> 3. There is, has been, and will be more, C++ class Class libraries which
> are powerful, easy to use, and free. I was using NIHs Class class about
> 5 years ago.

So what? Your pains start as soon as you want to combine libraries
from different sources which happen to use incompatible idioms (like
overloading global operators, ...).

> Define 'orthodox canonical class form'

J.O. Coplien's `Advanced C++ Programming' will tell you all about it.

> > [snip]
> 
> People, we should all be realistic. I have seen so many language
> assertions that are unfounded or un-reasearched that the credibility of
> the author goes in the toilet with my urine.

I've used C++ and Python for projects of comparable size/complexity
and based my statements on that experience.

That's not scientific research, and of course it depends on my style
of using the two languages.

Your experience might be different.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From rjroy at takingcontrol.com  Thu Jun  1 23:56:44 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Fri, 02 Jun 2000 03:56:44 GMT
Subject: Getting the directory size
References: <2B1262E83448D211AE4B00A0C9D61B030130118F@msgeuro1.creo.be> <3936560E.C289AB29@bigfoot.com>
Message-ID: <39372f36.696896859@news1.on.sympatico.ca>

On Thu, 01 Jun 2000 13:24:46 +0100, Steve Tregidgo <smst at bigfoot.com>
wrote:

>Pieter Claerhout wrote:
>> 
>> <code>
>[rest of code snipped]
>> def main():
>> 
>>         dir = sys.argv[1]
>> 
>>         print "Testing directorySize..."
>>         print "Directory: %s" %(dir)
>>         print "Size: %s %s" %(getDirSize(dir)[0], getDirSize(dir)[1])
>> 
>> if __name__ == '__main__':
>>         main()
>> </code>
>> 
>> Is there any way to fasten up this beauty, because depending on the
>> number of directories it has to walk through, it takes a long time. Is
>> there maybe a function in the win32 modules who does this for me,
>> but lots faster?
>> 
>I don't know about a dedicated function to do that, but you can double
>the speed by changing the last line from:
>
>    print "Size: %s %s" %(getDirSize(dir)[0], getDirSize(dir)[1])
>
>...to the pair of lines:
>
>    amount, units = getDirSize(dir)
>    print "Size: %s %s" % (amount, units)
>

if getDirSize returns a 2-tuple can simply do 

print "Size: %s %s" %getDirSize(dir)

or slice it if its a longer tuple

print "Size: %s %s" %getDirSize(dir)[:2]

:-)

Bob
>The original version walked the entire structure twice -- once to
>calculate the amount, and then again to give the units.  Not really what
>you were after, but faster nonetheless.
>
>> Kind regards,
>> 
>> Pieter Claerhout
>> 
>
>HTH a little,
>Steve
>
>-- 
>Steve Tregidgo
>Software Developer
>http://www.businesscollaborator.com
>



From thomas at xs4all.nl  Fri Jun 30 12:02:59 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 30 Jun 2000 16:02:59 GMT
Subject: Problem with sys.exit()
References: <395CC5BA.AE64BB75@sas.upenn.edu>
Message-ID: <slrn8lph51.451.thomas@centurion.xs4all.nl>

On Fri, 30 Jun 2000 11:07:22 -0500, Steven M. Castellotti
<scastell at sas.upenn.edu> wrote:

>Relevant portion of feeder.py:

>grand_total = grand_total + \
>  os.system('zcat %s | python authlog_parser.py' % (filename) )
>print "%i" % os.system('zcat %s | python authlog_parser.py' % (filename)
>)
>print "Grand Total: %s" % grand_total

>But my output looks something like this:

>3
>768
>Grand Total: 1024

>	...where the "3" is the actual total getting passed into sys.exit(),
>768 is what parser.py is recieving, and 1024 being the total.

Yes. Exactly how it should be ;-)

>From the library documentation:

system (command) 
    Execute the command (a string) in a subshell. This is implemented by
    calling the Standard C function system(), and has the same limitations.
    Changes to posix.environ, sys.stdin, etc. are not reflected in the
    environment of the executed command. The return value is the exit status
    of the process encoded in the format specified for wait(), except on
    Windows 95 and 98, where it is always 0. Note that POSIX does not
    specify the meaning of the return value of the C system() function, so
    the return value of the Python function is system-dependent.
    Availability: Unix, Windows.

And the documentation for wait():

wait () 
    Wait for completion of a child process, and return a tuple containing
    its pid and exit status indication: a 16-bit number, whose low byte is
    the signal number that killed the process, and whose high byte is the
    exit status (if the signal number is zero); the high bit of the low byte
    is set if a core file was produced. Availability: Unix.

Basically, if you want the number as passed to 'sys.exit()', you divide the
return value by 256 (or shift right by 8 bits, which is the same thing.)

And when you do that with your exitcode:

>>> 786/256
3
>>> 786>>8
3


Regards,
	Thomas.


From stevena at neosynapse.net  Fri Jun 30 15:03:55 2000
From: stevena at neosynapse.net (Steven D. Arnold)
Date: Fri, 30 Jun 2000 15:03:55 -0400
Subject: Class design: accessing "private" members
In-Reply-To: <962367319.379353174@news.libertysurf.fr>
References: <962352843.1667731268@news.libertysurf.fr>
 <8jhuvn01vs0@news2.newsguy.com>
Message-ID: <4.3.1.2.20000630145109.0201d9e0@phear.dementian.com>

At 02:15 PM 6/30/2000 +0200, Jerome Quelin wrote:

>Then, is it better/cleaner to access private members with accessors or not?
>Is it a matter of style? Or are there hidden caveheats using (or not using)
>accessors? It _seems_ cleaner to use accessors, but man, it's quite awkward.

If you always use accessors, then you can store or derive the value in any 
way you wish internally in your class.  You may completely change the way 
the value is obtained, and the class user doesn't need to change anything 
in their code.  This makes upgrading the functionality of your class very 
easy -- for example, you could add alternate methods by which the class can 
obtain values.  Furthermore, if you use accessors you can do a lot of 
tricks that are much more difficult without them, such as logging every 
time a value is set or accessed, or taking multiple actions when a value is 
set.  Even if you don't need to do these things right now, if you don't 
religiously use accessors, sooner or later you will want to do that and it 
will be more difficult to do so when the time comes.  In this sense, it's 
definitely cleaner and more scalable to always use accessors.

Having said all that, I admit a little hypocrisy: in short scripts on rare 
occasions, I do not use accessors.  Sometimes they're more trouble than 
this script is worth.  However, beware of that sort of thinking.  Little 
throwaway scripts very often end up hanging around a long time and it's 
good to craft every line of code carefully with that in mind.


steve


--
Steven D. Arnold           Que quiero sera         stevena at neosynapse.net
"If you have  built castles in the air, your work  need not be lost; that
is where they should be.  Now put the foundations under them!" -- Thoreau




From tdixon.no at spam.fwi.com  Fri Jun 16 12:21:12 2000
From: tdixon.no at spam.fwi.com (Tim Dixon)
Date: Fri, 16 Jun 2000 16:21:12 GMT
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com>
Message-ID: <394a5109.80394991@10.1.1.28>

On Mon, 05 Jun 2000 19:55:08 GMT, Rajarshi Ray <rajarshi at home.com>
wrote:

>Besides the gee-whiz aspect of such algorithms, do they have any value?
>After all, what makes pi pi is ALL of its digits in a particular
>sequence.

I can think of a great use for such an algorithm, if one existed (I
think this is a misunderstanding and a closer examination will reveal
that all they've found is that such an algorithm isn't theoretically
impossible or something like that).

The strongest possible encryption is a secret key one time pad.  A one
time pad is an encryption method by which each byte (or couple bytes)
is encrypted using a different key which is discarded and not used
again.  It's not possible to break this kind of encryption without
compromising the "key pad" itself because discovering one byte tells
you nothing about any of the other bytes.  Being unable to test a
sequence of bytes makes it impossible to tell whether you've cracked
the byte or not, and even if you did that, you'd still not have the
rest of the message.  An advantage is the the algorithm can be very
simple (a simple XOR against the key would suffice) because the
security is in the key, not the algorithm.

There are two problems:  one is that it's a secret key system and not
a public key system, so it's not useful for many applications.  It
doesn't work for something like SSL or PGP, for instance.  This is a
characteristic of secret key systems.

The second is the enormous effort involved in producing one time pads.
You need enough bits in your pad to encrypt the entire message;
essentially for a 16K message, your key would have to be 16K long as
well.  Naturally, you need to get the key to the recipient via some
secure means and communicate what portion of the pad is used by a
particular message.

Suppose, however, that you could get digit 'n' of pi directly (or e,
or any other inifinte, nonrepeating sequence).  All I have to do is
communicate which digit of pi the message key starts at, and I can
generate the rest of the key.  I still have the unbreakable nature of
the one time pad, but now I only have to keep secret the starting
point; I can generate the key on the fly.  Keeping *that* secure is
still nontrivial, of course. :-)

>
>Kirby Urner wrote:
>> 
>> Here's another "math through programming" challenge:
>> 
>> Recent number theory breakthrough:  you can get the
>> nth digit of PI without doing all the work needed
>> to compute all the digits up to the nth digit (same
>> with some other numbers -- not sure if we know all
>> the relevant criteria as to which).



From neilh at scintilla.org  Wed Jun 21 04:20:03 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 21 Jun 2000 08:20:03 GMT
Subject: Output from a python script to a calling program.
References: <3951f692.948648187@nntp.interaccess.com> <M7T35.3891$qz4.30140@news-server.bigpond.net.au> <395062db.976369718@nntp.interaccess.com>
Message-ID: <TU_35.4187$qz4.31727@news-server.bigpond.net.au>

> Yep, that did the trick. The question now is can I turn off buffering
> from inside the script?

   You can call flush on the outputs when you want output to appear.

   You can open the console with buffer size set to 0 and then assign this
file to sys.stdout and sys.stderr.

   Neil





From pehrens at nospam.ligo.caltech.edu  Wed Jun 21 20:04:03 2000
From: pehrens at nospam.ligo.caltech.edu (Phil Ehrens)
Date: 22 Jun 2000 00:04:03 GMT
Subject: Regex engines
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org> <8igpab$7tf$1@slb2.atl.mindspring.net> <3950B2BE.AF756B9F@hursley.ibm.com> <3dd7laes4x.fsf_-_@mems-exchange.org> <39514BD1.227B1131@nospam.com>
Message-ID: <8irl5j$438@gap.cco.caltech.edu>

Mo wrote:
>
>Did you know that there is a standalone version of henry's regexp
>engine ported to Java? It is included in the Jacl interp (Tcl in Java),
>but it is just a package, so you should be able to use it anywhere.
>In fact, it is just two file Regexp.java and Regsub.java. You should
>check it out.

It should be a small job to make a Tcl loadable standalone regexp
extension.  Just exposing the small set of functions is no big deal.
If somebody can point me to the version they would like to see
wrapped, I will do it.  Just post here.  No direct e-mails please.

-- 
Phil Ehrens <pehrens at ligo.caltech.edu>| Fun stuff:
The LIGO Laboratory, MS 18-34         | http://www.ralphmag.org
California Institute of Technology    | http://www.yellow5.com
1200 East California Blvd.            | ftp://ftp.no.pgpi.com/pub/pgp
Pasadena, CA 91125 USA                | http://slashdot.org
Phone:(626)395-8518 Fax:(626)793-9744 | http://freshmeat.net


From gmcm at hypernet.com  Fri Jun 23 22:43:42 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 24 Jun 2000 02:43:42 GMT
Subject: Java Objects vs Pythons?
References: <iUL45.23838$ds.694847@newsread2.prod.itd.earthlink.net>
Message-ID: <8F5CEEDFAgmcmhypernetcom@199.171.54.195>

Arint? wrote: 

>I want to do a wait object like in java.  

[snip]

>waitObject = ???<What type of Lock should I use
>def callbackNotify():
>    waitObject.acquire()
>    waitObject.notifyAll()
>    waitObject.release()
>
>...
>#main do some stuff
>waitObject.acquire()
>waitObject.wait()
>waitObject.release()

That's a threading.Condition.

>In the script I don't create any threads or anything, can I do a wait
>object without creating a thread (pretty sure you can't do that in
>java)?  I guess there kinda is a thread because there is the python
>script and then there is the C++ app with the callback(usually waiting
>for output to finish on a printer).  What kind of Lock would be best the
>Lock or RLock or other? 

Without threads, none of these will do you any good. I guess you want 
somebody to block while "waiting", but who (C++ or Python) or what 
"waiting" means in this context, I don't know...


From dana at oz.net  Sun Jun 11 13:14:29 2000
From: dana at oz.net (dana_booth)
Date: 11 Jun 2000 17:14:29 GMT
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21> <3942DC44.4A85EEA7@rubic.com> <39430A2E.99CE3FC@yahoo.com> <394396CB.8D2D53F0@rubic.com>
Message-ID: <8i0hdl$50e$0@216.39.141.21>

Jeff Bauer <jbauer at rubic.com> wrote:

JB: I've got MySQLdb.py and
JB: _mysqlmodule.so in my .../lib/python1.5/site-packages 
JB: directory, which is where build.py is supposed to put
JB: them.

That's another thing that I'd forgotten about... After running build.py,
further instructions say to 'make install', which I did, but only the shared
object ended up in site-packages. I manually copied the MySQLdb.py* files to
site-packages. Not a big deal, but a little thing like that might throw
someone...

I was also surprised that it built and installed on a Windows computer with
no fuss at all. While I'll be using a Linux computer to make the clients
with Tkinter, they'll actually be run on Windows. (with MySQL on OpenBSD :)
Therefore, I was ecstatic to get this module working, it'll make life soooo
much easier!

-- 

-----
Dana Booth <dana at oz.net>
Tacoma, Wa., USA

key at pgpkeys.mit.edu:11371


From darrell at dorb.com  Sat Jun 10 16:14:37 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sat, 10 Jun 2000 13:14:37 -0700
Subject: Type problem on Windows
References: <eviltofu-B4ED2B.12472010062000@news.earthlink.net>
Message-ID: <02d101bfd318$8091ae10$6401a8c0@home>

On Win2k:

>>> import socket
>>> s = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
>>> s.__class__
<class socket._socketobject at 88d6b8>
>>> s.__class__.__name__
'_socketobject'
>>>
>>> type(socket)
<type 'instance'>
>>> socket.__class__
<class socket._socketobject at 88d6b8>
>>>

sockets are platform dependent as expected :)

--Darrell Gallion


From: "Jerome Chan" <eviltofu at rocketmail.com>
> 
> On Unix and Mac, this works. On Win98, this bombs.
> 





From neilh at scintilla.org  Sat Jun 17 21:37:37 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 18 Jun 2000 01:37:37 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com> <39520785.690843718@nntp.interaccess.com>
Message-ID: <BJV25.1438$qz4.12296@news-server.bigpond.net.au>

> The point is not that I can copy and paste with out reindenting.
> The point is that sometimes I have to. let me show a generic example
>
> if a:
>    for b in B:
>       if c(b):
>           block statement.
>
> Assume that you are editing another part and you realise that you
> are about to essentially retype the block statement. So you do
>
> def fun():
>   block statement
>
> if a:
>    for b in B:
>       if c(b):
>            fun()
> you have to reindent the block statement. This is not copy and paste
> reuse. It is cut and paste refactoring, and you have to reindent.

   Your example does not /require/ any reindenting. It may look better
reindented but this is one case where refactoring does not have to lead to
reindenting.

   Neil





From dej at eastmail.com  Mon Jun 19 08:37:22 2000
From: dej at eastmail.com (dej (instructions in sig))
Date: Mon, 19 Jun 2000 21:37:22 +0900
Subject: ||| Help requested from newbie |||
References: <nkfrkss0a2ar68n165sok07pau6gegun71@4ax.com> <323d42ca.3ff86364@usw-ex0105-035.remarq.com>
Message-ID: <3u4skso873ok5r7mbpv7ic4lp72enk5ekr@4ax.com>

On Mon, 19 Jun 2000 01:07:41 -0700, Richard Chamberlain
<richard_chamberlainNOriSPAM at ntlworld.com.invalid> wrote:

>Hi David,
>
>The problem is in the function Analyze, specifically groups =
>getCharGroups(infile).
>
>If you look at the function getCharGroups it has no return
>statement, in these cases Python actually returns none.
>
>So although you are using the global groups all the while and
>appending words to the end as soon as the function completes it
>just assigns none to groups.
>
>for item in wordList doesn't work because it's trying to iterate
>over none.
>
>Replace that line with
>
>getCharGroups(infile)
>
>And it should work.
>
>Richard
>
>Got questions?  Get answers over the phone at Keen.com.
>Up to 100 minutes free!
>http://www.keen.com


Thank you! I heard from the author as well and offered the same
solution.

David

Without the number 0041 in the subject line of the message, 
any messages sent to dej at eastmail.com are discarded by the server.


From mkx at excite.com  Thu Jun 15 16:31:03 2000
From: mkx at excite.com (mkx at excite.com)
Date: Thu, 15 Jun 2000 16:31:03 -0400
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com> <Ra525.3904$227.93795@nnrp1.uunet.ca>
Message-ID: <4neikson9anst4npr4kvf7i6d03c4cb027@4ax.com>

On Thu, 15 Jun 2000 09:51:32 -0400, "Warren Postma"
<embed at NOSPAM.geocities.com> wrote:

>I have many developers in my office who would cry more loudly than I about
>Python's "required indentation", and yet these same developers are always
>telling me that all our company's programmers should tab and indent their
>code the same way, for "readability".   Few can see the contradiction there.

In some of my past experiences, these would be some of the same
developers who would balk at supplying documentation. 

In some respects, the "forced" whitespace helps provide Python some of
it's famed readability. It may not be necessary for my own code, but
it sure would have helped in some situations, where programmers seemed
to have gone out of their way to write in an incomprehensible fashion.

Sometimes the "code from hell" is just due to poor training or
discipline on the programmer's part, but I would swear that in some
cases it was on purpose (all humor aside)!

Even with my impeccable logic above (:), I am sure that this is one
argument that will never be won.


From alex at magenta.com  Fri Jun 30 07:02:40 2000
From: alex at magenta.com (Alex Martelli)
Date: Fri, 30 Jun 2000 13:02:40 +0200
Subject: Class design: accessing "private" members
References: <962352843.1667731268@news.libertysurf.fr>
Message-ID: <8jhuvn01vs0@news2.newsguy.com>

Jerome Quelin <jerome.quelin at insalien.org> wrote in message
news:962352843.1667731268 at news.libertysurf.fr...
> Hi all,
>
> When designing a class, I wrote some accessors (get_var() and set_var()
> methods) for my "private" members (private in the sense that I don't want
> somebody to access them directly from outside the class.
> I was wondering if it is good practice to acceed to those private members
from
> _inside_ my class, or should I use the accessors methods too?
> Well, ok, it would be slower (function calls are always more expensive),
but
> isn't it more OO-well-behaviored?
>
> -- File test.py
> class Spam:
>     def __init__(self, name="eggs"):
>         self.set_name(name)  # Which is the
>         self._name = name    # better/cleaner?
>     def set_name(self, name):
>         self._name = name
> -- End of file

By calling the accessors, rather than accessing the data
directly, you ensure that if somebody overrides the
accessor functions in a derived class you'll be calling
the derived-class versions of those functions.  If that is
what you want (one direction along which you want to let
inheritors of your class extend or tweak behaviour), then
it's probably worth the overhead you mention.  But if you
need to ensure that the actual data gets modified and/or
updated, then beware...!


Alex





From alex at magenta.com  Mon Jun 26 09:26:11 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 26 Jun 2000 15:26:11 +0200
Subject: boolean true and false values.
References: <slrn8l69kp.geg.hat@se-46.wpa.wtb.tue.nl> <memo.20000625230712.5257C@digitig.compulink.co.uk>
Message-ID: <8j7n6h01tnl@news2.newsguy.com>

Tim Rowe <digitig at cix.co.uk> wrote in message
news:memo.20000625230712.5257C at digitig.compulink.co.uk...
    [snip]
> > Yes, I know, but why is it not in the language ?
> > Quite a lot of algorithms rely on having sets. If you put them in a
> > list, there
> > is an order which is not always wanted.
>
> There is always an order, but you don't always know what it is. When you
> write your set class you just document that if the user iterates over the
> elements then the order in which they will be returned is undefined --
> that's what happens in the C++ STL. The fact that the implementation has a

Good general concept, un-good specific example.  C++'s std::set (originally
inspired from STL) is based on an ordering relationship, and when you
iterate
from .begin() to .end() you get the elements in increasing order according
to
just that relationship.  The current release of STL also has hash-based
containers (which didn't make it into the C++ standard library), but the
standard ones are sorted...


Alex






From bparsia at email.unc.edu  Mon Jun 19 20:47:09 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Mon, 19 Jun 2000 20:47:09 -0400
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl>
Message-ID: <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu>

Thomas Wouters <thomas at xs4all.net> wrote:

> On Mon, Jun 19, 2000 at 06:52:50AM -0400, Bijan Parsia wrote:
> > William Dandreta <wjdandreta at worldnet.att.net> wrote:
> 
> > > The biggest improvement came (about a factor of 6) when I changed the
> > > replace(x,y) function in string with
> > > joinfields(splitfields(x,y),''). Considering that this is exactly what the
> > > replace function in string does, I was quite surprised at the results.
> > > Essentially I reduced 3 nested function calls to 2.
> 
> > And that can make an enormous difference, particularly with earlier
> > versions. Remember that many "simple function calls" in fact involve
> > rather expensive polymorphic lookup and dispatching. In a tight loop
> > this can get very painful (Python doens't have, I believe even in the
> > latest versions, a method/function cache, so every lookup will go
> > through the whole rigamarole each time).
> 
> Indeed. Such a cache would have to be invalidated whenever the function
> object or one of the objects used to lookup the function, was altered (or
> rather, 'touched'.) The dynamic nature of Python makes this quite
> complicated, and probably quite expensive.

I'd love to hear more about why this would be the case. (Mere
"dynamicness" can't be it: Smalltalk is a s dynamic and most
implementations make quite effective use of various sorts of cache. So,
if your contention is correct, there must be some *special* aspect of
Python's dynamic nature that makes it' *especially* resistent. Note that
I am *very* skeptical about the existence of such a special aspect, but
I am quite willing to be enlightened! Merely being "more dynamic" or "so
very dynamic" which is what I usual read, doesn't cut any mustard.

Indeed, as my bet is that the majority of the time, in the majority of
code, things are quite "hands off" (i.e., no touching :)), I would be
surprised that this would prove a problem. In inner loops, indeed, it
could be quite effective, and would be no different than *manually*
copying the function to the local--a move, I hope you agree, can provide
a substantial speedup.

I confess to not being an expert in these matters, but I don't know I
understand why the *whole* cache would have to be invalidated if one
function object were "touched". Er..in mere other words...More please!
:)

> What's more, currently 'bound methods' (methods which have a 'self' to which
> they apply) are created on demand -- the unbound method exists on the class,
> but it has to be 'bound' to an instance before it can be used. This is done
> each time you access the method, though CPython is being smart and you have
> to try hard to actually see that ;-)

Hmm. Ok. I think. Maybe. Why it has to be do each time one accesses the
*instance*'s method I find a tad confusing. I mean ok, you change the
class def, you'll want to update the instances. But surely it's ok to
make class updates more expensive than method access?

> >>> class Minister:
> ...     def walk(self):
> ...             print "The minister walks silly."
> ... 
> >>> cleese=Minister()
> 
> >>> cleese.walk is cleese.walk
> 0
> >>> x,y=(cleese.walk, cleese.walk)
> >>> id(x), id(y)
> (134879776, 134879248)

Ewww! :) But *why* do this? walk hasn't changed. It's surely
determinable that it hasn't changed over the course of x,y=(cleese.walk,
cleese.walk), yes?

[snip]

I'll work on digesting the rest. Thanks.

-- 
Bijan Parsia
http://monkeyfist.com/
...among many things.


From pwilkinson at maxis.com  Thu Jun 15 14:29:36 2000
From: pwilkinson at maxis.com (Paul Wilkinson)
Date: Thu, 15 Jun 2000 11:29:36 -0700
Subject: tkinter vs. Perl/Tk
Message-ID: <vg925.1551$Qf6.60136@nuq-read.news.verio.net>

How does tkinter compare to Perl's implementation of Tk?  I'm a little
frustrated with Perl/Tk - I have strange focus problems that I've never
managed to tackle.  I never did manage to find any decent Perl/Tk
documentation.  Will I be happier with Python?

PEACE

OUT





From digitig at cix.co.uk  Thu Jun 15 15:28:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Thu, 15 Jun 2000 20:28 +0100 (BST)
Subject: 8 queen/game of life
References: <39485C09.306BF3DC@videon.home.com>
Message-ID: <memo.20000615202834.52185D@digitig.compulink.co.uk>

In article <39485C09.306BF3DC at videon.home.com>, krussell4 at videon.home.com 
(Kevin Russell) wrote:

> root wrote:
> 
> >    the matrix is fixed, which is not good. my solution use dictionary 
> > (i
> >    know this is not the right way). the key is a tuple of the matrix
> >    indices. the value is '0'. then base on game of life rule, update 
> > the
> >    dictionary and print for next generation.
> >
> >            m={}
> >            m[(0,0)] = 0, m[(0,1)] = 1,....
> 
> If it works, then it IS the right way.  After all, there's more than one
> way to... oops, sorry, wrong newsgroup.
> 
> As Trent Mick has pointed out, there's NumPy for doing serious
> matrix manipulation.  But the demands of the Game of Life and the
> 8-Queens problem are minimal enough that a list-of-lists solution
> or your dictionary solution are perfectly adequate.

Hmm. Game of Life is often considered computationally expensive. Though 
mainly because the sort of things people do tend to expand to fill the 
cycles available...


From gansevle at cs.utwente.nl  Thu Jun 22 07:58:44 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Thu, 22 Jun 2000 11:58:44 GMT
Subject: Dynamically add methods to classes?
References: <Sql45.1512$9W1.105781@news-west.usenetserver.com>
Message-ID: <8isv1d$tfn$1@nnrp1.deja.com>

In article <Sql45.1512$9W1.105781 at news-west.usenetserver.com>,
  Matthew Cline <matt at nightrealms.com> wrote:
> Is is possible to dynamically add methods to classes in Python?
>

If you take it one step furter and say: "Is it possible to dynamically
add (or change) baviour of classes in Python?", since the methods of a
class define the behaviour of that class, the answer is yes.

<SHAMELESS_PLUG>
get my dci module
(http://wwwhome.cs.utwente.nl/~gansevle/python/Dci-0.1.0.tar.gz) and use
it like this.

class A:
    def some_method(self):
        ...

class A1:
    def other_method(self):
        ....

class A2:
    def other_method(self):
        ....

...

import dci
a = A()
dci.inject(a, A1)
a.other_method()

# if you (or your code) decides that the 'other_method' of class A2
# should be used now, change the behavior.

dci.retract(a, A1)
dci.inject(a, A2)
a.other_method()

</SHAMELESS_PLUG>

> Thanks in advance.
>
>

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From pablo.prieto at dulcesol.es  Wed Jun  7 06:01:11 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Wed, 07 Jun 2000 11:01:11 +0100
Subject: [Fwd: PyOpinion: Does Python Programming Marginalize You?]
Message-ID: <393E1D67.E0D57892@dulcesol.es>

Pablo Prieto escribi?:
 
 Hi, evrybody there!.
 
 Being a programmer for a company which sells food, we make all our
 programs without relation to the internet. We have an Apache web server
 and several cgi-bin for showing info to our customers. But rather
light.
 I've been in C,C++,Pascal and Ada (this is not a curriculum, eh? ;))
and
 now in VB. We want to migrate to Linux workstations (servers are all
 linux now, except the database server, we are testing postgreSQL for
the
 migration).
 
 I'm evaluating Python now. Maybe in a month I kick off this list and
the
 Python installation, but, till now, I'm amazing of the languaje. Really
 good and nice to learns, being it's "advanced" power more or less
 "hidden" for learning when you are confortable with the basis.
 
 I'm being in love with this languaje (hope not to runaway, and stay
 forever). The web is important, but there's still a lot of traditional
 client-side work to do (I don't know about server side, I only buy it,
 and that's enough :)).
 
 And surely, could appear python-related semi-meta-x-languajes for
 scripting. That's what happened with java and  VB (javascript and
 vbscript), isn't it?
 
 BTW, Ada times were very exciting with all that people in uniforms in
 the meetings, and the great amounts of money in the horizont. At the
 end, the dream was over. But it was funny.
 
 Pablo, from Spain.



From nikolai.kirsebom at siemens.no  Wed Jun 21 03:05:49 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Wed, 21 Jun 2000 07:05:49 GMT
Subject: win32all-132 / Word Automation
References: <394e2b8d.515788975@news.mch.sni.de> <lAy35.3131$qz4.24329@news-server.bigpond.net.au> <394f269c.580059221@news.mch.sni.de> <ZVI35.3597$qz4.27900@news-server.bigpond.net.au>
Message-ID: <395068d0.58350693@news.mch.sni.de>

Hi again,

The result of
>>> print repr(doc.Variables.Item(1))

is

<win32com.gen_py.Microsoft Word 8.0 Object Library.Variable>

------

Nikolai



From paul at prescod.net  Wed Jun 21 12:23:27 2000
From: paul at prescod.net (Paul Prescod)
Date: Wed, 21 Jun 2000 11:23:27 -0500
Subject: boolean true and false values.
References: <3947f922.425016156@nntp.interaccess.com> <slrn8l1oc2.3c1.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <3950EBFE.AAA46ABE@prescod.net>

Albert Hofkamp wrote:
> 
> ...
>
> On the other
> hand, I use bool and its true and false values quite often in C++ rather than 0
> and 1 which leaves way to much room for mistakes like assigning 2 to it.

Python will never prevent you from assigning "2" to a variable anyhow.
There's the primary reason that there is little demand for a Python
boolean. It would prevent few mistakes and is thus not that useful. I
think it is a little bit more syntactically nice to say "true" and
"false" but it is not a large deal.
 
> Also, I am missing the data type 'set'. It can be faked with a dictionary, but it
> is just not the same. Any one else got this problem ?

Another thing that would be nice....

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"Music is the stuff between the notes." - Claude Debussy



From aahz at netcom.com  Wed Jun  7 10:14:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 14:14:09 GMT
Subject: ANNOUNCEMENT: The Python Imaging Library, version 1.1 (june 7, 2000)
References: <014a01bfd089$d32b4af0$0500a8c0@secret.pythonware.com>
Message-ID: <8hllbh$rqr$1@slb0.atl.mindspring.net>

In article <014a01bfd089$d32b4af0$0500a8c0 at secret.pythonware.com>,
PythonWare <info at pythonware.com> wrote:
>
>the pil team

Does this make you guys pils?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From neilh at scintilla.org  Tue Jun  6 20:43:32 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 07 Jun 2000 00:43:32 GMT
Subject: Future of PythonWin?
References: <w3Q_4.160138$55.3354931@news2.rdc1.on.home.com> <m03%4.6579$Hz.48365@news-server.bigpond.net.au> <EHa%4.165801$55.3514980@news2.rdc1.on.home.com>
Message-ID: <UUg%4.7105$Hz.52820@news-server.bigpond.net.au>

> I agree that MFC seems to be on the way out ...
> 2) What's next?
> - STL?  Yes, but that's not a GUI.
> - ATL & WTL?  Too MS only.
> - wxWindows?  Only has a tiny fraction of the number of users & code that
> the other options have.
> - TCL/TK?  I've heard this is slow, and it can't create native windows'
apps
> (users want the windows look & feel).
> - Java?  Again, too proprietary.

   It looks to me like the future of UIs (and I think this is the MS
direction as well) is HTML and related standards. MS has put a lot of effort
into HTML generation and bridges between HTML and code.

   Neil





From jvickroy at sec.noaa.gov  Mon Jun 12 13:40:49 2000
From: jvickroy at sec.noaa.gov (j vickroy)
Date: Mon, 12 Jun 2000 11:40:49 -0600
Subject: Testing for a file in python?
References: <39450724.231994953@nntp.interaccess.com>
Message-ID: <394520A1.77E47C1A@sec.noaa.gov>

Try the exists() method in the os.path module as follows:

>>> import os.path
>>> print os.path.exists ('my_file')
0
>>>

"Thaddeus L. Olczyk" wrote:

> How would you go about testing if a certain file exists in python?
> In most scripting languages you have the -f test:
>
> if -f filename do
>    whatever_you_were_going_to_do
>    fi
> else do
>    print "Error" file not found
>    fi
>
> stat doesn't work, it aborts if the file does not exist.



From akira.kiyomiya at autodesk.com  Mon Jun 26 13:56:11 2000
From: akira.kiyomiya at autodesk.com (Akira Kiyomiya)
Date: Mon, 26 Jun 2000 10:56:11 -0700
Subject: books????
References: <sla4ago0jev12@corp.supernews.com> <961961617.1744372225@news.rma.edu>
Message-ID: <8j85gu$e86@autodesk.autodesk.com>

I am using "Learning Python" from O'Reilly and "Python Essential Reference"
from New Riders.  They are both pretty good books for beginners.

Akira

"Robert" <rhicks at rma.edu> wrote in message
news:961961617.1744372225 at news.rma.edu...
> On Sat, 24 Jun 2000, SugarKing wrote:
> >Are there any good beginner books for Python? I'm itching to get my hands
on
> >any. Thanks
>
> There is:
>
> Learning Python
> Learning Python in 24 hours
> Python and Tkinter programming
>
> There should be links to these books in the library portion of the python
> website. I have not used the LP in 24 hours but have heard good things
about it.
>
> Robert




From digitig at cix.co.uk  Sat Jun  3 18:56:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sat, 3 Jun 2000 23:56 +0100 (BST)
Subject: How to find out height of a Canvas
Message-ID: <memo.20000603235615.14417A@digitig.compulink.co.uk>

How do I find out the height of a Tkinter Canvas? canvas.itemcget looks 
promising, but it needs two arguments (I get a Tcl error if I omit the 
second argument, or set it to None), and I can't find any documentation on 
what the second argument should be (or the first argument, for that 
matter; I'm assuming /that/ should be a string containing the name of the 
attribute).

TIA.


From NOSPAM_rdrew at cia-group.com  Mon Jun 12 17:10:38 2000
From: NOSPAM_rdrew at cia-group.com (Ray Drew)
Date: Mon, 12 Jun 2000 22:10:38 +0100
Subject: GUI editor
References: <8457258D741DD411BD3D0050DA62365907A1F9@huina.oceanic.com>
Message-ID: <8i3jpi$ddm$1@soap.pipex.net>

Doug Stanfield wrote in message
<8457258D741DD411BD3D0050DA62365907A1F9 at huina.oceanic.com>...
>Trid to find SpecPy but couldn't find references to it at the above site,
>the Vaults, or www.python.org
>
>Got any leads? It isn't bundled with the SpecTCL installer I grabbed.
>
>-Doug-
>

Try:

http://www.geocities.com/eric_brunel/SpecPython.en.html

Ray Drew
(I have no control over the footer, its added to all outgoing mail)




From scarblac-spamtrap at pino.selwerd.nl  Sun Jun  4 14:19:40 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 4 Jun 2000 18:19:40 GMT
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why -- or why  not? (was Re: Damnation!))
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <393326D3.400AA248@cosc.canterbury.ac.nz> <u1z2gl8zu.fsf@cs.uu.nl> <roy-C89718.12394104062000@netnews.nyu.edu>
Message-ID: <slrn8jl7bq.ikb.scarblac-spamtrap@flits104-37.flits.rug.nl>

Roy Smith wrote in comp.lang.python:
> piet at cs.uu.nl wrote:
> > As it is now in Python, you can have a==b and c==d both being true, but
> > a/c==b/d being false. Which would be a bad surprise for most people.
> 
> What values of a, b, c, and d would cause that?  Maybe I'm just being 
> dense, but i don't see it.

a = 1, b = 1.0, c = 2, d = 2.0

a==b and c==d, but a/c==0 and b/d==0.5.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Which is why you get 'em so cheap":
   In SPI's Universe, the sword is prohibited from use at any combat
   range.


From kens at sightreader.com  Tue Jun 13 06:59:35 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 13 Jun 2000 03:59:35 -0700
Subject: Python Productivity over C++
References: <20000609185450.91797.qmail@hotmail.com>
Message-ID: <39461417.F7BB78AC@sightreader.com>



Hung Jung Lu wrote:

> <snip>
> The biggest advatages of script languages are that they make dynamic
> reloading easier and they offer reflection (or introspection): the language
> is capable of looking into itself, and even perform meta-programming. I know
> you can make your C++ program to do this kinds of things, but you have to
> spend a tremendous amount of development time before you can reach some
> minimum features of dynamic reloading and reflection. (I have also used Java
> reflection features, but Python's reflection is way simpler and superior.)
> <snip>
> Hung Jung

I'm developing a python module that will generate C code (translated from a
C-like custom language edited within the python application), invoke the
compiler, and dynamically (re)load the resulting extension module.  With a
little help from Python, even C/C++ can be dynamic!

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000613/6ac8f188/attachment.vcf>

From frankc at colconsulting.com  Fri Jun  9 19:18:14 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 09 Jun 2000 19:18:14 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch> <8hrrfp$880$1@slb6.atl.mindspring.net>
Message-ID: <39417B36.B1A4F325@colconsulting.com>

Aahz Maruch wrote:
> 
> In article <8hqbvr$fgr$1 at sunnews.cern.ch>,
> Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote:
> >
> >First off, any claim of an order of magnitude increase in productivity
> >is pure nonsense. See Brooks for an explanation or Boehms and McConnell
> >for hard data.
> 
> That's half-true.  Lines of code per unit time is roughly constant per
> programmer (though I think Python breaks even that rule).  Of more
> interest is the extent to which one language has greater expressiveness
> than another for a given domain.  So if Python programs are generally
> one-tenth the size of C/C++ programs, there *is* an order of magnitude
> increase in productivity.

I would be careful with that bat. The more libraries you use, the more
editors that provide expansion and completion, the more generator type
tools available increase the LOC but reduce the time spent at the
keyboard. For any language.

Productivity is not measured by LOC alone, it needs to be measured by
QLOC (Quality lines of code). This is easily tracked with most revision
control systems. If developer A spends ten minutes putting out crap
code, then developer be must come along and spend even more time
debugging and correcting those "productive" lines of code.
 
> Where I think Python really kicks butt is in maintenance, though.
> Reading Python code is significantly easier than any other language I've
> seen; rewriting Python code is also significantly easier.

Agreed.

-- 
Frank V. Castellucci


From moshez at math.huji.ac.il  Wed Jun 28 02:44:09 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 28 Jun 2000 09:44:09 +0300 (IDT)
Subject: compilation
In-Reply-To: <Pine.LNX.4.21.0006270751280.2223-100000@fep132.fep.ru>
Message-ID: <Pine.GSO.4.10.10006280942550.27161-100000@sundial>

On Tue, 27 Jun 2000, Oleg Broytmann wrote:

> #! /bin/sh
> 
> if [ -z "$1" ]; then
>    echo "Usage: compyle file.py..."
>    exit 1
> fi
> 
> TEMPLATE="from py_compile import compile; compile('"
> 
> for file in $*; do
>    pgm=$TEMPLATE$file"')"
>    python -c   "$pgm" || exit 1
>    python -OOc "$pgm" || exit 1
> done


#!/usr/local/bin/python
import sys, py_compile

for file in sys.argv[1:]
	py_compile.compile(file)

It's a bit shorter <wink>
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From janne at nnets.fi  Mon Jun 19 18:14:27 2000
From: janne at nnets.fi (Janne Sinkkonen)
Date: 20 Jun 2000 01:14:27 +0300
Subject: How does __setitem__() work?
References: <slrn8kt5iu.2cn.bob@slim.instantappointment.com>
Message-ID: <m3og4xxqrg.fsf@kinos.nnets.fi>

bob at slim.instantappointment.com (Bob Erb) writes:

> Given the following class definition,
> 
> class Test:
>     
>     def __setattr__(self, attribute, value):
>         print 'setting', attribute, 'to', value
>         self.__dict__[attribute] = value
>     
>     def __setitem__(self, index, value):
>         print 'setting index', index, 'to', value
>         ## THE_BLANK


> >>> t.list[0] = 7
> 
> That doesn't happen; when does __setitem__ get
> called?

When you write t[0]=7. 

> My second question is, to pass the assignment in
> __setitem__ through, what code goes in 'THE_BLANK'
> above?

Something like self.list[index]=value.

> (What I'm trying to do is ensure that the value of
> a variable is a certain type. For instance, I want
> to disallow assignment of a string value to an
> integer variable. Is this anti-Python?)

No, but IMHO you'd be better off by writing a method called
Test.set<List>Item(index,value), where <List> is replaced by something
descriptive, or the whole method name is replaced by something more
oriented to your problem domain. If you also write an access method
(Test.getListItem() or some such) for your list, then the
implementation is completely hidden behind a well-defined interface.

To get the type checking working automagically probably requires (or
would be clearest if written as) two classes, one for the
functionality of Test, and one for the functionality of your
type-checking list. But all this is too much coding overhead just to
make the class interface "fancy". Unless, of course, you intend to use
your new list type very very often.

You should also think of whose stupidity you're trying protect your
class from. If the answer is yourself, consider skipping the type
checks and instead documenting the class interface carefully. 

Also make sure you're aware of 'assert'.

-- 
Janne


From t.harris at robinsons.co.uk  Fri Jun 23 12:08:15 2000
From: t.harris at robinsons.co.uk (Toby)
Date: Fri, 23 Jun 2000 17:08:15 +0100
Subject: Most Recently Used file list
Message-ID: <961776536.37.0.nnrp-06.c3ad12d2@news.demon.co.uk>

I am writing a Python 1.5.2 application which requires a "Most recently
Used" file list, are there any library routines to help me, or do I have to
code from scratch using the registry ? (the app runs under Windows NT 4.0)

Thanks in Advance

Toby




From chris at rpgarchive.com  Thu Jun 15 02:53:15 2000
From: chris at rpgarchive.com (chris)
Date: Thu, 15 Jun 2000 01:53:15 -0500
Subject: TCP packet size?
Message-ID: <39487D5B.FC96BE3F@rpgarchive.com>

Is there a way to ensure or predict packet size when using TCP?  I'm
developing a Multi-user application that functions much like a
muli-player game.  The applications are sharing xml data.  The xml data
can get very big 3000-6000 bytes.  Unfortunately, when the size of the
network message gets over about 2000, its to get broken up into smaller
packets.  So, when I send a 3000 byte message it gets cut up into 2-3
packets.  Anther problem I get is when I send two small messages back to
back, they are sometimes received together.  Is there something I'm
missing here?

I tried using headers and breaking up messages into smaller chunks, but
the size of the receiving packets doesn't always match what I sent.  I'd
appreciate any help or suggestions on how I can tackle this problem.
I'd be happy to send someone my network code.

Thanks

--
chris davis
chris at rpgarchive.com

http://www.rpgarchive.com
news and adventure database!

http://www.openrpg.com
open source role playing!


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/f60e6293/attachment.html>

From donn at u.washington.edu  Thu Jun  1 12:26:26 2000
From: donn at u.washington.edu (Donn Cave)
Date: 1 Jun 2000 16:26:26 GMT
Subject: RePost: Object deletion problem
References: <3936430F.CBB53754@earth.ox.ac.uk>
Message-ID: <8h62ri$1pl0$1@nntp6.u.washington.edu>

Quoth Nick Belshaw <nickb at earth.ox.ac.uk>:
| If someone could spare a moment to explain this little observation -
| I do not know whether it is significant/expected/unexpected....
|
| I am trying to track down a memory leak and noticed the following in a
| piece of test code whilst trying to reproduce some effects.
| On a Linux RedHat6.1 Box if I run the  code and hit the offered build
| button - I simply populate the screen with a number of windows. If I
| then hit the destroy button, the windows are removed and the __del__
| method of each child is called ok, but if I look at memory usage using
| 'Top' I see that no memory is released?
| However - if I increase the dummy-data count  from  30000 to 40000 or
| above, the memory IS released.
|
| Is this ok or might it be giving me a problem?

On UNIX (and other operating systems), the memory allocation procedures
that are used by normal applications are a layer on top of the system
functions (malloc() vs. brk()).  Storage for Python objects is allocated
and released at the malloc level, while the process memory as seen by
top is allocated by brk() (or whatever your platform's C library does.)

Classically, this brk memory is never released at all, it stays at its
high water mark until process exit.  In other words, malloc may call
brk, but free doesn't.  Process memory doesn't need to shrink, really,
because if the high end memory blocks are no longer in use they'll be
paged out as required anyway.  That paging is unnecessary and consumes
resources, though, so maybe your C library is actually shrinking the
process memory when a large region at the high end is free.

My explanation may be wrong on some of the technical details, since
I don't have extensive knowledge of memory allocation internals in
current UNIX platforms, but the key point is that process memory doesn't
directly reflect each malloc & free.

	Donn Cave, donn at u.washington.edu


From roy at popmail.med.nyu.edu  Sun Jun  4 12:39:41 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Sun, 04 Jun 2000 12:39:41 -0400
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why -- or why  not? (was Re: Damnation!))
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <393326D3.400AA248@cosc.canterbury.ac.nz> <u1z2gl8zu.fsf@cs.uu.nl>
Message-ID: <roy-C89718.12394104062000@netnews.nyu.edu>

piet at cs.uu.nl wrote:
> As it is now in Python, you can have a==b and c==d both being true, but
> a/c==b/d being false. Which would be a bad surprise for most people.

What values of a, b, c, and d would cause that?  Maybe I'm just being 
dense, but i don't see it.

Isn't having two division operators kind of like having two sheds?


From hanche at math.ntnu.no  Tue Jun  6 13:48:19 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 06 Jun 2000 13:48:19 -0400
Subject: Nth digit of PI
References: <bivnjsoq0amb8qjmqtgkbfjsl2df2n2tta@4ax.com> <393C053A.58DD8220@home.com>
Message-ID: <pcon1kylmzg.fsf@math.ntnu.no>

+ Rajarshi Ray <rajarshi at home.com>:

| Besides the gee-whiz aspect of such algorithms, do they have any
| value?

They appear to have considerable entertainment value.  And, for some
reason I could never fathom, every new supercomputer ever made is
supposed to crank out ever more digits of pi to prove their worth.

| After all, what makes pi pi is ALL of its digits in a particular
| sequence.

Nah.  Digits don't have anything to do with it.  What makes pi pi is
its definition, whatever your favourite definition might be.  For
example, the smallest positive x so that sum((-1)^n x^(2n+1)/(2n+1)!)
[summed from n=0 to infinity] is zero.  Digits be damned!

How-I-want-a-drink--alcoholic-of-course--after-the-heavy-chapters-
involving-quantum-mechanics-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From andyl at azaal.dircon.co.uk  Mon Jun 19 05:18:39 2000
From: andyl at azaal.dircon.co.uk (Andy Leighton)
Date: 19 Jun 2000 09:18:39 GMT
Subject: Python vs Perl
References: <8ikai5$v3d$1@nnrp1.deja.com> <394ebe95.255648282@news.online.no>
Message-ID: <slrn8krpbf.gc.andyl@azaal.dircon.co.uk>

On Mon, 19 Jun 2000 06:41:39 GMT, Thomas Weholt <thomas at cintra.no> wrote:
>This question has been asked several times before so you might find a
>better answer than mine searchin deja.com, but :
>
>Perl vs. Python :
>
>* Perl is harder to read than Python. This makes debugging and re-use
>of other peoples code harder.

Of course if you only know Perl, I would think Perl would be the easiest
to read.  Although it would be fair to say that Python is more transparent
than Perl.

>* Python is more object-oriented than Perl

In what way?  Admittedly Python programmers tend to use OO code more than
Perl programmers.  But I believe that is mainly due to culture.


I find Python more fun to develop and it is easier to refactor,

-- 
Andy Leighton => andyl at azaal.dircon.co.uk
"... January is your third most common month for madness" - _Sarah Canary_


From dparsavand at aol.com  Wed Jun 21 19:24:18 2000
From: dparsavand at aol.com (Dparsavand)
Date: 21 Jun 2000 23:24:18 GMT
Subject: Any other good tutorials?
References: <sl2i51h0e7f149@corp.supernews.com>
Message-ID: <20000621192418.00540.00001208@ng-fl1.aol.com>

Unfortunately, if you type "Python Tutorial" into most search engines, you get
many mirrors of Guido's tutorial (which I like, but it's nice to have
alternatives).  However, if you follow http://www.pythonlabs.com/ to
http://www.vex.net/parnassus/ and then choose the  Info/Books/Tutorials link
then Tutorials, you will get quite a few alternatives. I will include two more
that are not listed there:

http://www.networkcomputing.com/unixworld/tutorial/005/005.html

and a link that definitely should be listed, but isn't yet,
http://yhslug.tux.org/obp/thinkCS/thinkCSpy

Good luck, I hope you find something you like. - Dara


From Vladimir.Marangozov at inrialpes.fr  Tue Jun 13 01:00:35 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Tue, 13 Jun 2000 07:00:35 +0200
Subject: Recruiting python programmer
References: <8hsv8f$2dj$1@wanadoo.fr>
Message-ID: <3945BFF3.D617DE5F@inrialpes.fr>

Daniel Morton wrote:
> 
> I am looking to hire on a contract basis for 6 months a python programmer in
> Paris, France. Does anyone have someone to recommend or know where I can
> post the add? thanks.
> 
> Daniel

Mentioning the offer on the french python mailing list might be a good
try as well : mailto:python at aful.org

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


From matt at virtualspectator.com  Thu Jun 15 17:11:53 2000
From: matt at virtualspectator.com (matt)
Date: Fri, 16 Jun 2000 09:11:53 +1200
Subject: Python Productivity over C++ (Encapsulation)
In-Reply-To: <20000615205922.99304.qmail@hotmail.com>
References: <20000615205922.99304.qmail@hotmail.com>
Message-ID: <00061609124801.01053@localhost.localdomain>

Just sounds like you have trouble abstracting a process in the first place.

On Fri, 16 Jun 2000, Hung Jung Lu wrote:
> >Then why does Python support encapsulation that is just as strict as C++? 
> >Guido
> >van Rossum - exposed as a traitor to the people's revolution all along!
> 
> Ha. Then why does Guido explicitly say in the manual that these variables 
> are still accessible? Just to be nice, I guess? :)
> 
> >I can easily subvert encapsulation in many C++ classes simply with the line
> >"#define private public" before I include any header files. Perhaps this 
> >means
> >that C++ will in fact lead to the overthrow of evil and oppression in the 
> >world?
> 
> Sure. What do you think I did when it was 3:00 AM I needed to ship a product 
> the next day and the engine programmers weren't available to change and 
> recompile the libraries? (Well, it was not 3:00 AM nor the next day, but you 
> get the idea.) Subversion, of course. I'll let you figure out how to do 
> things without using "#define private public" yet still be able to access 
> private data members. In my world there are no variables that can't be made 
> public, ha! The thing is, if you have to constantly recurr to subversion to 
> add functionalities, there is a problem with the language.
> 
> Yeah yeah, of course people later were shocked (and one actually screamed) 
> at how I operated. But sorry, do you guys want the product shipped, or miss 
> the season and lose millions of dollars?
> 
> There have been just way too many times when I needed extra functionality 
> and the engine team just couldn't respond in time. Why? Because some of the 
> codes were written a long time ago, and it takes them (engine programers) a 
> long time to warm up and remember all the details. If I have to wait a few 
> weeks for the changes to come, we'd've missed the deadlines. By nature you 
> don't want to release a new version of the engine per each tiny 
> change/request, right? But then, how do I add functionalities when I need 
> them urgently? This kind of situations would've NEVER happened with Python, 
> where your hands are free to add more functionalities.
> 
> Then again, I guess you've never been handcuffed. :) Have you ever worked in 
> a corporate world? (Or am I again talking to a newbie without real-world 
> experience?)
> 
> >Also, remember that Java's encapsulation isn't exactly inviolable. It can't 
> >be.
> >Think about it.
> 
> Done that thinking. :)
> 
> regards,
> 
> Hung Jung
> 
> 
> ________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From hungjunglu at hotmail.com  Thu Jun 15 14:35:21 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Thu, 15 Jun 2000 11:35:21 PDT
Subject: Python Productivity over C++ (Encapsulation)
Message-ID: <20000615183521.59984.qmail@hotmail.com>

--- In python-list at egroups.com, Ken Seehof <kens at s...> wrote:
>"Frank V. Castellucci" wrote:
> > Courageous wrote:
> > While the last line seems rather toungue in cheek, there are other
> > things that C++ brings to the table such as encapsulation, access
> > control and type strictness.
>Encapsulation, access control and type strictness are often erroneously
>referred to as if they were features.  They aren't.  They are aspects of a
>design philosophy.  To make a system less error prone, constrain the 
>programmer
>in order to prevent the programmer from writing bugs.  ...
>You have to ask, "what is the actual value of these features?"  The answer 
>is
>that they prevent certain kinds of bugs from appearing.

Type strictness is fine. It has all to do with performance: small, fast, 
efficient machine code. Encapsulation is a totally different story.

One of the biggest lies in modern computer language theory is that 
encapsulation is there for a good technical reason.

And generations of students were lied to, and sadly, 99% believed in what 
the textbooks said. Some of them became programmers, and only after 15 or 20 
years in the field would they wake up one day to realize that, heck, they've 
been lied to all along. :) (Doesn't all this remind you of Nazis and 
Communism? Collective blindness is far more common that you think, ha!)

Encapsulation is not there for any technical reason. All the technical 
reasons are simply little white lies. Encapsulation has one and only one 
purpose: CORPORATE POWER CONTROL. Period.

Encapsulation is not a feature. It's an enslaving tool. Corporates like it, 
a lot lot, and demand for it to perpetuate (through Java.)

If one day Java becomes a legacy language and loses the battle against 
open-source languages like Python, it's exactly because people are waking up 
and liberating themselves from the corporate handcuffs.

I just don't see how those little white lies can perpetuate after Visual 
Studio 7.0 incorporates Visual Python. Sleepyheads will wake up, and the 
corporate tower will start to collapse. Not a bad thing. :)

Hung Jung

(Python desginers have left out the "private" keyword because... because 
they've got some brain.) :)

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From thomas at cintra.no  Wed Jun 14 08:39:55 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 14 Jun 2000 12:39:55 GMT
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net> <394b3e7d.284293552@news.online.no> <zGI15.8$Gp6.2456@dfiatx1-snr1.gtei.net>
Message-ID: <394c7c53.300124315@news.online.no>

On Wed, 14 Jun 2000 09:57:51 GMT, "jin choung" <jin.choung at gte.net>
wrote:

>howdy,
>
>ummm, is there any book that you reccommend for a rank beginner?  that
>learning python book is REALLY not for beginners.
>
>thanks.
>
>jin
>

Well, I disagree. Learning Python was fine. I don`t know what you
found so difficult. But since more people are agreeing with you I
might be the minority here. It`s at least the most basic book of the
books I got, which is "Learning Python", "Programming with Python" and
"Internet programming using Python".

Thomas


From akuchlin at mems-exchange.org  Thu Jun 15 15:22:30 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 15 Jun 2000 15:22:30 -0400
Subject: Regular Expressions and Threads?
References: <3947D2F7.264BDC6E@uswest.net> <8i8q6s$72t$1@nntp9.atl.mindspring.net> <39481F1C.419364F5@uswest.net> <8i979n$4ss$1@slb6.atl.mindspring.net>
Message-ID: <3dd7li696x.fsf@amarok.cnri.reston.va.us>

aahz at netcom.com (Aahz Maruch) writes:
> checking for the existence of a match, that could cause problems.  (As
> with anything else, the re module should be thread-safe, but individual
> re objects probably are not.)

No, individual re objects should be.  The MatchObject you get back
from a matching operation contains references to the string, groups,
and original regex object, but a RegexObject shouldn't ever change
after it's been created.  Tim Peters put some effort into making the
cache thread-safe.  Brian, can you post the relevant bits of code so
we can see what's going on?

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Surprisingly enough, Python has taught me more about Lisp than Lisp ever did.
  -- Glyph Lefkowitz, 03 May 2000


From intmktg at Gloria.CAM.ORG  Mon Jun  5 18:33:31 2000
From: intmktg at Gloria.CAM.ORG (Marc Tardif)
Date: Mon, 5 Jun 2000 18:33:31 -0400
Subject: threading with time limit
In-Reply-To: <8F4AB9E3Bgmcmhypernetcom@199.171.54.155>
References: <Pine.LNX.4.10.10006051515580.30679-100000@Gloria.CAM.ORG> <8F4AB9E3Bgmcmhypernetcom@199.171.54.155>
Message-ID: <Pine.LNX.4.10.10006051825270.30773-100000@Gloria.CAM.ORG>

> >For example, consider a situation where all the links from multiple URL's
> >must be retrieved at the same time. By setting a time limit, slow sites
> >don't slow down the whole process and appropriate error messages are
> >returned stating which sites wheren't completed in y seconds.
> 
> Then you need to use a method of retrieving that recognizes time limits. In 
> your case, you have 3 choices:
> 
> - a version of urllib where urlopen uses select with a timeout and aborts 
> if the timeout expires (Aahz may have done this?).
> - a version of urllib that uses non-blocking sockets (then no threads 
> needed).

What do you mean by "no threads needed"? If I use non-blocking sockets, I
can kill whichever connection takes too long to respond, but there's still
the matter of retrieving each url simultaneously. Considering I'm quite
new to network programming, please explain how else can I reach my
objective other than using threads or forking processing as you suggest
below.

> - use separate processes instead of separate threads (because you *can* 
> kill a process and properly release resources).
> 
> If none of those options are available to you, then you're doing roughly 
> the right thing - just ignore the thread if it's too slow.

In the code I provided, how can I ignore a thread if it's too slow? In
other words, how can I determine if a thread hasn't finished executing?
Would this be appropriate:

for thread in threadlist:
    if not thread.isAlive():
        # continue processing here



From sebv at netscape.net  Thu Jun 29 15:04:16 2000
From: sebv at netscape.net (sebv at netscape.net)
Date: Thu, 29 Jun 2000 19:04:16 GMT
Subject: Problem with execv
Message-ID: <8jg6j3$o84$1@nnrp1.deja.com>


Hi,

I'm trying to execute a simple 'ls' command from python, using
execv but I have a TypeError. I couldn't find any similar
problem reported somewhere else. Please give me a advice !
thanks,

Sebastian.

Python 1.5.2 (#2, Jun 28 2000, 16:52:22)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.execv('ls','')
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: illegal argument type for built-in operation




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


From tismer at tismer.com  Sat Jun  3 09:37:07 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sat, 03 Jun 2000 15:37:07 +0200
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use>
Message-ID: <39390A03.560F4225@tismer.com>


Russell Turpin wrote:
...
> I wish the default implementation for lists were deques, so
> that I didn't have to worry about which end of a list I add
> to or remove from.

Fine, but for that you would throw the O(1) behavior of
lists away?

> Arithmetically-and-all-for-doing-it-from-both-ends'ly yrs,
> Russell

Please note that there is an alternative to using doubly linked
lists which takes much less memory and gives you fast modification
of both ends:

Take a simple list, together with a start and an end index.
Give this list an interface that re-implements the list behavior
with this slightly rearranged management.
Your list now becomes like a cyclic structure with a moving
gap in between.
Ok, there are some copy operations when the list has to grow,
but with a liberal allocation strategy this should be no
big problem.

Such an extension could be implemented on top of existing
lists quite easily. You'd have random access, and cheap
modification of both ends. Well, an insert into the middle
is still expensive.

but-one-can't-have-all-at-once-ly y'rs - chris

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



From jcm at bigskytel.com  Sat Jun  3 23:03:21 2000
From: jcm at bigskytel.com (David Porter)
Date: Sat, 3 Jun 2000 21:03:21 -0600
Subject: User Input?
In-Reply-To: <cnj_4.3811$r7.919327@news1.rdc2.tx.home.com>; from borlak@home.com on Sun, Jun 04, 2000 at 02:43:20AM +0000
References: <cnj_4.3811$r7.919327@news1.rdc2.tx.home.com>
Message-ID: <20000603210320.A8586@bigskytel.com>

* Michael Morrison <borlak at home.com>:

> Is there an easy way to do user input in python?  I found a non-standard
> module, but it looked too complicated (not something I'd expect from the
> ease-of-use I've had so far).

> Example: python blah.py
> Name? (type in name)
> Address? (type in address)
> etc....

Looks like you want raw_input:

  name = raw_input("Name? ")
  address = raw_input("Address? ")

  print name, address

see: http://python.org/doc/current/lib/built-in-funcs.html


  David

-- 
Stop! -- e.e.cummings in a blender



From digitig at cix.co.uk  Tue Jun  6 18:20:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 6 Jun 2000 23:20 +0100 (BST)
Subject: What's in a name?
References: <393d24ee.0@news.proweb.co.uk>
Message-ID: <memo.20000606232051.11341C@digitig.compulink.co.uk>

In article <393d24ee.0 at news.proweb.co.uk>, ng at hardlight.couk.com (.:|:.) 
wrote:

> 
> "Courageous" <jkraska1 at san.rr.com> wrote in message
> news:392CBAB5.37BFCE21 at san.rr.com...
> > Can you name some situations
> > in reading and writing english (heh) where case is essential to being
> > understood?
> 
> river riding
> 
> River Riding

Years ago the New Musical Express had a front page banner that read 
"READING CAN DAMAGE YOUR HEALTH". For non-UK readers I may have to explain 
that Reading is an English town that hosts an annual rock festival. I 
never did find out if I was being warned against attending the 
festival or perusing the journal.


From matt at null.net  Thu Jun 15 19:08:33 2000
From: matt at null.net (Matthew Schinckel)
Date: Thu, 15 Jun 2000 17:08:33 CST
Subject: Parsing dates
References: <Pine.LNX.4.21.0006141949420.22989-100000@lade.src.no>
Message-ID: <20000615170833.560377@210.9.21.14>

"Morten W. Petersen" <morten at src.no> wrote:
> Where is the package that'll accept the most different "DateTime"
> formats? I'm thinking along the lines of all the different ways
> mail-clients format their "Date:" header. --
> 
> -Morten
> 
> ------------------------------------------------------------- How to
> reply to email: http://home.sol.no/~vidaandr/news/OBSquoting.html
> (norsk) http://home.sol.no/~vidaandr/news/FAQquoting.html (english)
> 
> Who is the bigger fool? The fool or the fool who follows him?

The fool who cares :-)

Best method for handling dates in Date: headers is to use the rfc822
module:



(Sorry for the screwed up formatting - my newsreader/editor thinks I
want to work in XML!)



import rfc822, time

mail_file = open(filename,'r')

message = rfc822.Message(mail_file)

if message.has_key('date'):

date = message.get_date('date')

print time.mktime(date)



If it didn't work, the last two lines are indented one level.

Matt.


From yawyi at gmx.de  Sat Jun  3 07:40:53 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Sat, 3 Jun 2000 13:40:53 +0200
Subject: How to free a port that has been used by a python server?
References: <etdg0qvtkel.fsf@w20-575-109.mit.edu>
Message-ID: <3938f8bd$0$25935@personalnews.de.uu.net>

Alex <cut_me_out at hotmail.com> wrote:
>
>######################################################################
>class Server (SocketServer.ThreadingMixIn,
>	      BaseHTTPServer.HTTPServer):
>    pass
>    
>if __name__ == '__main__':
>
>    port = 1729 # Called from interactive session, debug mode.
>
>    proxy =  Server (('127.0.0.1', port), Handler)
>    proxy.serve_forever ()
>######################################################################
>    
>The problem is, when I run this more than once in interactive mode, I
>get the error:
>
>socket.error: (98, 'Address already in use')
>
>Is there some way to free up port 1729 when proxy gets destroyed?  

Normally you should close the socket properly before
exiting. You could try to catch KeyboardInterrupt or whatever
signal you use to break out of serve_forever(), but it's
problematic with multithreaded servers.

Workaround:
-----
class Server (SocketServer.ThreadingMixIn,
	      BaseHTTPServer.HTTPServer):
    def server_bind(self):
        import socket
        self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
        BaseHTTPServer.HTTPServer.server_bind(self)
-----

Note that this doesn't free up the port, instead it allows more than
one process to bind to it, under certain circumstances (bind() still
failes if there's already a socket actively listening on that port).
See socket(7) (Linux) for details.

Johannes



From tanzer at swing.co.at  Wed Jun 21 11:55:42 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Wed, 21 Jun 2000 17:55:42 +0200
Subject: Removing items from a list in a loop 
In-Reply-To: Your message of "Tue, 20 Jun 2000 23:32:34 GMT."
             <3952fdf6.950540640@nntp.interaccess.com> 
Message-ID: <m134mqs-000wcEC@swing.co.at>

olczyk at interaccess.com (Thaddeus L. Olczyk) wrote:

> I would like to remove a bunch of items from a list *efficiently*.
> Essentially  I would like to do this:
> 
>    for x in list:
>         if Function(x):
> 	list.remove(x)
> 
> This does not work.

list = filter (lambda x : not Function (x), list)

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From loewis at informatik.hu-berlin.de  Tue Jun 27 04:31:35 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 27 Jun 2000 10:31:35 +0200
Subject: Coercing classes
References: <3953F500.ED22500D@pfluegel.fsnet.co.uk>
Message-ID: <p6qg0pzzfrs.fsf@informatik.hu-berlin.de>

Eckhard Pfluegel <Eckhard at pfluegel.fsnet.co.uk> writes:

> My question is the following: imagine you have two classes A and B where
> B is derived from A (B is a subclass of A). The object a is assigned to
> an instance of class A:
> 
> a = A()
> 
> Under which conditions on the structure of class B is it possible in
> Python to convert 'a' to an instance of B? In my application, B would
> only have one additional method defined and no additional attributes
> (than defined in A), so that a conversion would make sense.
> Are there any built-in functions for that purpose??

Variables in Python are not types. So if you want a to 'point to' an
object of B, just write

  a = B()

This throws away the A object, and a now is a B object. Perhaps you
also want to carry some state from the A object to the B object, so
you write

  a = B(a.name, a.age, a.citizenship)

(provided an A instance has all these attributes, and that B's
__init__ expects them as arguments). You can write your own conversion function:

def AtoB(a):
  b = B()
  b.name = a.name
  b.age = a.age
  b.citizenship = a.citizenship
  return b

which would allow you to write

  a = AtoB(a)

Or you could make that a method of A, writing

  a = a.toB()

It is also possible to change the class of the A object to B, by
assigning to a.__class__. However, this is rarely what you want.

Regards,
Martin



From thomas at xs4all.nl  Fri Jun 30 21:32:42 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 1 Jul 2000 01:32:42 GMT
Subject: Controlling an external program with Python
References: <slrn8lj6pa.l52.thomas@centurion.xs4all.nl> <3958C648.1D76@rd.francetelecom.fr> <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <8jc28s$8eb$0@216.39.151.169> <8jdf8j$8tm4$1@nntp6.u.washington.edu>
Message-ID: <slrn8lqih7.gr6.thomas@centurion.xs4all.nl>

On 28 Jun 2000 18:13:39 GMT, Donn Cave <donn at u.washington.edu> wrote:
>The MIT Kerberos 5 pty library I mentioned uses a getpty function that
>opens the master pseudoty and returns its file descriptor and the name
>of its slave terminal device.  Where it uses openpty() underneath, it
>doesn't supply any termios settings and closes the slave file descriptor
>immediately.  It doesn't use forkpty() for anything.  My reflex reaction
>is that's the right level of support for a C module:  export all possible
>semi-standard low level interfaces, and put convenience routines in a
>Python module.

I disagree, and I think some of the Python developers do, too, looking at
the way posixmodule is laid out (and the fact that the SGI module is
candidate for removal in the next Python release.)

Exporting all lowlevel calls to Python and letting the pty lib take care of
determining which are appropriate creates a horrible mess. I did it with the
new pty lib, that switches between os.openpty, sgi._getpty and the original
BSD 'try and get a tty' methods, and it blows chunks, especially if you take
into account that one method might be available but fail, but another might
succeed.

If there was one 'openpty' call in the os module, which might be emulated
from other kinds of calls in C, for whatever platform it is convenient, most
of what remains are #ifdefs (of which the posixmodule, for one, already has
quite a bunch) and the choice between os.openpty() and opening /dev/pty*
manually.

My reason for taking openpty, which returns an opened master/slave fd pair,
over something a little more sgi-like (open master, return master_fd + name
of slave tty) is very simple, and very selfish: I work mostly with BSDI
servers, FreeBSD and Linux workstations. All support openpty. And the fact
that Linux has it makes me believe it's the most used method of getting a
pty.

Both methods of returning a pty can emulate the other just as well; in fact,
the new pty lib does just that ;) it has the old master_open() and
slave_open(), and it has openpty().

I agree, though, that keeping forkpty() in the posixmodule might not be
worth it. It's badly documented, for instance on wether or not the new child
is a session leader or not, so it might end up much easier just using a
python version that utilizes pty.openpty(), and *is* properly documented.

>Some of these interfaces may be appropriate for the posix module, but
>I doubt it!  Aside from the narrow-minded question of whether any of it
>is covered by POSIX 1003.1, from the user side it's inevitable that on
>some platform that does build the posix module, the pty functions won't
>be supported.  They'll need another module that can be optionally
>supported, like termios, fcntl et al.

I'm not sure if 'POSIX defines it' is a criterium for adding a function to
the posix module, though it might be. In that case, it should be factored
out to a _pty module, together with things such as ttyname() possibly. But
I'm not sure if it's out of place in the posix module: Adding it to
posixmodule adds it to the os module, which means that other platforms can
write their own, entirely seperate implementation, without the need to add
more #ifdefs.

>The more we can steal, the better.  There's too much platform specific
>folklore in here to write a good C implementation from scratch with
>man pages.

Agreed. I do like the idea of an external, extensive pty-emulation library
or such, but I have two very simple problems with that: I do not have the
time to write it, and I do not have a need for it ;) openpty() is exactly
what I need. I'd be happy to move sgi._getpty into openpty() (if the sgi
module is indeed going to be deprecated/removed) and add other variants, as
a bunch of #ifdefs, but I don't feel like supporting non-UNIX operating
systems, and I don't feel like building an external project for this ;)

Besides, it's late, I shouldn't be posting at this hour :-P

Note-my-crafty-and-elegant-way-of-avoiding-Pythons-version-number<wink>-ly
	y'rs, Thomas.



From quinn at riyal.ugcs.caltech.edu  Sun Jun 25 15:23:36 2000
From: quinn at riyal.ugcs.caltech.edu (Quinn Dunkan)
Date: 25 Jun 2000 19:23:36 GMT
Subject: Reloading modules
References: <8j5atk$ijs$1@newsg4.svr.pol.co.uk>
Message-ID: <slrn8lcn7a.drb.quinn@riyal.ugcs.caltech.edu>

On Sun, 25 Jun 2000 17:09:36 +0100, Duncan Smith
<buzzard at urubu.freeserve.co.uk> wrote:
>>>> from Graph import *
>.
>.
>.
>>>> reload(Graph)
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>TypeError: reload() argument must be module
>
>What do I need to do to successfully reload a module?  I'm currently wasting
>a lot of time re-starting PythonWin each time I want to test a changed
>module and I can't find the answer.  Thanks

'from x import ...' doesn't actually bring the module into the
namespace, just its contents.  You can get mostly what you want
with 

del sys.modules['Graph']
from Graph import *'

If it's just interactive testing, then feel free to use nasty tricks like
that.  But for real code, you should generally stay away from them.


From yannig.gourmelon at wanadoo.fr  Sat Jun 10 04:43:09 2000
From: yannig.gourmelon at wanadoo.fr (Daniel Morton)
Date: Sat, 10 Jun 2000 10:43:09 +0200
Subject: Recruiting python programmer
Message-ID: <8hsv8f$2dj$1@wanadoo.fr>

I am looking to hire on a contract basis for 6 months a python programmer in
Paris, France. Does anyone have someone to recommend or know where I can
post the add? thanks.

Daniel




From frankc at colconsulting.com  Fri Jun 16 00:02:06 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 16 Jun 2000 00:02:06 -0400
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com>
Message-ID: <3949A6BE.81B500E1@colconsulting.com>

Ken Seehof wrote:
> 
> "Frank V. Castellucci" wrote:
> 
> > Courageous wrote:
> > >
> > >
> > > > As someone who uses and likes both C++ and Python, I think estimates
> > > > like that are excessively simplistic.
> > >
> > > Me, too. And yet I will claim that I've become *more* productive
> > > in Python after a brief 4-6 weeks of use than I am in C++ after
> > > 15 years of use, and that's no exaggeration. If Python executed
> > > as fast as other languages, I'd be tempted to *require* that our
> > > staff use it. And please don't take this wrong; I'm no language
> > > groupie -- I use C, C++, Java, Python, and Lisp -- all of them both
> > > lately and frequently.
> > >
> > > > I use C++ for commercial programming, because it gives me the power and
> > > > control I need.  Productivity isn't the best when measured in time to
> > > > get the first prototype running, but I've found nothing better when
> > > > measured in time to develop, deploy, maintain and support something a
> > > > customer will pay for.
> > >
> > > Shrug. If it weren't for execution speed, your C++ environment would
> > > be toast. Admit it. :)-
> >
> > While the last line seems rather toungue in cheek, there are other
> > things that C++ brings to the table such as encapsulation, access
> > control and type strictness.
> 
> All three of these items were intentially left out of python because python
> would not benefit from them, though C++ does.
> 
> "Frank V. Castellucci" might have written (but actually didn't):
> > A fish has certain definite advantages over a duck.  For example a fish has
> gills.

I didn't so it came from elsewhere.
 
> Encapsulation, access control and type strictness are often erroneously
> referred to as if they were features.  They aren't.  They are aspects of a
> design philosophy.  To make a system less error prone, constrain the programmer
> in order to prevent the programmer from writing bugs.  Of course, that's
> impossible, since it is always possible to write bugs anyway.  Because C/C++
> are extremely error-prone languages, they need stuff like encapsulation,
> access, control and type strictness.  Because python is less error prone, it
> doesn't need these "features".  Python is based on different philosophy: To
> make a system less error prone, make it simple.
> 
> You have to ask, "what is the actual value of these features?"  The answer is
> that they prevent certain kinds of bugs from appearing.  Now ask, "what bugs
> would a C++ programmer encounter if encapsulation, access, control and type
> strictness were removed from C++."  Undoubtedly it would be a mess.  Get a
> picture in your mind of these bugs.  Now ask a python programmer how often
> these kinds of bugs come up.  Answer: "can't remember that ever happenning!"
> 
> Look at the big picture.  Do encapsulation, access control and type strictness
> really make C++ really less error prone than python?

Somewhere in your conversation you went from the aspects of a particular
language, while true they do reduce errors because the language is still
low level enought to let you get in trouble, to implying that they were
put in there to make a system less error prone. You then finish with
justifying a position based on this posit.

I disagree with the postulation.

Before these articles of disagreement, C had (and still does)
structures. Systems built in C can be a mess, or not. That is based on
the talent and experience writing the code. Python does not prevent you
from writing "messy code", it just reduces the potential devastation
caused by messy code. Python has the ability to hide data, you just use
a different notation for the variable in the class. Python has the
ability for type strictness, but you have to code it by hand as it is
not a feature of the grammer.

My original response was to the (lite hearted) statement that the only
thing C++ brings to the table is execution speed. I will posit that
languages for a solution space should be used that appropriatley address
the design requirements. It ain't all C++, and it ain't all Python.

And the most original point of all these discussions was for
qualification of productivity between the two languages. To which I
still respond that if we measure productivity by the amount of time it
takes to produce identical functionality (in something a bit more
complex than "hello world") I would say Python.

I also believe that the larger the project, the more diminished the
return on this particular measurement. Complexity takes it's toll on any
language.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From Jonathan_Epstein at nih.gov  Wed Jun 28 10:28:01 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Wed, 28 Jun 2000 10:28:01 -0400
Subject: native Python text editor with Hebrew support?
Message-ID: <395A0B71.918A6673@nih.gov>

Hi,

I realize that this is getting a bit specialized, but I am looking for a
text editor written in Python which can handle some rudimentary facets
of Hebrew; namely, writing right-to-left and the Hebrew fonts (e.g., ISO
8859).

Ah, yes, I need the source code so that I can make some changes, and it
must run on Windoze.

Is anything available?  I have looked through Parnassus, and realize
that Scintilla and PythonWin might be candidates.

TIA,

-Jonathan


From matt at nightrealms.com  Mon Jun 26 03:44:24 2000
From: matt at nightrealms.com (Matthew Cline)
Date: Sun, 25 Jun 2000 23:44:24 -0800
Subject: String wrapper?
Message-ID: <B1D55.2657$Gh6.168003@news-west.usenetserver.com>

Is there any way to make a wrapper for a string object,
like UserDict is a wrapper for the dict type?

Thanks in advance.





From mwh21 at cam.ac.uk  Tue Jun 13 16:28:45 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jun 2000 21:28:45 +0100
Subject: no SyntaxError?
References: <Pine.LNX.4.10.10006131359020.17071-100000@localhost.localdomain>
Message-ID: <m3itvd2uma.fsf@atrus.jesus.cam.ac.uk>

"Robert W. Bill" <rbill at digisprings.com> writes:

> Just curious- I didn't find any references to this in a quick search of 
> the faq and elsewhere...
> 
> This snippet generates "SyntaxError: invalid syntax"
> 
> def test():
>     else: print "whoops"
> test()
> 
> However, the following snippet generates "0 1 whoops" when it seems like
> it should also generate "SyntaxError: invalid syntax" because there is no 
> if before the else. 
>  
> def test1():
>     for i in range(2):
>         print i,
>     else:
>         print "whoops"
> 
> test1()
> 
> Why is it that I do not get a "SyntaxError" in the second example?

Because it is valid Python!  An else block after a for loop is
executed if the loop exits normally (ie. by getting to the end of the
list), but not if the loop is exited via a break statement.  Same
applies to while loops.

A little used feature (they don't leap to my mind very readily,
certainly).

Cheers,
M.

-- 
  Imagine if every Thursday your shoes exploded if you tied them 
  the usual way.  This happens to us all the time with computers, 
  and nobody thinks of complaining.                     -- Jeff Raskin


From ian_maurer at my-deja.com  Wed Jun 28 11:39:50 2000
From: ian_maurer at my-deja.com (Ian Maurer)
Date: Wed, 28 Jun 2000 15:39:50 GMT
Subject: MD5 - Passwords with python
References: <6e7dj8.bvb.ln@iiab.soluma.de>
Message-ID: <8jd682$ea5$1@nnrp1.deja.com>

MD5 is with the latest distros.

>> import md5

See the module index in the documentation with
your distro or go online to:

http://python.org/doc/current/modindex.html

to get to MD5 directly:

http://python.org/doc/current/lib/module-md5.html

Happy Pythoning,
Ian

Open-Source Python Enhanced DOS prompt:
http://hammer.prohosting.com/~pyguy


In article <6e7dj8.bvb.ln at iiab.soluma.de>,
  "Holger M. F??ler" <fuessler at solutec.de> wrote:
> Hi!
>
> Is there a library function to encode passwords crypt-style, but with
md5 as
> hashing-algorithm for python?
> Do you know where I can find documentation about the format?
>
> Thanks
> Holger
>
>

--
Ian Maurer




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


From garabik at melkor.dnp.fmph.uniba.sk.spam  Thu Jun  1 03:28:04 2000
From: garabik at melkor.dnp.fmph.uniba.sk.spam (Radovan Garabik)
Date: 1 Jun 2000 07:28:04 GMT
Subject: OT: why tar is strange
References: <000601bfbfc5$95612f20$b52d153f@tim> <ja7lctey7i.fsf@helix.twistedmatrix.com> <39246112.4F01@seebelow.org> <20000518145435.C13281@xs4all.nl> <39264186.119D@seebelow.org> <slrn8ic65n.9nc.jajs@Pascual.es> <392EEEE3.2D37@seebelow.org> <393403D7.CAA8AB7D@sage.att.com> <_t1Z4.29279$WS3.338373@typhoon.we.rr.com>
Message-ID: <959844427.636599@cdwork.cvt.stuba.sk>

Stephen Hansen <stephen at cerebralmaelstrom.com> wrote:
 :     I don't think this is really a case of which one is more productive or
 : fast, but more on where you are at the time that you're doing it. If you're
 : sitting in a shell, then using 'tar zvxf' is just as fast as any GUI version
 : would be. But if you're already in a GUI and aren't working in a shell, then
 : starting up the shell and doing your tar'n isn't going to work for you then,
 : either.
 :     My problem with tar was not remembering the 'zxvf' part, but instead
 : remembering exactly how the file was either a) spelled, b) capitolized, or
 : c) punctuated(sp?). Even with an 'untar' script to do it, I just might
 : forget to capitolize a letter or put a '-' in the wrong place.

that is not a problem, just use TAB expansion.
tar xvzf a<TAB> and shell will fill the filename for you..
or list all matching filenames, then hit TAB second time to cycle through 
them... or type another letter, another TAB...

 :     Not that you can't use wildcards to make it quicker, I know, mind you..
 : but I don't think the shell approach would ever be quicker then be using
 : WinRAR (Am I the only one who doesn't use Winzip?) to right click on the
 : compressed-file and click 'Extract to...' or 'Extract to (name of file)\' in
 : that ever so friendly context menu. Two quick clicks :)

well, hitting <ENTER> <F5> <ENTER> in midnight commander is probably 
quickier :-)


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From cg at gaia.cdg.acriter.nl  Sat Jun  3 15:58:25 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 3 Jun 2000 21:58:25 +0200
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org>
Message-ID: <8hbo11$bo2$1@gaia.cdg.acriter.nl>

Jeff Collins  <Jeffery.D.Collins at aero.org> said:
>I'm currently porting a 1.5.2+ (soon to be 1.6) to the Palm.
>Hopefully, I'll be making an announcement soon.
>
Drool...

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From hjwidmai at foxboro.com  Tue Jun 13 01:12:01 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Tue, 13 Jun 2000 07:12:01 +0200
Subject: Tkinter: Entry widget with history/completion
Message-ID: <20000613071201.A7987@foxboro.com>

I'd like to have an Entry(-like) widget that has a history
and a completion mechanism. I see no problems doing the first
one, but I see no convenient way to do the latter, as I cannot
find any widget method that gives me the current index of the
cursor (my aim is to complete the phrase from index 0 upto the
cursor position, ignoring what is right to it). Did I miss
something? I certainly don't want to write a whole new widget,
just use Entry and extend it.

How would you recommend doing it?
 - By subclassing Entry
 - By creating a new Pmw MegaWidget

Both of them look like nice implementations to me.

I have another question that bothers me since long, just out of
curiosity.

The geometry manager methods on widgets [pack(), grid(), place()]
return None. To reduce the number of lines of code (which is
considered A Good Thing), wouldn't it be nice to be able to
write
  
  e = Entry().pack()

instead of

  e = Entry()
  e.pack()

I guess there is a reason for those methods to not return 'self'.
I just can't think of any ...

Thanks for any help,
-- 
Hans-Joachim Widmaier



From newsbunny at noether.freeserve.co.uk  Wed Jun 28 12:24:25 2000
From: newsbunny at noether.freeserve.co.uk (Martin P Holland)
Date: Wed, 28 Jun 2000 17:24:25 +0100
Subject: PythonQt Binding Is is available?
References: <8jd4hk$nda$1@bw107zhb.bluewin.ch>
Message-ID: <slrn8lk9lp.431.newsbunny@noether.freeserve.co.uk>

On Wed, 28 Jun 2000 17:05:35 +0200,
Markus Meng <meng.engineering at bluewin.ch> wrote it all in the subject

http://www.thekompany.com/projects/pykde/

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


From hgg9140 at seanet.com  Tue Jun  6 17:58:00 2000
From: hgg9140 at seanet.com (Harry George)
Date: 06 Jun 2000 14:58:00 -0700
Subject: Python <--> XML data binding.
References: <8hhlr6$eoh$1@slb3.atl.mindspring.net> <393C6944.2FBCE9B1@prescod.net>
Message-ID: <m3n1kysc9j.fsf@wilma.george.org>

I've also been writing a package along these lines.  Start with a
neutral representation (phase 1 used UML, phase 2 uses a python list
format, phase 3 will code directly from python class defs).  Generate

a) SQL data def and data manipulation code
b) UML XMI xml file
c) DTD and xml reader/writer


The truth is, I haven't done much on this recently, because
hand-coding xml reader/writers is utterly trivial once you have some
design patterns.  I generally structure the job as:

1 module per xml dtd.  In some cases this has resulted in 1000 LOC
modules, but each class in the file is self contained so maintenance
is no hassle.

The root object is XYZ_File, representing the xml file itself (e.g.,
file name, version).  Also has file-level tree decorations (e.g.,
cross references).

The various domain-specific classes are written in a neutral format,
with a "from_dom" and a "to_dom" method to handle translation.  Do a
bit of processing to get things in and out of raw XML formatting, but
retain the local semantics.

A good deal of the coding after that is transforming one OO rep
to another -- i.e., the semantics of moving from one domain to
another.  The xml just shows up in the input and output stages.  E.g.:

domain1.xml-->domain1.py-->domain1_to_domain2.py-->domain2.py-->domain2.xml

I'm finding that the best approach is to build the python OO
representations first, and then figure out how to serialize them to
xml.  DTD's are useful for inter-community transfer, but hardly
appropriate for data modeling. 

Paul Prescod <paul at prescod.net> writes:

> Werner Fouche wrote:
> > 
> > Does anyone know if a tool exist that will generate Python data binding code
> > for an XML DTD similar to what Breeze ( http://www.breezefactor.com ) does
> > for Java.
> 
> I'm working on such a library. Check back in a couple of weeks.
> 
> Nevertheless, the Breeze website vastly exaggerates the difficulty of
> doing object creation "by hand" using the output of a parser. I wouldn't
> even bother with my data binding technology if the data binding did not
> come "for free" as a side effect of other features I need.
> -- 
>  Paul Prescod
> Those who profess to favor freedom and yet depreciate agitation, are men
> who want crops without plowing up the ground. They want rain without
> thunder and lightning. They want the ocean without the roar of its many
> waters. - Fredrick Douglass
>  http://www.informamerica.com/Articles/Quotes.htm
> 

-- 
Harry George
hgg9140 at seanet.com


From pecora at anvil.nrl.navy.mil  Sun Jun 11 09:25:56 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Sun, 11 Jun 2000 13:25:56 GMT
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no>
Message-ID: <110620000925578570%pecora@anvil.nrl.navy.mil>

In article <3942B267.C769395 at stud.ntnu.no>, Peter Schneider-Kamp
<petersc at stud.ntnu.no> wrote:

> Peter Schneider-Kamp wrote:
> > 
> > I have to create a tuple (of two integer values) in a
> > C extension. At the moment I use the following (rather
> > ugly in my eyes) construction:
> 
> As an afterthought. Are there any reference counting things I
> have to deal with in this example?

Yes, usually when you _create_ a Python object in C you have to handle
the references or call routines that do that for you. Trouble is, it is
something that is not always clear on how to do it.  I would say you
need to INCREF your tuple after creation, unless you know the routine
that creates it already does that.  If you return the tuple, do
nothing.  If you do not return it, you should DECREF it.  How sure am I
of all this?  Not very.  I'm just learning.  Best to check out all
advice from the experts on this helpful NG and dig up all the examples
you can, especially using the same C-Python routines you do.  Sorry I
can't help more.  Do keep digging. Ref-counting is important.


From BgPorter at NOacmSPAM.org  Wed Jun  7 20:22:48 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Thu, 08 Jun 2000 00:22:48 GMT
Subject: I need a lesson.
References: <8hmn80$4vu$1@ionews.ionet.net>
Message-ID: <sHB%4.115735$R4.705084@news1.rdc1.nj.home.com>

"F. G. Brown" <fgbrown at ionet.net> wrote in message
news:8hmn80$4vu$1 at ionews.ionet.net...
> Greetings,
> ?????????????????????????????????????????????
>
> howfast.py
while 1:    # this will make your code loop...forever (see note [1] below)
    # Do yourself (and anyone who'll ever read your code)
    # and get in the habit /now/ of using meaningful variable names.
   distance = raw_input("What is the distance you traveled in miles?  ")
   time = raw_input("How many seconds did it require?  ")
   speed = (int(distance) * 3600.00) / int(time)
   print "You were going %d mph!" % speed
   if speed > 75:
      print "Hey you better slow down!"
   loopAgain = raw_input ("Would you like to do another? (y)es or (n)o")
   if loopAgain == 'y':
      print 'Here we go!'
   else:
      print 'See ya!'
      break    ## note [1] -- this gets you out of the loop!







From cgw at fnal.gov  Fri Jun  2 12:36:55 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: Fri, 2 Jun 2000 11:36:55 -0500 (CDT)
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
In-Reply-To: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD>
References: <6B00FB949906D211A88800104B8AE5200183D5A4@WIREHEAD>
Message-ID: <14647.58023.493846.225899@buffalo.fnal.gov>

Czerminski, Ryszard writes:
 > 
 > Is the behaviour illustrated below a bug or the Python's feature ?

I don't know why you're posting this to the NumPy discussion list,
since the
"math" module is part of the standard Python distribution, not NumPy.

Anyhow, the differing behaviors people are seeing are just coming from
differences in the system math libraries:

On Linux (Intel):

Python 1.6a2 (#9, May 22 2000, 12:34:51)  [GCC 2.95.2 19991024 (release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> from math import exp
>>> x = -706.
>>> while x > -900:
...    x = x - 1
...    print 'x, exp(x) = %g %g' % (x, exp(x))
... 
x, exp(x) = -707 8.99086e-308
x, exp(x) = -708 3.30755e-308
<snip>
x, exp(x) = -745 4.94066e-324
x, exp(x) = -746 0
<snip>
x, exp(x) = -899 0
x, exp(x) = -900 0
>>> 


Wheras on Irix (MIPS):
Python 1.6a2 (#8, Jun  1 2000, 20:01:55) [C] on irix646
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> from math import exp
>>> x = -706.
>>> while x > -900:
...    x = x - 1
...    print 'x, exp(x) = %g %g' % (x, exp(x))
... 
x, exp(x) = -707 8.99086e-308
x, exp(x) = -708 3.30755e-308
x, exp(x) = -709 0
x, exp(x) = -710 0
<snip>
x, exp(x) = -745 0
Traceback (most recent call last):
  File "<stdin>", line 3, in ?
OverflowError: math range error


>From the Irix man page for exp(3M):

     The exp functions return HUGE_VAL when the correct value would overflow,
     and return zero if the correct value would underflow. The -lm and -lmx
     versions set the value of errno to ERANGE for both underflow and
     overflow.

Since the Python math module sees errno set after the call to exp, it
raises an exception.

Whereas on Linux, exp(-very big number) simply returns 0 and does not
set errno.

On the one hand, Python's behavior makes sense because it simply
reflects the behavior of the system math libraries.  On the other
hand, these kinds of differences make it hard to write portable code -
you could test on Linux and think everything is OK, then run on IRIX
and get exceptions.  Maybe that's just the way life is when you are
using floating-point math... Tim Peters may have more to say on this
topic <exp(-900)wink>











From cg at gaia.cdg.acriter.nl  Fri Jun  2 11:26:04 2000
From: cg at gaia.cdg.acriter.nl (Cees de Groot)
Date: 2 Jun 2000 17:26:04 +0200
Subject: Palm Python?
Message-ID: <8h8jmc$t9i$1@gaia.cdg.acriter.nl>

Sorry if this is a FAQ - I haven't been following this group for a while,
at least not as closely as I'd like. But rumors are that a Palm version
of Python is pending, and as I'm looking for a language to do little Palm
experiments with, I am just curious to know whether this is true.

The reason I'm asking is that I want to see how Palms can integrate with
Jini networks. The KVM is neat, but far from anything you would call a 
good environment for Palm apps (the showstopper being that you cannot switch
to and from KVM applications - they always re-initialize), and I rather 
invest my time in something that I may use for production later on. I
don't like (Pocket)C, so it's either Java or Python for me...

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Forge your CipherSaber and list it: http://www.xs4all.nl/~cg/ciphersaber/


From M.Vitecek at sh.cvut.cz  Thu Jun 15 08:55:43 2000
From: M.Vitecek at sh.cvut.cz (Michal Vitecek)
Date: Thu, 15 Jun 2000 14:55:43 +0200
Subject: SV: Python Productivity over C++
In-Reply-To: <3947C453.CF24E30F@estinc.com>; from eric@estinc.com on Wed, Jun 14, 2000 at 05:42:05PM +0000
References: <20000614085924.80042.qmail@hotmail.com> <20000614150508.A2932@fuf.sh.cvut.cz> <3947C453.CF24E30F@estinc.com>
Message-ID: <20000615145543.A10869@fuf.sh.cvut.cz>

Eric Lee Green <eric at estinc.com> wrote:
>Michal Vitecek wrote:
>>  this is highly debatable - properly written C/C++ source is pretty well
>>  readable and self-documenting. similarly, badly written python source
>>  makes it as unreadable as in other languages - there's not a point in
>>  trying to prevent people from writing (un)readable source code.
>
>It's harder to write unreadable Python source because you can't write code
>like, hmm,
>
>opendir(FOO,\"%s\");@files=readdir(FOO);foreach (@files) { if
>(/^[^\\.].*\\.txt$/) { print \"%s/$_\\n\";};};closedir(FOO);

 you could notice i'm not advocating perl which can be most of the time
 someone's nightmare. i'm advocating C++ operators, braces and assignments
 inside if statements. you could save your time writing this perl example
 because it's irrelevant.

>since Python enforces proper indentation and limits you to one statement to
>line.
>
>Yes, you can write unreadable Python code. But you have to work at it.

 well - there are some (like me) who consider many more lines of source
 less easily understandable (i do not have 23" monitor). in the same
 manner i use &&, || to have a good view of the code ('and' 'or' inside an
 if statement can be easily overlooked).
 don't get me wrong - i like python - i just hate its enforcing me to some
 really uncommon syntax.

-- 
			fuf


------------------------------ na IRC -------------------------------------
 BillGates [bgates at www.microsoft.com] has joined #LINUX
 ...
 mode/#linux [+b BillGates!*@*] by DoDad
 BillGates was kicked off #linux by DoDad (banned: We see enough of Bill
          Gates already.)
 




From jimclark22 at msn.com  Fri Jun 16 04:09:32 2000
From: jimclark22 at msn.com (ZWang Productions)
Date: Fri, 16 Jun 2000 04:09:32 -0400 (EDT)
Subject: Are You Looking for a Merchant Account?
Message-ID: <20000616080932.768931D036@dinsdale.python.org>


    
I just visited your Web Site from a search engine.

We are sending you this email to tell you about what we do,
as it may be beneficial to both of us. 

We have an interesting concept! 

My company has taken notice of your online presence and
wanted to bring a few things to your attention.

We have spoken with hundreds of online merchants like
yourself, who, unfortunately, are not happy. The reasons
for their unhappiness varies, but all seem to have common 
issues.

I invite you to visit my company at www.zwangproductions.com
and, if you have any questions, please email me at jimclark22 at msn.com.

Should you receive this message in error, please e-mail me at the above 
address and in the subject line, please indicate "remove". 

Thank you for your time and consideration, and I look forward to
being of service to you.

				Sincerely,

				Jim Clark
				ZWang

PS. Unhappy with your Web  Hosting company visit www.wtz-productions.com

From kens at sightreader.com  Wed Jun 28 10:59:43 2000
From: kens at sightreader.com (Ken Seehof)
Date: Wed, 28 Jun 2000 07:59:43 -0700
Subject: Speed with Python
References: <Pine.GSO.4.10.10006280840140.27161-100000@sundial>
Message-ID: <395A12DF.9262B588@sightreader.com>

Moshe Zadka wrote:

> On 27 Jun 2000, [ISO-8859-1] Fran?ois Pinard wrote:
>
> > [Ken Seehof]
> >
> > > I do all GUI work in python (wxPython) and I have not noticed a difference
> > > in speed as compared to C++, probably because most time is spent in
> > > library calls anyway.
> >
> > I experimented some GUI writing in Python, and initially expected that things
> > would be slow.  Not at all: I was very pleasantly surprised by the speed.
>
> Indeed: in most GUIs, most time is spent in the user's head: IOW, the
> computer is too fast, and most time it is just sitting quietly, waiting
> for the user to make up his mind. Python's idle loop is almost as good
> as C++'s <wink>

> --
> Moshe Zadka <moshez at math.huji.ac.il>
> http://www.oreilly.com/news/prescod_0300.html
> http://www.linux.org.il -- we put the penguin in .com

On the contrary!  Python's idle loop is slightly more effective than C++'s. The
purpose of an idle loop is to be idle right?   Python's idle loop is slightly
slower and therefore more idle.  :-)  Kinda like in California where stop lights
are more effective.

Actually, in my case, I'm writing a graphical editor, so dragging objects around
and drawing lines as the mouse moves must be done as smoothly as possible.  For
extra smoothness, I'm using a double backplane to eliminate flicker and minimize
the amount of drawing work.  I probably wouldn't have bothered doing that if I had
been using C++.  The final result is much smoother than most C++ programs.

Conclusion: Given the same amount of development time, a python GUI is likely to
appear faster and smoother than the equivalent C++ GUI because the python
programmer has enough time to be clever.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!





From shouldbe at message.com  Wed Jun 21 17:09:29 2000
From: shouldbe at message.com (Arinté)
Date: Wed, 21 Jun 2000 16:09:29 -0500
Subject: strings and integers?
Message-ID: <8ir75b$qfi$1@news2atm.raleigh.ibm.com>

Is
x = "txt %(somevar)"%vars()
the only way to append text to a string?

Also, why do I get a len() of unsized object errors?  Do I need to rebuild
the python dlls? Download 1.6?

Thanx





















From jeremy at beopen.com  Fri Jun  9 11:43:54 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 09 Jun 2000 11:43:54 -0400
Subject: getting a URL
References: <8hop1b$c1b$1@nnrp1.deja.com>
Message-ID: <BBF4FD9CA76E9DF9.BC74420340D17532.6B7363F9B8212DF1@lp.airnews.net>

johnvert at my-deja.com writes:

> But nothing is printed.  The problematic line seems to be:
> 
> url = urllib.urlopen("http://www.server.org/foo.html")
> 
> But the URL I gave it in the actual program is correct.  I tried a few.
> If I run this on the shell (it's meant to be a cgi) it raises IOError
> sayin 'network unreachable'.  What could be wrong? 

The network is unreachable <wink>.  The error message means that the
operating system could not figure out how to deliver packets to the
address you used.  Where you able to load the URL using your Web
browser?  I would guess that your network connection had some problem,
or that the end site was unreachable.

Jeremy


From darrell at dorb.com  Sat Jun 17 02:39:34 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 16 Jun 2000 23:39:34 -0700
Subject: backslash in regular expression problem
References: <394A55BA.69269793@wolfson.co.uk>
Message-ID: <02de01bfd826$cf05bd80$6401a8c0@home>

Try to avoid assigning to "string" that will cause problems when you import
string
Especially when your using "\" try to use raw strings. Note the r"xxxx"

>>> s=r"\END\\"
>>> print re.sub('[\\\]','',s)
END
>>> print re.sub(r'[\\]','',s)
END

--Darrell Gallion


----- Original Message -----
From: "Gordon McLeod" <Gordon.McLeod at wolfson.co.uk>
> import re
> string="\END\\"
> x=re.sub('[\]','',string)
> print x
>
> and expect x to be 'END' but I get:
>
> Traceback (innermost last):





From MarkH at ActiveState.com  Tue Jun 20 07:52:25 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 20 Jun 2000 11:52:25 GMT
Subject: win32all-132 / Word Automation
References: <394e2b8d.515788975@news.mch.sni.de> <lAy35.3131$qz4.24329@news-server.bigpond.net.au> <394f269c.580059221@news.mch.sni.de>
Message-ID: <ZVI35.3597$qz4.27900@news-server.bigpond.net.au>

"Nikolai Kirsebom" <nikolai.kirsebom at siemens.no> wrote in message
news:394f269c.580059221 at news.mch.sni.de...
> Mark,
> Thanks for your quick response.  It might be obvious ... but not
> really to me.

Hrm - not to me either!


> >>> print doc.Variables.Item(1)
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File
> "win32com\gen_py\00020905-0000-0000-C000-000000000046x1033x8x0.py",
> line 11402, in __str__
>     return str(apply( self.__call__, args))

Ahhh!  for some bizarre reason, it appears the print itself triggers this.
(Note that the exception is in the __str__ method, which is being used to
print it.

What does:
print repr( doc.Variables.Item(1) )

yield?

Mark.





From dawks at tesco.net  Fri Jun  2 18:53:03 2000
From: dawks at tesco.net (phil)
Date: Fri, 02 Jun 2000 22:53:03 GMT
Subject: Extending 1.6 vs 1.52 - how runny is it?
Message-ID: <3938366a.8739787@news.tesco.net>

Hola,

i wrote a python extension .dll which is proving to be rather popular
(understatement really). I hear that 1.6 changes the extension rules,
but i can't find the documentation about this on the web site. Anyone
know where i can get the details? Can the two interfaces co-exist in
my .dll (i hope so) and if so, do i need to get the currently active
Python version to resolve the interface spec and if so, how?

Thanks in advance for any help.

phil.

 


From dimus63 at my-deja.com  Thu Jun 22 19:46:10 2000
From: dimus63 at my-deja.com (dimus63 at my-deja.com)
Date: Thu, 22 Jun 2000 23:46:10 GMT
Subject: Using anydbm.
References: <m3n1l980k3.fsf@underlevel.underlevel.net> <63c9jscp2qeratnaa1r3rgivh36l92g186@4ax.com>
Message-ID: <8iu8fv$ue9$1@nnrp1.deja.com>

I have the same thing on my computer (Mandrake6.1-7.1 mix):
Anydbm doesn't recognize Berkley dbm by some reason,
but anydbm on this computer works fine with gdbm files.

On other computer (which is sourceforge orbital)
anydbm works correctly with everything...

In article <63c9jscp2qeratnaa1r3rgivh36l92g186 at 4ax.com>,
  Tim Roberts <timr at probo.com> wrote:
> Jordan Katz <webmaster at underlevel.net> wrote:
> >
> >I'm having trouble getting anydbm to work correctly.  For example,
> >
> >  >>> import anydbm
> >  >>> db = anydbm.open('foo', 'c')
> >  >>> db['bar'] = 'baz'
> >  >>> db['bar']
> >  'baz'
> >
> >Works fine, but if I try to execute it again after restarting python,
> >I get the error:
> >
> >  Traceback (innermost last):
> >    File "<stdin>", line 1, in ?
> >    File "/usr/lib/python1.5/anydbm.py", line 83, in open
> >      raise error, "db type could not be determined"
> >  anydbm.error: db type could not be determined
>
> Good question; I have exactly the same problem.  I replaced "anydbm"
with
> "dumbdbm" and it worked fine, but I expected "anydbm" to work.  If you
get
> an answer privately, please post it for me.
> --
> - Tim Roberts, timr at probo.com
>   Providenza & Boekelheide, Inc.
>


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


From alex at magenta.com  Mon Jun 26 10:07:36 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 26 Jun 2000 16:07:36 +0200
Subject: Differentiating between WinNT and Win32
References: <eviltofu-F9E289.07322926062000@news.earthlink.net>
Message-ID: <8j7odv01uns@news2.newsguy.com>

Jerome Chan <eviltofu at rocketmail.com> wrote in message
news:eviltofu-F9E289.07322926062000 at news.earthlink.net...
> Is there a way to do this in Python? I went DejaDiving and found
> something called platform.py but the link to it on starship is dead.
> Anyone has any clues?

>>> from win32api import GetVersionEx
>>> GetVersionEx()
(4, 0, 1381, 2, 'Service Pack 5')

The 4th element in the tuple ("2" here) has the specific info
you require; it's 1 for Win/95 and Win/98, 2 for Win/NT (note
that all of these are "Win32"...).


Alex





From hei at adtranzsig.de  Thu Jun 22 05:00:05 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 22 Jun 2000 11:00:05 +0200
Subject: Conditional Compilation for bytecode?
References: <8iquvb$gtd$1@nnrp1.deja.com>
Message-ID: <8iskgf$i74$1@desig-bs01-s04.adtranzsig.de>

> I have some code that needs to run on Win32 and IRIX.
>
> I thought I read somewhere that python had a sort of conditional
> compilation that would generate different .pyc files depending on the
> directives in the .py file.
>
> i.e.
> #ifdef WIN32
> CreateProcess(...)
> #else UNIX
> fork()
> #endif
>

How about

if sys.platform = 'Win32':
  def fun():
      CreateProcess(...)
else:
  def fun():
      fork()

fun()

? When importing this, Python will create a function that doesn't check at
runtime.

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





From kupisch at hotmail.com  Sun Jun 18 16:25:09 2000
From: kupisch at hotmail.com (ratze)
Date: Sun, 18 Jun 2000 22:25:09 +0200
Subject: Pythonwin-wxPython
Message-ID: <394D3024.7E2B1A11@hotmail.com>

Hi Everyone,

I am relatively new to using Python on Windows NT and hence using
Pythonwin. I have a question that  I can't figure out myself.

My  problem is that when I am developing a wxPython application
everytime the the start of my app fails due to an exception I cannot
start it again and get 'Only 1 wxApp per process!'. How do I get around
this??
And by the way, is there any possibility to set the default path in the
open dialog?

Thanks in advance!



From db3l at fitlinxx.com  Thu Jun  8 22:36:21 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 08 Jun 2000 22:36:21 -0400
Subject: Pythonwin is terrible!
References: <8hp2ku$a25@news.or.intel.com> <vxW%4.1319$iJ6.12966@news-server.bigpond.net.au>
Message-ID: <uog5b4m3e.fsf@ctwd0143.fitlinxx.com>

"Neil Hodgson" <neilh at scintilla.org> writes:

> > 1. How do you kill an infinite loop program without killing the pythonwin
> > environment?
> 
>    You can try right-clicking on the little green python down in the lower
> right hand side of the screen and then choosing "Break into running code".
> This normally works, but sometimes it doesn't.

In particular, if the code uses broad exception clauses (for example,
plain "except") in order to attempt recoveries and continue executing,
then it's going to catch the KeyboardInterrupt exception this method
tries to generate just as any other exception, and continue execution.
If that execution is resulting in an infinite loop, well...

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


From folker at t-online.de  Thu Jun 15 05:34:34 2000
From: folker at t-online.de (Heinz Folker)
Date: Thu, 15 Jun 2000 11:34:34 +0200
Subject: Threads and GUI
Message-ID: <3948A32A.25025A54@t-online.de>

I want to create window like this:

tk=Tkinter.Tk()
widget=Test(tk)
thread.start_new_thread(widget.mainloop(), ())

I need a thread because I want to use the interpreter while the mainloop
is running. I want to interact with this window.
But this code doesn't work.

Thanks for help




From leonhardt at informatik.tu-darmstadt.de  Mon Jun 26 04:04:01 2000
From: leonhardt at informatik.tu-darmstadt.de (Thomas Leonhardt)
Date: Mon, 26 Jun 2000 10:04:01 +0200
Subject: SuSE Linux, python and setgid
Message-ID: <8j72mh$7q0$1@sun27.hrz.tu-darmstadt.de>

Hi,
I have got the problem that I should install a script on our server
which was written in python. The problem is that the script checks
wether the group id was set as specified by the file permissions.

But somehow the gid is not set when I execute the script. Is there
something like suidperl for python or smoethingelse special with
SuSE linux which causes the gid not beeing set??

Thanks for help
Thomas




From ge at nowhere.none  Thu Jun  8 17:09:32 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 08 Jun 2000 21:09:32 GMT
Subject: physical hardware accesss
References: <393f4dd2$0$2877@wodc7nh6.news.uu.net>
Message-ID: <gYT%4.1084$zm3.245488@ptah.visi.com>

In article <393f4dd2$0$2877 at wodc7nh6.news.uu.net>, Blu Dragoon wrote:
>If I wanted to would be able to probe a serial, parallel, or USB port
>with python.

>Open(/dev/cua1)
>Write(/dev/cua1)

How about:

 fd = os.open("/dev/cua1")
 os.write(fd,buffer)

-- 
Grant Edwards                   grante             Yow!  ... Um...Um...
                                  at               
                               visi.com            


From albert.brandl at lindeverlag.at  Wed Jun 14 09:07:19 2000
From: albert.brandl at lindeverlag.at (Albert Brandl)
Date: Wed, 14 Jun 2000 13:07:19 GMT
Subject: Where should I put my HTMLgen directory?
References: <7BD10B680501D411B9DF009027E06F32011ACA@EXCHANGE>
Message-ID: <39478128.22072047@news.inode.at>

>I am trying to install HTMLGen on my machine.
>[...]
>    import HTMLgen
>ImportError: No module named HTMLgen
>
>So obviously I am doing something wrong. I thought that Python traversed the
>Lib diretory recursively to find any modules. But no?
>
>Should I put all the *.py* files in the root of Lib? That seems a bit messy
>to me.

You could either add the HTML directory to your python path:

import sys
sys.path.append("c:\\Program files\\Python\\lib\\HTMLgen")

(make sure to use "\\" instead of "\"), or add empty files
called __init__.py to both the Lib and HTMLgen directory.
If these files are present, it's possible to import HTMLgen
with e.g.

from Lib.HTMLgen import HTMLgen

http://www.python.org/doc/essays/packages.html

explains the package mechanism in more details.

	Have fun!

		Albert


From piet at cs.uu.nl  Thu Jun 22 04:04:36 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 22 Jun 2000 10:04:36 +0200
Subject: Entering the python mode in emacs.
References: <3951a829.46476453@nntp.interaccess.com>
Message-ID: <uitv2m99n.fsf@cs.uu.nl>

>>>>> olczyk at interaccess.com (Thaddeus L. Olczyk) (TLO) writes:

TLO> Is it possible to configure emacs so that it automatically starts in
TLO> python mode when opening a file *.py?

You do this in the standard emacs way:

(setq auto-mode-alist
      (cons '("\\.pyw?$" . python-mode) auto-mode-alist))

or

(add-to-list 'auto-mode-alist '("\\.pyw?$" . python-mode))

It is also useful (at least on Unix) to have

(add-to-list 'interpreter-mode-alist '("python" . python-mode))
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From rhicks at rma.edu  Thu Jun 15 10:53:36 2000
From: rhicks at rma.edu (Robert)
Date: Thu, 15 Jun 2000 14:53:36 GMT
Subject: PythonWin Debugging tools???
References: <Zc425.166$M85.9384@news.uswest.net> <dw425.14$lY4.1297@news.uswest.net>
Message-ID: <Q5625.25$RB1.7081@news>

For PythonWin...not really...it is Mark's baby and as yet is not part of
"core" Python.

"Eric Lengvenis" <elengvenis at uswest.net> wrote in message
news:dw425.14$lY4.1297 at news.uswest.net...
> Never mind. I saw a different post that directed me to
>  http://starship.python.net/crew/mhammond/win32/Downloads.html
>
> There I downloaded the build 129 and patch. Now, the date is still April
> 13th, but all the functions are there.
>
> Does anybody find it strange that the Python.org site is not the
definitive
> site for current builds?
>
>
>
> Eric Lengvenis wrote in message ...
> >I am very new to Python.
> >I just picked up the book Python Programming on Win32.
> >Early in the book, the authors describe the debugging features available.
> >None of them are available in my version. I noticed that the date on my
> >interactive window is April 13th 1999 and  the date on the figures in the
> >book is June 14th 1999.
> >
> >Were the debugging features not included until this version? If so, how
do
> I
> >get it. I went all over python.org and it definitely installed the April
> >13th version. (I have installed a couple of times from different
> downloads.)
> >
> >Please advise. Where do I get the debugging tools???
> >
> >Eric
> >
> >
> >
>
>




From jcm at bigskytel.com  Sun Jun 25 19:43:19 2000
From: jcm at bigskytel.com (David Porter)
Date: Sun, 25 Jun 2000 17:43:19 -0600
Subject: smtplib
In-Reply-To: <39568BC4.96EBE8AE@acm.org>; from tuckerg@acm.org on Mon, Jun 26, 2000 at 07:46:28AM +0900
References: <000c01bfdec7$30f84b60$bdcb45c6@timaratz> <39568BC4.96EBE8AE@acm.org>
Message-ID: <20000625174319.D5190@novara.avenue>

* Gregory Tucker <tuckerg at acm.org>:
> 
> P.S. I am getting very little mail off the Python list. Is it something
> wrong with me?
> 

There must be something wrong with me too. My only thought is that the
newsgroup posts aren't making it to the list, but I haven't looked into it
at all.  

  david.



From shapr at uab.edu  Mon Jun  5 09:47:12 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 05 Jun 2000 13:47:12 GMT
Subject: Abstract Syntax Tree from Python2C search and transform?
Message-ID: <393BB5F8.C5DF0431@uab.edu>

I'm trying to design and write a search and transform module that works
specifically on the Python2C abstract syntax tree. The problem is I
can't seem to find a good approach.

I know this is a wide open question... but... can someone make some
suggestions?

I'd be happy to write this module if only I could figure out how.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From nascheme at enme.ucalgary.ca  Fri Jun 23 16:55:15 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Fri, 23 Jun 2000 20:55:15 GMT
Subject: Linux/NT python startup speed
References: <slrn8l3ifj.hv1.r2d2@mao.acc.umu.se> <3953bf64$0$7689@personalnews.de.uu.net>
Message-ID: <slrn8l7jpr.d81.nascheme@cranky.arctrix.com>

Johannes Stezenbach <yawyi at gmx.de> wrote:
>Two solutions were discussed:
>a) make winsound a .pyd
>b) use some obscure "delayload" feature of the M$ linker. There
>   is a ready to use python15.dll for Python 1.5.2 with this
>   feature somewhere out there (I don't remember where, though,
>   since I use Linux now)
>
>IMHO solution a) should be standard in Python 1.6 (it isn't in
>1.6a2).

To quote Guido:

    In 1.6, winsound is a separate subproject and hence a
    separately loadable DLL, so it doesn't slow down the main
    code.  Ditto for socket and a few others.

The problem should be solved.

  Neil

-- 
"Nothing cures insomnia like the realization that it's time to
get up." -- random fortune


From gregholmes at my-deja.com  Tue Jun 20 16:40:16 2000
From: gregholmes at my-deja.com (gregholmes at my-deja.com)
Date: Tue, 20 Jun 2000 20:40:16 GMT
Subject: Strange Tkinter trouble
References: <8inqbe$rgu$1@news.rz.uni-karlsruhe.de>
Message-ID: <8iokrd$r7d$1@nnrp1.deja.com>

I'm not sure if this is the problem, but you need a variable to point
to that root window if you are going to do anything with it.
For example:

from Tkinter import *
root=Tk()
w = Label(root, text="Hello, world!")
w.pack()
root.mainloop()

(this from An Introduction to Tkinter by Lundh)

Bjoern Giesler <un4e at rzstud1.rz.uni-karlsruhe.de> wrote:
> Here is the error message I'm getting:
>
> [snip]
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> from Tkinter import *
> >>> Tk()
> 'None','','Tk' # These are the values of {screen,base,class}Name below
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "/org/opt/python/lib/python1.5/lib-tk/Tkinter.py", line 887, in
> __init__
>     self.tk = _tkinter.create(screenName, baseName, className)
> TclError: couldn't stat "": no such file or directory
> >>>
> [snip]


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


From gward at mems-exchange.org  Thu Jun  1 17:01:41 2000
From: gward at mems-exchange.org (Greg Ward)
Date: Thu, 1 Jun 2000 17:01:41 -0400
Subject: [Python-Dev] What's that sound?
In-Reply-To: <393547A2.30CB7113@prescod.net>; from paul@prescod.net on Wed, May 31, 2000 at 12:10:58PM -0500
References: <39353CCD.1F3E9A0B@prescod.net> <14645.17611.318538.986772@anthem.python.org> <393547A2.30CB7113@prescod.net>
Message-ID: <20000601170140.A24794@mems-exchange.org>

On 31 May 2000, Paul Prescod said:
> Look closer, big-egoed-four-stringed-guitar-playing-one. It could just
> as easily be a J.
> 
> And you know what you get when you put P and J together?

Pajamas!

Yes, that's right, when Python takes over the world, we'll all have to
sleep BUCK NAKED!!!

(What was that about "zero noise" on python-dev?  Oh well, we must have
our fun...)

        Greg



From supersede at india.com  Sat Jun 24 22:50:02 2000
From: supersede at india.com (Extra Strength Flufferin)
Date: 24 Jun 2000 22:50:02 -0400
Subject: Python vs Perl
References: <8ikai5$v3d$1@nnrp1.deja.com>
Message-ID: <87ya3ubi1s.fsf@meowhost.meow.invalid>

This is the best explanation of the diffrences I've seen to date:
<URL:http://www.deja.com/getdoc.xp?AN=638435557>


From dburnett at mindspring.com  Wed Jun 28 09:24:04 2000
From: dburnett at mindspring.com (Dale Burnett)
Date: Wed, 28 Jun 2000 13:24:04 GMT
Subject: Ptyhon Thread Safety
Message-ID: <8jcu92$7op$1@nnrp1.deja.com>

What is the status of thread safety in Python?  I saw some information
stating that Python wasnt thread safe but not which version.


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


From jp at NOSPAMdemonseed.net  Sun Jun  4 12:49:17 2000
From: jp at NOSPAMdemonseed.net (Jason Petrone)
Date: Sun, 4 Jun 2000 11:49:17 -0500
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net> <3939FFB4.A7824367@mindspring.com>
Message-ID: <da1eh8.r4g.ln@demonseed.net>

David Broadwell <dbroadwell at mindspring.com> wrote:
>> Not that this exactly belongs in a python ng but I'm told the 
>> only usable onboard programming environment is quartus, an ansi 
>> forth system by Neal Bridges.

I've found the Scheme implementation LispMe(http://www.lispme.de) to be
usable onboard a palm.  It is very complete and has plenty of support for
things such as graphics and sound.  However, as much as I hate typing all 
those parentheses in Scheme on a keyboard, it is 10x worse in graffiti.

Before LispMe, I found myself using a basic interpreter when I wanted to code
on my palm.  It was usable and robust enough, but it was *basic*.

-jason


From cbbrowne at news.hex.net  Fri Jun 23 22:03:37 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Sat, 24 Jun 2000 02:03:37 GMT
Subject: SAP R/3 and Python?
References: <C59F2D5004BDD311BEC500508B6FF2AD08E816@dewdfx16> <39538247.28CFCB57@lemburg.com>
Message-ID: <ZFU45.219719$MB.3967328@news6.giganews.com>

Centuries ago, Nostradamus foresaw a time when M.-A. Lemburg would say:
>"Wegener, Joerg" wrote:
>> 
>> I think the best way to access R/3 is via a protocol called RFC (Remote
>> Function Call) or via BAPIs (Business APIs). IIRC there is something called
>> RFC-SDK which allows to write or generate code (I'm not sure here) which
>> gives access to RFCs via DDE/OLE. Use this in combination with WIN32-Python.
>> The downside is that you need much more knowledge about R/3 functions, but
>> this is the best and most reliable way.
>> 
>> Hope this helps
>
>Yes... at least it points into a direction which looks like a possible
>solution.
>
>Could it be that SAP (or a partner) already has such a Python
>interface available ? I wouldn't want to reinvent the wheel
>here and I assume this to be a highly non-trivial and time
>consuming task.

I'd suggest taking a look at:
<http://www.mcs.net/~garth/SAP/>

There are RFC interfaces both for TCL and for Python; it ought not to
be _overly_ nasty to adapt those to Python.
-- 
cbbrowne at hex.net - <http://www.hex.net/~cbbrowne/sap.html>
What would the world be like with no hypothetical situations? 


From jelizondo at my-deja.com  Thu Jun 22 10:11:00 2000
From: jelizondo at my-deja.com (jelizondo at my-deja.com)
Date: Thu, 22 Jun 2000 14:11:00 GMT
Subject: Carrage Returns and Tabs in Windows Applications
References: <8ir53n$l5n$1@nnrp1.deja.com> <paf45.4867$qz4.36531@news-server.bigpond.net.au> <slrn8l36j8.b3d.quinn@pyro.ugcs.caltech.edu>
Message-ID: <8it6p6$3b6$1@nnrp1.deja.com>

In article <slrn8l36j8.b3d.quinn at pyro.ugcs.caltech.edu>,
  quinn at pyro.ugcs.caltech.edu (Quinn Dunkan) wrote:
> On Thu, 22 Jun 2000 02:51:01 GMT, Mark Hammond
<MarkH at ActiveState.com> wrote:
> ><jelizondo at simpletel.com> wrote in message
> >news:8ir53n$l5n$1 at nnrp1.deja.com...
> >> Does anyone know if you can code Python to automate Carrage
Returns as
> >> well as Tabs in Windows Applications?
> >
> >Do you mean that as sson as you enter a ':', the editor inserts the
> >newline and the indent for you?
> >
> >If so, I know of no such editor.
>
> Any editor (vi for example) that lets you map one sequence of
keystrokes to
> another will do that, just map :^M to :^M^I  Hopefully most editors
will do
> autoindent (remember last indentaion level), too.
>
> If the original poster didn't mean editors, then sure:  all python
programs
> under windows will automatically insert tabs and CRs in text entry
widgets.
> Just click to show it where, press the tab or enter key, and voila!
>

Actually I may have worded my original message a little bit wrong.
Basicly what I'm wondering is if there is a way to automate a series of
commands in a windows application using python. For example.

os.system(<program> <file>)

then using python to automate a series of commands, sort of like batch
and/or script file.


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


From jwbnews at scandaroon.com  Wed Jun 14 11:20:00 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 14 Jun 2000 08:20:00 -0700
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com> <394a2fed.280565932@news.online.no>
Message-ID: <jwbnews-51ABCC.08200014062000@news.olympus.net>

RedHat 6.1 and 6.2 install Python, at least in the way we install it.

We don't use TkInter at the moment, so I don't know whether it is 
functional.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From mwh21 at cam.ac.uk  Wed Jun 14 08:11:06 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 14 Jun 2000 13:11:06 +0100
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net> <3946760F.C5E46FD3@muc.das-werk.de> <8i7r1s$l96$1@newshost.accu.uu.nl>
Message-ID: <m3aego31k5.fsf@atrus.jesus.cam.ac.uk>

m.faassen at vet.uu.nl (Martijn Faassen) writes:

>   * Pythoneers care less about inheritance hierarchies. When a class
>     supports an interface it supports an interface; inheritance often
>     isn't needed. 

To expand on this: it seems to me that the only reason to inherit in
Python is to inherit implementation; there's just no point in
inheriting to inherit an interface (subtyping).  I've lent my copy of
OOSC2 out, but that somewhere has a terrifying diagram listing (I
think) 13 different reasons to inherit; about two of them apply to
Python, IIRC.

simply-admire-simplicity-ly y'rs
Michael

-- 
  There are two ways of constructing  a software design: one way is to
  make it so  simple that there are obviously  no deficiencies and the
  other way  is to make  it so complicated  that there are  no obvious
  deficiencies.                                      -- C. A. R. Hoare


From bazza at ireland.com  Mon Jun 19 05:01:29 2000
From: bazza at ireland.com (Barry Kilcline)
Date: 19 Jun 2000 10:01:29 +0100
Subject: standalone programs
Message-ID: <8iknh9$2a3t$1@walton.maths.tcd.ie>

Folks

I am a novice python programmer with some experience in C++.  I'm 
trying to create a fully stand alone programme to put on systems 
where python is not installed.  For my sins, I'm working on a windows 
platform.  Can anyone give me an idiot's guide on how to do this, please?
I have tried using freeze but without much success. (I have VC++ installed)


Any help would be much appreciated

Regards
Bazza




From gansevle at cs.utwente.nl  Thu Jun 15 06:45:20 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Thu, 15 Jun 2000 10:45:20 GMT
Subject: Mapping a network drive
References: <2B1262E83448D211AE4B00A0C9D61B03013011E2@msgeuro1.creo.be>
Message-ID: <8iac3u$im8$1@nnrp1.deja.com>

In article <2B1262E83448D211AE4B00A0C9D61B03013011E2 at msgeuro1.creo.be>,
  Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> Hello to all,
>
> anyone an idea how I can map a drive using pythonwin, and see
> if it was successfull or not? I'm running python 1.5.2 on WinNT 4.
>
> Pieter
>
>
Path: news!uunet!ffx.uu.net!newsfeed.easynews.net!easynews.net!newsfeed.germany.net!newscore.gigabell.net!newsfeed00.sul.t-online.de!t-online.de!newsfeed.icl.net!newsfeed.icl.net!news.maxwell.syr.edu!nntp2.deja.com!nnrp1.deja.com!not-for-mail
Xref: news comp.lang.python:99709

This code I wrote using the 'dynwin' package.

# wnet.py -- wnet module

from dynwin import winwin, winclass, msgloop, windll
from dynwin.structob import struct_object
from dynwin.oracle import Oracle

mpr = windll.module ('mpr')

RESOURCETYPE_DISK       = 0x00000001

class NETRESOURCE (struct_object):
        oracle = Oracle (
                'NETRESOURCE',
                'Lllllllll',
                ('dwScope',
                 'dwType',
                 'dwDisplayType',
                 'dwUsage',
                 'lpLocalName',
                 'lpRemoteName',
                 'lpComment',
                 'lpProvider')
                )

def use(drive, host, user, passwd):
    nr = NETRESOURCE()
    nr.dwType = RESOURCETYPE_DISK
    lpLocalName = windll.cstring(drive)
    lpRemoteName = windll.cstring(r'\\%s\%s' % (host, user))
    lpProvider = windll.cstring('Microsoft Windows Network')
    nr.lpLocalName = lpLocalName.address()
    nr.lpRemoteName = lpRemoteName.address()
    nr.lpProvider = lpProvider.address()
    return mpr.WNetAddConnection2(nr,
                windll.cstring(passwd), windll.cstring(user), 0)

def unuse(drive):
    return mpr.WNetCancelConnection2(windll.cstring(drive), 1, 1)

--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From porter at et.byu.edu  Wed Jun 21 16:48:05 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Wed, 21 Jun 2000 14:48:05 -0600
Subject: Frozen socket.
In-Reply-To: <uln05qioe.fsf@ctwd0143.fitlinxx.com>
References: <Pine.GHP.4.21.0006161522100.3877-100000@don.et.byu.edu> <uln05qioe.fsf@ctwd0143.fitlinxx.com>
Message-ID: <Pine.GHP.4.21.0006211035450.5957-100000@don.et.byu.edu>

OK, I have been able to duplicate the problem that I was having, but I
still don't know what to do about it.  First things first, I suppose.

I made two small programs, a server and a client, and I got the client to
"freeze" in the exact same way.  My test server accepts three messages,
but after that, it freezes.  For some reason that I don't understand, the
client keeps sending messages (about 1000 of them, just like before), even
though the server is not receiving any.  After about 1000 messages, it
freezes, just like the program I am debugging.

Why is the client able to keep sending messages if the server is not
receiving them?  

In short, what is really happening, and how do I safeguard against it?

Here is the source for the two test programs I am talking about.
sample client:
from socket import *
import time 
sock = socket(AF_INET, SOCK_STREAM)
sock.connect("127.0.0.1",2778)
for i in range(5000):
   sock.send("This is a string of approximately the correct
length.          "
   time.sleep(0.001)
   print i

sample server:
from socket import *
s = socket(AF_INET, SOCK_STREAM)
s.bind(""<2778)
s.listen(5)
conn, addr = s.accept()
counter = 0
while 1:
   counter = counter + 1
   data = conn.recv(1024)
   if couter > 3:
      while 1:
         pass
   if data:
      print data

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------

On 16 Jun 2000, David Bolen wrote:

> "C. Porter Bassett" <porter at et.byu.edu> writes:
> 
> > I have looked around, but I haven't been able to figure out why a program
> > would choke on a send() command.  I don't even know if it is because of
> > something I am doing on my end or something happening on the server
> > end.  If it is my fault, how do I fix it?  If it is because of something
> > happening on the server side, how do I compensate?
> 
> Under default conditions, send() may block if there is no room to
> locally buffer the information you want to send.  It's not that it has
> "choked" - just that it's waiting for room to become available.  You
> can set a socket to non-blocking mode, but then you have to handle a
> failure to send() all of your data and keep retrying until it is
> accepted.
> 
> In most cases, this occurs because you have gotten far ahead of the
> receiver - enough so to back up the TCP window size across the network
> and your internal socket buffers.  As long as the systems are still
> actually sending data, it will eventually catch up and you can keep
> sending.
> 
> However, the fact that you say it happens regardless of how quickly
> you send out the strings makes me wonder if the other side is really
> receiving them at all.  Are you absolutely positive that you are
> performing the initial communication properly so that the receiving
> side is really receiving your data?  It may be just that it takes you
> about that many strings to fill up your internal queues while the
> remote system does not receive any data.
> 
> I suppose another possibility is something in the data is causing a
> problem for the remote system such that after receiving that initial
> amount it becomes busy and fails to continue receiving.
> 
> Do you have any other applications that perform this function that you
> might be able to compare/contrast with - having some sort of control
> or pre-existing behavior to compare against might give you some hint
> as to what is going on.  But send() blocking is normally just a
> symptom of something larger happening, and less likely to be a problem
> in the actual send() call itself.
> 
> --
> -- David
> -- 
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 




From tim_one at email.msn.com  Sun Jun 11 15:33:51 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 11 Jun 2000 15:33:51 -0400
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
In-Reply-To: <Pine.SGI.4.21.0006100937230.10317-100000@wendigo.bio.dfo.ca>
Message-ID: <000301bfd3db$f921be60$27a2143f@tim>

[George N. White]
> There are a couple issues in this thread:
>
> 1)  cross-platform consistency
>
> 2)  treatment of exceptions: underflow, overflow, invalid, ...
>
> For 1), Sun's freely distributable math library should eliminate the
> effects of differences in vendor-supplied math libraries

Except that, as you say later, "speed wins", and fdlibm is almost certainly
slower than the native libm.  I happened to be writing a math library for a
HW vendor at the time K.C.Ng was writing fdlibm, and by exploiting platform
quirks I was able to get better accuracy at 10x the speed.  There's really
no good portable & safe way to deal with the mass of 754 special cases
without casting tricks and mounds of conditionals.

> For 2), support for reliable reporting of f.p. exceptions when they
> occur conflicts with techniques (e.g., combined multiply+add: y=a*x+b)
> used to get better f.p. performance in new processor designs.  Since
> the marketplace favors speed over all else, the conflict will likely
> be resolved in favor of performance.  This is reflected in the lack of
> support for f.p. exceptions in language standards.  It is much faster
> and easier to simply store a 0, +/-Inf, or NaN and keep going.

Stangely enough, on a Pentium Inf and NaN inputs cause + - * / to take up to
10x longer than normal operands.  I've never seen that documented, BTW -- it
was an empirical observation after tracking down why an FFT took about 5x
longer than I thought it should (turned out that the test harness generated
Infs and NaN inputs by mistake).

> Until we have benchmarks that explicity test underflow and overflow
> behaviour, vendors are going to focus on performance for the simplest
> cases.

Yup.  The saving grace for Python is that it's so slow on fp arithmetic
anyway that nobody would notice if it took the time to do it right <0.6
wink>.

> In practical terms, users need to a) write code that doesn't rely
> on exception handlers

You can always get at them, albeit in wildly differing platform-specific
ways.  I'd like Python to abstract that and so make it usable via one
spelling.

> and b) push for support in language stds. for features needed to work with
> signed 0, NaN's and Inf's in (e.g.: don't optimize away expressions like
> 'x != x').

Note that the Numerical C Extension Group's 754 recommendations were adopted
pretty much in whole by C9X ("the next" iteration of the C std, and already
ISO's current official version).

> As it happens, just this week I encountered a subroutine that was
> returning NaN's from a calculation of the form: y*log(y) where y=exp(-x)
> and x was large enough to cause underflow, giving 0*(-Inf)=NaN.  I tried
> trapping the underflow, but there were _thousands_ of log(y) expressions
> used in a way that did not cause any difficulty before I reached the one
> that was returning a NaN.  In this case, trapping underflows gave so many
> false positives that it didn't help identify the real problem area in the
> code.

Why didn't you trap invalid operation instead?  Every 754 operation that
produces a NaN from non-NaN inputs (like 0 and -Inf) signals inv. op.

> Perhaps a better approach for Johann's specific example would be to use a
> special non-underflow function, exp_p(-x), that calls an error routine
> when x exceeds some threshold, or add a test before exp(-x) is called.

W. Kahan's pithy definition of an exceptional operation is any behavior that
someone may take exception to <wink>.

math-libraries-are-both-the-best-defined-and-least-agreed-upon-ly y'rs  -
tim





From cut_me_out at hotmail.com  Mon Jun  5 17:52:44 2000
From: cut_me_out at hotmail.com (Alex)
Date: 05 Jun 2000 17:52:44 -0400
Subject: C extension using Numeric arrays -- do I really need all this error
Message-ID: <etdpupvdccz.fsf@w20-575-109.mit.edu>

Hi.  I wrote a simple C extension using Numeric arrays today.  It takes
an integer array and the coordinates of an element in that array, and
increments it by one.  The code is below, if anyone's interested.

I was wondering, do I really have to be so prissy about error checking?
Is there anything I can do to make Numeric do more of the work?

Alex.

#include "Python.h"
#include "arrayobject.h"
#include <assert.h>

void *die (PyObject *exception, char *error_message) {
    PyErr_SetString (exception, error_message);
    return NULL;
}

void *die_value (char *error_message) {
    return die (PyExc_ValueError, error_message);
}

void *die_system (char * error_message) {
    return die (PyExc_SystemError, error_message);
}

PyObject *None () {
    Py_INCREF (Py_None);
    return Py_None;
}

long offset (int dimension, int *strides, PyObject *coordinates) {
  long int answer = 0, dimension_index;
  PyObject *coordinate;
  assert ((PySequence_Check (coordinates)) &&
	  (PySequence_Length (coordinates) == dimension));
  for (dimension_index = 0; dimension_index < dimension; dimension_index++) {
    coordinate = PySequence_GetItem (coordinates, dimension_index);
    answer += strides[dimension_index] * PyInt_AsLong (coordinate);
  }
  return answer;
}

void PyArrayObject_Print (PyArrayObject *array) {
  PyObject_Print ((PyObject *)array, stdout, 0);
  printf ("\n");
}

static PyObject *my_arrays_inc (PyObject *self, PyObject *args) {
  PyObject *indices, *increment;
  PyArrayObject *array;
  int dimension_index, cell_offset, coordinate;

  if (!PyArg_ParseTuple(args, "O!O", &PyArray_Type, &array, &indices)) {
    return NULL;
  }

  if (array->descr->type_num != PyArray_INT) {
    return die_value ("Array must be of type int");
  }

  if (!PySequence_Check (indices)) {
    return die_value ("Second argument should be a sequence.");
  }

  for (dimension_index = 0; dimension_index < PySequence_Length (indices);
       dimension_index++) {
    if (!PyInt_Check (PySequence_GetItem (indices, dimension_index))) {
      return die_value ("Second argument should be a sequence of integers.");
    }
  }

  if (array->nd != PySequence_Length (indices)) {
    return die_value ("Length of indices must correspond to dimension of array.");
  }
  
  for (dimension_index = 0; dimension_index < array->nd; dimension_index++) {
    coordinate = PyInt_AsLong (PySequence_GetItem (indices, dimension_index));
    if ((coordinate < 0) || (array->dimensions[dimension_index] <= coordinate))	 {
      return die (PyExc_IndexError,
		  "Values in indices must be nonegative and less than corresp. dimensions.");
    }
  }
  
  
  cell_offset = offset (array->nd, array->strides, indices);
  (int)(array->data[cell_offset]) += 1;

  return None ();
}


static PyMethodDef my_arrays_methods[] = {
  {"inc", my_arrays_inc, METH_VARARGS},
  {NULL, NULL} /* Marked as "Sentinel" in manual. */
};

void initmy_arrays () {
  (void) Py_InitModule ("my_arrays", my_arrays_methods);
}


From dworkin at ccs.neu.edu  Wed Jun 28 18:56:58 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 28 Jun 2000 18:56:58 -0400
Subject: tuple question
In-Reply-To: etoffi's message of "Wed, 14 Jun 2000 02:25:34 -0400"
References: <642A954DD517D411B20C00508BCF23B001296A2E@mail.sauder.com>
Message-ID: <vndvgyt9zyd.fsf@camelot-new.ccs.neu.edu>

etoffi <etoffi at softhome.net> writes:

> but what is a tuple and why is it used (as opposed to an array, etc).

It is a simple, immutable sequence.  Use it is situations where you
could use a list, but will never modify the contents of that list
and/or require that nothing else do so either.

This has at least two obvious initial uses:

- dictionary keys  
  (must be immutable)

- parameter passing  
  (it's nice to know that a function won't mangle what is passed to
   them as parameters, unless that's what you really wanted)

There are many other good places to use tuples, but that should give
you enough of an idea.

-Justin

 




From shapr at uab.edu  Tue Jun  6 14:32:30 2000
From: shapr at uab.edu (Shae Erisson)
Date: Tue, 06 Jun 2000 18:32:30 GMT
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <393D4359.F308AC05@uab.edu>

"Richard P. Muller" wrote:
> 
> Does Programming in Python Marginalize You?
> 
> I love Python. Of the 10-15 programming languages I know, Python is
> the only language which I actually look forward to using. I wake up in
> the morning, ask myself, "Will I be using Python today?" and if the
> answer is "yes" I know that it's going to be a good day.

I agree with you here. I'm a fanatic :)

> I'm also aware of the JPython module, which promises to enable
> us to write applets in Python instead of Java. But who among us has
> actually *used* JPython? Or, for that matter, who has come across a
> random web page on the net that uses JPython applets?

I have used JPython in the past, and will do so in future. JPython is a
wonderful tool when you must write for the JVM. I also think JPython is
easier for learning the Java API, it's quite spiffy to be able to create
frames and widgets and stick them together from a prompt.

> Java, but it didn't happen that way. Think how wonderful it would have
> been to write PythonScript functions in web pages rather than
> JavaScript.

I vaguely remember reading that Mozilla promises to allow Python and
Perl as client side scripting languages. I'll recheck that.. but I think
I saw it when reading about the ActiveState Komodo IDE

> What else should we be doing to make Python *the* programming language
> rather than *a* programming language?

I think IDLE should get lots more work. I think it could use CVS
integration, really good syntax highlighting (like using an incremental
parser), and lots of other features.

I also suggest that some way is found to use libraries from other
languages (like that other P language), much like Ruby uses Python
libraries. I don't know much about Ruby, but the idea of using a
well-rounded set of libraries from another language sounds extremely
useful. Several times I've run across libraries in P*** that I wished I
could use in Python.

<shameless plug>
Help me write the Refactoring Browser for Python :)
I think it would improve many things about medium to large Python
projects.
https://sourceforge.net/project/?group_id=4298
</shameless plug>
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From mtbros at tin.it  Thu Jun 22 18:19:55 2000
From: mtbros at tin.it (Mario Tola)
Date: Fri, 23 Jun 2000 00:19:55 +0200
Subject: Can I connect to a DDE server?
Message-ID: <8iu3eu$fnf$1@nslave1.tin.it>

Hi,
I'd like to know if it is possible (of course on Win) to send commands to a
DDE server (in my case it is ACDSee.exe) from python. And, if yes, how ?
I think yes as I have read on "Python Programming on Win32" that "PythonWin
uses DDE to
communicate between instances"... but I can't find documentation on how to
use it myself.
If someone could send me a small example...
Many thanks...
Mario






From fig at oreilly.com  Fri Jun  9 18:26:48 2000
From: fig at oreilly.com (Stephen R. Figgins)
Date: Fri, 09 Jun 2000 22:26:48 GMT
Subject: Gnome Applets and Python
Message-ID: <39416F2B.96AD2322@oreilly.com>

Any good suggestions on Gnome Applets and Python information?

Stephen



From thomas at xs4all.net  Tue Jun  6 12:15:46 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 6 Jun 2000 18:15:46 +0200
Subject: TypeError: unhashable type
In-Reply-To: <393d1c1a_2@news1.prserv.net>; from sh1232@yahoo.com on Tue, Jun 06, 2000 at 10:38:44AM -0500
References: <393d1c1a_2@news1.prserv.net>
Message-ID: <20000606181546.H469@xs4all.nl>

On Tue, Jun 06, 2000 at 10:38:44AM -0500, sh wrote:

> This works without a glitch in the command line interpreter:
>     segments = {}        # dictionary type
>     segment = []
>     # populate segment
>     segments[segment[0]] = segment[len(segment)-1]
> 
> BUT, when I run it from the DOS prompt: "python segments.py", I got error
> message:
>     TypeError: unhashable type.

> Any explanation is appreciated.  Thanks.

You ommited some code, namely the code that fills the list 'segment'. You
are probably filling it with unhashable objects (mutable objects, in general.)
Dictionary keys have to be hashable, which means they have to be immutable
(or at least always return the same hash value.)

If you populate 'segment' with lists, try populating it with tuples. Tuples
are immutable, and can be used as dictionary keys.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From rupole at compaq.net  Thu Jun  8 18:29:59 2000
From: rupole at compaq.net (Roger Upole)
Date: Thu, 08 Jun 2000 22:29:59 GMT
Subject: Win32net.NetServerEnum Unicode problem
References: <8hlnd2$lrm$1@nnrp2.deja.com> <erz%4.2026$wp.166067@pouncer.easynews.com> <8hoglp$54l$1@nnrp1.deja.com>
Message-ID: <H7V%4.7674$wp.595088@pouncer.easynews.com>

What this means is that server info level 102 is not supported by
NetServerEnum.
Try it with level 100 as the first parm.  BTW, the last 4 parms have
defaults, so
this function will accept anywhere from 2 to 6 parms.  If you pass the wrong
number of parms, it will explicitely tell you so.

                   Roger Upole

<jcashman55 at my-deja.com> wrote in message
news:8hoglp$54l$1 at nnrp1.deja.com...
> I have Hammond's Python on Win32 book, and it doesn't list another
> parameter relating to domain name.  Plus it doesn't seem to work.
> Complains that the system call level is not correct, which is usually
> when I don't have the right number of params or type of params.
>
> jim
>
> In article <erz%4.2026$wp.166067 at pouncer.easynews.com>,
>   "Roger Upole" <rupole at compaq.net> wrote:
> > It looks like you are missing the domain name, which should
> > be right after SV_TYPE_ALL.
> >         Roger Upole
> >
> > <jcashman55 at my-deja.com> wrote in message
> > news:8hlnd2$lrm$1 at nnrp2.deja.com...
> > > Having trouble with the following line:
> > >
> > >         entries, total, resume= \
> > > win32net.NetServerEnum('Dopey', \
> > > 102,win32netcon.SV_TYPE_ALL, 0, 4096)
> > >
> > > When I run code that contains this line, I get:
> > >
> > > TypeError: The object can not be converted to a
> > > Unicode object
> > >
> > > I have run some of the the other win32net calls
> > > without a problem.  Anyone have an idea?
> > >
> > > Thanks.
> > >
> > > Jim
> > >
> > >
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> >
> >
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From aahz at netcom.com  Fri Jun 16 11:15:38 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jun 2000 15:15:38 GMT
Subject: Newbie question : time zone differences
References: <bingdo19-D78F49.10323816062000@news-server.twcny.rr.com>
Message-ID: <8idgaq$jgb$1@slb7.atl.mindspring.net>

In article <bingdo19-D78F49.10323816062000 at news-server.twcny.rr.com>,
Bob Langdon  <bingdo19 at earthlink.net> wrote:
>
>I've installed the Pythion interpreter in my hme directory (hosting co.
>doesn't have it).
>
>My host is on the west cpoast but I'm on the east coast.
>
>How do I adds +3 hours to the time without going through every script?

You should be able to do this by setting the TZ variable to e.g. "EDT";
unfortunately, I think that if you do that, you'll have to remember to
change TZ twice a year (when Daylight Savings switches on and off).  I
just took a look at my Solaris box, and it has TZ set to "US/Pacific",
so that might work around the Daylight Savings problem.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"real love can't be explained by simplistic platitudes."  --piranha


From liang at cad.gatech.edu  Thu Jun  8 15:18:20 2000
From: liang at cad.gatech.edu (Shengquan Liang)
Date: 8 Jun 2000 19:18:20 GMT
Subject: Newbie: Pass Value
Message-ID: <8horhs$rfl$1@news-int.gatech.edu>

 in the main progam, i call a method 
populate (), and pupulate  call another
method assign() .

 like this: 
	     populate(assign())
 in the assign() method i assign values of fields in a form.
 populate() takes values from assign and put those values
 into a table and draw it out.
 so far so good.

 Now, i want to pass some input from user to the method assign()
 and let assign() do some processing.  Say the varible in assign()
 is post.   How do i pass the value of user input to "post"?

 e.g.       populate(assign(post= form1["InputNumber"]) )

in this statement i'm trying to assign a value collected from 
a HTML form field to "post".  but it seems that it is not successful.

any comments will be appreciated.


regards,

sqL


From shahn at cs.vu.nl  Thu Jun 22 03:52:18 2000
From: shahn at cs.vu.nl (Sander Hahn)
Date: Thu, 22 Jun 2000 09:52:18 +0200 (MET DST)
Subject: apply for objects
In-Reply-To: <Pine.BSF.4.10.10006211536090.1727-100000@chi.pair.com>
Message-ID: <Pine.GSO.4.20.0006220932160.9460-100000@keg.cs.vu.nl>

Hello Michal!

Thank you for your reply!

I wanted to call a method of an object in Python.
The way to call a function is:

>>> def f(x, y):
	print str(x) + str(y)
	
>>> apply(f, ('hello',), {'y': 6})
hello6

That works, however when you use apply on a method you get the following
error:

>>> class C:
	def m(x, y):
		print str(x) + str(y)
		
		
>>> c = C()

>>> apply(c.m, ('hello', ' bye now'), {})
Traceback (innermost last):
  File "<pyshell#12>", line 1, in ?
    apply(c.m, ('hello', ' bye now'), {})
TypeError: too many arguments; expected 2, got 3

I previously got errors about no __call__ or something weird...!?!
However i found out that you must use the im_func from the method to get
the corresponding function:

>>> apply(c.m.im_func, ('hello', ' bye now'), {})
hello bye now

So, my problem is solved for now :-)

Thank you for your friendly reply!

Regards,
Sander Hahn




From ddb at crystal.uwa.edu.au  Mon Jun 26 03:55:05 2000
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Mon, 26 Jun 2000 15:55:05 +0800
Subject: jpython casting to abstract java class
Message-ID: <39570C59.2CBCFAEB@crystal.uwa.edu.au>

Hi,
I have a small(ish) code fragment below which fails to 
cast a python-subclass of an abstract-java-base-class back to that
java-base-class when it is passed as input to another 
java (constuctor) method. 

There also seems to be that no checking that all the abstract 
methods of the abstract base class are implemented. 

Is there anything I can do to make this work (it works fine in
jpython1.0.3,
but is broken in jpython1.1 +latest CVS version + Finn Bock's errata-07, 
all with java 1.1.7)

thanks 
doug
-------------- next part --------------
<pre>
#!/usr/bin/env jpython
# file: testMyClass.py

import sys
sys.add_package('parser')

from parser import PyReadInterface,PyReader

class reader(PyReadInterface):    # PyReadInterface is an abstract class
        "@sig class reader extends PyReadInterface"
        def __init__(self,other):
                self.other=other
        def read(self,n):
                "@sig public String read(int n)"
                return String(self.other.read(n))
        def close(self):
                "@sig public void close(void)"
                self.other.close()


class MyClass:
        def __init__(self,fpin=sys.stdin):
                self.locals={}
                self.fpin=fpin
        def read(self,n):
                return self.fpin.readline()
        def close(self):
                pass
        def test(self):
                k=reader(self)
                m = PyReader(k) # reader should be reduced to PyReadInterface
                                # unfortunately it isn't and we get:
#TypeError:parser.PyReader(): 1st arg can't be coerced to parser.PyReadInterface
#                print k
#                print dir(k)
#                print k.__class__
#                print k.__class__.__dict__
#                print k.__class__.__name__
#                print k.__class__.__bases__
#                print k.__class__.__bases__[0]
#                l = k.__class__.__tojava__(k.__class__.__bases__[0])
#                print l
#                print l.__class__
#                print l.__class__.__name__
#                print super.__tojava__(k)
                print "didn't raise an exception! maybe its working now?"


if __name__ == '__main__':
        MyClass().test()
# endOfFile: testMyClass.py


/* file: parser/PyReadInterface.java */
package parser;
public abstract class PyReadInterface {
        public abstract void close();
        public abstract String read(int n) ;
}
/* endOfFile: parser/PyReadInterface.java */


/* file: parser/PyReader.java */
package parser;
// simple wrapper for a python class
// so that it can deliver characters to
// the tokenmanager
// python classes should inherit from PyReadInterface
public class PyReader extends java.io.Reader {
        public PyReadInterface py;
        public PyReader(PyReadInterface py) {
                this.py=py;
        }
        public int read(char[] buffer,int n,int m) {
                        String ss=py.read(m);
                        if (ss.length()==0) return -1;
                        ss.getChars(0,ss.length(),buffer,n);
                        return ss.length();
                }
        public void close() {
                py.close();
        }
}
/* endOfFile: parser/PyReader.java */
</pre>

From pierre.pompanon at wanadoo.fr  Tue Jun  6 07:26:50 2000
From: pierre.pompanon at wanadoo.fr (pierre.pompanon at wanadoo.fr)
Date: Tue, 06 Jun 2000 13:26:50 +0200
Subject: - Sondage -
Message-ID: <E12zHVS-0000Ra-00@aweb-pub.actionweb.fr>



Bonjour,

Voulez vous participer ? l?essor d?une Start up ?

Voulez vous devenir actionnaire sans d?bourser un centime ?

Rejoignez-nous sur le site 
http://lesondage.free.fr/html/sond02.php3

remplissez le questionnaire de participation et que l'aventure commence!!!

L'Equipe du sondage           





From bjorn at roguewave.com  Wed Jun  7 20:36:28 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 07 Jun 2000 18:36:28 -0600
Subject: I need a lesson.
References: <8hmn80$4vu$1@ionews.ionet.net>
Message-ID: <393EEA8B.46B3AF13@roguewave.com>

"F. G. Brown" wrote:
> 
> Greetings,
> I just recently started learning to program using Python, and to be honest
> when it comes to programming I'm almost dumb as a rock.  I've done some
> Qbasic, but I consider it a dead language as far as the future is concerned
> and wish to concentrate my efforts with Python.
> Here's the rub.  The following is a module called howfast.py that I can't
> get to repeat.  In Qbasic I would have used a goto at the end and it would
> have worked fine.  Of course there is no goto in Python and that where I can
> use anyone's help.  See if you can understand what I'm trying to accomplish
> below, it's a little scrambled towards the end, and tell me the error of my
> ways.  I have the book "Learning Python" but I'm still stuck.  If there are
> any questions I'll be glad to reply.
> Thanks,
> Fred
> ?????????????????????????????????????????????
> 
> howfast.py
> 

Welcome to Python!

Something like:

d = 'y'
while d == 'y':
    a = input("What is the distance you traveled in miles?  ")
    b = input("How many seconds did it require?  ")
    c = a*3600.00/b
    print "You were going %d mph!" %c
    if c > 75:
        print "Hey you better slow down!"
    d = raw_input ("Would you like to do another? (y)es or (n)o")
    if d == 'y':
        print 'Here we go!'
    else:
        print 'See ya!'

will work. Here are some other suggestions (free of charge ;-)

 - do yourself a favor and use more than one space to indent
   (it's much easier to read with e.g. four space indents).
 - name your variables distance, seconds, speed, and answer
   instead of a, b, c, and d (when you come back to your
   code in a couple of days you'll be glad you did :-)

-- bjorn



From cjensen at bioeng.ucsd.edu  Thu Jun 29 18:30:10 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 29 Jun 2000 15:30:10 -0700
Subject: Which member is it?
Message-ID: <395BCDF2.CAFD7A34@bioeng.ucsd.edu>

class: Foo:
  def __init__(self):
    self.bar1 = [1,2,3]
    self.bar1 = [1,2,3]
    self.bar2 = [1,2]
    self.bar2 = [1,2,3,4]

tmp = Foo()

If I make reference to tmp.bar1, which do I get or, does it matter?
If I make reference to tmp.bar2, which do I get?  it will make a
difference here.

In my code, it seems tmp.bar1 doesn't seem to cause problems.  However,
tmp.bar2 does cause some funny problems, especialy when making calls to
C modules.  What exactly does python do when you reference same name
with different bindings?  Thanks.

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


From bwarsaw at python.org  Wed Jun  7 01:30:54 2000
From: bwarsaw at python.org (Barry A. Warsaw)
Date: Wed, 7 Jun 2000 01:30:54 -0400 (EDT)
Subject: Python under VMware/NT?
References: <r3h%4.23$cT1.480@news>
Message-ID: <14653.56846.80327.424290@anthem.python.org>

>>>>> "CR" == Chris Ryland <cpr at emsoftware.com> writes:

    CR> Has anyone tried running Python in a virtual machine under
    CR> VMware/NT, whether in Linux or NT/2K/9x?

I very briefly ran Linux under NT-native VMware and of course ran
Python inside the Linux vm.  Worked fine.  I had the general
impression that networking was much slower for the entire OS but spent
no time benchmarking or otherwise investigating.  I do plan to play
with NT running inside Linux-native VMware pretty soon, but I don't
expect it to be all that different than running in native NT.

-Barry



From aahz at netcom.com  Fri Jun  9 13:50:37 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jun 2000 17:50:37 GMT
Subject: readlines() question
References: <8hra5t$mfi$1@pulp.srv.ualberta.ca>
Message-ID: <8hrapd$e2g$1@nntp9.atl.mindspring.net>

In article <8hra5t$mfi$1 at pulp.srv.ualberta.ca>,
L Stewart  <liam at gpu.srv.ualberta.ca> wrote:
>Just out of curiosity, when something like the following is used,
>
>contents = open('file').readlines()
>
>is 'file' closed after readlines() finishes? If not, is it possible to
>somehow get either the file descriptor or the file object of the opened
>file? Thanks,

Yes, it's closed.  The anonymous object created by open() gets
deallocated after the assignment; part of the cleanup is to silently
call close().
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From mwh21 at cam.ac.uk  Tue Jun 13 11:26:05 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Tue, 13 Jun 2000 11:26:05 -0400
Subject: importing module through variable
References: <8i5e9p$nm0$1@zingo.tninet.se> <20000613160606.W20700@xs4all.nl> <20000613162155.A12328@beta.telenordia.se>
Message-ID: <642A954DD517D411B20C00508BCF23B0012968E9@mail.sauder.com>

=?iso-8859-1?Q?Andr=E9_Dahlqvist?= <andre at beta.telenordia.se> writes:

> > You want the __import__ function:
> > 
> > module = __import__(var)
> 
> I misphrased the question. Inside my module I have a dictionary that I
> need to access. And the name of this dictionary is stored in a
> variable. If I do:
> 
> import the_module
> 
> How can I then access the dictionary inside this module, if the name of
> the dictionary is stored in a variable?

getattr(the_module,var)

Python is gloriously regular in some ways...


> Or is there some equivalence to "from the_module import
> the_dictionary" when you're using __import__?

Probably.  But you want the above, if I understand you properly...

Cheers,
M.

-- 
41. Some programming languages manage to absorb change, but 
    withstand progress.
  -- Alan Perlis, http://www.cs.yale.edu/homes/perlis-alan/quotes.html



From dcalvelo at pharion.univ-lille2.fr  Sat Jun 10 16:22:49 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 10 Jun 2000 20:22:49 GMT
Subject: OpenDX Python Wrapper -- any interest?
References: <20000515112219.C1304991@vislab.epa.gov> <8hrdpr$ahj$1@nnrp2.deja.com>
Message-ID: <8hu82p$33p$2@netserv.univ-lille1.fr>

Randall Hopper <aa8vb at yahoo.com> wrote:
>      A while back I wrote a Python wrapper for OpenDX, complete with
> callbacks to Python and exceptions thrown across the DXL layer.  Before I
> devote any time toward preping, packaging, and posting it, I thought I'd
> first see if any others are interested in this tool (e.g. to build Python
> apps which use DX for graphics/visualization under-the-hood).

Count me in, if large figures (>8) are needed.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From msoulier at nortelnetworks.com  Thu Jun 29 07:34:09 2000
From: msoulier at nortelnetworks.com (Michael P. Soulier)
Date: 29 Jun 2000 11:34:09 GMT
Subject: compilation
References: <Pine.LNX.4.21.0006270751280.2223-100000@fep132.fep.ru> <8jaht1$rui$5@bmerhc5e.ca.nortel.com> <8jdi5s$om4$1@nnrp1.deja.com>
Message-ID: <8jfc7h$so4$1@bmerhc5e.ca.nortel.com>

In article <8jdi5s$om4$1 at nnrp1.deja.com>, Fr?d?ric van der Plancke wrote:
>
>When the interpreter finds a xxx.pyc file on your Python path when
>you type 'import xxx', it loads that directly instead of recompiling
>xxx.py  (provided xxx.py is older, and other soundness conditions)
>
>Is that what you want ?

	Umm, sort of. I guess the top-level script still has to be a .py file.
That's fine, since if you're really wanting performance you could put
everything in the modules, let python compile them into pyc's, and then just
get the ball rolling from the .py file. 
	I'll have to benchmark this and see how much of an improvement it offers. 

	Mike

-- 
Michael P. Soulier, 1Z22, SKY  Tel: 613-765-4699 (ESN: 39-54699)
Optical Networks, Nortel Networks 
"...the word HACK is used as a verb to indicate a massive amount
of nerd-like effort."  -Harley Hahn, A Student's Guide to UNIX


From aa8vb at yahoo.com  Thu Jun  8 14:46:35 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 08 Jun 2000 14:46:35 -0400
Subject: CPAN for python?
In-Reply-To: <3941bf74.166708937@nntp.interaccess.com>; from
 olczyk@interaccess.com on Thu, Jun 08, 2000 at 03:46:23PM +0000
References: <393fbb15.100053125@nntp.interaccess.com>
 <393ECB55.4AC016A4@uab.edu> <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us>
 <20000608073337.A324967@vislab.epa.gov>
 <3941bf74.166708937@nntp.interaccess.com>
Message-ID: <20000608144635.B328070@vislab.epa.gov>

Thaddeus L. Olczyk:
 |Randall Hopper:
 |
 |>Wonder if Python could have something similar.  Folks could post URLs to a
 |>"getit.py" (or setup.py) script, and users could literally cut-and-paste
 |>this URL into a Python install tool that would just "do the install".  That
 |>is, don't bother them with where to FTP or surf, how to get the files,
 |>where to put the packages, what the name of the distdir in the tarball is,
 |>etc.  Just install the package and do what it takes to make it happen.
 |>That'd be cool.
 |
 |You've got to be kidding! 
 |Haven't you heard of "melissa" or "ILoveYou"?

Hey, I didn't mean to get your dander up on this.  Far be it from me to
suggest solutions on par with MSWindows joke security where users just
click on attachments to nuke their machine.  [[ Executable .vbs attachments
look like text attachments, the Windows Scripting Host enabled by default,
users having no idea where an e-mail with attachment really came from, most
installations having no concept (or having negated) the concept of a
priviledged user -- no, nothing so silly. ]]

Relevent differences: in this case, the user proactively "pulls" a
port/module dist/getit.py from a "trusted" host.  None of this insecure
push nonsense.  And if it's not from a trusted host, it's up to the user to
make the call whether they'll download and install the module dist.

I credit Python folks with more intelligence than Outlook users mindlessly
click on e-mail attachments, not really knowing what's hinding underneath.

-- 
Randall Hopper
aa8vb at yahoo.com



From alan.gauld at gssec.bt.co.uk  Thu Jun 29 09:31:15 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Thu, 29 Jun 2000 14:31:15 +0100
Subject: OO Approach to file parsing with Python?
References: <3958294F.4B17E6F7@attglobal.net>
Message-ID: <395B4FA3.ACB6D4DA@gssec.bt.co.uk>

Jon McLin wrote:
> Our company has decided to try an initial deployment of Python in a file-parsing
> role.  We need to translate some customer-provided structured ASCII files into
> our standard schema for use by legacy applications.    A procedural
> implementation is obvious, but completely non-reusable, and I'd like to be able
> to leverage this in the future.  (different inputs, the same outputs.)  Can
> someone point me to examples of OO approaches to this type of problem,
> preferrably implemented in Python?

I assume you've checked out the parser objects in the 
standard library? I've never used them coz the simple 
stuff I've done has been easy using regex's...

Alan G.


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


From alex at magenta.com  Mon Jun 12 07:42:16 2000
From: alex at magenta.com (Alex Martelli)
Date: Mon, 12 Jun 2000 13:42:16 +0200
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <8i2ift01vtm@news2.newsguy.com>

Steve Mullarkey <stevemul at ozemail.com.au> wrote in message
news:8F4E9C028stevemulozemailcomau at 139.130.250.4...
> I have been a 'C' programmer for the past 15 years and a "C++"
> programmer for the past 3 years. I'm not a genius but would modestly (?)
> class myself as reasonably competent.
>
> In the past week I have become aware of Python. I've bought a book and
> read this newsgroup every day.
>
> I have read in several places productivity claims of 5 to 10 times over
> 'C' and "C++". I would like to ask for some feedback from "C++"
> programmers who have moved to Python as to whether these estimates are
> realistic.

I have not "moved" from C++ to Python: most of my professional
development (what I get paid for) is still C++.  For personal
programming projects, such as my bridge studies, I've moved most
stuff to Python (everything that used to be in Perl, and a good
fraction of what used to be in C or C++).


> In "C++" I have :-
>
> 1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
> Profilers, Programmers Editor including Class Browser, etc..

My favourite editor has always been vi (now, and since a few years,
in the free gvim version), and it's just as good for Python as it is
for C++.  Otherwise, I find PythonWin and IDLE adequate substitutes
for Microsoft Visual Studio.  ObWarning: I don't do much GUI's, in
either language, so I have no experience of how the GUI builders
compare; Hammond and Robinson's "Python Programming on Win32" seems
to do an adequate job of presenting a huge variety of different
approaches you can use -- if I did have a huge investment in a GUI
builder I'm sure I could keep it (a GUI shell in VB, Delphi, or
whatever you prefer, with most or all of the real work in Python).


> 2. I use the STL (Standard Template Library) which gives powerful string
> and container classes.

I suspect that here we see the usual confusion between STL (the
prototype experimental research thingy which provided inspiration
for what is now in the C++ standard and has since kept changing)
and the Standard C++ Library.  Matters only to pedants like me, I
guess, but one of the little things giving me joy is that I just
barely managed to catch this confusion about to be propagated in
Eckel's otherwise-excellent "Thinking in C++" 2nd edition, and it
was in time to have him correct it (he's such a wonderful writer,
I hope he really gets to grips with the "Thinking in Python" book
he mentions as a possibility!).

Anyway, back to your point 2: Python's dictionaries, lists, and
strings, give you basically all you get from Standard C++'s
containers, and then some.  You may have to do some adaptation
for certain constructs (e.g., remap a std::multimap<> in terms
of a dictionary with lists as entries), and some of the best
features of C++'s <algorithm> are missing (std::next_permutation,
for example), but I notice you don't mention those so I guess
you won't miss them (there is sure to be something around the
net usable as a substitute, anyway:-).


> 3. I use incremental compiling and linking. This reduces wait times to
> very small amounts.
>
> 4. I can also use background compiling and linking.
>
> 5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
> etc..

No compiling and linking at all surely beats even those "very
small amounts", and the good stuff you can find on the net for
Python is a LOT.

> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated.

For many people, I suspect a key issue is that they never really
understood _100%_ of C++.  It IS a hugely complex language, and
an ENORMOUS investment is needed to fully grasp 90% of it; just
as much effort and expense, to move up to 99%; and as much again,
to get to 99.9%.  I'm about at this last level, so I'm RARELY bitten
by things I just didn't understand perfectly, but I'm unwilling
to do it all over again to get to 99.99% (particularly since, at
this level, existing implementations still deviate substantially
from the Standard).  But I also do consulting and mentoring, and
see presumably-experienced C++ programmers baffled by issues such
as
    MyClass my(5);   // declares a variable
    MyClass mx();    // declares a function, NOT a variable
or
    std::copy(a.begin(), a.end(), b.begin());    // does NOT extend b!
etc, etc, as well as 'deeper' issues such as partial template
specialization, order-of-initialization issues, dominance rules,
the distinctions between overriding, overloading, and hiding, &c.

Judging from my experience, over 90% of "experienced C++ programmers"
have a 90%-or-less grasp of the C++ language.  This is quite
understandable given its complexity and difficulty, but surely
explains why such programmers would be much more productive in
a language that is simple enough that they CAN grasp it well with
much more modest effort (and, mostly, carefully designed so that
not-fully-grasped issues don't rear up and bite you when you
least expect it:-).


However, avoiding C++'s traps and pitfalls is not really where the
big gains came for me, since I'm reasonably well-versed in avoiding
them (and have the scars to prove it:-).  Rather, for me personally
the key issues have to do with Python's intrinsically-dynamic nature.


Exploratory programming, ease of reuse, very easy refactoring, are
three key issues.  In C++, you have to start with a substantially
"right" class structure; you can do minor tweaks, but you can't really
go very far "just playing around with various possibilities"; and when
you see a refactoring would be warranted, it's most often a lot of
work.  Plus, to get any substantial reuse value out of your work, you
have to invest a lot in making it _reusable_ in the first place, or
in reworking it with reusability in mind -- very worthwhile investment,
but still hard to do under production deadlines.  Python's more fluid,
dynamic style helps on all of these fronts: you play around more, get
more possibilities explored, refactor more frequently and effectively,
and get reuse without huge up-front investments in reusability.


> I am genuinely interested
> to ask "C++" programmers :-
>
> 1. What productivity increase do you achieve ?

I've never measured it precisely, but it does lie somewhere
between a factor of 2 and one of 10.  It's probably less for
areas one knows so well (having written a dozen separate programs
or subsystems for them) that one could recode them in one's
sleep, while it tends to the higher end for areas where the
exploratory-programming and refactoring advantages loom large.

> 2. How long did you use Python before you achieved increased
> productivity ?

A couple of weeks (versus a decade of C++, and C before that).
One of the most amazing characteristics of Pythin is how
amazingly FAST one becomes productive with it.


Alex





From steffen.ries at sympatico.ca  Thu Jun 29 07:34:34 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Thu, 29 Jun 2000 11:34:34 GMT
Subject: re.split question
References: <m3k8fiosky.fsf@gondolin.beleriand> <8jdpg7$6ri$1@news.acns.nwu.edu>
Message-ID: <m3og4kd8l4.fsf@gondolin.beleriand>

mrw065 at merle.acns.nwu.edu (Mitchell R Whorlow) writes:

> How about using re.findall?

Hmm, hadn't thought about that one, but it looks like it gives me I
was expecting from re.split.

> >>> s = r"""id=joe, ou=org\, unit, o=some org"""
> >>> print s
> id=joe, ou=org\, unit, o=some org
> >>> patt = re.compile(r"""([^, ]+=[\w ]+(?:\\,)?[^,]+)""")

This regex finds only one \, the following does what I want:

patt = re.compile(r"([^, ]+=(?:(?:\\,)|[^,])+)")

> >>> patt.findall(s)
> ['id=joe', 'ou=org\\, unit', 'o=some org']
> 
> I think this does what you want, but taking into account the regex I used,
> you still may prefer your solution.

Actually, I found the proper method in ldapmodule, so I don't have to
mess around with regex...

thanks to everybody who helped.
/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From m.faassen at vet.uu.nl  Fri Jun  9 09:50:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 9 Jun 2000 13:50:04 GMT
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <8hqbvr$fgr$1@sunnews.cern.ch>
Message-ID: <8hqsmc$eek$1@newshost.accu.uu.nl>

Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote:
> stevemul at ozemail.com.au (Steve Mullarkey) writes:

>>I have read in several places productivity claims of 5 to 10 times over
>>'C' and "C++". I would like to ask for some feedback from "C++"
>>programmers who have moved to Python as to whether these estimates are
>>realistic.

> I'm a C++ programmer (leave out the quotes ;) ).

> First off, any claim of an order of magnitude increase in productivity is
> pure nonsense. See Brooks for an explanation or Boehms and McConnell for
> hard data.

That's likely true. I do suspect there are more anecdotal examples of
"I did the same project in Python and I finished in half the time!" compared
to C++ than the other way around, but these aren't very scientific. Usually
the person already had experience with the problem domain before reaching
for Python, for instance. Still, Python is far more dynamic than C++ and
probably fits certain problem domains more naturally. I have a hard
time seeing Zope in C++, for instance.

> <snip enumeration of tools>

>>I don't want to start a flame war but, given the above, I just can't see
>>where the productivity increase is generated. I am genuinely interested
>>to ask "C++" programmers :-

>>1. What productivity increase do you achieve ?

>>2. How long did you use Python before you achieved increased
>>productivity ?

> I want to use Python for scripting tasks. I dislike 'ordinary' scripting
> languages as they lack the ability to recover from errors; a problem that
> Python, being a real programming language, does not have.

> Up to now, I've been using C++ for scripting. I'm not proficient in Python
> yet, so writing a Python script takes far longer than writing a C++ utility
> and the Python code itself is larger than the equivalent C++ code.

But then you must be very good at C++. :) (Hi Wim!) My experience was that I
was faster than C++ after a couple of days already. Note that I haven't had
any real experience with post-STL C++, which may or may not matter.

When you write a class in C++ and you want to do it *well*, you have to
think about a horrible amount of stuff; constructor, destructor, copy
constructor, assignment operator, possibly other operators, public, private,
protected. Without the STL you'll have to do lots of manual memory 
management as well; reallocating arrays, etc. If you want to write a
generic class there's a whole set of extra things to consider.
I found that this took a long time in C++, while whipping up a decent
class in Python is a piece of cake in comparison. Even generic Python
classes.

It's perhaps just a difference in philosophies as well; C++ is a statically
typed language and its proponents put a lot of emphasis on this kind of
correctness. In Python you might come up with lots of correctness rules
as well, but people don't, as the whole approach is different. People
rarely care about things like public/private, even. This also means a lack
of getters/setters in many cases; those will only be created if they
are deemed necessary. This means encapsulation can be broken easily,
but Python programmers don't care about this so much -- in practice often
it is more of a problem to strictly preserve encapsulation (and type safety)
than it is to fix any problems when and if they arise.

> What I expect from Python, is that I can reuse scripting utilities,
> something that I do not do for C++ utilities. This could, perhaps, yield
> a 10% increase in productivity, but I'm not writing scripts all day, so
> my overall productivity increase will be negligible.

I think that in quite a few problem domains more increase of productivity
can be seen, but it's extremely hard to measure and probably also
depends on the individual.

> Most of all, however, I enjoy the incremental coding (I come originally
> from a Smalltalk background). I *like* Python, which is all excuse I need
> to use it.

Right -- it's just plain fun. In my opinion, it's more fun than C++, where
the fun is really in dealing with its the dark side. :) So in many cases, 
I can have more fun coding in Python than in C++, while my productivity is
at least not adversely affected (and in many cases I believe positively
affected). Since run-time speed often doesn't matter, this means I'd rather
use Python in many cases.

Regards,

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


From alan.gauld at gssec.bt.co.uk  Mon Jun 19 07:28:06 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 19 Jun 2000 12:28:06 +0100
Subject: python is hard to learn without c
References: <9yS15.2877$Gp6.138923@dfiatx1-snr1.gtei.net> <memo.20000615202832.52185B@digitig.compulink.co.uk> <1eca95m.7c2pcf1kp9klfN%bparsia@email.unc.edu>
Message-ID: <394E03C6.F77841A6@gssec.bt.co.uk>

Bijan Parsia wrote:
> Another point, there are many different things one might 
> expect out of a book entitled *Learning Python* 
>         1) Learning Python while/as a means to learning *programming 
>         2) Learning Python while/as a means to learning *OOP in
>         3) Learning Python as yet another oop language under your 

This is a very good point. In trying to find a title for 
my forthcoming book we have been round several loops 
trying to find a title which correctly implies its aims. 
Each batch of reviews has suggested new wording!

In my case we're aiming for (1) above and the current 
working title is:
"Learning to Program using Python"

It could still change but feedback as to whether that really 
implies what I want would be useful(by email so as not to 
bog down the group?)


> Books which do a good job of meeting an expectation which falls under
> one of these classes might well do a terrible job meeting other types.

Absolutely, which is why it's important to set the right 
expectation. My book does not teach all of Python, it 
doesn't even use the most efficient Python techniques 
for some things, rather it tries to teach techniques 
that are applicable in nearly all languages, with Python 
merely being a conveniently concise/simple demonstration tool.

Alan G.

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


From ivanlan at home.com  Tue Jun  6 12:22:12 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Tue, 06 Jun 2000 10:22:12 -0600
Subject: [XML-SIG] Pyexpat
References: <393C2455.1B9685B@home.com>
Message-ID: <393D2534.9EFE4763@home.com>

Hi All--

[cc'd to Python-List in order to get this into FAQTS]

Ivan Van Laningham wrote:
> 
> Hi All--
> I've "installed" the PyXml distribution (0.5.4) and downloaded pyxie
> (from Sean McGrath) which uses pyexpat.  Despite having the pyexpat.dll
> in my winnt directory, it refuses to import:
> 
> >>> import pyexpat
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: No module named pyexpat
> >>>
> 
> I know that when I get the answer to this, I'll be deeply embarassed
> since it is almost certainly something stupid that I'm doing/not doing
> (have we left anything out?), but I am prepared.
> 

Tom Passin replied, yesterday:

> Subject: Re: [XML-SIG] Pyexpat
> Date: Mon, 5 Jun 2000 22:10:16 -0400
> From: <tpassin at home.com>
> To: "Python xml-sig" <xml-sig at python.org>
> 
>  I had trouble getting pyexpat to import too, on Win95/98.  Right now I have
> it in both C:\Program FIles\python\dlls and C:\Program
> Files\Python\xml\parsers.  One of them works, but after all these months, I
> forget which one did the trick.  You need it in a python directory, rather
> than just a windows path directory, I think.

There were a couple of other suggestions, but they didn't work.  PyXml
installed into c:\Python\Lib\xml on my system, but failed to install
pyexpat.dll.  I copied it into c:\Python\Lib\xml\parsers, without
renaming it, and that fixed the problem.

The install should copy pyexpat.dll to the correct place, especially
since the install package comes with a pre-built one.  It should also be
in the FAQ, shouldn't it?

Thanks, everyone, for your help.

<luckily-i-didn't-*promise*-to-be-embarassed>-ly y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From bawolk at ucdavis.edu  Fri Jun  9 09:20:52 2000
From: bawolk at ucdavis.edu (Bruce Wolk)
Date: Fri, 09 Jun 2000 06:20:52 -0700
Subject: tkinter -- why doesn't this work?
References: <jl66rj3cqa.fsf@endor.bbb.caltech.edu>
Message-ID: <3940EF34.1A04405C@ucdavis.edu>

I'm a newbie too, but my guess is that Frame widgets do not receive key events.
If you use root.bind, it should work.

Bruce

Michael Vanier wrote:

> Hi,
>
> Here's a trivial tkinter script meant to catch and print key events in a
> frame.  It doesn't work on my system (RH Linux 6.1, pentium pro, python 1.6a2
> or python 1.5.2), but if I modify it to catch (say) button presses or
> anything else but key bindings it does work.  Any ideas?
>
> TIA,
>
> Mike
>
> --------------------------------------------------------------------------
> #! /usr/bin/env python
>
> from Tkinter import *
>
> root = Tk()
>
> def callback(event):
>     print dir(event)
>     print "key: %s" % event.char
>
> frame = Frame(root, width=200, height=200)
> frame.bind("<Key>", callback)
> frame.pack()
> root.mainloop()
> --------------------------------------------------------------------------
>
> --------------------------------------------------------------
> Mike Vanier     mvanier at bbb.caltech.edu
> Department of Computation and Neural Systems, Caltech 216-76
> GNU/Linux: We can't lose; we're on a mission from God.



From shouldbe at message.com  Fri Jun  2 09:09:29 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 2 Jun 2000 08:09:29 -0500
Subject: Escapes???
Message-ID: <8h87u1$sga$1@news2atm.raleigh.ibm.com>

I am trying to write this escape seq to a device from python
"H\x1b\x23\x12art\x1b\x23\x01\n"
The \x1b\x23\x12 is supposed to be a command tell the device to do something
special with "art", but python is making it look like this \x1b\x23\x12a
"rt".  Am I missing something on how to work with escape sequences?  Is
there another way to write hex values in a string.  I rather us \x00 because
that is what are testers are familiar with.

Thanks
jamarijr at hotmail.com




From dsavitsk at e-coli.net  Tue Jun 20 01:28:43 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Tue, 20 Jun 2000 05:28:43 GMT
Subject: Current date and time
References: <fOC35.10326$dF.445003@news1.rdc1.il.home.com> <20000620010356.A892@better.net>
Message-ID: <fiD35.10341$dF.446338@news1.rdc1.il.home.com>

thanks, i knew it couldn't be that hard ...

"William Park" <parkw at better.net> wrote in message
news:20000620010356.A892 at better.net...
> On Tue, Jun 20, 2000 at 04:54:35AM +0000, dsavitsk wrote:
> > what is the easiest way to get the current date and or time?
> > something like
> > >>> print Now()
> > is what i am hoping to find.
> > thanks
> > doug
>
> >>> import time
> >>> time.ctime(time.time())
> 'Tue Jun 20 01:02:03 2000'
>
> --William
>




From kens at sightreader.com  Fri Jun 16 01:23:36 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 22:23:36 -0700
Subject: Waffling between Python and Ruby
References: <FVN15.3523$227.83704@nnrp1.uunet.ca> <8i8bi2$1s8$1@news.wrc.xerox.com>
Message-ID: <3949B9D8.CC3EC5B0@sightreader.com>

It's happened to me many times.  Usually when cutting and pasting from other
peoples code (the other code having tabs, where my code does not).  Over the
past two years this has cost me at least 80 to 90 seconds of development time.
The solution is simple: make tabs illegal.  The biggest time saver, of course,
is that it would stop this utterly rediculous whitespace debate.

In any case, indentation/block rules don't seem to be the most important issue
when selecting a language for a task.

Mark Jackson wrote:

> "Warren Postma" <embed at NOSPAM.geocities.com> writes:
>
> > Having something
> > look correct on the screen, but not work because of misuse of tabs versus
> > spaces, or incorrect indentation, is something that still continues to bug
> > me.
>
> Indulge my curiosity - how many times has this actually happened in
> your six months of using Python?
>
> --
> Mark Jackson - http://www.alumni.caltech.edu/~mjackson
>         You can never solve all difficulties at once.
>                                 - Paul Dirac

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/acba56b0/attachment.vcf>

From dworkin at ccs.neu.edu  Wed Jun 28 23:58:49 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 28 Jun 2000 23:58:49 -0400
Subject: (Serious?) package namespace problem (and a proposal)
In-Reply-To: hzhu@knowledgetrack.com's message of "Wed, 28 Jun 2000 22:42:33 GMT"
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com>
	<vndwvj9bgzh.fsf@camelot-new.ccs.neu.edu>
	<slrn8ll00c.25h.hzhu@rocket.knowledgetrack.com>
Message-ID: <vnd7lb99lza.fsf@camelot-new.ccs.neu.edu>

hzhu at knowledgetrack.com (Huaiyu Zhu) writes:

> What it hurts is to import BBB/AAA.py instead of AAA.py.  I'd be interested
> in how do you "Don't do it".

My apologies.  I saw that you were doing something messy, and ended up
with a mess.  I responded without looking any deeper than that.

It seems that you're being bitten by two things.
 
One of them is case-insensitivity, which is broken but unavoidable
when on Windows.
 
The other is relative importing, which doesn't currently have any
user-modifiable way to specify that you really don't want to import
from the package directory.

> Now could any one that think the problem can go away suggest a solution that
> does not need to change file names and still works on Windows.
 
Sadly, I can't.  As far as I can see, you simply have to avoid naming
submodule files after top level modules that could possibly be
imported by anything in that package.  Feh.

-Justin

 



From pgf at globalreach.net  Sat Jun 10 02:50:12 2000
From: pgf at globalreach.net (Phil Fraering)
Date: 10 Jun 2000 01:50:12 -0500
Subject: Damnation! (was: Re: Python3k extended grammar)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <87n1kuowrf.fsf@globalreach.net>

Guido van Rossum <guido at python.org> writes:

> Python is not a democracy.  Voting doesn't help.  Crying may... :-)

What about large numbers of small unmarked bills with serial
numbers not in sequence?

-- 
Phil Fraering         "One day, Pinky, A MOUSE shall rule, and it is the
pgf at globalreach.net    humans who will be forced to endure these humiliating
/Will work for tape/   diversions!"
                      "You mean like Orlando, Brain?"


From darrell at dorb.com  Sat Jun  3 02:32:38 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 2 Jun 2000 23:32:38 -0700
Subject: Escapes???
References: <03de01bfcd24$6cc01e70$6401a8c0@home>
Message-ID: <041101bfcd25$b47ba7b0$6401a8c0@home>

Bug fixes :) 

hexData1='''
0016 4210 0100 0004 0000 0101 c042 0100
0242 0500 000f
'''
hexData2='''
0x0016 0x4210 0x0100 0x0004 0x0000 0x0101 0xc042 0x0100
0x0242 0x0500 0x000f
'''
hexData3=r'''
\x0016 \x4210 \x0100 \x0004 \x0000 \x0101 \xc042 \x0100
\x0242 \x0500 \x000f
'''
hexData4=r'''
\x00\x16 \x42\x10 \x01\x00 \x00\x04 \x00\x00 \x01\x01 \xc0\x42 \x01\x00
\x02\x42 \x05\x00 \x00\x0f
'''
hexData5=r'''
0x0016 \x42\x10 0100 \x0004 \x00\x00 \x01\x01 \xc0\x42 \x01\x00
\x02\x42 \x05\x00 \x00\x0f
'''

import re, string, struct
bigEndian={1:"c", 2:"c", 3:">h", 4:">h", 5:">l", 6:">l"}
def binConvert(s, sizeMap={1:"c", 2:"c", 3:"h", 4:"h", 5:"l", 6:"l"}):
    """
    Convert s to binary
    s can be formated like:
    0xabcd \x11 \x9999 abcd

    sizeMap maps the size of each number to a struct format string
    for it's conversion. bigEndian was used for regression so that
    r"\x00\x16" == r"\x0016"
    """
    res=[]

    temp = re.findall(r"(?i)[^ \\x\n\t]+(?!x)",s)
    fmt=sizeMap[len(temp[0])]
    lastLen=len(temp[0])
    for v in temp:
        if len(v) != lastLen:
            lastLen=len(v)
            fmt=sizeMap[lastLen]

        if fmt != 'c':
            res.append(struct.pack(fmt, eval("0x"+v)))
        else:
            res.append(struct.pack(fmt, eval(r'"\x%s"'%v)))

    return string.join(res,'')

def hexConvert(s, size=1, outPutStyle="0x%s "):
    """
    Convert binary string s to a hex format
    size        ==> how large should each hex value be
    outPutStyle ==> how should each value be formated
    """
    res=[]

    while s:
        temp=s[:size]
        s=s[size:]
        l=[]
        for i in temp:
            l.append("%x"%ord(i))
        val=string.join(l,'')
        res.append(outPutStyle%val)

    return string.join(res,'')

def regress(show=0):
    lastB=None
    for k,v in globals().items():
        if re.match("hexData", k):
            print 'Testing:', k
            b=binConvert(v, bigEndian)
            h=hexConvert(b)
            if show:
                print b
                print h
                print '-'*50
            assert(b==binConvert(h, bigEndian))
            if lastB:
                assert(b==lastB)
                lastB=b
            else:
                lastB=b

regress(1)



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000602/5d2a8330/attachment.html>

From darcy at vex.net  Fri Jun  2 08:28:47 2000
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 2 Jun 2000 12:28:47 GMT
Subject: SQL: don't use bind variables (was Re: Using strings with ' in them  in SQL-queries)
References: <39342055.1301760@news.c2i.net> <393510AB.D62AE9A5@infercor.no> <8h38ak$s1f$1@slb6.atl.mindspring.net> <3937864A.DB218804@infercor.no>
Message-ID: <8h899v$rch$2@news.tht.net>

Paul Boddie <paulb at infercor.no> wrote:
>   db.execute("insert into test (id, name) value (1, :var)", {"var" : var})

> (Now I look again, I don't know where the 'query' method came from, but I assume
> that the thread's initiator meant 'execute'.)

He's probably using PyGreSQL which has two distinct interfaces, PyGreSQL
Classic (or PyGres) and PyGreSQL DB-API which is in the 3.0 version
just released.  He is using the classic interface which, btw, I am not
planning to deprecate.  Both will continue to be available in PyGreSQL
as long as I am maintaining it.

There's-more-than-one-way-to-skin-a-dead-parrot-ly y'rs

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


From wjdandreta at worldnet.att.net  Tue Jun 20 20:56:16 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Wed, 21 Jun 2000 00:56:16 GMT
Subject: Enums and Python
Message-ID: <QoU35.15352$Xx5.796205@bgtnsc06-news.ops.worldnet.att.net>

Is there a Python type like the C enum type?

It can be very useful in making code more readable.

Bill




From cut_me_out at hotmail.com  Tue Jun 13 08:48:48 2000
From: cut_me_out at hotmail.com (Alex)
Date: 13 Jun 2000 08:48:48 -0400
Subject: Newbie questions
References: <q9abks07oq5q327ijvf6ejgf91kh5o5aci@4ax.com>
Message-ID: <etdn1kpn3v3.fsf@w20-575-109.mit.edu>

> pi = 3.141592654
> radius = input("Enter is the radius: ")
> circ = 2 * pi * radius
> print circ

Perhaps the third line should be 'circ = 2 * pi * float (radius)'

As it is, you are multiplying a float by a string, I think.  That should
be giving you a type error.

By the way, you can get pi from the math module:
import math
print math.pi

It's good not to have constants encoded directly in to your program, in
case they ever change. :)

Alex.


From cbbrowne at news.hex.net  Wed Jun 21 19:59:43 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Wed, 21 Jun 2000 23:59:43 GMT
Subject: Python complements C++ for productivity
References: <NEBBIOGBDMDPEANLFKGHKEGJCEAA.jschmitt@vmlabs.com> <395128A4.3D70A4E4@san.rr.com>
Message-ID: <PFc45.211883$MB.3849912@news6.giganews.com>

Centuries ago, Nostradamus foresaw a time when Courageous would say:
>John Schmitt wrote:
>> 
>> When I first opened Pythonwin, I used the Python/Perl phrasebook.  Very
>> useful to me.  I don't know the STL very well (I don't even know what the
>> SWIG is) so I couldn't write one, but how hard would be to write a Python
>> <-> C++/STL phrasebook?
>
>Well, while there is much more to the STL these days than
>in its early years, in many ways the STL is about containers,
>and iterating over them. While in C++, this is a complicated
>subject matter, in Python, it's not.
>
>There are really only "two and a half" container classes
>in Python:
>
>Dictionaries,
>Lists (which are analagous to C++ vectors),
>and Tuples (which are immutable lists).
>
>I haven't seen any of these phrase books, but Python is just
>soooo easy. There's really not much to it.

The appalling thing is that I can't really see all that much that you
gain for the added complexity.
-- 
cbbrowne at ntlug.org - <http://www.hex.net/~cbbrowne/lsf.html>
Rules of the Evil Overlord #4. "Shooting is not too good for my
enemies." <http://www.eviloverlord.com/>


From aahz at netcom.com  Thu Jun  1 17:44:50 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Jun 2000 21:44:50 GMT
Subject: Styling Temporary Code
References: <3936A9BB.FBEF9201@visionart.com>
Message-ID: <8h6lgi$8me$1@slb0.atl.mindspring.net>

In article <3936A9BB.FBEF9201 at visionart.com>,
Pete Shinners  <pete at visionart.com> wrote:
>
>one thing i do frequenctly in C is to place temporary
>"debug/doublecheck/sanity" type code while i'm developing
>it. i always put this code in with no indenting, so it's
>easy to spot, work with, and cleanup. it looks something
>like this...
>
>
>int myfunc(int nullargs)
>{
>    int newval = do_fancy_stuff(nullargs);
>cout << "newval = " << newval << endl;
>    return more_fancy(newval);
>}
>
>
>ok, in this 5 line example it doesn't really sparkly,
>but when i use it in big sections of code, it can't
>be beat. call it the mark of the beast, chase me with
>pitchforks, i simply love it and i don't want to change!

This is one of the few areas where I think Python probably could use
some work.  A fair number of people on this newsgroup say that you
should just use a decent editor.

One person already suggested that you use assert.  Note that in Python
1.5.2 and later, you can turn off assert by setting the PYTHONOPTIMIZE
variable in addition to using -O on the command line.  You can also
express more complicated code with

if __debug__:
  ....

Both assert and "if __debug__" get optimized out with optimization
turned on.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From dcalvelo at pharion.univ-lille2.fr  Tue Jun 27 13:45:29 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 27 Jun 2000 17:45:29 GMT
Subject: Lista de correo sobre Python
References: <2B1262E83448D211AE4B00A0C9D61B0301301248@msgeuro1.creo.be>
Message-ID: <8jap7p$u3e$2@netserv.univ-lille1.fr>

Pieter Claerhout <PClaerhout at creo.be> wrote:
: Can somebody help? I forgot my Spanish glasses...

I suppose that on the other you do speak Dutch :)
And anyway, it's about starting an eGroups spanish Python list.

: Pieter

: On Mon, Jun 26, 2000 at 09:47:13AM +0000, JD wrote:
:> Un foro sobre Python en:
:> python_dev-subscribe at egroups.com
:> 
:> Muchos usuarios dispuestos a resolver tus dudas, a ayudarte y deseando que
:> compartas tus conocimintos.

A forum about Python at [snip]
Many users willing to clear your doubts, help you and waiting for you to 
share your knowledge.

: ?Cual es la justificacion para comenzar una lista nueva en espagnol cuando
: ya existe una?

What reasons are there to start a new spanish list when one already exists?

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From balrog at webtv.com  Sat Jun 17 18:04:57 2000
From: balrog at webtv.com (balrog)
Date: Sat, 17 Jun 2000 22:04:57 GMT
Subject: I would have wanted to know about Ruby
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net> <20000617153652.A2727@dophnic.yi.org> <8igpab$7tf$1@slb2.atl.mindspring.net>
Message-ID: <394BF56B.A27234D9@webtv.com>

> >> Y'know, those of you in the Ruby world who make a point of going around
> >> trashing other languages are probably doing more to hurt the acceptance
> >> of Ruby over the long haul than anything else.  It's *rude*.  I was
> >> almost interested enough in Ruby at one point to check it out, but I
> >> have to say that the Attitude I've seen so far is even worse than the
> >> Perlers.

Well I won't tell you how to pick your languages.  

I would have been glad to know about Ruby when I was looking at Python, 
Tcl & Perl some time back.


From neelk at brick.cswv.com  Sat Jun 10 09:07:04 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 10 Jun 2000 13:07:04 GMT
Subject: wishlist Python3K: subclass ints
References: <Pine.GSO.4.10.10006091052200.21387-100000@sundial>
Message-ID: <slrn8k4hap.cna.neelk@brick.cswv.com>

Moshe Zadka <moshez at math.huji.ac.il> wrote:
> On 9 Jun 2000, Aahz Maruch wrote:
> 
> > Oh, I see.  Hmmmm....  That makes a lot of sense, yes.  I'm still having
> > a little trouble wrapping my head around how this works with
> > user-defined numeric types, but I'll let the people who are into
> > metaclass programming deal with it.  ;-)
> 
> Oh, it doesn't really. I mean, the best you can do (in a type/class
> dichotomy solved world), is inherit from "number", and change any of the
> methods: but that wouldn't help you at all to implement constructive
> reals the "right way".

Well, if we added multimethods, then it wouldn't be a problem (plug,
plug).
 

Neel


From abc at spud.dinkbrain.org  Sat Jun 17 12:20:41 2000
From: abc at spud.dinkbrain.org (Mahogany Rush)
Date: Sat, 17 Jun 2000 16:20:41 GMT
Subject: identifying operating system
References: <39499924.3BC84686@engineering.uiowa.edu>
Message-ID: <slrn8kn92f.es.abc@sparc5.localnet>

On Fri, 16 Jun 2000 02:50:44 GMT, Juerg Tschirren <jtschirr at engineering.
uiowa.edu> wrote:
 >> 
 >> Is there a way how I can identify the operating system my Python script
 >> is running on (e.g., Linux, Windows, Mac, ...)?
 >> 
 >> Juerg


Python 1.6a2 (#1, May 14 2000, 08:15:17)  [GCC 2.8.1] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.platform
'sunos5'





From frankc at colconsulting.com  Fri Jun 16 08:31:36 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 16 Jun 2000 08:31:36 -0400
Subject: Python Productivity over C++
References: <m130IXh-000wcEC@swing.co.at> <3948745B.941@esatclear.ie> <394875B7.CAAA3C1F@san.rr.com> <3948B766.4C256CAE@colconsulting.com> <39491494.671DC3D9@sightreader.com> <3949A6BE.81B500E1@colconsulting.com> <3949A423.13371179@sightreader.com>
Message-ID: <394A1E28.480F7319@colconsulting.com>

Ken Seehof wrote:
> 
> "Frank V. Castellucci" wrote:
> 
> > Ken Seehof wrote:
> > >
> > > "Frank V. Castellucci" wrote:
> > >
> > > > Courageous wrote:
> > > > >[snip]
> > > > > Shrug. If it weren't for execution speed, your C++ environment would
> > > > > be toast. Admit it. :)-
> > > >
> > > > While the last line seems rather toungue in cheek, there are other
> > > > things that C++ brings to the table such as encapsulation, access
> > > > control and type strictness.
> > >
> > > All three of these items were intentially left out of python because python
> > > would not benefit from them, though C++ does.
> > >
> > > "Frank V. Castellucci" might have written (but actually didn't):
> > > > A fish has certain definite advantages over a duck.  For example a fish has
> > > gills.
> >
> > I didn't so it came from elsewhere.
> 
> I made it up.  That was a deliberate witty attempt to put words in your mouth :-)

Well the words in my mouth did make me witty, when I tried it out on my
son he said "Your a dimwit", so find something better for me please as I
am loosing ground here.
 
> > > Encapsulation, access control and type strictness are often erroneously
> > > referred to as if they were features.  They aren't.  They are aspects of a
> > > design philosophy.  To make a system less error prone, constrain the programmer
> > > in order to prevent the programmer from writing bugs.  Of course, that's
> > > impossible, since it is always possible to write bugs anyway.  Because C/C++
> > > are extremely error-prone languages, they need stuff like encapsulation,
> > > access, control and type strictness.  Because python is less error prone, it
> > > doesn't need these "features".  Python is based on different philosophy: To
> > > make a system less error prone, make it simple.
> > >
> > > You have to ask, "what is the actual value of these features?"  The answer is
> > > that they prevent certain kinds of bugs from appearing.  Now ask, "what bugs
> > > would a C++ programmer encounter if encapsulation, access, control and type
> > > strictness were removed from C++."  Undoubtedly it would be a mess.  Get a
> > > picture in your mind of these bugs.  Now ask a python programmer how often
> > > these kinds of bugs come up.  Answer: "can't remember that ever happenning!"
> > >
> > > Look at the big picture.  Do encapsulation, access control and type strictness
> > > really make C++ really less error prone than python?
> >
> > Somewhere in your conversation you went from the aspects of a particular
> > language, while true they do reduce errors because the language is still
> > low level enought to let you get in trouble, to implying that they were
> > put in there to make a system less error prone. You then finish with
> > justifying a position based on this posit.
> >
> > I disagree with the postulation.
> 
> My point is that encapsulation, access control and type strictness are relevant in a
> comparison between C++ and python if they produce a specific advantage in C++ over
> python that would make C++ preferable for some task.  My assumption is that
> encapsulation, access control and type strictness produce robustness in C++, which
> would be relevant if C++ were more robust than python as a result, which it isn't.

I have found that encapsulation and type strictness are useful in class
libraries. If we consider invarient state in line with B. Meyer, I have
some insurance that the inner workings will continue to do so without
fear of change. And I know that the user can change this by changing the
code, but that is the same everywhere no? I enjoy and employ development
by contract, if you use the class then you agree to it's interface, and
you agree to use it as such.

But more to your point, I won't argue over robustness advantages because
the argument goes both ways and therefore cancels itself out. In my
dimwitted view anyway.
 
> > Before these articles of disagreement, C had (and still does)
> > structures. Systems built in C can be a mess, or not. That is based on
> > the talent and experience writing the code. Python does not prevent you
> > from writing "messy code", it just reduces the potential devastation
> > caused by messy code. Python has the ability to hide data, you just use
> > a different notation for the variable in the class. Python has the
> > ability for type strictness, but you have to code it by hand as it is
> > not a feature of the grammer.
> 
> Python doesn't /prevent/ messy code, but all other things being equal, it is easier
> to write python code that is less messy because it's a simpler language. Typical ways
> to be messy in C++ include: inconsistent choice of integer types and other basic
> types, poor indentation practices, excessive abstraction layers, class inheritance
> messiness, preprocessor messiness, etc.  Python has fewer basic types, no choice
> about indentation, less need for abstraction layers, cleaner inheritance system and
> less need for inheritance, and no need for a preprocessor.

I would say that the enumeration of what in C++ allows one to get messy
are vast. I would also say that experience has a tendency to cancel
these things out as "newbie" mistakes, or a poor design.

Like with any language, Python can be messy. There are not less ways to
be messy in Python, there are just different ways. As I am pretty much a
neophyte in the Python ways (but trying to come up to speed), a review
of any code I have written by someone like yourself would probably
elicit the statement "Well, thats a wee bit messy lad". 

> Python is less messy /and/ messiness is less devastating in python.
> 
> > My original response was to the (lite hearted) statement that the only
> > thing C++ brings to the table is execution speed. I will posit that
> > languages for a solution space should be used that appropriatley address
> > the design requirements. It ain't all C++, and it ain't all Python.
> 
> Agreed. So the question becomes: What are the relevant characteristics of the problem
> spaces and corresponding characteristics of the candidate languages?  The best answer
> I have so far is that python is a high level language and C++ is a low level
> language, and that the two languages work well together.

I would add to this that the protection the interpreter sandbox offers
has value, although the 'C/C++' or other language extensions are not
protected there. 

> > I also believe that the larger the project, the more diminished the
> > return on this particular measurement. Complexity takes it's toll on any
> > language.
> 
> In my experience, complexity takes less of a toll on python because it is a higher
> level language.  I find that when I write a module in python, I tend to make it
> behave as a component (i.e. having minimum dependency issues).  More often than not,
> these components can be separately posted on Parnassus and used (hopefully) in
> unrelated projects.  Component programming is extremely good for making robust large
> scale projects.

I would be curious to know how large scale Python applications fare
(defects per release, etc.) but I (in my newbie Python suit) am not yet
aware of the really big projects (Zope? - ignorance disclaimers apply).
 
> In C++, which I've been using for much longer, I tend not to componentize quite as
> well.  As a result, my code tends to hang together on a large scale in python quite a
> bit better than in C++.  In addition, the small scale advantages hold true for large
> scales projects as well.  So IMHO, based on experience, pythons productivity
> advantages relative to C++ tend to increase as the project gets larger and more
> complex.  The advantages of C++ tend to be more significant on a small scale.

Just remember you are unique, just like everyone else. <grin> (Now thats
witty, although I can't claim it).

But seriously, your habits are not neccessarily the standard are they?
You could do a better job with your C++ work, if you elect to.
 
> Specific features, such as the requirement that one use semicolons, don't amount to
> much unless they produce characteristics that differentiate between languages.  So
> the real question is, what characteristics does C++ bring to the table, and for what
> kind of jobs do these characteristics make C++ the better choice.  IMHO, C++ brings
> speed and legacy to the table, so I use C++ for high speed components within a python
> project, and for integration of legacy software.  Maybe it brings other things as
> well, but I don't know what they would be.

As I said earlier, I believe you can be more productive in Python
(although I acutally need to be so first before I commit totally to that
opinion). But I also know from experience that in addition to speed, C++
can be used in a very clean way to take advantages - To increase the
robustness of the deliverables. If done right, it reduces maintenace
cost, as well as more resiliant to change. I am not saying Python isn't,
just that C++ is also.

The net of all of this will be that there are different views for the
two from different people.
 
> The question of which language scales better is very important.  If C++ scales
> better, python should perhaps be used primarily as a scripting language within C++
> applications.  On the other hand if python scales better, C++ should primarily be
> used for high speed components within python applications.

I don't know if I posed it more as a question or a tenet of large
systems. See my large system question earlier in this missive.

-- 
Frank V. Castellucci


From dieter at handshake.de  Thu Jun  8 17:48:22 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 08 Jun 2000 23:48:22 +0200
Subject: Solaris 2.6 Build Fails
References: <393EACCE.495F8421@usa.alcatel.com>
Message-ID: <x7vgzjn8t5.fsf@lindm.dm>

Sara Fowler <Sara.Fowler at usa.alcatel.com> writes on Wed, 07 Jun 2000 15:13:02 -0500:
> I downloaded the source for the Python Interpreter from
> http://www.python.org/download/download_source.html.  When trying to
> build it on Solaris 2.6 (SunOS 5.6),
> using gcc for Solaris 2.6, I get the following errors:
> 
> ---------------------------------------------------------------------
> ../posixmodule.c: In function `posix_listdir':
> ../posixmodule.c:944: `DIR' undeclared (first use in this function)
> ../posixmodule.c:944: (Each undeclared identifier is reported only once
> ../posixmodule.c:944: for each function it appears in.)
> ../posixmodule.c:944: `dirp' undeclared (first use in this function)
> ../posixmodule.c:945: parse error before `struct'
> ../posixmodule.c:958: `ep' undeclared (first use in this function)
> make: *** [posixmodule.o] Error 1
> ----------------------------------------------------------------------
Did you run "./configure" before "make"?

We compiled Python on Solaris 2.4/5/6/7 without any problems.

"DIR" usually is defined in "sys/dir.h" or "ndir.h" or "sys/ndir.h".

If you did run "configure", then maybe you "gcc" is not
properly installed.


Dieter


From bingdo19 at earthlink.net  Fri Jun 16 10:20:51 2000
From: bingdo19 at earthlink.net (Bob Langdon)
Date: Fri, 16 Jun 2000 14:20:51 GMT
Subject: €€ Newbie question : time zone differences
Message-ID: <bingdo19-D78F49.10323816062000@news-server.twcny.rr.com>

Hi,

I've installed the Pythion interpreter in my hme directory (hosting co. doesn't have it).

My host is on the west cpoast but I'm on the east coast.

How do I adds +3 hours to the time without going through every script?

Do I need to reinstall the interpreter and configfure during the install?

All I can figure is that "import time" refers to the timemodule.c file.

Do I ediut that? if so, where?

TIA!

Bob Langdon
bingdo19 at earthlink.net


From ge at nowhere.none  Mon Jun 12 12:48:13 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 12 Jun 2000 16:48:13 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <Cl%05.2665$cQ.22917@news-server.bigpond.net.au>
Message-ID: <hv815.298$i06.89347@ptah.visi.com>

In article <Cl%05.2665$cQ.22917 at news-server.bigpond.net.au>, Neil Hodgson wrote:

> I think that was part of my problem with Perl. While I can use
> Unix, its at a low level of competence - I count myself lucky
> to survive a session with vi without pulling the power cord to
> stop the torture. Coming from an RSX/11M+, Mac, OS/2, Windows
> background I found the syntax choices in Perl, which probably
> appear obvious to native Unixers, bizarre and difficult to
> reason with. 

I've been using Unix for 15 years, and I too find Perl syntax
bizarre and difficult.

-- 
Grant Edwards                   grante             Yow!  Do you need
                                  at               any MOUTH-TO-MOUTH
                               visi.com            resuscitation?


From mhammond at skippinet.com.au  Sun Jun 18 00:08:42 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Sun, 18 Jun 2000 04:08:42 GMT
Subject: Can't find win32all-132
References: <Tvr25.13782$7o1.355124@news2.rdc1.on.home.com> <8ifann$pte$1@phys-ma.sol.co.uk>
Message-ID: <eXX25.1537$qz4.13098@news-server.bigpond.net.au>

"John Parkey" <jparkey at nospamrapidbs.com> wrote in message
news:8ifann$pte$1 at phys-ma.sol.co.uk...
> Tom,
>
> go to
>
> http://starship.python.net/crew/mhammond/downloads/win32all-132.exe

Thanks John - this reminds me that I forgot to mention I have update the
"new" starship, so everything should be back to normal.

Mark.





From speed at ?.com  Mon Jun 19 18:07:17 2000
From: speed at ?.com (Jake Speed)
Date: Mon, 19 Jun 2000 22:07:17 GMT
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news and  links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net> <394E96D2.5787F2A3@gmx.de>
Message-ID: <skt6clie7f158@corp.supernews.com>

Michael.Scharf at gmx.de (Michael Scharf) wrote in
<394E96D2.5787F2A3 at gmx.de>: 

>"Andrew M. Kuchling" wrote:
>>     In the same thread, David Bolen and Huaiyu Zhu write a script to
>>     verify that a class implements all the methods of a template
>>     class: 
>>         http://x53.deja.com/getdoc.xp?AN=635109618
>> 
>
>Trying to extract the code or even read the code in the daja.com
>representation seems impossible. Either I am an old-fashioned ascii
>formatting fetishist or just to stupid to figure out, how to display
>a deja article in simple plain ascii. Looking at the source of
>this article: the entire script is put into one line.....

It's at the bottom of the article,
 'View Original Usenet Format'

http://www.deja.com/getdoc.xp?AN=635109618&fmt=text

-Speed!


From shapr at uab.edu  Thu Jun  1 12:08:28 2000
From: shapr at uab.edu (Shae Erisson)
Date: Thu, 01 Jun 2000 16:08:28 GMT
Subject: "non-essential difficulty" - syntax highlighting?
References: <D15A35708943D3119BA70060979C3C663828D5@PETRIFIED> <8ge7ka$ouo$1@slb6.atl.mindspring.net> <39351648.ED1F1288@uab.edu> <8h37ve$djt$1@slb7.atl.mindspring.net> <39352F52.B5A5B462@uab.edu> <8h4egl$pkc$4@newshost.accu.uu.nl>
Message-ID: <393690B2.ED517A8C@uab.edu>

Martijn Faassen wrote:
> 
> Shae Erisson <shapr at uab.edu> wrote:
> [snip]
> > The only other way I can think of to get around the case problem and not
> > get rid of case sensitivity is to have an IDE that's smart enough to
> > figure out when a variable or name is being created versus when it's
> > being modified and change the highlighting accordingly.
> 
> What about simply imitating the various shells around (I think there are
> even Python shells that do this..I think even IDLE does this, partially),
> and do name completion? If I type in some characters and they happen to
> be the first characters of a name known to the system, it'll spell it
> out completely (and of course provide a list in case of ambiguity).
> 
> If you get the case wrong in the first few letters, nothing will happen
> (or the ambiguity detector will kick in and offer suggestions). If you
> get the case right and it's unique, it'll just fill in the rest of
> the word in the case spelling you want.
> 
> That way newbies can be taught about this feature and won't have as many
> problems hopefully, and experienced programmers can do whatever they prefer.

I have a vague idea that I could hack together a bearable incremental
parser and much superior syntax highlighter for a python buffer by doing
parser.suite(line) everytime you leave a line, and then sewing the
pieces together manually.

There would obviously be more to it than that, ie. you'd have to special
case things like if x: lines and some others, but I think it would work.

I was thinking of doing it after the Python Refactoring Browser is
finished, but if someone wants to join me in coding this up, I can
change priorities. I would guess it would take a month or two.

Any suggestions, comments or offers of help?
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From phd at phd.russ.ru  Tue Jun 27 03:52:36 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 27 Jun 2000 07:52:36 +0000 (GMT)
Subject: compilation
In-Reply-To: <8j84cf$6bj$1@bmerhc5e.ca.nortel.com>
Message-ID: <Pine.LNX.4.21.0006270751280.2223-100000@fep132.fep.ru>

On 26 Jun 2000, Michael P. Soulier wrote:
> 	Hey people. I tried the FAQ, and I can't find this. I'm a Perl 
> programmer exploring Python, and I'm wondering how to do a simple compile pass 
> on python source to check syntax. In Perl, I'd just do a perl -cw sourcefile
> and it would do a quick compile pass with warnings on. I know I can start the
> interpreter and import the file as a module, but that's not always desirable.
> Is there a simple command-line way to do this?

#! /bin/sh

if [ -z "$1" ]; then
   echo "Usage: compyle file.py..."
   exit 1
fi

TEMPLATE="from py_compile import compile; compile('"

for file in $*; do
   pgm=$TEMPLATE$file"')"
   python -c   "$pgm" || exit 1
   python -OOc "$pgm" || exit 1
done

   (I know I call python in the loop instead of looping inside python; I do
this by purpose).

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From emile at fenx.com  Wed Jun 14 08:02:22 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 14 Jun 2000 05:02:22 -0700
Subject: Case-sensitivity: why -- or why not? (was Re: Damnation!)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <87em66owf6.fsf@globalreach.net> <960673850.532727042@news.pacbell.net> <394768BA.1BCAD2B7@infercor.no>
Message-ID: <0df101bfd5f8$6c2c1fe0$1906a8c0@fc.fenx.com>

----- Original Message -----
From: Paul Boddie <paulb at infercor.no>

<snip>

> What are the problems with case insensitivity? Being aware
that some names which
> might look very different actually refer to the same
thing, because it is the
> "pattern" of case in these names which is the only thing
distinguishing them.
>

<snip>

> Paul


This past weekend, I inherited a web site project laid
out for NT using VBSript.  I re-hosted to linux with
Chili!Soft.  My major problem so far has been case
issues related to winxx/*nix file system.  In particular,
quoted references to files were inconsistent in the code.
AUTH.asp vs auth.asp vs AUTH.ASP. ;(  Windows liked this
just fine.  Linux of course didn't.  These problems
will continue to exist regardless of which way python
moves as long as the *nixen distinguish file names
based on case.


Emile van Sebille
emile at fenx.com
-------------------






From ssthapa at midway.uchicago.edu  Sat Jun 17 14:02:15 2000
From: ssthapa at midway.uchicago.edu (Suchandra Thapa)
Date: Sat, 17 Jun 2000 18:02:15 GMT
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>
Message-ID: <slrn8kmc4l.o8h.ssthapa@ntcs-ip48.uchicago.edu>

John Schmitt <jschmitt at vmlabs.com> wrote:
>Hi.
>
>I'd love to take the diversion of writing Make using Python.  The things
>that are keeping me from doing this are, in order:
>
>In a smaller way, I would like to have Bash, sed, and grep as Python
>programs rather than compiled C.

	I think the software carpentry project is funding a set
of tools to replace make/autoconf/misc. dev tools in python.  The
proposals that they've received and the winning proposals are on 
their web page (unfortunately I don't have the address but a 
search should turn it up).


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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From jack at oratrix.nl  Fri Jun  9 17:38:09 2000
From: jack at oratrix.nl (Jack Jansen)
Date: Fri, 09 Jun 2000 23:38:09 +0200
Subject: Stackless Python for the Mac? 
In-Reply-To: Message by Jeff Senn <senn@maya.com> ,
	     09 Jun 2000 15:10:52 -0400 , <zoouk6v7.fsf_-_@SNIPE.maya.com> 
Message-ID: <20000609213814.01734E2673@oratrix.oratrix.nl>

Recently, Jeff Senn <senn at maya.com> said:
> 
> Has anyone contemplated (or started...or finished) a port of stackless
> to the Mac?  Other than syncronizing the sources, there shouldn't be
> any complications, right?

There may be more complications, as the MacPython mainloop is a bit
funny because of the frequent calls to WaitMextEvent to give cpu-time
to other programs (MacOS is still in the basis non-preemptively
scheduled).

I've cc'd Just on this message, because he did a port of microthreads
to macpython, and I seem to vaguely remember that this was related to
stackless python. But I could be all wrong, i didn't really have the
time to follow what happened. Just?
--
Jack Jansen             | ++++ stop the execution of Mumia Abu-Jamal ++++
Jack.Jansen at oratrix.com | ++++ if you agree copy these lines to your sig ++++
www.oratrix.nl/~jack    | see http://www.xs4all.nl/~tank/spg-l/sigaction.htm 



From moeller at dophnic.yi.org  Sat Jun 17 16:36:52 2000
From: moeller at dophnic.yi.org (Derek Moeller)
Date: Sat, 17 Jun 2000 15:36:52 -0500
Subject: Python 3000 -> No, but there's 'Ruby'
In-Reply-To: <8igkq3$v9l$1@slb7.atl.mindspring.net>; from aahz@netcom.com on Sat, Jun 17, 2000 at 07:50:27PM +0000
References: <394c2b15.634411296@nntp.interaccess.com> <394BB8DE.AE1D6A7A@webtv.com> <8igkq3$v9l$1@slb7.atl.mindspring.net>
Message-ID: <20000617153652.A2727@dophnic.yi.org>

On Sat, Jun 17, 2000 at 07:50:27PM +0000, Aahz Maruch wrote:
> [x-posted to comp.lang.ruby, followups to comp.lang.ruby]
> [posted & e-mailed]
> 
> In article <394BB8DE.AE1D6A7A at webtv.com>, balrog  <balrog at webtv.com> wrote:
> >
> >Half-seriously, Ruby is nicer looking & acting;  it's part Python,
> >part Eiffel / Sather, with the quick & easy string handling of Perl.
> 
> This is mainly addressed to comp.lang.ruby; I'm just including
> comp.lang.python because that's where this started:
> 
> Y'know, those of you in the Ruby world who make a point of going around
> trashing other languages are probably doing more to hurt the acceptance
> of Ruby over the long haul than anything else.  It's *rude*.  I was
> almost interested enough in Ruby at one point to check it out, but I
> have to say that the Attitude I've seen so far is even worse than the
> Perlers.
> 
> It'd be real nice if the level-headed Ruby people restrained the
> hotheads.

We seem to harp on the shortcomings of C, C++, Perl, Ruby, and a few other
languages in c.l.py. Seeing as how this fellow mentioned some things that
he felt he didn't like about Python, and what he did like about Ruby, I can't
see the large difference between that and Pythoners spieling about their
increased productivity compared to C++/C, the hairiness of Perl, whatever Ruby's
got against it, etc.

Or is what we say the truth, and what this man says merely befuddled and perhaps
angered myth? It isn't good to hold double standards.

Now, I can't say I agree with posting that bit about Ruby as a follow-up to
mentioning Python-3k, as this was not what the original poster asked for.

-- 
Derek Moeller

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \



From thomas at xs4all.nl  Sun Jun 25 13:08:21 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 25 Jun 2000 17:08:21 GMT
Subject: Reloading modules
References: <8j5atk$ijs$1@newsg4.svr.pol.co.uk>
Message-ID: <slrn8lcf3p.2g7.thomas@centurion.xs4all.nl>

On Sun, 25 Jun 2000 17:09:36 +0100, Duncan Smith
<buzzard at urubu.freeserve.co.uk> wrote:
>>>> from Graph import *

>>>> reload(Graph)
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>TypeError: reload() argument must be module

>What do I need to do to successfully reload a module?  I'm currently wasting
>a lot of time re-starting PythonWin each time I want to test a changed
>module and I can't find the answer.  Thanks

You do want reload(), but you have to give it a module object (not the name,
but the imported object.) 'from Graph import *' doesn't bind the module
object to a local name, but apparently something else in your module is
named Graph, and that object isn't a module.

If you *have* to use both reload() and from Graph import *, do something
like this:

import sys
from Graph import *
...
...
reload(sys.modules['Graph'])
from Graph import *
...
...

The 'import *' after the reload is necessary to re-bind all local names from
the Graph module to the new values in the reloaded Graph module. If you
leave it out, the reload() is pretty useless. The sys.modules trick avoids
having to import Graph seperately, and getting it overwritten by
Graph.Graph.

But personally, I'd myself a lot of pain and not use 'import *'. It bites in
more way than one ;-)

Just-say-no-ly y'rs,
Thomas.


From wjdandreta at worldnet.att.net  Sun Jun 18 23:22:06 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Mon, 19 Jun 2000 03:22:06 GMT
Subject: Why is Python so slow ?- revisited.
References: <CC59868A0A76D311B1F50090277B321C1F33C6@VRTMAIL>
Message-ID: <ylg35.9510$C44.538934@bgtnsc05-news.ops.worldnet.att.net>

Mike Fletcher wrote in message ...
>
>Also, don't discount the expense of an extra function call.  Function calls
>were (and still are to a lesser extent) extremely heavy in Python five
years
>ago.


I did discount it because in assembly language a function call is little
more than a jump. I thought that even if it were extremely inefficiently
implimented in Python, it couldn't reduce speed by a factor of 6. In
addition, there are 3 function calls in each replace. Logic says eliminating
one improve speed by 1/3.

But on second thought, the one I eliminated was a call to a python.py
library module, the other 2 function calls are to functions in strop, a
compiled C 'module'. I'll have to test it to be sure.

Bill




From vigenereNOviSPAM at leland.stanford.edu.invalid  Mon Jun  5 05:01:39 2000
From: vigenereNOviSPAM at leland.stanford.edu.invalid (cheesecake)
Date: Mon, 05 Jun 2000 02:01:39 -0700
Subject: Python GUI requires tcl?
References: <8hbeol$obo$1@samba.rahul.net> <8hbjl0$611$1@news1.xs4all.nl> <393968C8.CBFAAF1@bogusaddress.com> <3939817F.3FDBD89B@uab.edu>
Message-ID: <2bb81b1b.87acce59@usw-ex0103-019.remarq.com>

I've found that Tkinter is pretty good for GUI. But under
PythonWin, GUI Apps don't quit properly. And under the command
line Python interpreter, I get the error message when
_tkinter.dll tries to load, although the exception is caught and
fixed with FixTk.py (which I changed to fit my installation).

I feel like there's currently a shortage of widgets, but people
do put their own on the Internet, so the number is growing.

TD

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



From spamfranke at bigfoot.de  Sun Jun 18 17:22:36 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Sun, 18 Jun 2000 21:22:36 GMT
Subject: Function that knows the names of its actual parameters
References: <394a6395.7217524@news.btx.dtag.de> <jwbnews-9D73CE.13205916062000@news.olympus.net>
Message-ID: <394d399b.4366388@news.btx.dtag.de>

On Fri, 16 Jun 2000 13:20:59 -0700, "John W. Baxter" <jwbnews at scandaroon.com> wrote:

>I would think that this method should point the way (you can have some 
>positional arguments as well, along with named keyword ones, plus the 
>catchall):
>
>>>> def m(**args):
>...   print args
>... 
>>>> m(a=5, b='ten')
>{'b': 'ten', 'a': 5}
>

Huh? Are we talking about the same thing? Keyword arguments either overwrite
existing ones or supply new ones along with the call. What I am looking for is
more or less (and arguably sensible) syntactic sugar for building a subdict
of locals() given non-quoted variable names:

magic(a, b, ...) <-> {"a": locals("a"), "b": locals("b"), ...}

This is obviously not meaningful for expressions as arguments. I don't making
this look like a function call is a good thing, but I do think the functionality is
quite useful sometimes (Python, HTML, SQL, etc. generation...).


even-though-I-don't-know-Perl-very-much-I'm-sure-it-has-a-special-syntax-for-that-kind-of-thing-ly y'rs
Stefan





From moshez at math.huji.ac.il  Wed Jun 14 01:39:35 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 14 Jun 2000 08:39:35 +0300 (IDT)
Subject: Two newbie questions
In-Reply-To: <39468B55.FE9488D5@concentric.net>
Message-ID: <Pine.GSO.4.10.10006140839080.25977-100000@sundial>

On Tue, 13 Jun 2000, Manus Hand wrote:

> Following my own post.  How gauche.
> 
> Anyway, a slightly more legible alternative to get a random number
> between -sys.maxint and sys.maxint:
> 
> import random
> yourNumber = random.randrange(sys.maxint) * random.choice([1, -1])

Note that this is wrong: the chance to get 0 is twice the chance to get 1.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From sdhyok at email.unc.edu  Wed Jun 14 16:32:29 2000
From: sdhyok at email.unc.edu (Daehyok Shin)
Date: Wed, 14 Jun 2000 16:32:29 -0400
Subject: Numeric python
References: <Pine.LNX.4.10.10006141123440.26946-100000@intel3.wag.caltech.edu>
Message-ID: <8i8q4s$g8t$1@news2.isis.unc.edu>

Try www.sourceforge.com.

"Michael Todd Feldmann" <feldmann at wag.caltech.edu> wrote in message
news:Pine.LNX.4.10.10006141123440.26946-100000 at intel3.wag.caltech.edu...
> Could anyone point me in the right direction to get Numeric python?  I
> have a linux box and tried
>
> http://www.python.org/download/download_linux.html
>
> and went to
>
> http://andrich.net/python/
>
> which states that "This site is closed." since things may be outdated.  I
> am guessing there is somewhere else to find this and other RPM files.
>
> Thanks a bunch!
>
> Mike
>
> --------------------------------------
> Michael Todd Feldmann
> Beckman Institute (139-74)
> California Institute of Technology
> Pasadena, California 91125 USA
> Phone:(626)395-2722, FAX:(626)585-0918
> email: feldmann at wag.caltech.edu
> --------------------------------------
>
>
>






From moshez at math.huji.ac.il  Fri Jun  9 04:17:10 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 11:17:10 +0300 (IDT)
Subject: Why is Python so slow?
In-Reply-To: <r7%%4.1754$iy.137191@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <Pine.GSO.4.10.10006091112080.21387-100000@sundial>

On Fri, 9 Jun 2000, William Dandreta wrote:

> The Python version was 1.2.

Ummmmm.....I'm not sure when 1.2 was current, but it was probably more
then 3 years ago. How about benchmarking vs. something a little more
recent?

> Why does it take Python 50 times longer than C?

If you code C in Python, you'll get 50 times difference. If you code good
Python in Python, you can reach speeds which are quite close for text
analysis.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From thor at localhost.localdomain  Wed Jun  7 12:14:09 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 7 Jun 2000 16:14:09 GMT
Subject: Dutch.comp.python.org (was : Re: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com)
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net>
Message-ID: <slrn8jt4ej.64h.thor@localhost.localdomain>

On 7 Jun 2000 13:50:50 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>>>
>>> Da's goed, maar het is echt nogal maf dat zoveel mensen op deze
>>> nieuwsgroep Nederlands blijken te spreken. Hoe komt dat toch?
>>
>>*Blijken* te spreken. Misschien zijn we allemal bots die van een
>>enkele (nederlands) programmeur komen. Of misschien is er maar een
>>enkele mens in deze nieuwsgroep, die honderde adressen gebruikt om
>>de indruk te wekken dat Python heel populair is. Of misschien is
>>er een vertalingsprogramma dat alle teksten gewoon naar het Engels
>>vertaald, maar af en toe naar het Nederlands.
>>
>>Er zijn dus merdere rationale verklaringen, dus geen behoefte om
>>aan kosmische krachten te geloven!
>
>Okay, okay, a joke's a joke, but if people are actually going to write
>full paragraphs in Dutch, it'd be nice if someone wrote translations for
>the small minority of us who don't know Dutch.
>--

No, if every ones speak Dutch here, then why not : 
Dutch.comp.python.org 

or even

HowILoveDutch.comp.python.org

And now my own rant in Spanish:

Python era un lenguaje casi perfecto, al menos hasta que empezaron
a meterle tonterias sintacticas. Ahora estoy empezando a ver cual
es el problema de python: los holandeses. Al parecer estan tan
excitados con la idea de python que no pueden evitar soltar la
parida en Holandes. La verdad es que los holandeses son muy imaginativos:
- Las cafeterias de drogas
- El trafico de diamantes ( De Beers estara detras de Sierra Leona)
- Tomates trangenicos. 
...

Bueno, ya en serio, si realmente no se pueden aguantar, que 
creen un Dutch.comp.python.org. 

-- 
MGA


From richard_chamberlain at ntlworld.com  Sat Jun 10 03:57:40 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sat, 10 Jun 2000 08:57:40 +0100
Subject: Tkinter: treeview widget
References: <393E3782.FF7A5DCA@proceryon.at>
Message-ID: <Lpm05.3258$jX6.60871@news2-win.server.ntlworld.com>

A couple more:

You could also look at:

http://www.magicnet.net/~gcash/tree.html
or
http://www2.zyvex.com/OpenChem/index.htm

I also notice that Pmw is threatening to have one in the Todo list, so
crossed fingers.

Richard

Horst Gassner <horst at proceryon.at> wrote in message
news:393E3782.FF7A5DCA at proceryon.at...
> Hi all!
>
> I am looking for a treeview widget. Does anyone know if such a widget is
> available somewhere?
> At the moment I am using PMW which has a very good collection of meta
> widgets but unfortunately it does not contain a treeview widget.
>
> Thanx
> - Horst




From pj at sgi.com  Mon Jun 12 15:55:59 2000
From: pj at sgi.com (Paul Jackson)
Date: 12 Jun 2000 19:55:59 GMT
Subject: spaces in filenames
References: <000201bfd350$0d818660$f6a2143f@tim>
Message-ID: <8i3f8f$lb01s$1@fido.engr.sgi.com>

So, Tim, you sister has the soul of an artist, but is:
  1) stuck in some bureaucracy
  2) with you for a brother.
Pray tell, what horror did she commit in her previous
life to warrant this?  <g>

-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From david_ullrich at my-deja.com  Thu Jun 29 11:11:48 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Thu, 29 Jun 2000 15:11:48 GMT
Subject: what is lambda?
References: <8jd8uu$gnh$1@nnrp1.deja.com> <8jdaht$i43$1@nnrp1.deja.com> <erq65.11249$227.232310@nnrp1.uunet.ca>
Message-ID: <8jfov2$ckr$1@nnrp1.deja.com>

In article <erq65.11249$227.232310 at nnrp1.uunet.ca>,
  "Warren Postma" <embed at NOSPAM.geocities.com> wrote:
>
> >    You should note there's varying opinions on whether
> > lambda is a good thing - if you're a beginner at programming
> > as well as with Python it's not the first thing you should
> > worry about.
>
> It also (in my case) expanded the vocabulary I use internally (my own
> pseudo-code-language) to include concepts that are not easily
expressable in
> less dynamic languages. In other words, lambda was a kind of
> who's-programming-who keyword. Rather than expanding only the
language, it
> also opened up my brain a bit wider.

     Oh, I feel the same way, seeing a lambda something was
very, I can't come up with exactly the right word but learning
how it worked was _definitely_ <right word> for me as well.
The idea that we can define a function as the value of an
expression (ie without a function definition) is not only keen,
it also serves to clarify a little just what a function _is_.

DU

> Warren
>
>


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


From elengvenis at uswest.net  Tue Jun  6 19:28:21 2000
From: elengvenis at uswest.net (Eric Lengvenis)
Date: Tue, 6 Jun 2000 18:28:21 -0500
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee> <393bc9f8.3018710@news.iol.ie>
Message-ID: <9Qf%4.520$t56.6546@news.uswest.net>

I


From jparkey at nospamrapidbs.com  Sat Jun 17 03:49:29 2000
From: jparkey at nospamrapidbs.com (John Parkey)
Date: Sat, 17 Jun 2000 08:49:29 +0100
Subject: Can't find win32all-132
References: <Tvr25.13782$7o1.355124@news2.rdc1.on.home.com>
Message-ID: <8ifann$pte$1@phys-ma.sol.co.uk>

Tom,

go to

http://starship.python.net/crew/mhammond/downloads/win32all-132.exe

jp

"Tom" <tom-main at REMOVEME.home.com> wrote in message
news:Tvr25.13782$7o1.355124 at news2.rdc1.on.home.com...
> A week or two ago Mark Hammond announced that win32all-132 (based on
1.5.2)
> was now available to all.  But I can't find it.  Anywhere.
>
> Anyone else?
>
> Tom.
>
>




From mal at lemburg.com  Fri Jun  9 11:57:38 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 09 Jun 2000 17:57:38 +0200
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4>
Message-ID: <394113F2.8D6D0833@lemburg.com>

Steve Mullarkey wrote:
> 
> I have been a 'C' programmer for the past 15 years and a "C++"
> programmer for the past 3 years. I'm not a genius but would modestly (?)
> class myself as reasonably competent.
> 
> In the past week I have become aware of Python. I've bought a book and
> read this newsgroup every day.
> 
> I have read in several places productivity claims of 5 to 10 times over
> 'C' and "C++". I would like to ask for some feedback from "C++"
> programmers who have moved to Python as to whether these estimates are
> realistic.

Let's see...
 
> In "C++" I have :-
> 
> 1. Good tools, Borland C++ Builder, MS VC++, Memory Leak Tools,
> Profilers, Programmers Editor including Class Browser, etc..

IDEs:
* Python comes with an IDLE
* PythonWare sells a brand new product called PythonWorks which
* PythonWin is another Win32 based product

Memory Leaks:
* These can only occur if you are using cyclic references; Python1.6
  will probably have cycle GC, so starting with 1.6 leaks are
  probably only in C code.
 
Profilers:
* Python comes with a profiler.

> 2. I use the STL (Standard Template Library) which gives powerful string
> and container classes.

Python doesn't really have a need for these, since coding
polymorphic code which only relies on available interfaces
is no problem at all (Python uses name binding, not
static typing as C++).

> 3. I use incremental compiling and linking. This reduces wait times to
> very small amounts.
> 
> 4. I can also use background compiling and linking.

Python compiles on-the-fly at run-time and only if needed.
You can even program interactively and have the program
compile its own code.

The typical C cycle editor-compiler-linker-debugger boils
down to a editor-run cycle -- Python provides run-time
exceptions which show you exactly where errors occurred.
 
> 5. I have 3rd party libraries to handle Regular Expressions, Date/Time,
> etc..

Regular Expressions are included in Python. For Date/Time
there is a 3rd party extension called mxDateTime (see my
Python Pages), for other things take a look at the 
Parnassus Vaults (via www.python.org).

Best of all: Python comes with batteries included (tm)
Just scan the library documentation at www.python.org to
get an impression of what is there.
 
> I don't want to start a flame war but, given the above, I just can't see
> where the productivity increase is generated.

Should be obvious: Python reduces dev-time, enhances
productivity, reduces LOCs, scales well (from 5-100k LOCs
if needed -- which is roughly equivalent to 25-250k LOCs
in C++ from my own experience).

Besides, you can continue programming in C++ and then
glue everything together using Python. It offers great
extension and embedding capabilities. I have turned to
writing all object oriented code in Python and only
resort to C for low-level high performance routines.
Works well :-)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From shouldbe at message.com  Fri Jun 23 08:47:22 2000
From: shouldbe at message.com (Arinté)
Date: Fri, 23 Jun 2000 12:47:22 GMT
Subject: strings and integers?
References: <8ir75b$qfi$1@news2atm.raleigh.ibm.com> <394FF466.51A5B77B@sightreader.com> <8it0j0$1du6$1@news2atm.raleigh.ibm.com> <S_s45.81332$WS3.720017@typhoon.we.rr.com>
Message-ID: <u%I45.23523$ds.686205@newsread2.prod.itd.earthlink.net>

Well I just tried str() an it works with no errors (so far;-)).  This is
where python is embedded in the app, so I import 3 things:  sys,poss,and
PossDevice.  poss is the C++ application, PossDevice is another layer
between the C++ app and the running scripts.  PossDevice also imports poss
and sys.

regards




From nascheme at enme.ucalgary.ca  Sat Jun  3 01:04:01 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: 3 Jun 2000 05:04:01 -0000
Subject: Copyright changes with move from CNRI to BeOpen?
In-Reply-To: <000601bfcd16$7308e2c0$b42d153f@tim>
References: <000601bfcd16$7308e2c0$b42d153f@tim>
Message-ID: <20000603050401.12049.qmail@cranky.arctrix.com>

In comp.lang.python, you wrote:
>[Harry Tanovich]
>> Will the copyright move to a more GPL style thing now?
>
>The details are still being worked out, but I'll tell you this:  if BeOpen
>had showed any sign of wanting to make the Python license materially more
>restrictive, I would not have accepted the job there (GPL is fine for many
>things, but Python isn't one of them).  To the contrary, the BeOpen
>PythonLabs team would like to *reduce* the legal folderol that accreted
>under CNRI's tenure.  However, at this precise moment in time, I'm
>unemployed, so don't take this as an Official Statement from anyone -- least
>of all me <wink>.
>
>if-i-wanted-to-speak-for-myself-i'd-hire-me-ly y'rs  - tim
>
>
>PS:  Note that the JPython license has always been stiffer than CPython's,
>     and that may not be changeable.
>
>
>


-- 
"All truth passes through three stages: first, it is ridiculed; next it is
violently attacked; finally, it is held to be self-evident." -- Schopenhauer



From david_ullrich at my-deja.com  Sun Jun 11 14:25:13 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Sun, 11 Jun 2000 18:25:13 GMT
Subject: No strptime in Windows?
References: <8hu4tc$6m0$1@nnrp1.deja.com>
Message-ID: <8i0lhu$ped$1@nnrp1.deja.com>

      Last night I saw two replies to this which are
not yet visible here at the office due to mysteries
of usenet and deja.com. Thanks, but I'm still puzzled -
if anyone can figure out what I'm misunderstanding I'd
appreciate it:

       Both replies explained that strptime must not
be available on the Windows platform. I surmised as
much. But the problem is I'm not aware that there's
anything like strftime() in Windows either. (I
certainly could be wrong about that. But eg the
Delphi DateToStr function doesn't take advantage of
any Windows functions, it does it all by hand in
Pascal code...)

       Maybe I'm not clear on exactly what "the
platform" is. I've taken that to mean "Windows".
But maybe when people say I'm using the Windows
platform that's shorthand for saying the platform
I'm using is "Windows C"? (No, that can't be it,
because surely there's no such thing as "Windows
C".) Or the "platform" is the specific C compiler
that was used? (No, that can't be it either...)

     I'm so confused. You'd think I'd be used to
that by now.

      Reminds me of when I'd hear VB guys wonder
why they couldn't do something, saying that they
could set the pen width to 0 in C. I didn't follow
that, seemed to me that there couldn't possibly be
a notion of "pen width" in C. Years later I realized
that what they actually meant was something about
the pen width in some specific C implementation.
If anyone has an analogous clarification here that
would be great - exactly what is it that's missing
strptime, so that my Python doesn't have one available
to wrap?

    (When I asked the question I was assuming that
most of the time.this/that functions were done "by
hand" in the Windows version of Python.)

DU

In article <8hu4tc$6m0$1 at nnrp1.deja.com>,
  david_ullrich at my-deja.com wrote:
>     The docs say that time.strptime may not be available
> on all platforms. It doesn't seem to be available under MS
> Windows. I don't see why not - there's nothing Windological
> about strptime but the same applies to the other time
> functions that _are_ included.
>
>     (I'm getting the time from somewhere on the internet.
> What I get is something like "Sat Jun 10 18:34:10 2000"
> and I have to subtract five hours. Ended up more or less
> writing my own strptime, at least I _think_ I did...)
>
>     Also: Is Monday really day 0? I woulda thought
> Sunday was day 0. (Doesn't seem to matter in the
> subtract-five-hours thing, I get the same results
> either way. Evidently mktime() ignores the day-of-week
> if it's inconsistent with the day-of-month field?)
>
> DU.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From olczyk at interaccess.com  Fri Jun  9 14:28:38 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Fri, 09 Jun 2000 18:28:38 GMT
Subject: Problem with script and typing
References: <Pine.GSO.4.10.10006092056410.28247-100000@sundial>
Message-ID: <394735dc.93980406@nntp.interaccess.com>

On Fri, 9 Jun 2000 20:59:48 +0300 (IDT), Moshe Zadka
<moshez at math.huji.ac.il> wrote:

>On Fri, 9 Jun 2000, Thaddeus L. Olczyk wrote:
>
>> 
>> Here goes ( extraneous pruned ):
>> 
>> project =[]
>> 
>> class Project:
>>        def IsNull(self):
>> 	return self.name==""
>> 	project.append(self)
>        ^^^^^ This line is never reached ???????
>
A remnant of cut and paste it was never meant to be there.
>> def FindProject(x):
>>     for project in projects:
>>         if project.name == x:
>>             return x
>>         null=Project("")
>>         projects.remove(null)
>          ^^^^^ projects are compared by identity, unless you supply an
>          __cmp__ method
>
I'm not sure exactly what you are trying to say here. The Project
constructor is supposed to insert self into the list. So if you are
trying to say the null may not be removed , I don't think so because
they are the same instance. If you mean that it would be easier to
put
           def __cmp__method(self,rhs):
                 return self.name == rhs.name
in the Projects clas, you are right. I didn't know.

>> def is_project(s):
>>     if s[len(s)-3:] == 'prj':
>>         project=FindProject(s)
>> ## error here
>>         if project.IsNull():
>>             project=Project(s)
>>         project.Scan()
>> ----------
>> error message AttributeError: 'None' object has no attribute 'IsNull'
>
>You snipped too much: how are we supposed to know what you put in the
>"projects" global variable?
>
>> 1) I've banged my head against this type of typing problem before.
>>     How do I get around it?
>
>Put correctly typed objects in containers
>
Actually it is the return x which whould have read return project.

>> 2) Is there a better way of handling null objects? ( Most languages I 
>>     know have objects which are explicitly null )
>
>None is an object which is explicitly null. What more do you need?

Thanks.


From "Marcin.Kolbuszewski " at nrc.ca  Mon Jun 19 13:01:05 2000
From: "Marcin.Kolbuszewski " at nrc.ca (Marcin Kolbuszewski)
Date: Mon, 19 Jun 2000 12:01:05 -0500
Subject: accessing access from Python on windows
Message-ID: <394E51D1.71382D39@nrc.ca>

Hi there!

I am running python 1.5.2 on Windows NT and am trying to 
access an Access database through ODBC. No lick though...:

Traceback (innermost last):
  File "C:\PROGRA~1\Python\Tools\idle\ScriptBinding.py", line 131, in
run_module_event
    execfile(filename, mod.__dict__)
  File "D:\python\toc.py", line 3, in ?
    import dbi, odbc
ImportError: No module named dbi

Where may I get the dbi and odbc modules?

marcin Kolbuszewski
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Marcin.Kolbuszewski .vcf
Type: text/x-vcard
Size: 320 bytes
Desc: Card for Marcin Kolbuszewski  
URL: <http://mail.python.org/pipermail/python-list/attachments/20000619/8f56680b/attachment.vcf>

From gmcm at hypernet.com  Mon Jun 12 15:09:31 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 12 Jun 2000 19:09:31 GMT
Subject: Threads -- play well with others? If not, what instead?
References: <8i34mr$9hti$1@nntp6.u.washington.edu>
Message-ID: <8F519FCF0gmcmhypernetcom@199.171.54.194>

Russell E. Owen wrote:

>David Beazley's excellent "Python Essential Reference" says in the 
>section on threads: "In addition, many of Python's most popular 
>extensions such as Tkinter may not work properly in a threaded 
>environment."

>I assume it's true, but it was quite a bombshell. I was hoping to write 
>a networked GUI client, hence:
>- read data from a socket and fill in a GUI display
>- accept input from the user and write data to the socket
>I assumed I'd use two threads, one for input, one for output. Now I have 
>no idea what to do. Any suggestions?

It's probably less true now (1.5.2) than when David wrote that.

If you keep all the GUI in one thread (the main thread) it should work 
fine. That means having some kind of inter-thread queue. Use idle time or a 
timer in the GUI thread, and check the queue when it fires.

- Gordon


From moshez at math.huji.ac.il  Thu Jun 29 01:53:20 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 29 Jun 2000 08:53:20 +0300 (IDT)
Subject: list.pop([i]) and list.remove(x) w/ for loops
In-Reply-To: <8jdfb6$ugh$1@netnews.upenn.edu>
Message-ID: <Pine.GSO.4.10.10006290852230.27161-100000@sundial>

On 28 Jun 2000, David White wrote:

> I was getting some unexpected behavior when using list.remove(x), wondering if
> someone can tell me what's going on.
> 
> I have a list of 2 kinds of objects, such as [a, a, b, b, a, b].  I wan't to
> delete all the a objects out of the list while preserving the order of the b
> objects.  My code looks like this:
> 
> for o in olist:
>     if (o.type == a):
>        olist.remove(o)

Short answer: never ever modify a list you're iterating on. Make a copy
first.

running-faster-is-pointless-if-you-get-the-wrong-answer-ly y'rs, Z.

--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From moshez at math.huji.ac.il  Tue Jun 13 00:52:54 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 13 Jun 2000 07:52:54 +0300 (IDT)
Subject: Python Productivity over C++
In-Reply-To: <slrn8ka8kq.1r0.jajs@Pascual.es>
Message-ID: <Pine.GSO.4.10.10006130752300.20303-100000@sundial>

On Tue, 13 Jun 2000, Juanjo [iso-8859-1] ?lvarez wrote:

> |Python is a simply wonderful prototyping environment, IMO.
> |Now only if it had an optimizing compiler...
> 
> I think every Python fan dream with that... 
> 
> It's really so difficult?

Short answer: yes

Long answer: but it's being worked at nonetheless
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From matt at null.net  Fri Jun 16 21:24:24 2000
From: matt at null.net (Matthew Schinckel)
Date: Fri, 16 Jun 2000 19:24:24 CST
Subject: identifying operating system
References: <39499924.3BC84686@engineering.uiowa.edu>
Message-ID: <20000616192425.305204@210.9.20.69>

Juerg Tschirren <jtschirr at engineering.uiowa.edu> wrote:
> Is there a way how I can identify the operating system my Python
> script is running on (e.g., Linux, Windows, Mac, ...)?
> 
> Juerg

If it is only for compatibilty reasons, you can use os.name.

This will give you one of: ['posix', 'nt', 'dos', 'mac', 'os2', 'ce'].

Matt.


From jbauer at rubic.com  Wed Jun 14 09:20:27 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Wed, 14 Jun 2000 08:20:27 -0500
Subject: Tag file for Visual Slick Edit
References: <39476E54.4673C04B@proceryon.at>
Message-ID: <3947869B.45D66B07@rubic.com>

Horst Gassner wrote:
> Does anyone use Visual Slick Edit for Python development?
> If yes, does anyone konw if it is possible to get a so
> called tag-file for Python?

Horst,

I use XEmacs almost exclusively, and vi when I'm feeling
masochistic <0.3 wink>. However, once upon a time I edited
Python code in Windows with Visual SlickEdit.  If you
traverse the ftp section of http://www.slickedit.com
you'll find a user contributed python.zip file.

Jeff Bauer
Rubicon Research



From phd at phd.russ.ru  Wed Jun 28 12:47:26 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 28 Jun 2000 16:47:26 +0000 (GMT)
Subject: postgreSQL successes
In-Reply-To: <8isjlssf3j3ivrg7bn4ln567pobgo0eb3l@4ax.com>
Message-ID: <Pine.LNX.4.21.0006281641540.16176-100000@fep132.fep.ru>

On Wed, 28 Jun 2000 robin.escalation at ACM.org wrote:
> Does anyone have any success stories (or gotchas) related to
> implementing a PostgreSQL database with Python that they'd like to
> share? I am hoping to use it as a back-end to a b2b web site.

   I used Postgres in one of projects. Pure python CGIs, Postgres 6.4
(later upgraded to 6.5), PyGreSQL. Postgres was a little bit slow, but
fast enough for our task. What is more important, Postgres has a very good
community - skillful and helpful. I leaved the project, but I still
subsribed to their mailing lists.

   Postgres is still alive on my development machine. I use it to debug
portable solutions - I test programs on both Postgres and MySQL (I found
these two more than enough for usual tasks). Recently I upgraded to 7.0.
Pretty good, faster.

   I run Zope on the same computer, and use Postgres to test ZSQL. I am
satisfied. I will develop news engine for Zope, and will use Postgres as
reference point (along with MySQL).

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From wfouche45 at hotmail.com  Mon Jun  5 21:58:14 2000
From: wfouche45 at hotmail.com (Werner Fouche)
Date: Mon, 5 Jun 2000 21:58:14 -0400
Subject: Python <--> XML data binding.
Message-ID: <8hhlr6$eoh$1@slb3.atl.mindspring.net>

Does anyone know if a tool exist that will generate Python data binding code
for an XML DTD similar to what Breeze ( http://www.breezefactor.com ) does
for Java.

Thanks,

Werner




From hzhu at mars.localdomain  Sun Jun 11 20:56:28 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Mon, 12 Jun 2000 00:56:28 GMT
Subject: Where are the math functions?
Message-ID: <slrn8k8dfv.tu.hzhu@mars.localdomain>

I'm quite puzzled by not being able to find most of the common special
functions, such as (incomplete) gamma, beta and bessel and erf.

This arrises from effort to provide statistical functions for the MatPy
package.  Many stat functions require incomplete gamma/beta and some
variants.

Yes there is random.gamma but that's generating random numbers not the gamma
function.

In this day and age there must be some open source C++ implementations
somewhere that can be plugged into Python.  Any leads or volunteers?

I've even checked the Octave source but it turns out to use severl fortran
programs in a style that's quite beyond my willingless to examine.


-- 
Huaiyu



From mfletch at tpresence.com  Fri Jun  2 19:41:53 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 2 Jun 2000 19:41:53 -0400 
Subject: Escapes???
Message-ID: <CC59868A0A76D311B1F50090277B321C1F338B@VRTMAIL>

In case you don't find some compiler switch that fixes it, here's a
simplistic mechanism that does the hex escaping (doesn't currently do the \n
etceteras, that's an exercise for the reader I suppose)...  Note the use of
raw strings in the test case!

import string

def dehex( rawstring ):
	set = string.split( rawstring, '\\x' )
	initial = set[0]
	for element in set[1:]:
		initial = initial + chr(string.atoi( element[:2], 16 ))
		initial = initial + element[2:]
	return initial

test = r"H\x1b\x23\x12art\x1b\x23\x01\n"

print `dehex( test )`

Enjoy,
Mike

-----Original Message-----
From: kern at caltech.edu [mailto:kern at caltech.edu]
Sent: Friday, June 02, 2000 4:21 PM
To: python-list at python.org
Subject: Re: Escapes???


In article <8h8jg3$d66$1 at slb6.atl.mindspring.net>,
	aahz at netcom.com (Aahz Maruch) writes:
> In article <8h87u1$sga$1 at news2atm.raleigh.ibm.com>,
> Arint? <shouldbe at message.com> wrote:
>>
>>I am trying to write this escape seq to a device from python
>>"H\x1b\x23\x12art\x1b\x23\x01\n"
>>The \x1b\x23\x12 is supposed to be a command tell the device to do
something
>>special with "art", but python is making it look like this \x1b\x23\x12a
>>"rt".  Am I missing something on how to work with escape sequences?  Is
>>there another way to write hex values in a string.  I rather us \x00
because
>>that is what are testers are familiar with.
...



From tim_one at email.msn.com  Sat Jun  3 23:01:46 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 3 Jun 2000 23:01:46 -0400
Subject: Copyright changes with move from CNRI to BeOpen?
In-Reply-To: <200006022311.TAA16891@tor.dhs.org>
Message-ID: <000501bfcdd1$384702a0$1d2d153f@tim>

[Tim]
> ... folderol ...

[Blake Winton]
> Do you read Miriam-Webster's Word Of The Day too, or is this just a
> strange coincidence?

I think it more likely that they read my posts.  Maybe, in this case, using
Guido's time machine as well!  Wouldn't be the first time something like
this has happened on Guido's wedding day ...

getting-married-without-a-time-machine-is-like-going-to-las-
    vegas-without-money-ly y'rs  - tim





From arta at NOSPAMx-stream.nl  Mon Jun 12 14:20:35 2000
From: arta at NOSPAMx-stream.nl (A[r]TA)
Date: Mon, 12 Jun 2000 20:20:35 +0200
Subject: Testing for a file in python?
References: <39450724.231994953@nntp.interaccess.com>
Message-ID: <uHK0DsJ1$GA.106@net025s>

> How would you go about testing if a certain file exists in python?
> In most scripting languages you have the -f test:
>
> if -f filename do
>    whatever_you_were_going_to_do
>    fi
> else do
>    print "Error" file not found
>    fi
>
> stat doesn't work, it aborts if the file does not exist.

try:
    open('filename')
except:
    print 'File does not exist'

A[r]TA


--
-Life can kick your ass, but all you got to do is stand up and move on...




From root at rainerdeyke.com  Wed Jun 14 21:55:46 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 15 Jun 2000 01:55:46 GMT
Subject: boolean true and false values.
References: <3947f922.425016156@nntp.interaccess.com>
Message-ID: <CIW15.1255$OZ.357618@news1.rdc2.tx.home.com>

Thaddeus L. Olczyk <olczyk at interaccess.com> wrote in message
news:3947f922.425016156 at nntp.interaccess.com...
> Does python have true and false values.
> I would like to do something like
>
> class A:
>      def _init_(self):
>            self.processed=false
>      def process(self):
> self.processed=true
>
> at the moment I am using
> false=(1==0)
> true=(1==1)
> but would like something more elegant, or at least less sloppy.

false = 0
true = 1


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




From hnowak at cuci.nl  Fri Jun  9 18:40:33 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Sat, 10 Jun 2000 00:40:33 +0200
Subject: "Announce": Python Snippets site on hold
Message-ID: <200006092242.e59Mg9K29454@hera.cuci.nl>

Hi,

As some of you might have noticed, I decided to stop working on my snippets 
site. The main reason is that I just don't have the time anymore to scan 
the newsgroup for interesting code, much less collecting, editing and 
adding it.

However:
1. The site will stay online for a while; after all, it contains 390 
snippets, so it may as well stay as a reference point for users. (That is, 
if Blake Winton permits it... it's his server.)
2. There is such a thing as Python FAQTS (the Python knowledge base); the 
idea is roughly similar to that of the snippets site. I offered them my 
snippets (published and unpublished); it's up to them if you will actually 
see them in the knowledge base.

I'd like to thank everyone who has provided me with code and support during 
the time that the site existed.

Blessed be,

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



From un4e at rzstud1.rz.uni-karlsruhe.de  Tue Jun 20 09:07:58 2000
From: un4e at rzstud1.rz.uni-karlsruhe.de (Bjoern Giesler)
Date: 20 Jun 2000 13:07:58 GMT
Subject: Strange Tkinter trouble
Message-ID: <8inqbe$rgu$1@news.rz.uni-karlsruhe.de>

Hi,

I'm having strange troubles with our new file server's Tkinter installation
and was hoping that anyone that anyone can help me.

I've recompiled Python myself, and have done so several times before -- and 
I'm not sure what I've done wrong. 

Here is the error message I'm getting:

[snip]
Python 1.5.2 (#2, Jun 20 2000, 10:54:20)  [GCC 2.95.2 19991024 (release)] on
linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from Tkinter import *
>>> Tk()
'None','','Tk' # These are the values of {screen,base,class}Name below
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/org/opt/python/lib/python1.5/lib-tk/Tkinter.py", line 887, in
__init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: couldn't stat "": no such file or directory
>>> 
[snip]

My system is SuSE Linux 6.4 (glib-2.1 based), with Tcl/Tk 8.0.4 (also
self-compiled).

Has anyone seen this before?

Thanks in advance,
		--Bj?rn


From shapr at uab.edu  Sat Jun 17 07:52:29 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sat, 17 Jun 2000 11:52:29 GMT
Subject: whitespace denotation - rebuttal
References: <394ac7ea.609024015@nntp.interaccess.com>
Message-ID: <394B66EE.F7C75F38@uab.edu>

"Thaddeus L. Olczyk" wrote:
> 
> After using Python for a while I've come to the conclusion that
> using whitespaces to determine the begining and ending of blocks
> is bad for three reasons.
<snip>
> By having
> the code depend on indentation, and the editor do the indentation
> for you, you give the editor responsiblity in how the code will flow.

I am an advocate for specific indentation as a requirement.
Also, typing backspace or tab allows you to change the indentation,
so... No. My editor does not have complete control over my code flow.

> 1) The first is rather stylistic and may not appeal to most, but I
> have often found it useful. When debugging code I often put in
> code to help me figure out whats happening. I also avoid indenting
> such code so that it easy to remember where it is later when I want to
> take it out. An example ( assuming I use begin and end to denote
> blocks ):
<snippage>
> You can immediately see code that you used for debugging and remove
> it. However right now writing code that way makes it illegal.

You should read up on the -O command line switch for the Python
interpreter. It allows you to do at least two quite spiffy things that I
know of:
First, assert statements are compiled right out.
Second, any "if __debug__:" statements are compiled right out as well.
This means there is a better way to do debug code in python that even
allows conditional compilation of that code.
(random note: the -v command line switch can also be quite fun, but for
completely unrelated reasons)

> Initially the line in error was indented correctly, but someplace
> along the line the extra tab was introduced. A lot of effort then had
> to go in to finding this bug. With begin...end or braces this is less
> likely to happen.

Unit testing can really make your life easier, no matter what language
you use. Check out http://pyunit.sourceforge.net
PyUnit is great, thanks to Steve Purcell.
You may also want to take advantage of tabnanny and friends to check
your files for inconsistent indentation.

> 3) Finally. Python is a language where it is almost easy to write
> extensible programs. To illistrate what I mean consider this example
> of a python program I recently wrote:
> There is a C++ project which involves many project files and workspace
> files 

This looks like a C++ problem to me. How does this apply to Python?

> That
> means that I have to reindent the block. So as I reindent I have to
> try to remember how the indenting went. Something I stink at.

Try IDLE. Ctrl-] indents the entire block you have selected, Ctrl-[
dedents it.
It's quite easy.
PythonWin has an option to check/display all or illegal whitespace, and
both IDLE and PythonWin allow Tabify and Untabify. (convert spaces to
tabs and back, your choice of # of spaces per tab).

In conclusion, I don't think Python is for everybody, but it is for me.
If you like begin and end statements and you don't like whitespace,
Ruby, Rebol or J may be more your style.
On the other hand, I think you might like Python more if you check out
some of the nifty bits I've mentioned. Seriously, I'd like to help if I
can.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
 squirrels furnished by Paul E Black, the world is safe.


From shensey at xnet.com  Wed Jun 14 07:27:33 2000
From: shensey at xnet.com (Scott Hensey)
Date: Wed, 14 Jun 2000 06:27:33 -0500
Subject: Can't get NumPy to Install
Message-ID: <upqeksgfc39indo8mg3dsc4vfj8oghdr2d@4ax.com>

I'm having trouble installing NumPy. Distutils-0.8.2 seemed to install
without any errors, but when I tried to install Numerical-14 I get the
following error message.  I'm running RH 6.1.

[root at Scottsky Numerical]# python setup.py build
Traceback (innermost last):
  File "setup.py", line 29, in ?
    ext_modules = [('_numpy',
  File "/usr/local/lib/python1.5/site-packages/distutils/core.py",
line 72, in setup
    dist = klass (attrs)
  File "/usr/local/lib/python1.5/site-packages/distutils/dist.py",
line 196, in __init__
    raise DistutilsSetupError, \
distutils.errors.DistutilsSetupError: invalid distribution option
'install_path'

Thanks for your help.


From eviltofu at rocketmail.com  Sat Jun 10 12:47:26 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 10 Jun 2000 16:47:26 GMT
Subject: Type problem on Windows
Message-ID: <eviltofu-B4ED2B.12472010062000@news.earthlink.net>

I am using Pythonwin with win32-129, MacPython 1.5.2c1, and Linux Python 
1.5.2.

While working on some socket stuff, I check the input parameters to some 
function which expects a socket. The code I use is:


def sendData(self,data,s):
  if type(s)!=socket.SocketType:
    raise TypeError,"S "+str(type(s))+" is not a socket."
  ... more code ...


On Unix and Mac, this works. On Win98, this bombs.

I get the following Exception Message


S <type 'instance'> is not a socket.


Running under the Python Interpreter I do:


import socket
x = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
print (x)


Windows98 gives me <type 'instance'>
Unix and Mac gives me <type 'socket'>


Is there a workaround for this? And should this be reported as a bug?


From jbusler at my-deja.com  Tue Jun 13 18:46:34 2000
From: jbusler at my-deja.com (jbusler at my-deja.com)
Date: Tue, 13 Jun 2000 22:46:34 GMT
Subject: Can't Install Modules on RedHat rpm 1.5.2 devl
Message-ID: <8i6dka$17a$1@nnrp2.deja.com>

Seems my problem is not unique, but I haven't
found the answer on the Usenet.

I've installed developer's version of python 1.5.2
using the rpm.  I'm now trying to compile a new
shared module but can't seem to get all the
various Makefile's and sedscripts to work.

Upon install, I had a Makefile and
Makefile.pre.in.  I edited the Setup file as I am
accustomed to and tried makesetup.  Which
complained with:
sed: can't read Makefile.pre: No such file or
directory

So, then I try:
make -f Makefile.pre.in boot

Which gives me Makefile.pre and a sedscript.  Now,
makesetup works and the Makefile contains the
necessary compile statements for the modules I
need -- however, upon "make" or "make
sharedinstall" I get:
make: *** No rule to make target
`/usr/lib/python at VERSION@/config/Makefile', needed
by `sedscript'.  Stop.


In fact, all the locally set variables which were
correct in the original Makefile (like VERSION,
CC, RANLIB, etc) are now filled with @BLAH@ stuff
rather than something useful.  I presume this has
to do with the sedscript -- but something doesn't
seem to be working quite right.

Ideas?


	Jennifer



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


From scarblac-spamtrap at pino.selwerd.nl  Mon Jun  5 07:48:41 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 5 Jun 2000 11:48:41 GMT
Subject: Questions regarding design decisions in listobject.c
References: <393ABA63.18E2F46@san.rr.com> <20000604230608.U469@xs4all.nl> <393ACA13.FD2416F1@san.rr.com>
Message-ID: <slrn8jn4qi.lau.scarblac-spamtrap@flits104-37.flits.rug.nl>

Courageous wrote in comp.lang.python:
> Well, yeah, but [0:3] as "element zero through three, inclusive"
> is about as intuitive as you can get.

When I see s[:3], my intuition says it takes 3 elements, not 4...


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From neilh at scintilla.org  Tue Jun 27 11:07:18 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Tue, 27 Jun 2000 15:07:18 GMT
Subject: Microsoft's  C# (Sharp) & .NET -- A Heads Up
References: <qJ165.8938$nZ5.153196@typhoon.southeast.rr.com>
Message-ID: <Gq365.6003$Sn2.40878@news-server.bigpond.net.au>

> In case you haven't heard Microsoft is coming out with a massive new Web
> stratagy that, in part, uses a new language called C# (That sharp, like in
> music).  ...

   I read the intro and skimmed the language reference today. It doesn't
look like a bad language but it doesn't have anything that jumps out as a
good reason for moving to it. I'd call it a safer superset of C++. There are
some C++ features missing but apart from templates they are not very
important. And its the lack of a template or generic feature that I find
puzzling. The language strives for compile time type safety while allowing
the use of the generic 'object' type, but has no provision for creating new
type safe containers. Even Sun have realised that Java needs a template
equivalent.

   One thing I like is the type unification between basic types and objects.
Unlike Java's use of explicit box classes - Integer holding an int, etc, C#
allows arbitrary boxing of the basic types and structs as needed or by
casting: (object)value. You can not subclass the basic types.

   Its a lower level language than Python, and misses most of the dynamicity
that makes Python so good. OTOH, it will run faster.

    Neil




From fgbrown at ionet.net  Fri Jun  9 18:06:24 2000
From: fgbrown at ionet.net (F. G. Brown)
Date: Fri, 9 Jun 2000 17:06:24 -0500
Subject: I need a lesson - howfast.py Revisited
Message-ID: <8hrphd$6m7$1@ionews.ionet.net>

Greetings!
First off I want to thank everyone for their responses to my post a few days
back (2000 06 07).  I was able to learn much from all of your suggestions.
At the same time I can see that it's going to take alot of effort on my
part, to be proficient in using Python, so you might be hearing from me now
and then when I hit a wall.  If anyone has a suggested challenge for me to
work on next, I'll give it a try.
Thanks again, Bjorn, Brent, Scott, David, Brett, & Moshe.
Regards,
Fred

Below is what I finally came up with using everyone suggestions.
I next am going to see if I can make it an executable file in a DOS
and Windows environment.  I seem to remember reading that I
can download a program for doing that somewhere.
Time to go hunting!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!

howfast.py

decide = 'y'
while decide == 'y':
     distance = input("What is the distance you traveled in miles?  ")
     time = input("How many seconds did it require?  ")
     speed = distance*3600.00/time
          print "You were going %d mph!" %speed
     if speed > 75:
          print "Hey you better slow down!"
     decide = raw_input ("Would you like to do another? (y)es or (n)o")
     if decide == "y":
          print 'Here we go!'
else:
          print 'See ya!'





From ivanlan at home.com  Tue Jun 20 13:19:58 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Tue, 20 Jun 2000 11:19:58 -0600
Subject: Roman Number Representation
References: <S9Q1YMA7B5T5Ew1U@jessikat.demon.co.uk>
Message-ID: <394FA7BE.F9AE2509@home.com>

Hi All--

Robin Becker wrote:
> 
> I could have sworn I had seen a discussion of Roman Numbers somewhere on
> clp, but can't find it. Anyone got any clues?
>

I'm sure if you search DejaNews, you can find it.

I probably started the discussion, but I certainly contributed.  I do a
*teeny* (and bad) bit of Roman numeral code in TYPython, but it's not
very useful.

I also have archived a bunch of posts from the Historia-Matematica list
covering pretty much every aspect of Roman numerals; I've been meaning
to stick these together with some glue and (after obtaining HM's
listowner's permission) work them into a web page, with examples in
Python.  RSN, uh-huh. ...

Roman numerals are fascinating. ...

<eagerly-awaiting-knuth-vol-IV>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From kens at sightreader.com  Fri Jun 16 01:39:44 2000
From: kens at sightreader.com (Ken Seehof)
Date: Fri, 16 Jun 2000 01:39:44 -0400
Subject: Python class browser
References: <XKQ15.1$oL1.1552@paloalto-snr1.gtei.net>
Message-ID: <642A954DD517D411B20C00508BCF23B001296E3D@mail.sauder.com>

Brian wrote:

> Anyone know of a class browser for Python?  Preferrably written using
> wxPython?
> Thanks!

There's a good one that's part of PythonWin.  I haven't looked at the
code though.  PythonWin and wxPython are very similar, so it shouldn't be
too difficult to port.

Let me know if you find a solution.  I'm gonna need it too in a couple
months.

Thanks,

Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me
spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000616/20832ebf/attachment.vcf>

From shapr at uab.edu  Wed Jun  7 18:24:21 2000
From: shapr at uab.edu (Shae Erisson)
Date: Wed, 07 Jun 2000 22:24:21 GMT
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com>
Message-ID: <393ECB55.4AC016A4@uab.edu>

"Thaddeus L. Olczyk" wrote:
> 
> Is there anything like CPAN for python?

Many times it has been brought up, but it seems the buck has been passed
to dist-utils.

If you did it IETF[1] style, I think people would go with it.


[1] IETF motto: 'rough consensus, running code'
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From donn at u.washington.edu  Tue Jun 13 15:36:43 2000
From: donn at u.washington.edu (Donn Cave)
Date: 13 Jun 2000 19:36:43 GMT
Subject: Client with select.select()
References: <emt7cXW1$GA.180@net025s> <#F2#YuI1$GA.106@net025s> <8i36rt$42mq$1@nntp6.u.washington.edu>
Message-ID: <8i62gb$afa$1@nntp6.u.washington.edu>

Quoth "A[r]TA" <arta at NOSPAMx-stream.nl>:
[quoting myself]
|> I think you also may have the wrong idea about the write-set parameter
|> to select().  The test with select() is ``will I block?''  That is,
|> if read-set returns a file descriptor, you may read from that descriptor
|> without blocking - either there's data, or end of file or whatever may
|> cause a non-blocking state on a blocking descriptor.  The same is true
|> of the write set, mutatis mutandi.  The reader on the other end will
|> probably ignore what you're writing, you just won't block.  Sockets
|> are usually in this state.  I guess it's a good thing to check anyway,
|> but by no means is there any implication that the other end of the socket
|> has requested data.  ("block" means "wait".)
|
| But if you want to write a telnet-client. The server requests for data,
| isn't he?
| Then you can check with the write-set and send some data.
| Or am I missing something?

There is absolutely no way to tell, from your end, whether there is
a read waiting on the other end.  But otherwise, yes, that makes
sense.  Maybe we could write it like this:

    sfd = sock.fileno()
    while 1:
        rfds = [sfd, pty]
        efds = rfds
        wfds = []
        if ptydata:
            wfds.append(sock)
        if socketdata:
            wfds.append(pty)
        rfds, wfds, efds = select.select(rfds, wfds, efds, timeout)
        if sfd in rfds:
            data = sock.recv(16384)
            if data:
                socketdata = socketdata + data
            else:
                   sock.close()
                   raise EOFError
        [and likewise if pty in rfds, read into ptydata]
        if sfd in wfds:
            nbytes = sock.send(ptydata)
            ptydata = ptydata[nbytes:]
        [and likewise if pty in wfds, write from socketdata]

So I check first whether more socket output can be accepted.  I don't
think you will really find this done very often, though, because only
in relatively dire circumstances would select say "no", and even then
it may be OK to simply block.  It's more important to check the amount
actually sent, in any case, because it may be less than the full string
you gave send() and the rest will have to be re-sent.

When I learned to use sockets and pipes, I had already worked with the
somewhat pipe-like VMS MBX device.  It can notify you specifically when
the process on the other end posts a read.  I loved that.  Haven't seen
anything like it since.

	Donn Cave, donn at u.washington.edu


From ivanlan at home.com  Wed Jun 14 23:04:32 2000
From: ivanlan at home.com (Ivan Van Laningham)
Date: Wed, 14 Jun 2000 21:04:32 -0600
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net>
Message-ID: <394847C0.E237B43C@home.com>

Hi All--

Jerome Chan wrote:
> 
> Has anyone done any extreme programming in python? Care to share your
> experiences?
> 
> <http://www.extremeprogramming.org/index.html>
> 

Not yet, but the company I just started with recently hired me to do
Python programming (& other stuff), and we're all gearing up to do XP.

I'd like to hear others' experiences, too, if there are any.  Otherwise,
I'll let you know.

<i-think-XP-means-you-have-to-like-your-colleagues-though>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Axent Technologies, Inc.
http://www.pauahtun.org/
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
Author:  Teach Yourself Python in 24 Hours



From ame at swipnet.se  Thu Jun 15 12:43:42 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Thu, 15 Jun 2000 18:43:42 +0200
Subject: Python and the PIL...
References: <8iajbv$m1m$1@hunter.orchestream.com>
Message-ID: <j41iks0k3v6cg4lvbf24hbqv0pphne7p9q@4ax.com>

On Thu, 15 Jun 2000 13:48:08 +0100, "Tobe" <tbradshaw at orchestream.com>
wrote:

>I have some code that relies on the Python Imaging Library which I'm trying
>to use on my NT box but I'm finding a few probs.. namely Image.thumbnail(())
>produces garbage..
>
It has probably with the binary/text mode thing that NT has. If you
don't say anything then you have a text file and then all <lf>
magically will change to <cr><lf>

So you need to remind Python that you're working with binary files.

f = open("binfile.jpg","wb")

if-your-using-StringIO-then-I-don't-know-ly y'rs

// Anders



From rumjuggler at cryptarchy.org  Thu Jun 29 14:40:05 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Thu, 29 Jun 2000 18:40:05 GMT
Subject: ANN: a better getopt.py
References: <4.3.2.7.2.20000629140130.00e2d5c0@localhost>
Message-ID: <s56nlsgfmq7vupe2rqug6upsuk9sm718tn@4ax.com>

On Thu, 29 Jun 2000 16:25:02 +0100, Frank Stajano
<fstajano at uk.research.att.com> wrote:

There are other command-line parsers at

http://members.home.com/gindikin/dev/python/cmdline/cmdline.py
and
http://members.home.com/rumjuggler/python/arglist.py

"arglist" in the second case is a rather inappropriate name.

>This is to announce an unofficial new version of the getopt.py module that 
>helps parsing command line options. I wrote it because it scratched an 
>itch; if it's useful to others too, so much the better.
>
>http://www.cl.cam.ac.uk/~fms27/getopt/
>
>This is a backwards compatible replacement that does not take away or 
>change any existing functionality. It adds a new function getoptdict() 
>which does the same things as getopt() but returns the result in a 
>dictionary as opposed to a list of pairs. From my humble experience, the 
>caller usually likes random (not sequential) access to the supplied 
>options, so the dictionary appears to be the best match.
>
>The only problem comes up with repeated options (e.g. "-v for verbose, -v 
>-v for very verbose, -v -v -v for really very verbose indeed"). This of 
>course doesn't quite fit in the dictionary model, since each key must be 
>unique. The getoptdict() lets you choose what to do; you can REJECT 
>repeated options (default behaviour) if you abhor them like I do; but you 
>can also APPEND them to a list or KEEP_EARLIEST or KEEP_LATEST.
>
>Another minor update concerns exceptions: the lonely GetoptError has been 
>turned into a small class hierarchy. This allows the caller to distinguish 
>the various errors if fine-grained control is desired. At the same time it 
>is still possible to intercept all errors by simply calling the single 
>GetoptError which is at the root of the hierarchy.
>
>Note that the base file was the getopt.py that came with Python 1.6 alpha 
>2. Therefore this module won't work with Python < 1.6, because the base 
>file uses 1.6 features such as the index() method for a string.
>
>This modification to getopt.py is released as open source under the same 
>licence as the rest of Python. I, and my employer, disclaim all warranties 
>just like Stichting Mathematisch Centrum and CNRI do. Feedback and bug 
>reports to fstajano at uk.research.att.com.
>
>   <P><A HREF="http://www.cl.cam.ac.uk/~fms27/getopt/">getopt.py</A> 1.3 - 
>Compatible replacement for getopt that returns a dictionary instead of a 
>list of pairs.  (29-Jun-2000)
>
>
>  Frank (http://i.am/filologo.disneyano/)  http://www.uk.research.att.com/~fms/
>



From hzhu at knowledgetrack.com  Fri Jun 23 14:49:03 2000
From: hzhu at knowledgetrack.com (Huaiyu Zhu)
Date: Fri, 23 Jun 2000 18:49:03 GMT
Subject: MatPy-0.2.9 released
Message-ID: <slrn8l7cem.q3.hzhu@rocket.knowledgetrack.com>

OK, I've released version 0.2.9 of MatPy - Matrix package for Python at
http://MatPy.sourceforge.net/.  Feedbacks on features, bugs and where it
runs (or not) are appreciated. 


                       Description:
MatPy is a Python package for numerical matrix and vector computation and
plotting with a MatLab-like interface. It currently consists of wrappers
around the Numeric, Gnuplot and SpecialFuncs packages. It may eventually be
implemented directly in C/C++, or as interface to Octave.


                      Release Notes:
Added statistical distributions and special functions.
File names changed to work around Windows problem with cases.
Added demo in matpy.
Many bug fixes.
					    
                          
                      Changelog:
	  
Version 0.2.9, 2000-06-22
-------------------------

- Add Stats.distribs
- Add test_probs, test_sfuncs.  Rewrite test_efuncs.
- efuncs.py: Rewrite with __doc__.  Added gamma ...
- Add demo in matpy.py
- Improved documentation on installation.
- Update tests/__init__.py
- efuncs allow complex operations.
- Bug fix: approx_real, rdotmul, rdotdiv allow scalar argument.
- efuncRC: use complex only when one argument is complex.
- Add test_transpose.  Bug fix: diag, flipud, fliplr
- Update __all__ to include Stats, gnuplot
- Move aux.py to formats.py (aux is reserved word in Windows)
- Move matpy.py to matpy as Windows can't distinguish it from MatPy.py
- Add STATUS


Enjoy!


-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From sblakey at freei.net  Fri Jun  9 01:44:57 2000
From: sblakey at freei.net (Sean Blakey)
Date: Thu, 8 Jun 2000 22:44:57 -0700
Subject: Newbie: Simple question about string slices
In-Reply-To: <8hpsl6$mve@newshost.fujitsu.com.au>; from Ray.Smith@fujitsu.com.au on Fri, Jun 09, 2000 at 02:47:02PM +1000
References: <8hpsl6$mve@newshost.fujitsu.com.au>
Message-ID: <20000608224457.A1413@freei.com>

On Fri, Jun 09, 2000 at 02:47:02PM +1000, Ray Smith wrote:
> Hi there,
> 
> I'm completely new to Python and was reading the standard tutorial for
> Python and came across string slices., The tutorial says:
> 
> "Strings can be subscripted (indexed); like in C, the first character of a
> string has subscript (index) 0. There is no separate character type; a
> character is simply a string of size one. Like in Icon, substrings can be
> specified with the slice notation: two indices separated by a colon."
> 
> Then later:
> 
> "Here's a useful invariant of slice operations: s[:i] + s[i:]equals s."
> 
> Now,
> if I have a string defined as
> 
> s = "abcdefg"
> 
> I assumed s[2:2] to be equal to 'c',
> but when I tried it it equals '' (empty).
> The way I read this is give me the slice from index 2 to index 2.
> 
> I also "feel like" s[:2] + s[2:] should be equal to 'abccdef'
> 
> Simple testing shows me
> s[2:3] = 'c', and
> s[:2] + s[2:] = "abcdef"
> 
> My question is:
> If given a slice s[a:b]
> can I assume index "b" means upto but not including the index "b"?
> 
> eg.  s="abcdef"
> 
> s[:3] = "abc", meaning from index 0, upto but not including index 3
> s[3:] = "def", meaning from index 3 to the end of the string
> 
> I guess another way to ask my question is:
> Why is s[a:a] always '' (empty) ?
> 
> Hope someone out there in Python land can make sense of my question!
> 
> Thanks
> 
> Ray Smith
> 
> 
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list

Probably the easiest way to make sence of slice semantics is to think of
the slice indexes as counting the borders between values.  For example,
s[0:] is the slice from before the first character on. s[2:2] is the
"space" number two, which is between the elements of index 1 and two.

You have probvbly seen magic acts where a person is locked into a
segmented box.  The magician then places blades between the various
segments.

Think of a python sequence as beinglike this box.  Slice indexes do not
count the sub-boxes, but the spaces between the boxes where blades can
be inserted.  You can take your slice from above the head to just above
the ankles (body[0:-1]), or from the neck to below the feet (body[1:]),
or from the neck to the ankles (body[1:-1]).  If you insert two blades
at the neck (body[1:1]), you end up with nothing between them.

Slice indexes and direct indexes are different.  Perhaps it would help
to visualize your string like this:

0   1   2   3   4   5   6
| a | b | c | d | e | f |
  0   1   2   3   4   5

With this in mind, let's try to explain your examples:
s[2:2] == ''
The slice from seperatror 2 to seperator two is a zero-length string.
This is the slice starting at the ceperator between 'b' and 'c', and
ending at the same place.

s[:2] + s[2:] == s	(a very nice feature of python, BTW)
Everything up to seperator 2, plus everything after seperator two.

In summary, adn to actually get around to your question, s[:b] meens up
to seperator b, which is just before index b.  Similarly, s[b:] means
everything after seperator b, which is just before index b.

Perhaps this will help, perhaps this will just confuse you more.
    -Sean
-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
A debugged program is one for which you have not yet found the conditions
that make it fail.
		-- Jerry Ogdin



From aahz at netcom.com  Sat Jun 10 12:29:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 10 Jun 2000 16:29:57 GMT
Subject: namespace question
References: <eviltofu-8E2F45.12073510062000@news.earthlink.net>
Message-ID: <8htqe5$hul$1@slb7.atl.mindspring.net>

In article <eviltofu-8E2F45.12073510062000 at news.earthlink.net>,
Jerome Chan  <eviltofu at rocketmail.com> wrote:
>
>import socket
>
>socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
>
>how do I call a function from the socket module later on?

Most people would suggest that you name the variable "sock".  Which
might be a Bad Idea if you're writing a laundry application.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"I love you."
"Duhhhh!"  --SFJ


From nospam at nospam.hotmail.com  Fri Jun  9 23:45:24 2000
From: nospam at nospam.hotmail.com (Mark A.)
Date: Sat, 10 Jun 2000 03:45:24 GMT
Subject: Python and Mac OS X
References: <eviltofu-A14D38.19591923052000@news.earthlink.net>
Message-ID: <oRi05.4858$hp4.81363@newsread1.prod.itd.earthlink.net>

There is a version of Python available.  Go to http://softrak.stepwise.com
and search for python.  I believe this is a recompiled version of the UNIX
source code.  Keeping the core interpreter up to date with the other
sources should be much easier in MacOS X since there shouldn't be a
significant amount of changes necessary to make it run under the BSD layer
of MacOS X (my guess would be that the main changes will be in the compile
and make files).

"Jerome Chan" <eviltofu at rocketmail.com> wrote in message
news:eviltofu-A14D38.19591923052000 at news.earthlink.net...
> Does Python work with Mac OS X?
>




From mwh21 at cam.ac.uk  Thu Jun 22 07:16:34 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 2000 12:16:34 +0100
Subject: Dynamically add methods to classes?
References: <Sql45.1512$9W1.105781@news-west.usenetserver.com>
Message-ID: <m3og4u3qzx.fsf@atrus.jesus.cam.ac.uk>

Matthew Cline <matt at nightrealms.com> writes:

> Is is possible to dynamically add methods to classes in Python?

Perfectly possible:

>>> class C:
...  def a(self): print "a"
... 
>>> c=C()
>>> def b(self): print "b"
... 
>>> c.a()
a
>>> C.b=b
>>> c.b()
b

Is that what you meant?

Cheers,
M.

-- 
  In many ways, it's a dull language, borrowing solid old concepts
  from many other languages & styles:  boring syntax, unsurprising
  semantics, few  automatic coercions, etc etc.  But that's one of
  the things I like about it.                 -- Tim Peters, 16 Sep 93


From mogul at primus.ca  Wed Jun  7 22:58:09 2000
From: mogul at primus.ca (Robin Porter)
Date: Wed, 7 Jun 2000 22:58:09 -0400
Subject: Please help!!!!
Message-ID: <7WD%4.1099$0l2.8049@news2.tor.primus.ca>

I have been trying to run some basic Python code in PYTHON version 1.5.2 in
a Win9x OS and have been having one problem consistantly when trying to run
it.  These are the steps I take each time.

1.)Open PYTHON idle (GUI)
2.)Open a new window
3.)Open the file I want to run (PAYpython.py)
4.)Click EDIT
5.)Click RUN
6.)A new screen appears and prompts me to enter the number of hours worked.

(Up until this point all seems fine)

7.)I input the number 50
8.)Press enter and expect to be asked the rate of pay.

**I GET NOTHING**

Absolutely nothing happens after this point.....

Could someone please tell me what is wrong.  I have re-installed Python
1.5.2.  I have tried other code and all seems to lead me to the same
problem.  Please help!

Thank you,

Robin

The code in PAYpython.py is....

TaxRate  = 0.23
MinPay   = 300.0
RegHours = 40.0

TaxPayable = 0
Overtime = 0
OvertimeHours = 0.0

HoursWorked = float(raw_input("Please enter the hours worked per week? "))
PayRate = float(raw_input("Please enter the rate of pay? "))

if HoursWorked > RegHours:
    OvertimeHours = HoursWorked - RegHours
    Overtime = (OvertimeHours) * (PayRate*1.5)

GrossPay = ((HoursWorked-OvertimeHours)*PayRate)+Overtime

if GrossPay > MinPay:
  TaxPayable = (GrossPay-MinPay)*TaxRate

NetPay = (GrossPay-TaxPayable);

print "The gross pay is %f" % GrossPay
print "The net pay is %f" % NetPay





From akuchlin at mems-exchange.org  Mon Jun  5 23:31:16 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 05 Jun 2000 23:31:16 -0400
Subject: SSL and POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com>
Message-ID: <3daegzpjsr.fsf@newcnri.cnri.reston.va.us>

jsantaniello at my-deja.com writes:
> cgi I'm trying to access requires POST. Are there
> any particular strangenesses to using POST that
> anyone knows about? Does someone have a

There were bugs in the SSL support in the urllib.py included in
1.6a2; they've been fixed in the CVS tree.  You can download
the CVS version from SourceForge:
http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/python/dist/src/Lib/urllib.py?cvsroot=python

--amk


From fiona at sitegnome.com  Sat Jun 24 03:11:21 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 24 Jun 2000 07:11:21 -0000
Subject: [FAQTS] Python Knowledge Base Update -- June 24th, 2000
Message-ID: <20000624071121.12925.qmail@synop.com>

Hello Python users,

Below are the entries I've entered into http://python.faqts.com today.

regards, Fiona


## New Entries #################################################


-------------------------------------------------------------
Has anyone created typemaps for SWIG for Microsoft __int64 to bind them to Python long ints?
http://www.faqts.com/knowledge-base/view.phtml/aid/3961
-------------------------------------------------------------
Fiona Czuczman
Mark Hammond

Not specifically - but note that 1.5.2 and later has LONG_LONG support,
meaning you just pass the __int64 directly to Python - eg:

>From pywintypes:
PyObject *PyLong_FromI64(__int64 ival)
{
 return PyLong_FromLongLong(ival);
}
PyObject *PyLong_FromUI64(unsigned __int64 ival)
{
 return PyLong_FromUnsignedLongLong(ival);
}

BOOL PyLong_AsI64(PyObject *val, __int64 *lval)
{
 *lval = PyLong_AsLongLong(val);
 return *lval != -1 || !PyErr_Occurred();
}

BOOL PyLong_AsUI64(PyObject *val, unsigned __int64 *lval)
{
 *lval = PyLong_AsUnsignedLongLong(val);
 return *lval != (unsigned __int64)-1 || !PyErr_Occurred();
}

Should be pretty easy to wrap with SWIG...


-------------------------------------------------------------
Is it possible to configure emacs so that it automatically starts in python mode when opening a file *.py?
http://www.faqts.com/knowledge-base/view.phtml/aid/3963
-------------------------------------------------------------
Fiona Czuczman
Piet van Oostrum

You do this in the standard emacs way:

(setq auto-mode-alist
      (cons '("\\.pyw?$" . python-mode) auto-mode-alist))

or

(add-to-list 'auto-mode-alist '("\\.pyw?$" . python-mode))

It is also useful (at least on Unix) to have

(add-to-list 'interpreter-mode-alist '("python" . python-mode))


-------------------------------------------------------------
Is is possible to dynamically add methods to classes in Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/3964
-------------------------------------------------------------
Fiona Czuczman
Ken Seehof, Remco Gerlich

class some_class:
   pass
   
instance = some_class()

def some_method(self):
   print "Hello!"
   
some_class.method = some_method

instance.method()

eg:

>>> class C:
...  pass
...
>>> x = C()
>>> def f(self, z):
...  print self, z
...
>>> C.f = f
>>> x.f(12)
<__main__.C instance at 190d450> 12
>>>


-------------------------------------------------------------
Can anyone tell me how I would go about disabling a pmw combobox?
http://www.faqts.com/knowledge-base/view.phtml/aid/3960
-------------------------------------------------------------
Fiona Czuczman
Richard Chamberlain

In the ToDo list for Pmw disabling is mentioned so hopefully we'll have 
a solution from the source before long. In the meantime...

This is a bit of hack (in the derogatory sense of the word) but hey!

So you basically want to import the myComboBox class and create an 
instance just as you would with Pmw. I've added a couple of methods - 
disable and enable suprisingly enough.

Included below is some test code.

It works in the same way as you mentioned, by unbinding the events, and 
rebinding them to enable it.

Richard


from Tkinter import *
import Pmw
root = Tk()
root.title('Disabling ComboBox Hack')
Pmw.initialise()

class myComboBox(Pmw.ComboBox):
    def disable(self):
        # Rebind things to my empty handler
        self.component('arrowbutton').bind('<1>',self.handler)
        self.component('arrowbutton').bind('<3>',self.handler)
        self.component('arrowbutton').bind('<Shift-
3>',self.handler)
        self.component('entryfield_entry').configure
(state='disabled',fg='grey')

    def enable(self):
        # bind the events back up to the original methods
        self.component('arrowbutton').bind('<1>',self._postList)
        self.component('arrowbutton').bind('<3>',self._next)
        self.component('arrowbutton').bind('<Shift-3>',
self._previous)
        self.component('arrowbutton').configure(takefocus=1)
        self.component('entryfield_entry').configure
(state='normal',fg='black')

    def handler(self,event):
        # so it doesn't propagate the event
        return('break')

# Now test it

comboentries = ("Pmw Should", "have a nicer way", "to
disable", "a combo box")
theEnabler = Button(root, text='Enabled',padx=20, pady=10)
theEnabler.pack(expand=1, fill=BOTH, padx=8, pady=8)

combobox = myComboBox(root, label_text='Disabling Combobox:',
labelpos='wn',
                        listbox_width=24, dropdown=1,
                        scrolledlist_items=comboentries)
combobox.pack(fill=BOTH, expand=1, padx=8, pady=8)

combobox.selectitem(comboentries[0])

def disable(event):
    if theEnabler['text']=='Enabled':
        combobox.disable()
        theEnabler.configure(text='Disabled')
    else:
        combobox.enable()
        theEnabler.configure(text='Enabled')
theEnabler.bind('<Button-1>',disable)
root.mainloop()


## Edited Entries ##############################################


-------------------------------------------------------------
I'm new to Python, where should I start?
Can you give me an overview of the Python Documentation?
Do I need books to learn Python?
http://www.faqts.com/knowledge-base/view.phtml/aid/1356
-------------------------------------------------------------
Nathan Wallace, Fiona Czuczman
Tom Funk,Simon Brunning,Dparsavand, Phil Austin

The handbook is part of the on-line documentation available at the
Python web site and with the Python installation -- and it's free.

You may not need to go to a bookstore if you peruse the following page:

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

The tutorial is quite complete:

  http://www.python.org/doc/current/tut/tut.html

If you work through the tutorial, it should carry you pretty far along
in your quest.  I found it to be *quite* useful.

The Library Reference discusses the modules that ship with Python. 

  http://www.python.org/doc/current/lib/lib.html

The Language Reference is a bit more abstract, and *much* more dry.  
However, it does completely describe the core Python language
constructs, grammar and syntax.   It's often referred to as material for
"Language Lawyers."

  http://www.python.org/doc/current/ref/ref.html

I like the Module Index:

  http://www.python.org/doc/current/modindex.html

It allows you to jump straight to the module of your choice.

If you're using Win32 (as I do), then you may find the MS HTML Help 
version to be useful (it's my favorite).  If you use Win32, you might 
want to check out:

  http://www.orgmf.com.ar/condor/pytstuff.html

Best of all, these very complete works of non-fiction are FREE.... gotta 
love that!  

I own five Python books, but I still find myself referring back to the 
Python documentation regularly.

---

As one newbie to another, I can recommend
<http://www.idi.ntnu.no/~mlh/python/programming.html> for an
introduction to programming, and
<http://www.idi.ntnu.no/~mlh/python/instant.html> for a bit more on
Python. After that, you might want to look at
<http://starship.python.net/crew/amk/grimoire/html/> for a
'cookbook' of useful techniques. If none of these take your fancy,
there are other links at the python site -
<http://www.python.org/doc/Intros.html>.

If you have any questions look at the FAQs -
<http://www.python.org/doc/FAQ.html>

If you follow http://www.pythonlabs.com/ to
http://www.vex.net/parnassus/ and then choose the  Info/Books/Tutorials 
link then Tutorials, you will get quite a few alternatives.

Also, have a look at:

http://www.networkcomputing.com/unixworld/tutorial/005/005.html

http://yhslug.tux.org/obp/thinkCS/thinkCSpy

The current numpy tutorial (supercedes the one that's cited on 
Parassus):

http://numpy.sourceforge.net/

Richard P. Muller's python short course:

http://www.wag.caltech.edu/home/rpm/python_course/

Konrad Hinsen's Python for Science tutorial

http://starship.python.net/crew/hinsen/






From mlauer at trollinger-fe.rz.uni-frankfurt.de  Fri Jun  9 18:26:58 2000
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 10 Jun 2000 00:26:58 +0200
Subject: Using python with Corba
References: <eieJ4.8408$H7.704991@brie.direct.ca> <871z4awiz0.fsf@parus.parus.no> <39410efe.18494828@nntp.interaccess.com> <8hqgbc$iga$1@pegasus.csx.cam.ac.uk>
Message-ID: <39416f32$1@nntp.server.uni-frankfurt.de>

Duncan Grisby (dgrisby at uk.research.att.com) wrote:
> In article <39410efe.18494828 at nntp.interaccess.com>,
>  Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:

> >>> Anyone know TAO can be used in Python ?
> >>
> >>Don't think so. But a Python ORB can interoperate with TAO.

> >They are supposed to but (sigh ) most orbs have many parts of the
> >standard unimplemented yet. Do you know if Fnorv,ILU, OmniORb
> >or ORBit (didn't know ORbit supported python) can interoperate with
> >TAO?

> omniORB can definitely interoperate with TAO. I am not aware of any
> problems at all. ORB interoperability is now at the stage that there
> are very few difficulties between ORBs, and those that remain are
> obscure cases which are unlikely to crop up in real systems. I don't
> imagine you'll encounter any problems with any of the Python ORBs.

I played around with all CORBA bindings and found omniORB to operate best.
It's also
closest to the CORBA Python language binding document. omniORBpy is really
great. 

:M:

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




From alexandre.ferrieux at rd.francetelecom.fr  Tue Jun 27 11:20:40 2000
From: alexandre.ferrieux at rd.francetelecom.fr (Alexandre Ferrieux)
Date: Tue, 27 Jun 2000 17:20:40 +0200
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com>
Message-ID: <3958C648.1D76@rd.francetelecom.fr>

Paul Duffin wrote:
> 
> > existence of the 'pty' unix package, which is roughly "Expect minus
> > Tcl", and especially its 'nobuf' tool:
> >
> >         nobuf cmd [args...]
> >
> > starts cmd as a child, strongly held between ptys, so that the overall
> > command behaves properly (ie in unbuffered or line-buffered mode) betwen
> > pipes. If all you want is that, Expect is surely an overkill.
> >
> 
> Where can I get the 'pty' package from ?

Last time I checked it was a pty-4.0.tar.gz .
I find it currently with FTPSearch at
ftp://ftp.sunsite.org.uk/Mirrors/wuarchive.wustl.edu/languages/c/unix-c/kernel/pty4.tar.gz
.

Surely more recent/portable variants must exist...

> > But ptys are no rocket science anyway. The nobuf above should be doable
> > in pure Python within 20 loc...

> While ptys are not rocket science they are also not particularly portable,
> Expect handles all of this for you.


In unix, AFAIK the only nonportable part is the *naming* of the pty
devices. That kind of decision is fairly trivial to make in
sh/awk/perl/python/tcl...

> Windows is also a real problem as it
> does not have ptys and needs to be approached in a completely 
> different way.

Yes but Expect for Windows is a totally different distrib too :-(

Hence the suggestion to consider writing a specific Python module (if
there isn't already - I've just arrived in town): for unix it just opens
the proper devices and substitutes the popen2() pipes with the masters;
for windows more trickery is needed (namely, behaving as a debugger to
trap IO calls!), but should be doable in win32all, shouldn't it ?

For the Mac, er.. ;-)

-Alex


From kens at sightreader.com  Tue Jun 20 18:13:53 2000
From: kens at sightreader.com (Ken Seehof)
Date: Tue, 20 Jun 2000 15:13:53 -0700
Subject: Python complements C++ for productivity
References: <NEBBIOGBDMDPEANLFKGHKEGJCEAA.jschmitt@vmlabs.com>
Message-ID: <394FECA1.7C43EF55@sightreader.com>

Excellent idea.  Why didn't I think of that.  It would certainly make STL look
clumpy.  Of course python has no need for templates since it has nice
heterogenous container types (list and dictionary) and the ability to create
more of them easily.  Not having to tell the compiler what you are doing (i.e.
dynamic typing) is a big ease-of-use advantage.

John Schmitt wrote:

> When I first opened Pythonwin, I used the Python/Perl phrasebook.  Very
> useful to me.  I don't know the STL very well (I don't even know what the
> SWIG is) so I couldn't write one, but how hard would be to write a Python
> <-> C++/STL phrasebook?  I imagine it would be useful for C++ programmers
> who want to move to Python and useful to Python programmers who have to move
> some stuff to C++.
>
> John
>
> > -----Original Message-----
> > From: python-list-admin at python.org
> > [mailto:python-list-admin at python.org]On Behalf Of Don Bashford
> > Sent: Tuesday, June 20, 2000 2:01 PM
> > To: python-list at python.org
> > Subject: Python complements C++ for productivity
> >
> >
> > I've been using C++ for about 10 years, and Python less than a year.
> > I recently had the experience of implementing an algorithm for
> > Delaunay triangulation in both Python and C++.  I was working from my
> > reading of someone's Ph.D. thesis that had parts of the algorithm in
> > pseudo-code, and part in English.
> >
> > I started out in Python.  The work was pleasant and went fairly
> > quickly.  It was pleasing that a lot of the Python I wrote ended up
> > looking very close to the pseudo-code in the thesis.  The interpreted
> > nature of Python was very helpful, since it made it easier to inspect
> > results as I went along.  I also benefited from a Python graphics
> > package written by a colleague that hepled me visualize what the
> > algorithm was doing.  I think in the early phases, development in
> > Python was faster than it would have been in C++.
> >
> > But the resulting code was slow---about 100 times slower than the
> > thesis author's C code.  I started re-implementing in C++.  The
> > similarity of the basic "class" ideas in the two languages made the
> > re-write very easy and fast.  In some cases, I just dropped the Python
> > code into the C++ file and edited it.   In the translation process,
> > the stricter type rules of C++ helped me find some bugs that I might
> > never have found in the Python version, and it also resulted in
> > improved design.
> >
> > The only real unpleasentness in the whole process was when I had my
> > implementation partly in Python (higher level) and partly in C++
> > (lower, more number crunching levels).  Then I was using SWIG to
> > "wrap" the C++ parts for use from Python, but getting proper handling
> > of translations like Python lists to C++ containers was quite a pain
> > (although I learned a lot about SWIG typemaps in the process).
> >
> > In short, I found Python to be a great prototyping language, with
> > pretty easy movement to C++, which is a great language when you want
> > strict type checking, data hiding and performance.
> > --
> > http://www.python.org/mailman/listinfo/python-list

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000620/15a4c713/attachment.vcf>

From aahz at netcom.com  Tue Jun 27 13:55:49 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 17:55:49 GMT
Subject: more python evangelising...
References: <200006271406.AAA06218@mbuna.arbhome.com.au> <D9465.71358$7o1.1704294@news2.rdc1.on.home.com>
Message-ID: <8japr5$1qb$1@nntp9.atl.mindspring.net>

In article <D9465.71358$7o1.1704294 at news2.rdc1.on.home.com>,
Tom <tom-main at REMOVEME.home.com> wrote:
>
>But you're falling into the old trap of thinking that one language is best
>for everything.

I think you'll find that very few Pythonistas think that way.

>Python is excellent for what it was designed for, but this does not include
>large scale app development.

Really?  Then what's your reaction to all the people building
large-scale applications with Python?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From bob at slim.instantappointment.com  Sun Jun 18 12:34:23 2000
From: bob at slim.instantappointment.com (Bob Erb)
Date: 18 Jun 2000 16:34:23 GMT
Subject: Help: 'AttributeError: wm_title' Error starting IDLE
References: <slrn8knc4g.3u3.bob@slim.instantappointment.com>
Message-ID: <slrn8kpuk5.qa.bob@slim.instantappointment.com>

Problem fixed. Hacked out all instances of 'wm_' in idle/*.py.
Works ok now.

On 17 Jun 2000 17:06:39 GMT, Bob Erb <bob at slim.instantappointment.com> wrote:
>I can't run IDLE. It bombs with 'AttributeError: wm_title'.
>Complete traceback is below.
>
>- Linux 2.2.10
>- Python 1.5.2
>- IDLE 0.5
>
>Any ideas? All suggestions appreciated. Thanks.
>
>Traceback (innermost last):
>  File "/usr/lib/python1.5/site-packages/idle/idle.py", line 3, in ?
>    PyShell.main()
>  File "/usr/lib/python1.5/site-packages/idle/PyShell.py", line 720, in main
>    shell = PyShell(flist)
>  File "/usr/lib/python1.5/site-packages/idle/PyShell.py", line 288, \
>in __init__
>    OutputWindow.__init__(self, flist, None, None)
>  File "/usr/lib/python1.5/site-packages/idle/OutputWindow.py", line 15, \
>in __init__
>    apply(EditorWindow.__init__, (self,) + args)
>  File "/usr/lib/python1.5/site-packages/idle/EditorWindow.py", line 169, \
>in __init__
>    self.saved_change_hook()
>  File "/usr/lib/python1.5/site-packages/idle/EditorWindow.py", line 431, \
>in saved_change_hook
>    self.top.wm_title(title)
>AttributeError: wm_title
>


From trentm at activestate.com  Sat Jun 17 19:25:45 2000
From: trentm at activestate.com (Trent Mick)
Date: Sat, 17 Jun 2000 16:25:45 -0700
Subject: writing (Gnu)MAKE in Python
In-Reply-To: <slrn8kmc4l.o8h.ssthapa@ntcs-ip48.uchicago.edu>
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com> <slrn8kmc4l.o8h.ssthapa@ntcs-ip48.uchicago.edu>
Message-ID: <20000617162545.C19446@activestate.com>

On Sat, Jun 17, 2000 at 06:02:15PM +0000, Suchandra Thapa wrote:
> 	I think the software carpentry project is funding a set
> of tools to replace make/autoconf/misc. dev tools in python.  The
> proposals that they've received and the winning proposals are on 
> their web page (unfortunately I don't have the address but a 
> search should turn it up).
> 

You are mostly right. The Software Carpentry Project is a project to create
replacements for the standard suite of software development tools
(make/autoconf/bug tracking/etc.) with a focus on Python implementation
and/or scriptability of the tools, cross-platform support, etc.

THere were a number of proposals submitted a couple of months ago for four
different categories.  Four of the most promising in each section were
invited to submit a more detailed proposal/design document. This is due by
the end of this month. The plan is then to select the most promising in each
category and use the Project's funding to help develop the tools and document
them.

The category most applicable tothis thread is SC-Build.

Trent


-- 
Trent Mick
trentm at activestate.com



From pollet at liliput.lct.jussieu.fr  Tue Jun 20 08:50:12 2000
From: pollet at liliput.lct.jussieu.fr (Rodolphe Pollet)
Date: 20 Jun 2000 12:50:12 GMT
Subject: Computer Algebra System
Message-ID: <8inpa4$8iv$1@vishnu.jussieu.fr>

Is anyone aware of a Computer Algebra System (CAS) ---
i.e a mathematical software that is able to derivate, integrate,
simplify, find limits, ... analytically --- in python ?
It should compete with Mathematica or Maple.

-- 
Rodolphe POLLET
Laboratoire de Chimie Theorique, Case 137,    |   pollet at lct.jussieu.fr
Universite Pierre et Marie Curie, T 22-23,    |   tel:   01.44.27.42.11
4, Place Jussieu, 75252 PARIS CEDEX 05, FRANCE|   Fax :  44.27.41.17


From ddb at crystal.uwa.edu.au  Fri Jun 30 04:16:59 2000
From: ddb at crystal.uwa.edu.au (Douglas du Boulay)
Date: Fri, 30 Jun 2000 16:16:59 +0800
Subject: jpython casting to abstract java class
References: <39570C59.2CBCFAEB@crystal.uwa.edu.au> <395820B1.417B380E@crystal.uwa.edu.au> <Pine.LNX.4.10.10006280845130.3063-100000@vangogh.ulb.ac.be>
Message-ID: <395C577B.9FA62815@crystal.uwa.edu.au>

Well thanks, for providing that example of yours Chris.
It appears that I was struck by the dreaded system path versus 
classpath problem again. 

My directory hierarchy was  
<pre>
top  ----  A  -------parser
    \     *.py src      *.java src
     \                  (parser module)
      \----B
          working 
          directory
</pre>

I had CLASSPATH = A  while working in directory B but 
calling  a *.py source file from A which imported some 
components of my java "parser" package from the A/parser directory.
For some reason in jpython1.1 that must have got loaded via the 
system path, and not via the CLASSPATH as it did for jpython1.0.3.
Moving the parser directory somewhere else and sticking the new 
directory containing it on the CLASSPATH, fixes the problem.

I thought I was immune from this problem by working in directory
B, but apparently I was wrong.

doug

P.S. for the record, the discrepancy arose because the 
PyInstance.javaProxy object was not considered to be an instance of 
the underlying abstract java class I was subclassing. 
Maybe someone will fix that oneday?


From eduard.hiti at t-online.de  Wed Jun 14 15:34:09 2000
From: eduard.hiti at t-online.de (Eduard Hiti)
Date: Wed, 14 Jun 2000 21:34:09 +0200
Subject: Problem assigning timevalue to Excel - COM
References: <8i7pp7$1vl$1@newstoo.ericsson.se>
Message-ID: <8i8msk$k1o$11$1@news.t-online.com>

> I haven't found out what type and/or format to use when assigning a value
to
> a cell in MS Excel having a predefined time format like this:
> 1997-04-03  12:15:00
>
> When reading - via COM - from a cell with a manually entered content, I
get
> a 'time object' back.
> This 'time object' is recognized as being of 'TimeType' and can be
converted
> to an int.
>
> How can I convert an int into something that can be assigned to the Value
> property of the desired cell and be accepted according to the format as of
> above?
>

I recall the module mxDateTime being mentioned to have support for COM date
values. I don't have a URL ready, but it is a fairly popular module, so you
should have no trouble finding it on python.org or parnassus.

Bye,

    Eduard Hiti






From nut at netidea.com  Wed Jun  7 19:38:52 2000
From: nut at netidea.com (Ian Main)
Date: Wed, 7 Jun 2000 16:38:52 -0700
Subject: Getting link arguments for embedding Python in C
Message-ID: <20000607163852.A9122@gimp.org>

Hello all,

I'm working on a project that embeds the perl interpreter (an XML
application framework - http://entity.evilplan.org), and am finding
that I'm having a hard time getting it to reliably build on a variety
of systems.

The general problem is not knowing what link arguments are required
by libpython.a.  For example, as far as I can tell, there's no way to
know if thread support was built in, and if so, to use -lpthread, or
-lc_r (though I'm doing my own test now).  Or if the readline library
is required (-lreadline) etc.

Is this information kept anywhere after python is installed on the
system ?

Also, why is there no libpython.so built by default on platforms that
can use them ?  This would be much handier for us embedders (so I can
link in the interpreter on demand - it's not always required).

Thanks,


	Ian




From unjp at rz.uni-karlsruhe.de  Fri Jun 30 12:07:37 2000
From: unjp at rz.uni-karlsruhe.de (Christian Schneider)
Date: Fri, 30 Jun 2000 18:07:37 +0200
Subject: Tk_FindPhoto...
Message-ID: <395CC5C9.1DE04483@rz.uni-karlsruhe.de>

Hi there,

has anyone already worked with the PhotoImage class and thereby obtained
the image data not only from files but through  C code. I know how to do
this in Tcl/Tk via the Tk_FindPhoto / Tk_PhotoPutBlock commands using
the Tk_PhotoImageBlock struct and now i'm looking for a solution to
solve this with Python.

Thanks,
Christian



From aaron.ginn at motorola.com  Fri Jun  2 15:59:30 2000
From: aaron.ginn at motorola.com (Aaron Ginn)
Date: 02 Jun 2000 12:59:30 -0700
Subject: Sorting a list
References: <snk8g856mf.fsf@motorola.com> <393801C7.960D6E86@uab.edu>
Message-ID: <snitvr6ehp.fsf@motorola.com>

Shae Erisson <shapr at uab.edu> writes:

> Aaron Ginn wrote:
> > 
> > I'm a python newbie, so please be kind! :)
> > 
> > I'm trying to sort a list.  Fairly simple, but I'm not getting
> > at all what I expect.  When I attempt to sort the list, the
> > return value is None, and I can't understand why at all.
> <snip>
> > Why is sort returning an empty list?
> 
> because in this case, sort is expected to have the side effect of
> sorting the list, rather than returning a sorted list.
> 
> >>> a = [3,2,1]
> >>> a.sort()
> >>> a
> [1, 2, 3]
> -- 
> sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
>    bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt

Shae,

Thank you for making that distinction.  That is much clearer now.

Aaron

-- 
Aaron J. Ginn

Opinions mine, not Motorola's


From scarblac-spamtrap at pino.selwerd.nl  Sat Jun  3 18:37:01 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 3 Jun 2000 22:37:01 GMT
Subject: float does not parse '0.000000E+00'
References: <39396D65.D2196E66@inc.it>
Message-ID: <slrn8jj22h.hgn.scarblac-spamtrap@flits104-37.flits.rug.nl>

Paolo Redaelli wrote in comp.lang.python:
> 1st of all, cheers to everyone!
> 
> I need to parse numbers coming from a file (made by a fortran program)
> the problem comes when parsing 
> When I made some tests on the shell they went fine, like this:
> 
> =======================
> >>> a = "0.000000E+00"
> >>> float(a)
> 0.0
> >>> 
> ============================
> 
> but if I run a program that does x=float(x) where x = "0.000000E+00 "
> (being a string) here's the output. 
> 
> ===========================
> [paolo at aldebaran Telaio]$ ./pyplot.py
> x(<type 'string'>)='0.000000E+00' y(<type 'string'>)='0.000000E+00'
> Traceback (innermost last):
>   File "./pyplot.py", line 198, in ?
>     t = Telaio()
>   File "./pyplot.py", line 172, in __init__
>     x = float(x)
> ===========================

What is the exact error? You snipped the last line, with the message.

Maybe one your 'from module import *' commands redefines float?


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"Early to rise, early to bed, makes a man healthy, wealthy and dead." -- TP


From JamesL at Lugoj.Com  Thu Jun  8 22:01:39 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 08 Jun 2000 19:01:39 -0700
Subject: Testing -- not a new message in a while
References: <slrn8k0hv9.pq3.kc5tja@garnet.armored.net>
Message-ID: <39405003.CFDA87FC@Lugoj.Com>

There are hundreds per day; check with your Usenet news provider.

"Samuel A. Falvo II" wrote:
> 
> Just testing; I haven't seen any new posts in this newsgroup for quite a
> while.  Thanks for your patience.
> 
> --
> KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
> Samuel A. Falvo II          |
> Oceanside, CA               |


From bradlosh at laddservices.net  Mon Jun 26 09:57:01 2000
From: bradlosh at laddservices.net (Brad Losh)
Date: Mon, 26 Jun 2000 09:57:01 -0400
Subject: Pick or Universe Module
Message-ID: <3957612D.B631299D@laddservices.net>

has anyone seen a python way to talk to a Pick or Unverse DBMS.  if
there is nothing I would like to start something, just not sure how.

Brad Losh



From phd at phd.russ.ru  Thu Jun  1 04:27:06 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 1 Jun 2000 08:27:06 +0000 (GMT)
Subject: ANN: bzip2 module
In-Reply-To: <20000601003920.A18615@dsl-only.net>
Message-ID: <Pine.LNX.4.21.0006010822300.22762-100000@fep132.fep.ru>

On Thu, 1 Jun 2000, eito wrote:
> I would like to announce about a bzip2 module. This is my
> first attempt to write Python module. I really don't know
> how to package, announce, etc... I would appreciate if
> somebody can tell me what is the good way to do this kind
> of thing.

   Packaging - learn to use Distutils (distribution utilities). Currently
it is separate package, but will be included in python 1.6.

   Announcing - watch the best announcements here. Marc Lemburg's
announcements are probably the best of the best. Search some of them on
Deja.com. They are constracted the following way (and after I saw them few
times I started to mimick my announcements the same way):

1. What is it (short).

2. What's new since last announcement (short).

3. Where to get; copyright; licensing terms.

4. Any longer details you want - long description, long what's new, etc...

Oleg.            (All opinions are mine and not of my employer)
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From avle at rogn.ii.uib.no  Sat Jun 17 09:39:54 2000
From: avle at rogn.ii.uib.no (Helge Avlesen)
Date: 17 Jun 2000 15:39:54 +0200
Subject: Python vs. Matlab
Message-ID: <724s6sa0k5.fsf@rogn.ii.uib.no>

I have become almost addicted to Matlab after using it extensively for
some years, however, the licencing feels very strict after getting
used to free and very good quality software like Emacs, Linux, etc.

My main use of Matlab is reading in large binary files into
arrays(output from simulations), do some processing with various
matrix routines, then extract sections of the 3D arrays for plotting.

Spotted Python some months ago, and liked it!  Can Python, Numpy and
the many plotting extensions be used as a replacement for Matlab, is
Python slower/faster etc.?  

It would be nice to hear some experiences!

-- 
Helge Avlesen


From eddthompson at ssi.parlorcity.com  Wed Jun  7 21:59:37 2000
From: eddthompson at ssi.parlorcity.com (Blu Dragoon)
Date: Thu, 8 Jun 2000 01:59:37 -0000
Subject: GUI editor
Message-ID: <393f4568$0$25329@wodc7nh0.news.uu.net>

Is there a gui editor written in python for tkinter?
The reason I ask is that I use python on 6 different platforms.
And tkinter seems to be the most prevailant of the gui's.
Or should I go ahead a take the PIL? and is there any gui editor
for PIL too.



Thanks





From mwh21 at cam.ac.uk  Wed Jun  7 03:19:19 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 07 Jun 2000 08:19:19 +0100
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <001101bfcfde$0a6a6f20$48a2143f@tim> <393D3993.84B507C9@bam.com>
Message-ID: <m3ln0ivtzc.fsf@atrus.jesus.cam.ac.uk>

Bill Scherer <scherbi at bam.com> writes:

> > [Michael Hudson]
> > > ...
> > > Tangentially, does anybody know of any implementations of JavaScript
> > > (not necessarily connected to a browser) that let you type interactive
> > > commands into a listener and genereally muck about with the language?
> 
> There's the NJS Javascript Interpreter:
>      http://www.bbassett.net/njs/

Ta.  It doesn't actually have a listener though does it?  I can't find
mention of one in the docs or command line help though...  still, beats
typing javascript:... urls, I think.

-- 
  Ya, ya, ya, except ... if I were built out of KSR chips, I'd
  be running at 25 or 50 MHz, and would be wrong about ALMOST
  EVERYTHING almost ALL THE TIME just due to being a computer!
                                              -- Tim Peters, 30 Apr 97


From robin at jessikat.demon.co.uk  Wed Jun 14 10:10:08 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 14 Jun 2000 15:10:08 +0100
Subject: Problem assigning timevalue to Excel - COM
References: <8i7pp7$1vl$1@newstoo.ericsson.se>
Message-ID: <bNM3ULAAJ5R5Ewn3@jessikat.demon.co.uk>

In article <8i7pp7$1vl$1 at newstoo.ericsson.se>, Bror Johansson
<bror.johansson at esavionics.se> writes
>I haven't found out what type and/or format to use when assigning a value to
>a cell in MS Excel having a predefined time format like this:
>1997-04-03  12:15:00
>
>When reading - via COM - from a cell with a manually entered content, I get
>a 'time object' back.
>This 'time object' is recognized as being of 'TimeType' and can be converted
>to an int.
>
>How can I convert an int into something that can be assigned to the Value
>property of the desired cell and be accepted according to the format as of
>above?
>
>TIA /BrorJ
>
>
the time values used by excel are double floats. The integer part
represents days, the mantissa the day fraction. There are two systems
based on
1==1/jan/1900 or 1==1/jan/1904; this because M$ morons didn't know that
1900 wasn't a leap year.
-- 
Robin Becker


From gmcm at hypernet.com  Tue Jun  6 15:15:19 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 06 Jun 2000 19:15:19 GMT
Subject: PyRun_SimpleString error handling
References: <393D062D.8413AC36@ncsa.uiuc.edu>
Message-ID: <8F4B9A166gmcmhypernetcom@199.171.54.194>

heiland at ncsa.uiuc.edu (Randy Heiland) wrote in 
<393D062D.8413AC36 at ncsa.uiuc.edu>:

>I'm trying to figure out how to detect/return an error when using
>PyRun_SimpleString.

As the Python/C API Reference Manual, Section 2. The Very High Level Layer 
says:

"""
int PyRun_SimpleString (char *command) 
Executes the Python source code from command in the __main__ module. If 
__main__ does not already exist, it is created. Returns 0 on success or -1 
if an exception was raised. If there was an error, there is no way to get 
the exception information. 
"""

In other words, PyRun_ stuff is pretty much only good for side effects. If 
you want to get results or poke around, you have to get your hands dirty.

>Here's my simple pgm:
>
>#include <Python.h>
>#include <import.h>
>#include <graminit.h>
>#include <pythonrun.h>

BTW, only the first include is required.

- Gordon


From akuchlin at mems-exchange.org  Wed Jun 28 12:46:48 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 28 Jun 2000 12:46:48 -0400
Subject: backslash in regular expression problem
References: <394A55BA.69269793@wolfson.co.uk>
Message-ID: <3d8zvpwy6f.fsf@kronos.cnri.reston.va.us>

Gordon McLeod <Gordon.McLeod at wolfson.co.uk> writes:
> I am trying to follow the advice of the Python Library Reference Manual,
> section 4.3 which says:
> 
>      However, if you want to include a literal backslash in a
>      regular expression represented as a string literal, you have
>      to quadruple it or enclose it in a singleton character class.

I think the manual is simply wrong on this point; try
re.compile(r'[\\]').

--amk


From phil at geog.ubc.ca  Wed Jun  7 19:54:14 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 07 Jun 2000 16:54:14 -0700
Subject: Calling python from C++.
References: <393eb924.99556671@nntp.interaccess.com>
Message-ID: <m3wvk1vyhl.fsf@brant.geog.ubc.ca>

olczyk at interaccess.com (Thaddeus L. Olczyk) writes:

> Is it possible for C++ to call Python routines and vica-versa?
> I have to write apps in C++ that have a GUI. They have to run under
> almost any platform, so I thought that a python front end might do the
> trick.
> TIA.

Take a look at http://CXX.sourceforge.net/

You'll want the current CVS snapshot, as there have been several
bugs fixed in CXX-4.2

Regards, Phil


From shapr at uab.edu  Sun Jun 11 19:11:49 2000
From: shapr at uab.edu (Shae Erisson)
Date: Sun, 11 Jun 2000 23:11:49 GMT
Subject: was: Pythonwin is terrible! now incremental parsing
References: <8hp2ku$a25@news.or.intel.com> <3940af8b.236270458@news.online.no> <8hrhqa$afm$1@news1.xs4all.nl> <G2g05.688$cQ.9254@news-server.bigpond.net.au>
Message-ID: <39441CDB.8A6E05EA@uab.edu>

Mark Hammond wrote:
[snip explanation of the difficulties of correct code completion]

> So if anyone can come up with generic code that can do this, all current
> and future IDEs for Python could benefit.
> 
> Mark.

If the python parser could be hacked up into an incremental parser, that
would solve all of the problems.

I have no idea how to do it, but Ka-Ping Yee's IPC8 Developers Day Notes
mention that Guido would like to have such a thing. You can find the
notes in the 'purpose statement' of the compiler-sig. The compiler-sig
is the place for this kind of thing to happen.

If someone wants to write an incremental parser for Python, I would
suggest looking at the javax.swing.text.* part of Java's Swing package,
the EditorKit functionality is the kind of framework that would allow
easy use of an incremental parser.
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre
   bRing mE fIve sQuirrels aNd nO oNe wIll gEt hUrt
13:00pm up 1 season, 1 squirrel, load average: 1+1j 0.5+2j 0.2+0.5j


From bjorn at roguewave.com  Wed Jun  7 13:27:58 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 07 Jun 2000 11:27:58 -0600
Subject: 1.6 regular expression copyright
Message-ID: <393E861E.E02A4849@roguewave.com>

I was just looking over some of the sources in the 1.6 alpha-2 release,
and I noticed that the sre files have a different copyright than what is
in the README file.  Is this by accident or by design?  (and if it is by
design, what are the practical consequences of this?)

not-a-copyright-lawyer'ly y'rs
-- bjorn



From sabren at manifestation.com  Tue Jun  6 07:36:49 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Tue, 6 Jun 2000 07:36:49 -0400 (EDT)
Subject: Local Databases with Python
In-Reply-To: <Fc3%4.37636$WS3.443234@typhoon.we.rr.com>
Message-ID: <Pine.BSF.4.10.10006060734120.9878-100000@chi.pair.com>

On Tue, 6 Jun 2000, Stephen Hansen wrote:

>     As for accessing MSAccess with ADO and DAO -- is that a COM derivation?
> To distribute an application that uses this technology, what all do I need
> to include in it to go about doing this? I'm not sure if its safe to assume
> that all machins have the nessecary files, drivers, and so on installed --
> and, if not, is it easily obtainable on the internet? For free? And what
> about redistribution of said nessecary files? (Beyond win32all and the
> python part)

If you don't care about whether or not it has SQL, you might want to
look at MetaKit:  http://www.equi4.com/metakit/  .. It works with C++
directly, and/or through python bindings.. I haven't had a chance to
use it yet, but from what I've seen, it looks quite powerful.

Cheers,

- Michal
------------------------------------------------------------------------
Zike Interactive    http://www.zike.net/    http://zike.sourceforge.net/
------------------------------------------------------------------------





From email at junked.nospam.com  Sat Jun 10 16:39:48 2000
From: email at junked.nospam.com (email at junked.nospam.com)
Date: 10 Jun 2000 21:39:48 +0100
Subject: windows programming
Message-ID: <8hu92k$28ae$1@walton.maths.tcd.ie>

I know that this is probably not really relevant to this newsgroup but I am 
new to Python (2 wks - looks cool so far!) and am trying to code a windows 
application using it. Part of the application is a graphing output, DISLIN 
looks good for this (any comments?). My problem is that I want to be able to 
give the user the option to print the graph on a user selected printer. Is 
there anywhere online that I can find a tutorial for this sort of thing (even 
if its not specifically in Python), or look at some sample code that involves 
these topics? 

thanks

-- 
David



From sugaking at gis.net  Sat Jun 24 16:48:40 2000
From: sugaking at gis.net (SugarKing)
Date: Sat, 24 Jun 2000 15:48:40 -0500
Subject: books????
Message-ID: <sla4ago0jev12@corp.supernews.com>

Are there any good beginner books for Python? I'm itching to get my hands on
any. Thanks




From tim_one at email.msn.com  Wed Jun  7 16:03:45 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 7 Jun 2000 16:03:45 -0400
Subject: math.exp underflow (was: exp behaviour: is it a bug ?)
In-Reply-To: <vmebt1dms3s.fsf@buffalo.fnal.gov>
Message-ID: <001301bfd0bb$7ccdff40$912d153f@tim>

[posted & mailed]

[Konrad Hinsen]
> I'd prefer a portable solution to one that reflects the system math
> library.

Konrad, make up your mind <wink> :  "portable" and "reflects the system" are
at each other's throats.

> I had the same problem under AIX, and I solved it by patching
> Python. I just wonder if someone has the patience to work out a
> patch that would create the same behaviour independently of what the
> math library does. Not me, unfortunately...

C gives almost no control over this, so it's not "a patch", but a whole
bunch of #ifdef'ed patches contributed by individual platform experts.  If
my new employer lets me, I'd like to work on setting up a usable framework
to make such patches possible and pleasant (& see below).


[Charles G Waldman]
> What do you think the consistent behavior should be?  Should
> math.exp(-10000) return 0, or raise some exception?

IEEE-754 has been around since the mid-80's and almost all platform HW
supports it directly now.  The next version (ISO's current version, in
fact!) of C ("C9X") mandates new features for good control over the 754
environment ("current" C offers *no* control).  So that's the only realistic
way to go.  In the 754 spirit, math.exp(-10000) should return +0 and set
"the underflow flag"; the user should also be given a runtime-settable
option to say whether or not "setting the underflow flag" also raises an
exception, and functions to query and set and clear the states of individual
754 flags (both the "did such-&-such a condition occur?" and the "when
such-and-such a condition occurs, should it raise an exception?" flags).
Guido and I agreed earlier that the "overflow", "divide by 0", and "invalid
operation" conditions should raise exceptions by default; "inexact" and
"underflow" should not; & all of those should be changeable by the user at
runtime.

python-*could*-be-an-outstanding-754-platform-ly y'rs  - tim





From alex at coolest.com  Mon Jun 26 00:00:01 2000
From: alex at coolest.com (Alexander Schonfeld)
Date: Mon, 26 Jun 2000 13:00:01 +0900
Subject: Python class tree viewer (with wxPython)
References: <3951AA42.13155B89@coolest.com> <yBGXTvAmEcU5Ew62@jessikat.demon.co.uk>
Message-ID: <3956D541.4471726E@coolest.com>

Which algorithm is that? :)

I wish I had payed more attention in alg analysis now.

Alex.

Robin Becker wrote:
> 
> In article <3951AA42.13155B89 at coolest.com>, Alexander Schonfeld
> <alex at coolest.com> writes
> >Hi,
> >
> >I wrote (I mean hacked a demo...for wxOGL) to show class trees of bunches of
> >python files.  You can then move them around and take screen shots and bask in
> >the glory that is your python classes.
> >
> >To use:
> >       python pyinherit.py d:\cvs\Zope2\lib\python\Products\LoginManager
> >       python pyinherit.py [specific .py files and dirs... whatever you want]
> >Dirs are recursively traversed with 'walk'... looking for py files.
> >etc...
> >
> >It requires wxPython (a gui lib) from:
> >       http://wxpython.org/
> >
> >Please expand on this if it looks interesting, and email me a copy.
> >
> >Thanks,
> >
> >Alex.
> >
> ...
> Wow! Now all we need is to apply that algorithm which tries to draw the
> diagram decently in planar form.
> --
> Robin Becker

--
--
--> Software ---> http://Coolest.com
-----> Art -----> http://Sk8r.net
-------------------------------------


From jhe at webde-ag.de  Mon Jun 26 05:18:03 2000
From: jhe at webde-ag.de (Jürgen Hermann)
Date: Mon, 26 Jun 2000 11:18:03 +0200
Subject: adding mail attachments
References: <8ivdl4$90k$1@supernews.com>
Message-ID: <8j771j$qj4$1@pollux.ip-plus.net>

"geoff munday" <geoffm at nojunkmailqsiuk.com> schrieb im Newsbeitrag
news:8ivdl4$90k$1 at supernews.com...
> Can anyone advise where to look for instructions or examples??

As an additional note, I discovered that MimeWriter does NOT add a
mime-version header (IMHO it should). That caused problems with my mailer,
refusing to parse the multi-part message correctly. So be sure to add this
to your code:

mail = MimeWriter.MimeWriter(mailtext)
...
mail.addheader("MIME-Version", "1.0")





From MarkH at ActiveState.com  Fri Jun 30 00:53:35 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 30 Jun 2000 04:53:35 GMT
Subject: win32print - adding some APIs?
References: <MPG.13c414ba715d9c249896b9@news-enteract>
Message-ID: <jJV65.1831$Tb7.13240@news-server.bigpond.net.au>

"Alan Miller" <ajm at enteract.com> wrote in message
news:MPG.13c414ba715d9c249896b9 at news-enteract...

> What I'm wondering is what it would take to extend win32print?  As it

Someone to do it :-)

> If nobody else is interested I can maybe take a hack at the source, but
> my current C++ experience is pretty much limited to modifying existing
> code (appropriate here) and I'm not quite sure what the procedure would
> be for getting changes incorporated....

The procedure is pretty simple:
* Make the changes.
* Mail them to me

:-)

Obviously, you get bonus points for a test/demo script, and ensuring that
the "autoduck" documentation is as good as you can make it without
actually running autoduck.

Mark.




From cplee at bigfoot.com  Tue Jun 20 02:12:20 2000
From: cplee at bigfoot.com (ChungpingLi)
Date: Mon, 19 Jun 2000 23:12:20 -0700
Subject: how to get file list via ftp connection
Message-ID: <394F0B44.33998BAA@bigfoot.com>

Hi friends
I have a problem
I use ftplib to access a ftp site but I do not
know how to copy all file from the ftpsite
anyone please help me with this
Andrea





From panu at ws23.pca.fi  Mon Jun 26 08:56:49 2000
From: panu at ws23.pca.fi (Panu A. Kalliokoski)
Date: Mon, 26 Jun 2000 12:56:49 GMT
Subject: Garbage collection versus reference counting
Message-ID: <lqI55.485$U3.22441@news.kpnqwest.fi>

I read the python FAQ at www.python.org and came across the question about
implementing automatic storage reclaim in python via garbage collection
instead of reference counting. The most important argument against
garbage collection was code that depends on reference-counting-based
cleanup actions.

This seems quite weird to me, as garbage collection and reference counting
by no means make it impossible to use the other. Is there something wrong
with the approach that both reference counting and garbage collection
are kept, and garbage collection is used for the objects that do not get
collected otherwise? Better still, all memory management could be left
for (generational) garbage collection, but other resources such as file
handles would be cleaned up by reference counting (or gc if that fails).

Is there something I've missed here?



From MarkH at ActiveState.com  Fri Jun 30 01:02:01 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 30 Jun 2000 05:02:01 GMT
Subject: pythonservice on NT
References: <8jfe0g$3pl$1@nnrp1.deja.com> <8F628CD32gmcmhypernetcom@199.171.54.194>
Message-ID: <dRV65.1835$Tb7.13237@news-server.bigpond.net.au>

markh at activestate.com - but if you think I speak for them, you dont know
Dick!
"Gordon McMillan" <gmcm at hypernet.com> wrote in message
news:8F628CD32gmcmhypernetcom at 199.171.54.194...
> Dale Burnett wrote:
>
> >Do all services run under one pythonservice process?
>
> No.

Gordon is (of-course :-) correct - no 2 Python services run in the same
process.  But it is true that all Python services generally share the same
executable (which is what I think you meant to ask!)

> >If not then is
> >there a way to change the name pythonservice in NT TaskManager to
> >something more descriptive?
>
> class MailFetchService(win32serviceutil.ServiceFramework):
>     _svc_name_ = "MailFetchService"
>     _svc_display_name_ = "Fetch SMTP mail on regular schedule"

Actually, that only changes the name in the "services" control panel, and
not the name in the task manager.  You can simply rename
"PythonService.exe" to anything else you like, and re-register the .exe.
Your service will also need special registration code to determine the
name of the .exe it wants to use.

It has been designed to work this way, so it wont be hard! However, I cant
recall the exact details - use the source, Luke :-)

Mark.





From grzegorz at ugrad.cs.ualberta.ca  Wed Jun 21 12:49:41 2000
From: grzegorz at ugrad.cs.ualberta.ca (Grzegorz Dostatni)
Date: Wed, 21 Jun 2000 10:49:41 -0600
Subject: Tkinter - image not found.
Message-ID: <Pine.BSO.4.10.10006211042360.18399-100000@gsb080.cs.ualberta.ca>

Hello. 
I'm trying to put an image on a frame.  What I am doing is something like
this:

label=Tkinter.Label(container)	
pic = Tkinter.PhotoImage(file="SAR.gif")
label["image"]=pic
label.pack()

This is essentially it.  Problem is, this code works when tested, but will
not work from my program.
Somehow the pic does not hold a valid reference after that second line.
(I think it loads a picture, but where does it go?)

Here's the trace:
Traceback (innermost last):
  File "main.py", line 18, in ?
    main()
  File "main.py", line 15, in main
    interface.Interface(cnx)
  File "interface.py", line 57, in __init__
    self.root=self.CreateRoot(cnx,lst[0])
  File "interface.py", line 166, in CreateRoot
    self.SetUpTopLevel(cnx, dict, frame)
  File "interface.py", line 227, in SetUpTopLevel
    self.TestPhoto(cnx,dict,frame)
  File "interface.py", line 255, in TestPhoto
    label["image"]=pic
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 630, in
__setitem__
    self.configure({key: value})
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 623, in
configure
    self.tk.call((self._w, 'configure')
TclError: image "1065616" doesn't exist


I'm out of ideas.  Any help will be greatly appreciated..
Greg

container is the top level widget.

########################################################################
SCSI is not magic. There are fundamental technical reasons why it
  is necessary to sacrifice a young goat to your SCSI chain now and
  then.                                                       -- Anon. 

"If you want to know what God thinks about money, just look at the people
He gives it to." 
-- Old Irish Saying 

Time is the best teacher. Unfortunately, it kills all its students. 



From hzhu at localhost.localdomain  Thu Jun 29 16:38:03 2000
From: hzhu at localhost.localdomain (Huaiyu Zhu)
Date: Thu, 29 Jun 2000 20:38:03 GMT
Subject: (Serious?) package namespace problem (and a proposal)
References: <slrn8lknve.1v9.hzhu@rocket.knowledgetrack.com> <m3k8f9cqcc.fsf@greebo.nodomain.de> <slrn8ll5vu.25h.hzhu@rocket.knowledgetrack.com> <m3pup0k8g7.fsf@greebo.nodomain.de>
Message-ID: <slrn8lnd34.1p5.hzhu@localhost.localdomain>

To summarize the thread I started:

1. The problem was caused by
 (1) relative import (or, rather, lacking support for global import)
 (2) Windows case insensitivity (only incidentally)
 (3) having a module name that is identical to global package name
 
   It is not caused by
 (1) having a '.' in sys.path
 (2) using import *

Thanks to Gordon McMillan for the clear explanation.

2. The immediate solution is to rename all module names to avoid global
package or module names.  This is not a long term solution for Python [1].

3. Another solution is to use the imputils module.  The docs take quite some
time to read.

4. The long term solution is to introduce a syntax for specifying global
imports.  Proposals include:

import .AAA
import magicroot.AAA  # magicroot in ["rootpath", "pythonroot", "python"]
import global AAA
global import AAA
import global.AAA
from global import AAA

The first is not explicit enough.  The second introduces a new keyword
unnecessarily.  The third introduces a new syntax.  The last three look
promising.  The last two fit completely within current syntax and will not
change the default behavior in any working code.

[Another proposal is to require relative imports to use 'import .AAA' but
that breaks most of existing code.]

5. There is one issue I mentioned that has not been followed up: The global
package name space cannot be the same as sys.path as it could contain '.' or
anything else.  It seems one additional attribute in sys is necessary:

sys.globalpath = ["/usr/lib/python1.5", "/usr/lib/python1.5/site-packages"]

Once these proposals are implemented, Python's module namespace would be
well encapsulated: Only top level package names need to be distinguished.
The module names within a package will not be affected by any name outside
the package.


-----------------
[1] Problems with avoiding using identical names:

Problem 1.  With this advice one has to avoid any existing or future global
package or module names that _might_ need to be imported in later
implementations.  Once a package gets widely used its module names cannot be
easily changed.  New package names also need to avoid using names of any
modules inside any existing or future packages.  For windows compatibility
"identical" is also case insensitive.

Such restrictions are simply too severe and unwarrantied.  And they go
against the elegance of Python exhibited in other aspects.  It is not
necessary to prefix variable names with indicators of their project names
because module name spaces provide good protection.  It would be a tragedy
(what else can I say) that one has to prefix module names with package
indicators because packages do not provide similar protection.

Problem 2. This is unlike variable names where: (1) there is a global
keyword, (2) global variables can still be qualified by module names.  In
contrast blocked package names are simply inaccessible, even if one is
willing to use a keyword or qualifier.

Huaiyu


From yawyi at gmx.de  Fri Jun  2 17:01:29 2000
From: yawyi at gmx.de (Johannes Stezenbach)
Date: Fri, 2 Jun 2000 23:01:29 +0200
Subject: Escapes???
References: <8h87u1$sga$1@news2atm.raleigh.ibm.com> <8h8jg3$d66$1@slb6.atl.mindspring.net> <8h8lkt$rb7$1@nnrp1.deja.com> <8h8qqs$l5a$1@slb6.atl.mindspring.net>
Message-ID: <39382a71$0$25922@personalnews.de.uu.net>

Aahz Maruch <aahz at netcom.com> wrote:
>In article <8h8lkt$rb7$1 at nnrp1.deja.com>, Arint?  <jamarijr at my-deja.com> wrote:
>>You must be using a different version or something.  I use the python
>>gui thing and did this
>>print "\x1ba"
>>got a funny circle
>>print "\x1b a"
>>[] a   < not an exact represetation.
>>But, it still points out the fact that python is picking up the 1 b and
>>a.
>
>Ah.  Python appears to allow three-digit hex numbers.  I guess you'll
>have to paste the string together, character by character.

Expecting this to be a bug in the Python string parser, to my surprise
I found that the Python language reference says in section 2.4.1
about string literals (section numbers from 1.5.2 docs):

"""... an unlimited number of hex digits is taken to be part
of the hex escape (and then the lower 8 bits of the resulting hex
number are used in 8-bit implementations).
"""

Using the information from section 2.4.2 about string literal
concatenation you can write:
   print "\x1b" "a"


I'm curious to know which platform has characters with more
than 8 bits, except for Unicode in Python 1.6, of course, but
there's \u for that purpose -- with \x1234 being equivalent to
\u1234 in u"" strings:
>>> "\x1234"
'4'
>>> "\u1234"
'\\u1234'
>>> u"\x1234"
u'\u1234'
>>> u"\u1234"
u'\u1234'
>>> 

Are more-than-two-digit hex escapes obsolete? Were they ever
useful? Are they part of devilish plan to confuse former
C programmers? Is \u vs. \x a symptom of TMTOWTDI?

Johannes



From aahz at netcom.com  Tue Jun 27 11:49:25 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jun 2000 15:49:25 GMT
Subject: float literal too large
References: <ulsntz2kra.fsf@runner.princeton.idaccr.org>
Message-ID: <8jaie5$vpf$1@nntp9.atl.mindspring.net>

In article <ulsntz2kra.fsf at runner.princeton.idaccr.org>,
Victor S. Miller <victor at idaccr.org> wrote:
>
>I just wrote a python program to read output produced by a Fortran
>program that somebody had written.  Much to my surprise I got the
>exception:
>
>float() literal too large.
>
>The literal in question was '0.23489798479873E-2325' (this was
>generated on a Cray computer).  

I'm guessing the exponent is too large (not the mantissa).  In case you
don't hear from someone who actually knows something about floats in
Python, I suggest you try different exponents to see if that is in fact
the problem.  If so, I suggest you create a string operation to manually
underflow this to zero (or possibly NumPy would help).
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From pgf at globalreach.net  Sat Jun 10 02:52:53 2000
From: pgf at globalreach.net (Phil Fraering)
Date: 10 Jun 2000 01:52:53 -0500
Subject: Damnation! (was: Re: Python3k extended grammar)
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <39260037.146C3467@san.rr.com> <slrn8ic24u.sc3.wtanksle@dolphin.openprojects.net> <39260B99.86CBF73C@san.rr.com>
Message-ID: <87itviowmy.fsf@globalreach.net>

Courageous <jkraska1 at san.rr.com> writes:

> I appreciate what you're trying to say, but if Python ever
> becomes a mainstream programming language, I can guarantee
> you that it will outgrow Guido.

I'm new to Python, but I was under the impression that

a) It was a mainstream programming language, and

b) That the language has a consistent core managed by one
person with a clear vision is an asset; the language doesn't
have to be huge (in grammar and vocabulary) to be useful.

While I am scared about such things as an all-uppercase future,
it's just one of the risks we have to take.



-- 
Phil Fraering         "One day, Pinky, A MOUSE shall rule, and it is the
pgf at globalreach.net    humans who will be forced to endure these humiliating
/Will work for tape/   diversions!"
                      "You mean like Orlando, Brain?"


From aahz at netcom.com  Mon Jun 26 17:36:34 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jun 2000 21:36:34 GMT
Subject: stopping threads
References: <3957C1DC.E7C77E2E@sec.noaa.gov>
Message-ID: <8j8id2$v3h$1@nntp9.atl.mindspring.net>

In article <3957C1DC.E7C77E2E at sec.noaa.gov>,
j vickroy  <jvickroy at sec.noaa.gov> wrote:
>
>This may not be a Python question per se, but ...
>
>Why is there no stop_execution() function/method in either the thread or
>the threading modules?  It seems that if I want to stop an executing
>(sub)process, I must launch it using os.fork() which is platform
>specific.

I'm currently reading up on Java threads, and while such a function
exists, the authors recommend very strongly against doing so.  Instead,
your thread should watch for some condition and return/exit from its
main function/method when that condition becomes true.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

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


From olczyk at interaccess.com  Mon Jun 19 14:30:18 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 19 Jun 2000 18:30:18 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com> <3eomks8pah3at7h9ponni0r7s8ekt4ukh9@4ax.com> <39520785.690843718@nntp.interaccess.com> <3p0sksgnqhso5dsp1cmf5srffbeh42ov0g@4ax.com>
Message-ID: <394e5f98.844462828@nntp.interaccess.com>

On Mon, 19 Jun 2000 06:30:20 -0500, "Bob Hays, Computer Geek"
<rhays at interaccess.com> wrote:

>
>Good for you - you've been perfect and never, due to time constraints,
>checked something in that had a debug line in it.  Congrats!
>
Gee and guess what? I actually put in revision logs that fairly
describe my changes most of the time ( some times it can be difficult
or impossible, if you limit yourself to roughly ten lines, which you
should ). It's the little things like this ( as sports announcers like
to point out ) which seperate the amateur from the proffessionals.

>>>>2) You can easily insert a tab introducing a bug.  Consider the
>>>>following code ( similar to a bug I recently fixed ):
>>>{Example skipped}
>>>>Initially the line in error was indented correctly, but someplace
>>>>along the line the extra tab was introduced. A lot of effort then had
>>>>to go in to finding this bug. With begin...end or braces this is less
>>>>likely to happen.
>>>
>>>Yep, this happens.  Its a tradeoff - speed vs. occasional extra work.
>>What increases in speed do you get by having to type braces?
>
>The code can be typed in much faster; its SLOWER to type in the
>braces.
>
Are you saying that you should choose a programming language based on
how long it takes to type in things? If so then you should be using
perl not python. The experts all agree that it is much faster to type
in perl than python.

Me, I prefer a language which helps me write good code. Faster to type
or slower to type doesn't make much difference to me. Of course if you
want something that is faster to type, you can always go to perl.

>>>In addition, I had an example, in C++, where someone had commented out
>>>a block of code, but then had an embedded inline comment inside that
>>>block, and, to top it all off, the less-than-desired commented out
>>>block would still compile - the guy worked on that bug for a day
>>>before coming to me.  I loaded it into emacs and the color coding told
>>>him immediately what had happened; he was very ashamed and angry, but
>>>I told him sh*t happens:-).
>>>
>>But that's the point. His code was illegal, it wouldn't compile. Easy
>>to catch with emacs, and the compiler will catch it too. The little
>>typo I show is easy to make and perfectly legal code. Its hard to
>>accidentally bump against a keyboard and make a similar nasty typo.
>>EVen something that looks like it might work like }{ would have to be
>>put in just the right spot.
>
>No, you didn't read carefully.  His could *WOULD* compile, but it ran
>somewhat wrong (obviously) and he wasted a sinful amount of time and
>energy trying to find the problem.
>
Your right I didn't read all the way in. ASAIK there is no way
to commant out a block of code, so that it contains a comment,
which causes part of the code not to be commented, and still compile.
You can do everything up to still compile, but it should not compile.
Care to give an example?

>>>>3) Finally. Python is a language where it is almost easy to write
>>>>extensible programs. To illistrate what I mean consider this example
>>>>of a python program I recently wrote:
>>>
>>>Here's were we will agree to disagree, I think.  I think its way
>>>faster and easier to extend Python stuff than C++ (and Java) due to
>>>the lack of compile/link cycle and due to inheritance inside of
>>>Python.  Yep, you have to be careful how you indent, but if you have a
>>>good editor it will cut/copy/paste as-is without reindenting.  I see
>>>more problems due to bad cut/copy/paste reuse in all programming
>>>languages than I do from mis-indenting so far.
>>>
>>I think you misunderstood what I said. I think that Python is not now
>>easily extensible do to the  problems with whitespaces. If that were
>>fixed then Python would be easily extensible. 
>{lots removed}
>
>But I do understand.  Good reuse would be to not copy but to reuse via
>inheritence or reinvocation - copy/paste reuse is always fraught with
>problems, and that's part of what OOP should help avoid (although I've
>always thought structured programming (SP) could deal with this
>problem also - its code size that limits SP).

No you don't. Did you afctually try to read the part I snipped? Guess
not. You need to cut/paste code ( aka move ) when refactoring. Many
examples in Fowlers Refactoring involve moving code. Just one, page
144 moving the overDraftCharge from Account to AccountType. Often
refactoring can't be done without moving code.

Example:

you have this bit of code

...
if x in xList:
      x.dosomething()
      x.dosomethingelse(y)
      if x.meetscondition(y,z):
          z.fixthings(y)
...

Some place far away, you realise that you are about to type
...
for a in xList:
      a.dosomething()
      a.dosomethingelse(b)
      if a.meetscondition(b,c):
          c.fixthings(b)
...
So what do you do?
Do you just type in the code? ( Great example of reuse )
Do you create a function and type in the code?

def foo(x,y,z):
      x.dosomething()
      x.dosomethingelse(y)
      if x.meetscondition(y,z):
          z.fixthings(y)

and change the code to

if x in xList:
   foo(x,y,z)

and 

for a in aList:
    foo(a,b,c)

Or do you move the code from the first routine to the function?



From emile at fenx.com  Thu Jun 15 10:27:12 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 15 Jun 2000 07:27:12 -0700
Subject: TEXT MANIPULATION (in CSV format)
References: <20000615133826.11516.qmail@web1502.mail.yahoo.com> <3948E40C.2425B170@bam.com>
Message-ID: <130601bfd6d5$d29c4ce0$1906a8c0@fc.fenx.com>

You'll need to be more careful than Bill pointed
out in his reply.  If you know your source data
is clean, you should have no problems.  Otherwise,
proceed cautiously.  In particular, if you must parse
CSV, watch out for embedded commas, eg: name fields
like "Abe Lincoln, Pres" or address lines like
"123 Main, Apt C".  You'd be better off if you
could avoid CSV files by seeing if, for example,
tab delimited files could be provided instead.

It-can-be-parsed-I-just-don't-like-to-ly y'rs,

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: Bill Scherer <scherbi at bam.com>
To: Robin Porter <robinporter72 at yahoo.ca>
Cc: <python-list at python.org>
Sent: Thursday, June 15, 2000 7:11 AM
Subject: Re: TEXT MANIPULATION (in CSV format)


>
>
> Robin Porter wrote:
>
> > I am wondering how difficult it would be to use PYTHON
to go through a
> > CSV (Comma separated value) formatted file, take two of
the fields and
> > merge them into one field.
> >
> > An example would be:
> >
> > Field 1             Field 2
> >
> > "K"           ,       "1234"
> >
> > I want to change to:
> >
> > Field 1
> >
> > K1234
> >
>
> Easy:
>
> -----------------------
> #untested...
> import string
>
> split = string.split
> strip = string.strip
>
> filename = "/some/path/and/file.csv"
>
> newFile = open(filename + '.new', 'w')
>
> for line in open(filename).readlines():
>     f1, f2 = map(strip, split(line, ",",  1))
>     newFile.write("%s%s\n" % (f1, f2))
>
> newFile.close()
> -----------------------
>
> This will work fine if your original file is not huge.
> If your file is huge, you'll want do your readlines in
chuncks so as not
> to swamp memory.
>
>
>
> > Thanks in advance for your assistance.
>
> Your welcome.
>
>
> --
> William K. Scherer
> Sr. Member of Applications Staff
> Verizon Wireless
>
>
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From wfouche45 at hotmail.com  Wed Jun  7 00:02:15 2000
From: wfouche45 at hotmail.com (Werner Fouche)
Date: Wed, 7 Jun 2000 00:02:15 -0400
Subject: Python <--> XML data binding.
References: <8hhlr6$eoh$1@slb3.atl.mindspring.net> <393C6944.2FBCE9B1@prescod.net>
Message-ID: <8hkhge$j93$1@nntp9.atl.mindspring.net>

Paul,

I should have mentioned that Breeze reads the extended DTD specification
called DT4DTDs developed by Extensibility -- their XML Authority Editor
can convert many XML schema language dialects to DT4DTDs. DT4DTDs
extend standard DTDs with the data types found in W3C XML Schema
language specification.

It would be easy to support DT4DTDs once you can process standard DTDs.
This is ideally what I am looking for since I want to be able to model XML
document types in XML Authority and then generate Python code from the
specification. In this way the Python code and the XML document definition
can easily be kept in sync.

The biggest benefit of a tool such as what you are developing would be to
quickly and reliable translate industry standard XML document specifications
(say OFX 2.0) to crisp Python code! I am looking forward to the fruits of
your current endevour.

Regards,

Werner

"Paul Prescod" <paul at prescod.net> wrote in message
news:393C6944.2FBCE9B1 at prescod.net...
> Werner Fouche wrote:
> >
> > Does anyone know if a tool exist that will generate Python data binding
code
> > for an XML DTD similar to what Breeze ( http://www.breezefactor.com )
does
> > for Java.
>
> I'm working on such a library. Check back in a couple of weeks.
>
> Nevertheless, the Breeze website vastly exaggerates the difficulty of
> doing object creation "by hand" using the output of a parser. I wouldn't
> even bother with my data binding technology if the data binding did not
> come "for free" as a side effect of other features I need.
> --
>  Paul Prescod
> Those who profess to favor freedom and yet depreciate agitation, are men
> who want crops without plowing up the ground. They want rain without
> thunder and lightning. They want the ocean without the roar of its many
> waters. - Fredrick Douglass
>  http://www.informamerica.com/Articles/Quotes.htm
>




From sh1232 at yahoo.com  Tue Jun  6 11:38:44 2000
From: sh1232 at yahoo.com (sh)
Date: Tue, 6 Jun 2000 10:38:44 -0500
Subject: TypeError: unhashable type
Message-ID: <393d1c1a_2@news1.prserv.net>

This works without a glitch in the command line interpreter:
    segments = {}        # dictionary type
    segment = []
    # populate segment
    segments[segment[0]] = segment[len(segment)-1]

BUT, when I run it from the DOS prompt: "python segments.py", I got error
message:
    TypeError: unhashable type.

Any explanation is appreciated.  Thanks.
- sh




From moshez at math.huji.ac.il  Tue Jun  6 13:46:57 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 6 Jun 2000 20:46:57 +0300 (IDT)
Subject: Spurious coercing? [was: Re: 'if foo' calls getattr?!?]
In-Reply-To: <oqg0qqdb35.fsf@titan.progiciels-bpi.ca>
Message-ID: <Pine.GSO.4.10.10006062045150.1650-100000@sundial>

On 6 Jun 2000, [ISO-8859-1] Fran?ois Pinard wrote:

> And I would presume that if __cmp__ happened to be found, a call to
> __coerce__ would be tried before going any further?

Yes, indeed. The reason is that Python tries very hard *not* to be smart:
the opposite of smart, in this case, being predictable. If you know
__coerce__ is always called before binary operators take place, your code
can be that much simpler.


--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From gward at python.net  Fri Jun 30 00:26:20 2000
From: gward at python.net (Greg Ward)
Date: Fri, 30 Jun 2000 00:26:20 -0400
Subject: ANNOUNCE: Distutils 0.9
Message-ID: <20000630002620.A2329@beelzebub>

                     Python Distribution Utilities
                              release 0.9
                             June 29, 2000

The Python Distribution Utilities, or Distutils for short, are a
collection of modules that aid in the development, distribution, and
installation of Python modules.  (It is intended that ultimately the
Distutils will grow up into a system for distributing and installing
whole Python applications, but for now their scope is limited to module
distributions.)

The Distutils are a standard part of Python 2.0; if you are running 2.0
(or one of the 1.6 alpha releases that preceded it), you don't need to
install the Distutils separately.  This release is primarily so that you
can add the Distutils to a Python 1.5.2 installation -- you will then be
able to install modules that require the Distutils, or use the Distutils
to distribute your own modules.

More information is available at the Distutils web page:

    http://www.python.org/sigs/distutils-sig/

and in the README.txt included in the Distutils source distribution.

You can download the Distutils from

    http://www.python.org/sigs/distutils-sig/download.html

Trivial patches can be sent to me (Greg Ward) at gward at python.net.
Larger patches should be discussed on the Distutils mailing list:
distutils-sig at python.org.



From richard_chamberlain at ntlworld.com  Sun Jun  4 06:35:21 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Sun, 4 Jun 2000 11:35:21 +0100
Subject: How to find out height of a Canvas
References: <memo.20000603235615.14417A@digitig.compulink.co.uk>
Message-ID: <eaq_4.3871$JO2.84543@news2-win.server.ntlworld.com>

Hi Tim,

from Tkinter import *
root=Tk()
canvas=Canvas(root,height=250,width=100)
canvas.pack()
print canvas ['height']

canvas.itemcget has two arguments, the first is the tag or id of a
particular widget on the canvas and the second is the option you want back.

Richard

Tim Rowe <digitig at cix.co.uk> wrote in message
news:memo.20000603235615.14417A at digitig.compulink.co.uk...
> How do I find out the height of a Tkinter Canvas? canvas.itemcget looks
> promising, but it needs two arguments (I get a Tcl error if I omit the
> second argument, or set it to None), and I can't find any documentation on
> what the second argument should be (or the first argument, for that
> matter; I'm assuming /that/ should be a string containing the name of the
> attribute).
>
> TIA.




From piet at cs.uu.nl  Sun Jun  4 11:52:59 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 04 Jun 2000 17:52:59 +0200
Subject: Division considered un-Pythonic (Re: Case-sensitivity: why -- or why  not? (was Re: Damnation!))
References: <Pine.GSO.4.21.0005181437530.2583-100000@z.glue.umd.edu> <392456A9.993FB15C@concentric.net> <39245ED7.9C3A4FAB@prescod.net> <Pine.GSO.4.21.0005191815350.22081-100000@y.glue.umd.edu> <vrya5615eg.fsf_-_@bor.IRO.UMontreal.CA> <cphfbtvk9p.fsf@cj20424-a.reston1.va.home.com> <8g58j4$11ei$2@thoth.cts.com> <8g6abe$s5m$17$1@news.t-online.com> <cppuqg4hsu.fsf_-_@cj20424-a.reston1.va.home.com> <8grnj1$sb3$13$1@news.t-online.com> <slrn8j5k1c.8a7.wtanksle@dolphin.openprojects.net> <393326D3.400AA248@cosc.canterbury.ac.nz> <u1z2gl8zu.fsf@cs.uu.nl> <eERZ4.2206$r7.561638@news1.rdc2.tx.home.com>
Message-ID: <wzhfb9o338.fsf@sunshine.cs.uu.nl>

>>>>> "Rainer Deyke" <root at rainerdeyke.com> (RD) writes:

RD> <piet at cs.uu.nl> wrote in message news:u1z2gl8zu.fsf at cs.uu.nl...
>> I want to add another reason for a/b to mean floating division:
>> 
>> As it is now in Python, you can have a==b and c==d both being true, but
>> a/c==b/d being false. Which would be a bad surprise for most people.

RD> But, with floating point numbers, you can have (a + b) - b == a be false.

I would argue that this is a different kind of unequalness, it has to do
with accuracy. In the integer div case it is a semantic difference.

RD> The only way to avoid this sort of situation is to avoid floating points
RD> entirely.  I don't want my programs to become infested with floats just
RD> because of one innocent division.

Therefore Python should offer a div operator for integer division. (Or use
int()).
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From mfletch at tpresence.com  Fri Jun 16 06:37:03 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 16 Jun 2000 06:37:03 -0400
Subject: Python segfault
Message-ID: <CC59868A0A76D311B1F50090277B321C1F33BF@VRTMAIL>

Not really insight, but Python's grammar is such that you create nested sets
of expressions when you create what looks like a set of simple adds...

2+2+2+2 -> (((2+2)+2)+2)

So you get a rather deep expression tree in your examples ;) .  I'm guessing
something doesn't like creating (or evaluating) this, though whether it's
the parser I don't know.

Enjoy,
Mike

-----Original Message-----
From: Peter Schneider-Kamp [mailto:petersc at stud.ntnu.no]
Sent: Friday, June 16, 2000 2:30 AM
To: perly at xnet.com
Cc: python-list at python.org
Subject: Re: Python segfault


Pat Szuta wrote:
> 
> I don't know if anyone has seen this,  but python segfaults when given a
> rather large input.
> 
> [pat at perly pat]$ perl -e "print '2+2+' x 10000" > input
> [pat at perly pat]$ python < input

Strange. Looks like the limit is exactly at 8192. That makes 16384
operations the parser (or whatever crashes) can hold.

Not that the problem is not the length of the input itself:

$ perl -e "print '22222222 ' x 200000 > input
$ python < input

works just fine. I think the problem is the add (or mul) operation.
So the problem might be in the expression parser.

Is there anyone with some more insight around?

else-I-will-have-to-check-it-myself-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de

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



From fstajano at uk.research.att.com  Tue Jun 27 13:34:25 2000
From: fstajano at uk.research.att.com (Frank Stajano)
Date: Tue, 27 Jun 2000 17:34:25 GMT
Subject: Q: memory ownership in python extensions
Message-ID: <3958e1ee.13405033@localhost>

This is a very basic question from someone who has been using Python
for quite a while but only just started writing an extension module.

In the C code, I malloc some memory, perform some processing that
writes a string in it, then turn this C string into a Python string
via Py_BuildValue. The question is: Who is responsible for freeing
that memory? Example code fragment follows.

             ...
	ciphertext = (byte*) malloc(pLen+1); /* who frees this? */
	if (!ciphertext) {
		PyErr_NoMemory();
		return NULL;
	}
	encrypt_block(key, kLen, plaintext, pLen, ciphertext);
	return Py_BuildValue("s#", ciphertext, pLen);

It would be logical and symmetrical for the code who malloc'ed to be
the one responsible for free'ing. For this to happen, Py_BuildValue
would have to take a copy. But I strongly suspect that, for efficiency
reasons, Py_BuildValue does not take a copy and just reuses the memory
passed to it; in which case the free would be done by the Python
system when the reference count of the Python object goes to 0.

Which one of these is the actual behaviour of Py_BuildValue is
something that isn't at all apparent from Chapter 1.9 of "Extending
and embedding" version 1.5.2 (hint hint). I also couldn't find any
reference to this issue in the FAQ by searching for malloc or
Py_BuildValue. Feel free to embarrass me by pointing me at the bit I
missed...



From thomas at xs4all.nl  Mon Jun 26 14:45:13 2000
From: thomas at xs4all.nl (Thomas Wouters)
Date: 26 Jun 2000 18:45:13 GMT
Subject: compilation
References: <8j84cf$6bj$1@bmerhc5e.ca.nortel.com>
Message-ID: <slrn8lf95c.fu7.thomas@centurion.xs4all.nl>

On 26 Jun 2000 17:37:19 GMT, Michael P. Soulier
<msoulier at nortelnetworks.com> wrote:
>
>	Hey people. I tried the FAQ, and I can't find this. I'm a Perl 
>programmer exploring Python, and I'm wondering how to do a simple compile pass 
>on python source to check syntax. In Perl, I'd just do a perl -cw sourcefile
>and it would do a quick compile pass with warnings on. I know I can start the
>interpreter and import the file as a module, but that's not always desirable.
>Is there a simple command-line way to do this?

No. (other than 'python -c "import <myscript>"')

It's not that big a deal, though, because the only thing done is syntax
checking -- and not code validity, as perl -cw does. 99% of Pythons' errors
are runtime errors, especially since true syntax errors are generated right
away, preventing a legion of other errors from popping up at the same time
;-)

Python also doesn't give warnings, unless you use python -t, in which case
it'll give a warning about inconsistent indentation (mixing tabs & spaces).
In short, a 'python -cw' wouldn't be too useful anyway. You're probably
better off with some decent unit testing ;-)

(Then again, contrary to my C and Perl code, both a fair bit, my Python code
almost never has syntax/programming errors, and most of my programs are
simple enough to do the unit testing 'by hand' ;)

perl-ly y'rs,
	Thomas.


From pbt at pipeline.com  Sat Jun 17 04:10:03 2000
From: pbt at pipeline.com (Philip Thorne)
Date: Sat, 17 Jun 2000 08:10:03 GMT
Subject: Ret status technique popen, popen2 (NT)
References: <3949a528.1161643@news.pipeline.com> <2Kq25.372$qz4.3272@news-server.bigpond.net.au> <hgC25.714$qz4.6314@news-server.bigpond.net.au>
Message-ID: <394b2fed.3375663@news.pipeline.com>

Thanks, I'm on W95a for the weekend so have given it a try on that
with mixed results. The zop test script works well but I haven't
manged to get output from anything else ("dir") as the pipe appears to
close immediately. e.g.

=== start ===
Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import piper
>>> p=piper.PipedCommand("c:\\utils\where lha.exe")
>>> print p.closed()
1
>>> print p.exitstatus()
0
>>> from os import popen
>>> w=popen("c:\\utils\where lha.exe")
>>> print w
<open file 'c:\utils\where lha.exe', mode 'r' at 79aa00>
>>> e=w.readlines()
>>> print e
['C:\\UTILS\\LHA.EXE\012', 'C:\\UTILS\\TOOLS\\LHA.EXE\012']
>>> p.close()
=== end ===

I'll try again tomorrow when I'm a little more alert.
Philb


"Neil Hodgson" <neilh at scintilla.org> wrote:

>    Thanks to some comments from David Bolen, piper has been made more
>robust. I've also fiddled with the code and added some comments.
>
>http://www.scintilla.org/piper.zip
>
>    Neil
>




From alan.gauld at gssec.bt.co.uk  Mon Jun 19 07:20:29 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Mon, 19 Jun 2000 12:20:29 +0100
Subject: python is hard to learn without c
References: <7QG15.2637$sI4.335806@dfiatx1-snr1.gtei.net>
Message-ID: <394E01FD.8138F16E@gssec.bt.co.uk>

jin choung wrote:
> anyone else feel this way?

Yes, Even O'Reilly agree. There is an article on the 
ora.com website specifically saying so.

It indirectly recommends Ivan Lanningham's book:
Teach Yourself Python in 24 hours.

You also might try downloading my online tutor which
is aimed at complete beginners with a view to getting 
you to the point where you can follow the official 
tutorial.

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

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


From thomasm at legend.co.uk  Tue Jun 13 06:55:05 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Tue, 13 Jun 2000 11:55:05 +0100
Subject: No Boolean type in "types"
Message-ID: <394612a2.0@katana.legend.co.uk>

Hello,

Can someone explain me why the module "types" have no boolean type defined ?

true = (0==0)
false = (not true)

BooleanType = type(0==0)

Thank you

Thomas





From parkw at better.net  Fri Jun  2 17:32:30 2000
From: parkw at better.net (William Park)
Date: Fri, 2 Jun 2000 17:32:30 -0400
Subject: Bad Math Module Notation re:  atan2
In-Reply-To: <010620001459577031%pecora@anvil.nrl.navy.mil>; from pecora@anvil.nrl.navy.mil on Thu, Jun 01, 2000 at 02:59:57PM -0400
References: <010620001459577031%pecora@anvil.nrl.navy.mil>
Message-ID: <20000602173230.A840@better.net>

On Thu, Jun 01, 2000 at 02:59:57PM -0400, Lou Pecora wrote:
> Maybe you've all kicked this around before, but I've just picked up
> Python in the last month or so.  The notation for atan2(x,y) is 
> 
>  atan2(x,y)= arctangent( x/y)
> 
> Given the usual association of x with the abscissa and y with the
> ordinate in mathematics, science and engineering it is easy to mistake
> atan2(x,y) as arctangent (y/x) (x and y reversed!).  Or in another
> view, according to the majority of mathematically trained people
> atan2(x,y) as defined appears to deliver the arccotangent.  
> 
> Am I mising something?

If you look at it as 'atan2(y,x) = arctangent(y/x)', then it makes
sense.

--William



From gmcm at hypernet.com  Thu Jun 29 12:51:43 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 29 Jun 2000 16:51:43 GMT
Subject: Module Name Problem
References: <eviltofu-C678D0.18293328062000@news.earthlink.net> <8F61CAE59gmcmhypernetcom@199.171.54.194> <eviltofu-E613EC.21014728062000@news.earthlink.net>
Message-ID: <8F628AC42gmcmhypernetcom@199.171.54.194>

Jerome Chan wrote:

>In article <8F61CAE59gmcmhypernetcom at 199.171.54.194>, gmcm at hypernet.com 
>(Gordon McMillan) wrote:

[snip]

>> Not if you want access to os within maxvu.
>> 
>> - Gordon
>
>Will this be fixed in 1.6?

No.

- Gordon


From olipt at mayo.edu  Thu Jun 15 14:37:13 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Thu, 15 Jun 2000 13:37:13 -0500
Subject: SWIG + NumPy - any examples ?
In-Reply-To: <8ib5eu$6ck$1@nnrp1.deja.com>
References: <8ib5eu$6ck$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.10.10006151334070.27249-100000@us2.mayo.edu>

> If anyone has done this, I would appreciate an example so I can
> get started.

This is what I did in fftw-numpy 

Taken from dfftw.i in fftw-numpy-0.6:

%init %{
        import_array();
%}

%typemap(python,in) fftw_complex * {
  /* Make sure input is a NumPy array and a Complex Double */
  PyArrayObject *arr;
  if ((arr = (PyArrayObject *) PyArray_ContiguousFromObject($source,
PyArray_CDOUBLE, 0, 0)) == NULL) return NULL;
  $target = (fftw_complex *)arr->data;
  $source = (PyObject *)arr;
}

%typemap(python,freearg) fftw_complex * {
  Py_XDECREF($arg); 
}  




From moshez at math.huji.ac.il  Wed Jun 14 22:48:17 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 15 Jun 2000 05:48:17 +0300 (IDT)
Subject: Which linux distros have Python by default?
In-Reply-To: <aTS15.654$i06.206149@ptah.visi.com>
Message-ID: <Pine.GSO.4.10.10006150546090.5976-100000@sundial>

On Wed, 14 Jun 2000, Grant Edwards wrote:

> Nah, I wrote a TCL program once and swore never again.  Since
> this is just a GUI front-end for some command line utilities,
> the user who doesn't have Python/Tkinter can get along fine
> without it.

Suggestion: write a Python module to wrap the command line utilities, and
have two UIs available: a Python/Tkinter GUI and a cmd-based high-level
command line utility.

Exercise for the advanced reader: write a Python/Curses, a Python/GNOME
and a Python/KDE GUIs too <0.9 wink>
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From abc at spud.dinkbrain.org  Sun Jun 11 00:33:10 2000
From: abc at spud.dinkbrain.org (Mahogany Rush)
Date: Sun, 11 Jun 2000 04:33:10 GMT
Subject: Wrapper for pcap?
References: <8huolp$olv$1@bob.news.rcn.net>
Message-ID: <slrn8k65c5.eh.abc@sparc5.localnet>

On Sat, 10 Jun 2000 21:04:05 -0400, Z 3 Penguin <z3penguin at penguinpowered.com>
 wrote:
 >> -----BEGIN PGP SIGNED MESSAGE-----
 >> Hash: SHA1
 >> 
 >> Does anyone have/Has anyone seen a Python wrapper for libpcap?
 >> 
 >> 

Try http://www.vex.net/parnassus/

Click on networking, then sockets.




From richard_chamberlainNOriSPAM at ntlworld.com.invalid  Wed Jun 28 10:02:26 2000
From: richard_chamberlainNOriSPAM at ntlworld.com.invalid (Richard Chamberlain)
Date: Wed, 28 Jun 2000 07:02:26 -0700
Subject: Problem with os module
References: <3959FF57.85E1A503@mc2.fr>
Message-ID: <192c2800.183750f0@usw-ex0105-038.remarq.com>

Hi Aline,

What you're doing (from os import *) is importing the contents
of os into your current namespace.

So path is available without the os. at the beginning.

It's not a good idea to do this though because if os uses any
method or variable names that you are using they'll get wiped
over.

So with that in mind you could either do:

from os import *
if path.isfile("c:\\Test"):

(not a good idea)

or:

import os

if os.path.isfile("c:\\Test"):

(much better)

b.t.w. you'll notice above that I've used c:\\Test - the reason
is that \ has a special meaning within a string. For instance \t
is a tab and \n a newline, \\ is in fact a single \. So your
paths should look something like "c:\\my folder\\my file".

>> I'm new on the list, and I'm french

I think we'll cope.

Richard



Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From fiona at sitegnome.com  Sat Jun 10 09:09:25 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sat, 10 Jun 2000 23:09:25 +1000
Subject: "Announce": Python Snippets site on hold
References: <200006092242.e59Mg9K29454@hera.cuci.nl>
Message-ID: <39423E05.DEE93153@sitegnome.com>

> 2. There is such a thing as Python FAQTS (the Python knowledge base); the
> idea is roughly similar to that of the snippets site. I offered them my
> snippets (published and unpublished); it's up to them if you will actually
> see them in the knowledge base.

Hi Hans,

It seems that my mail the other day might not have reached you....

FAQTS will be very happy to incorporate your snippets into the python
knowledge base.

I wrote :

> Hi Hans,
> 
> Thanks for the offer!  We'd be really happy to incorporate your work
> into the Python knowledge base.
> 
> It would be great if you could send a zipfile to Nathan -
> nathan at synop.com .  He's responsible for the entire FAQTs site and will
> be able to put all your content into the database directly.

cheers, Fiona Czuczman


From rumjuggler at cryptarchy.org  Sun Jun  4 15:34:20 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 04 Jun 2000 19:34:20 GMT
Subject: deque vs list: performance notes
References: <3933484B.B471017A@san.rr.com> <8h34v4$7ac$1@nnrp1.deja.com> <393530D7.D495D3F9@san.rr.com> <39353A47.10EC15FB@do.not.use> <39390A03.560F4225@tismer.com> <39392A62.BE04911@do.not.use> <393931E4.4283BD7F@tismer.com> <39395BE1.B52756D7@san.rr.com> <39398EEE.C18EC987@tismer.com> <3939903D.79BD8D2D@san.rr.com> <393AA136.36C34915@tismer.com> <393AAC0F.FFB4B65@san.rr.com>
Message-ID: <7sbljso0tvth02de6ql5at6100pns97f3h@4ax.com>

On Sun, 04 Jun 2000 19:18:03 GMT, Courageous <jkraska1 at san.rr.com>
wrote:

>it would also be possible to allow my deque
>implementation to be concatenated in as well, but I didn't do it,
>because I didn't want to force the module dependency.

Surely that code could be wrapped in #ifdef/#endif?

-- 
Barnabas T. Rumjuggler

That happens to me often: I'll find an old coffee on the bookshelf or
a banana on the mantelpeice or myself halfway up the ladder to the
attic and not know how the old coffee, the banana, or I arrived at
those places.
 -- Russell Hoban, _The Medusa Frequency_


From Stuart.Ford at Boeckh.com  Tue Jun 27 18:54:17 2000
From: Stuart.Ford at Boeckh.com (Stuart Ford)
Date: Tue, 27 Jun 2000 17:54:17 -0500
Subject: Can anyone solve this?
Message-ID: <DD768E0BEA95D211B27A0008C74C86250A6B8E@BEKEXCH1>


<rdudfield at my-deja.com> wrote in message
news:<8j737m$t5a$1 at nnrp1.deja.com>...
> Hi,
> 
> I've got a problem here which I just can't figure out.
> 
> 
> a = [1,2,3]
> b = ["c",32]
> c = [1,2]
> 
> inputlist = [a,b,c]
> 
> Inputlist can have any number of lists in it.  The lists in it can have
> any number of elements in them.
> 
> I need the code to give every possible combination of the lists in
> inputlist.
> 
> This is what the answer should be for the inputlist above.
> 
> outputlist = [[1, 'c', 1], [1, 'c', 2], [1, 32, 1], [1, 32, 2], [2, 'c',
> 1], [2, 'c', 2], [2, 32, 1], [2, 32, 2], [3, 'c', 1], [3, 'c', 2], [3,
> 32, 1], [3, 32, 2]]
> []
> 
> 
Try this routine on... it's a kludge, but it works..

def combiner(inlist):
    outlist = []
    indent = 0
    command = ''
    for arg in inlist:
        command = command + indent*'  ' + 'for xx' + `indent` + ' in
inlist[' + `indent` + ']:\n'
        indent = indent + 1
    command = command + indent*'  ' + 'outlist.append(['
    for x in range(0, indent):
        if x == indent-1:
            command = command + 'xx' + `x`
        else:
            command = command + 'xx' + `x` + ','
    command = command + '])\n'
    exec command
    return outlist

Stuart



From shapr at uab.edu  Mon Jun  5 10:59:56 2000
From: shapr at uab.edu (Shae Erisson)
Date: Mon, 05 Jun 2000 14:59:56 GMT
Subject: Newbie Questions
References: <8hgdpf$2fn$1@nnrp1.deja.com>
Message-ID: <393BC70C.FDD0C18B@uab.edu>

arun_rajappa at my-deja.com wrote:

> c) What is wxPython ? How does it compare with Tkinter ??

wxPython is a GUI toolkit that's based on wxWindows:
It's at http://alldunn.com/wxPython/
-- 
sHae mAtijs eRisson (sHae at wEbwitchEs.coM) gEnius fOr hIre


From dalke at acm.org  Fri Jun  2 12:15:59 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 2 Jun 2000 10:15:59 -0600
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h7blm$j1b$1@slb2.atl.mindspring.net> <slrn8jeqfv.n4g.morpheus@teleute.rpglink.com>
Message-ID: <8h8meo$a57$1@slb7.atl.mindspring.net>

I said:
>>Packages compiled with GPL code
>>restrict their use to only GPL-compatible licenses.


Steve Lamb replied:
>    That's a new one to me.  I've never heard of the GPL requiring
> things that I compile with a GPLed compiler to be distributed also
> under that license.  I've only seen it where the distributed /code/,
> if modified and recompiled must also be distributed under the GPL.


My apologies, I messed up with a few words.  How's this:

  "Packages compiled along with GPL code restrict their being linked
                     ^^^^^                              ^^^^^^^^^^^^
   to only GPL-compatible licenses.

So not only am I not a lawyer, but I'm not a wordsmith.

                    Andrew
                    dalke at acm.org





From robin at jessikat.demon.co.uk  Sat Jun  3 08:39:25 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 3 Jun 2000 13:39:25 +0100
Subject: mixed solution: unicase (unique allowed case) (was: Re: Case-sensitivity: )
References: <87n1l3u7q1.fsf@deneb.cygnus.argh.org>
 <Pine.GSO.4.10.10006031335390.17165-100000@sundial>
Message-ID: <rtBAaUA9xPO5Ew7E@jessikat.demon.co.uk>

In article <Pine.GSO.4.10.10006031335390.17165-100000 at sundial>, Moshe
Zadka <moshez at math.huji.ac.il> writes
>On 2 Jun 2000, Florian Weimer wrote:
>
>> There's the convention to use "Foo" for classes and "foo" for
>> instances.  I don't think people want to give up that.
>
>OK, I've heard this one once too many times. 
>
>What's wrong with:
>
>>>> class Foo:
>...     pass
>... 
>>>> aFoo = Foo()
>
>Certainly much clearer?
>--
>Moshe Zadka <moshez at math.huji.ac.il>
>http://www.oreilly.com/news/prescod_0300.html
>http://www.linux.org.il -- we put the penguin in .com
>
>

not to me. I still prefer c = C(), for f in F etc
-- 
Robin Becker


From akuchlin at mems-exchange.org  Sat Jun 10 19:22:59 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 10 Jun 2000 19:22:59 -0400
Subject: Type problem on Windows
References: <Pine.GSO.4.10.10006102005580.6989-100000@sundial>
Message-ID: <3dvgzh2ka4.fsf@newcnri.cnri.reston.va.us>

Moshe Zadka <moshez at math.huji.ac.il> writes:
> The answer is yet again "don't do that". Never ever check type sanity in
> Python. If the object is good enough, your function will work with it. And
> if it isn't, well, it will raise a TypeError or AttributeError itself.

In Python 1.6, optional SSL support has been added to the socket
module.  An SSL socket is represented by a different type, so
explicitly checking for the socket type will mean that callers can't
use SSL sockets to run the protocol being implemented.  Another
argument in being a bit sloppy with the type-checking...

--amk




From gmol at my-deja.com  Mon Jun 12 23:33:56 2000
From: gmol at my-deja.com (gmol at my-deja.com)
Date: Tue, 13 Jun 2000 03:33:56 GMT
Subject: dynamic classes/objects and the like
Message-ID: <8i4a33$67m$1@nnrp1.deja.com>

Hi all,
I am actually planning on using some dynamic classes and objects in my
application..looking at

http://starship.python.net/crew/gandalf/DNET/classes/DynaObj.html

it seems just a tad dated(1.4)...although I can't think of how different
something else would be , there isn't a more recent generally accepted
way to do this kind of thing is there?

Are there any examples of more large applications that rely on on
dynamic classes / objects that I could look at to avoid a possible
hitherto unseen pitfalls?

Thanks a tonne for help if you can offer any...



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


From _delete_this_mlerutte at square.nl  Wed Jun 28 03:12:20 2000
From: _delete_this_mlerutte at square.nl (Maurice le Rutte)
Date: Wed, 28 Jun 2000 09:12:20 +0200
Subject: JPython and Java integration
References: <8j7h3m$5v18j$1@fu-berlin.de> <39579a03.17964799@news.clara.net> <3959891B.23073A4@myriad.com>
Message-ID: <8jc8id$8pmr$1@ID-26372.news.cis.dfn.de>

Paul Hill wrote in message <3959891B.23073A4 at myriad.com>...
> [...]
>"Laurence V." wrote:
>> If your product is a development tool for Java programmers, it
>> makes more sense to use one of the Java interpreters as
>> your scripting engine. Learning curve = 0. JPython learning curve =
>>?high? for people who don't know it.
>
>The assumption you are making is that the target audience for the
>script writers are Java programers. There are many applications
>where some quisi-end-user or less experienced programmer is just
>adding a few extra business rules, or connecting a few objects >together.
In that case the scripting language of choice would be the
>one that looks the 'cleanest' to the these users.


Exactly, the target audience is not the programmer community and it is very
unlikely that they will know Java beforehand.

I want to use a scripting language because it doesn't require compilation
and it doesn't require the classes to be unloaded or the server to be
restarted after a change is made.

Maurice le Rutte.




From scoota at my-deja.com  Tue Jun 27 11:15:18 2000
From: scoota at my-deja.com (scoota at my-deja.com)
Date: Tue, 27 Jun 2000 15:15:18 GMT
Subject: Python COM for Linux?
Message-ID: <8jagdk$det$1@nnrp1.deja.com>

Thanks for the advice Warren,
could you point me some more details/examples of your publish/subscribe
solution, it sounds just what I'm looking for.  I'm especially
interested in the client side as I need the client script to be as
minimal and legible as possible.
Scoota


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


From didier.willame at ieee.org  Fri Jun 30 12:12:56 2000
From: didier.willame at ieee.org (Didier WILLAME)
Date: Fri, 30 Jun 2000 16:12:56 GMT
Subject: Tkinter focus problem
Message-ID: <395CC8DE.AA663D4@ieee.org>

Hello all,

I've slight problems with focus in Tkinter, running under win NT.

With the following script I'm able to create 'nodes' (left button) in a
canvas. The color of a 'node' toggles if the mouse is putting on.
Moreover, I would like display "Hello World!" when I press on the key
'1' and, when the mouse is putting on this 'node'.

But I'm unable to bind a key event (line 25), although I can run the
script with a button event (line 26 instead of line 25).
I should probably give the focus to the 'node' object inside the
entryNode function (line 57).


Thank you for your time,
Didier



# ----------
 1 import Tkinter
 2 import Canvas
 3
 4 class Graph( Tkinter.Frame ):
 5
 6   # ----------
 7   def __init__( self, master=None ):
 8     Tkinter.Frame.__init__( self, master )
 9     self.pack()
10     self.createCanvas()
11     return
12
13   # ----------
14   def createCanvas( self ):
15     self.editingArea = Tkinter.Canvas( self )
16 
17     # --- To create a node with the left button.
18     self.editingArea.bind( '<Button-1>', self.mkNode )
19
20     # --- To toggle the node color when passing with the mouse.
21     self.editingArea.tag_bind( 'node', '<Any-Enter>', self.entryNode
)
22     self.editingArea.tag_bind( 'node', '<Any-Leave>', self.leaveNode
)
23
24     # --- To active a nodes.
25     self.editingArea.tag_bind( 'node', '<KeyPress-1>', self.srcNode )
26 #    self.editingArea.tag_bind( 'node', '<Button-2>', self.srcNode )
27
28     self.editingArea.pack( { 'side' : 'top' } )
29
30     return
31
32   # ----------
33   def mkNode( self, event ):
34     "Create a new node at (x,y)"
35
36     x = event.x
37     y = event.y
38
39     new = self.editingArea.create_oval( x-5, y-5, x+5, y+5 )
40     self.editingArea.itemconfigure( new, { 'tag'     : 'node'  } )
41     self.editingArea.itemconfigure( new, { 'outline' : 'black' } )
42     self.editingArea.itemconfigure( new, { 'fill'    : 'red'   } )
43
44     return
45
46   # ----------
47   def entryNode( self, event ):
48     self.editingArea.itemconfigure( 'current', { 'fill' : 'blue' } )
49    return
50
51   # ----------
52   def leaveNode( self, event ):
53     self.editingArea.itemconfigure( 'current', { 'fill' : 'red' } )
54     return
55
56   # ----------
57   def srcNode( self, event ):
58     print 'Hello World!'
59     return
60
61 # ----------
62 graph = Graph()
63 try:
64   graph.mainloop()
65 except KeyboardInterrupt:
66   graph.quit()
67 # ----------
-------------- next part --------------
A non-text attachment was scrubbed...
Name: didier.willame.vcf
Type: text/x-vcard
Size: 453 bytes
Desc: Card for Didier WILLAME
URL: <http://mail.python.org/pipermail/python-list/attachments/20000630/8470d485/attachment.vcf>

From dag at orion.no  Tue Jun 20 19:01:05 2000
From: dag at orion.no (Dag Sunde)
Date: Wed, 21 Jun 2000 01:01:05 +0200
Subject: "TypeError: read-only buffer, None" returned from file.write...
Message-ID: <MIS35.664$PQ4.13634@news1.online.no>

I have the following code-snippet that retuns a string:

   result = m_XMLDoc.transformNode(xslDoc)

...and the string shows the expected result when printed,
but when I try to save the string to a file with the following
code:

  f = open( outFile, 'w')
  f.write(sBuffer)
  f.close()

I get this error:

Traceback (innermost last):
  File "f:/apache/cgi-bin/getfinalconfirmation.py", line 142, in
saveConfirmedData
    f.write(sBuffer)
TypeError: read-only buffer, None

Can anyone give me a hint?

TIA...

Dag Sunde





From hgg9140 at seanet.com  Tue Jun  6 19:44:02 2000
From: hgg9140 at seanet.com (Harry George)
Date: 06 Jun 2000 16:44:02 -0700
Subject: PyXML 0.5.5: fix for pyexpat.c
Message-ID: <m3k8g2fk8t.fsf@wilma.george.org>

The command:
  python setup.py build
will work if you change the "New" to "NEW" on line 474

static xmlparseobject *
newxmlparseobject( char *encoding, char *namespace_separator){
        int i;
        xmlparseobject *self;
        
        /*self = PyObject_New(xmlparseobject, &Xmlparsetype);*/
        self = PyObject_NEW(xmlparseobject, &Xmlparsetype);
        if (self == NULL)
                return NULL;

  
-- 
Harry George
hgg9140 at seanet.com


From pduffin at hursley.ibm.com  Thu Jun 22 11:56:04 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 22 Jun 2000 16:56:04 +0100
Subject: Python module for Dbase files?
References: <_6A35.14056$Xx5.689560@bgtnsc06-news.ops.worldnet.att.net> <g3buks8bdup5b8o4sqsg77n14q7p9vmisp@4ax.com> <9aU35.15328$Xx5.793661@bgtnsc06-news.ops.worldnet.att.net> <7ns0ls4tmiv2c1rfpl8nmmthlf45sdgf53@4ax.com> <Uw945.16309$Xx5.881927@bgtnsc06-news.ops.worldnet.att.net> <DXk45.79957$WS3.716413@typhoon.we.rr.com> <Rfq45.15293$C44.879496@bgtnsc05-news.ops.worldnet.att.net>
Message-ID: <39523714.1273966F@hursley.ibm.com>

William Dandreta wrote:
> 
> Hi Stephen,
> 
> >I have absolutely *no* idea if it will run with Python 1.2, or under DOS,
> >but when I was looking for database support awhile back, someone pointed me
> >to MetaKit, and I absolutely *love* it...
> 
> I checked the website for MetaKit. There are no DOS binaries and I searched
> the site for DOS but nothing useful showed up so I lost interest.
> 

The source is available at
	http://www.equi4.com/pub/mk/mk-2.01/metakit-2.01.tar.gz


From keating at mech.uq.edu.au  Fri Jun 23 22:38:10 2000
From: keating at mech.uq.edu.au (Tony Keating)
Date: 24 Jun 2000 02:38:10 GMT
Subject: Controlling an external program with Python
Message-ID: <slrn8l87is.dpi.keating@mars.rcode.com.au>

I have a small project that I'm trying to develop in Python. Basically it's
a GUI front end to a interactive text based program. So what I want to do is
start the program, then send various strings to the program, read some output
from the program then send some more strings to the program. Anyone care to
point me in the right direction for this sort of stuff?

Tony.

-- 
Tony Keating <keating at mech.uq.edu.au> [http://www.mech.uq.edu.au/~tonyk]
Department of Mechanical Engineering, University of Queensland, AUSTRALIA


From petersc at stud.ntnu.no  Thu Jun  8 01:38:26 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Thu, 08 Jun 2000 07:38:26 +0200
Subject: Dutch.comp.python.org
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <393793B0.83E25F53@ecs.soton.ac.uk> <8h8ej8$7c0$2@newshost.accu.uu.nl> <m33dmpj1b9.fsf@chinon.cnrs-orleans.fr> <8hljvq$mun$1@slb7.atl.mindspring.net> <slrn8jt4ej.64h.thor@localhost.localdomain> <8hmmvu$ets$2@newshost.accu.uu.nl> <3dr9a9hr2w.fsf_-_@newcnri.cnri.reston.va.us>
Message-ID: <393F3152.234B894A@stud.ntnu.no>

Andrew Kuchling wrote:
> 
> Still, at least they're not Austrian.  [1]

Which explains why Python is such a smooth and understandable
language. In the last days it felt like the dam was
beginning to break though.

> [1] While we're drifting off topic... I'm going to be in Berkeley
>     early next week.  Any Python or Linux-related goings-on around
>     there Sunday or Monday?

Oh, depends on where Berkely is ;-]

> [2] My dad is from a town near Salzburg whose name eludes me at the
>     moment.  It *does* explain why I wear shorts a good deal, doesn't
>     it?

Again depends. Do you wear sandals in combination to white socks?

having-more-Austrian-than-Dutch-speaking-friends-ly y'rs Peter

P.S.: But mostly the Dutch are better to understand.
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From paul.robinson at quantisci.co.uk  Fri Jun  2 09:30:14 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Fri, 02 Jun 2000 14:30:14 +0100
Subject: Status of List
References: <010501bfcc92$ff4b9c20$1906a8c0@fc.fenx.com> <3937B59C.3AFE081E@bam.com>
Message-ID: <3937B6E6.25ECBA46@quantisci.co.uk>


Bill Scherer wrote:
> 
> Things have been very quite since yesterday from my point of view.  I'm
> just glad I'm not the only one.
> 

It's not exactly busy from where I am either - 13 posts yesterday and
this will be number 5 today...

	Paul.

-----------------------------------
Business Collaborator Team
Enviros Software Solutions
http://www.businesscollaborator.com
-----------------------------------



From rmartin at objectmentor.com  Sun Jun 18 21:39:04 2000
From: rmartin at objectmentor.com (Robert C. Martin)
Date: Sun, 18 Jun 2000 20:39:04 -0500
Subject: ?Extreme Programming in Python?
References: <eviltofu-A13457.22304614062000@news.earthlink.net> <394847C0.E237B43C@home.com> <3948719B.96B9367A@san.rr.com>
Message-ID: <skqufugte7f44@news.supernews.com>

Courageous <jkraska1 at san.rr.com> wrote in message
news:3948719B.96B9367A at san.rr.com...
>
> > Not yet, but the company I just started with recently hired me to do
> > Python programming (& other stuff), and we're all gearing up to do XP.
>
> Well, if I were joking, I'd say that python makes you
> so productive that XP was overkill. :)-

XP is not so much about productivity as it is about creating a team of
programmers that communicate well with their customers, and run at a fast,
but sustainable pace.


--

Robert C. Martin    | "Uncle Bob"              | Training Courses:
Object Mentor Inc.  | rmartin at objectmentor.com | OOD, Patterns, C++, Java,
PO Box 85           | Tel: (800) 338-6716      | Extreme Programming.
Grayslake IL 60030  | Fax: (847) 548-6853      | http://www.objectmentor.com

"One of the great commandments of science is:
    'Mistrust arguments from authority.'" -- Carl Sagan




From pablo.prieto at dulcesol.es  Tue Jun  6 08:19:44 2000
From: pablo.prieto at dulcesol.es (Pablo Prieto)
Date: Tue, 06 Jun 2000 13:19:44 +0100
Subject: Newbie question.
Message-ID: <393CEC60.D527C260@dulcesol.es>

> Now, I'm not saying that's the best way to provide this interface.
> But it fixes your example.  Another idea:
> 
>    class Environment:
> 
>      def Set( val ):
>        self.ays = val
> 
>      def Get():
>        return self.ays
> 
> Now:
> 
>    env = Environment()
>    env.Set( "NUMBER 9" )
>    print env.Get()
> 
> --
> Randall Hopper
> aa8vb at yahoo.com

Yes. The class solution looks pretty fine to me. I'll use it.
Great Thanks a lot.

Yours sincerely,
Pablo, from Spain.



From tim_one at email.msn.com  Fri Jun  9 16:46:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 9 Jun 2000 16:46:37 -0400
Subject: Python for Windows -- what to install?
In-Reply-To: <dpbt1ay6ec.fsf@mraz.iskon.hr>
Message-ID: <000001bfd253$ce6f9180$5b2d153f@tim>

[posted & mailed]

[Hrvoje Niksic]
> I'd like to install Python for my younger brother (age 16) on the
> Windows machine that he uses.

Python is suitable for brothers of all ages <wink>.

> I've never installed Python on Windows, I'm not sure what components
> I need to prepare.

Relax!  The presumption is that Windows users won't be able to compile
anything, so *everything* comes in one box.

> I'd like him to be able to use the stuff that is normally distributed
> with Python 1.5.2, plust IDLE under Tkinter.
>
> I assume I'll need `py152.exe' for the basic functionality.

That contains all of the above -- jusr run it.  I strongly recommend one
deviation from the defaults, though:  when it asks for the Python
installation directory, force it to "c:\python15" (where "c:\" is the root
of his boot drive, almost certainly exactly "c:\").  The default is some
monster with embedded spaces in its name, and that just makes life harder.
Python 1.6 will suggest a space-free path by default.

> Do I need `win32all.exe' for the above mentioned things to work?

No.  Those are Windows-specific *extensions* to Python (for things like
getting at Microsoft's Win32 API).  Make your lives easy and don't install
these at first.

> If so, does it supersede py152.exe?

No, it's an add-on.

> Do the above files include documentation, or do I have to download it
> separately?

It's all in py152.exe.

A probable problem and a suggestion:  depending on which version of Windows
he's running and the phase of the moon, the Scriptics Tcl installer (which
comes with and is invoked by py152.exe -- one big reason that's bigger than
the Unix Python distributions) may not set up Tcl correctly.  The cure for
that is to get rid of pre-existing tcl*.dll and tk*.dll files, and/or to set
the PATH in his autoexec.bat to include Tcl's bin directory.  After Tcl is
working, IDLE will work too, and at that point you should visit python.org
and grab a more-recent version of IDLE (IDLE has improved a lot since the
1.5.2 release).

he's-in-for-a-treat-ly y'rs  - tim





From rogerha at ifi.uio.no  Mon Jun 26 15:19:25 2000
From: rogerha at ifi.uio.no (Roger Hansen)
Date: 26 Jun 2000 21:19:25 +0200
Subject: Shoud I use Java or Python?
References: <00062612550704.00594@dagomachine>
Message-ID: <ya0d7l4uu6a.fsf@trymheim.ifi.uio.no>

* Davide Del Vento
> 
> Hello world!
> I am a physicist, with experience in "C" language (not too much C++)
> and no previous web-based experiences. Now I have to develop some
> applications based on a client-server queries on databases and
> remote (but not only) big images manipulation. It should work
> indepentently of client platform and it should be simply ported from
> windows to linux server.
> 
> This project will be used by scientist that will be interested in
> use some of the ESA's SMART-1 images, see
> http://sci.esa.int/home/smart-1/index.cfm for details.
> 
> I think that using java will be fine (I will call "exec" in order to
> run other application that will do the eavy work on server), but
> what's about python? I use linux and I found it installed on my PC,
> so I visited the web-site (and the newsgroup), but I don't
> understand if there are some advantage in writing phyton application
> instead in writing Java ones or not (for my case, obviously).

Many discussions about Java vs. Python have found place here. Try
searching at deja.com. Especially the "java vs. python" thread had
some fine articles.

I would of course recommend Python (suprise, suprise!), since python 
code looks nicer, are easier to code, more maintainable and easy to
extend. (This are my subjective views.)

There is also a comparison at 
<URL: http://www.python.org/doc/essays/comparisons.html >


Good luck


R


From jcm at bigskytel.com  Fri Jun  9 23:12:12 2000
From: jcm at bigskytel.com (David Porter)
Date: Fri, 9 Jun 2000 21:12:12 -0600
Subject: Porting CPython
In-Reply-To: <8hrbph$a4n$1@nnrp1.deja.com>; from perl_monkey@my-deja.com on Fri, Jun 09, 2000 at 06:08:01PM +0000
References: <8hrbph$a4n$1@nnrp1.deja.com>
Message-ID: <20000609211211.B6218@bigskytel.com>

* perl_monkey at my-deja.com <perl_monkey at my-deja.com>:

> I am currently porting CPython to IBM's OS/390 Unix System Services.  I
> wanted to point out a few issues that I've had regarding this port, and
> ask for some advice from anybody currently involved in ports or anyone
> who has ported python in the past.  OS/390 is a mainframe platform, and
> Unix System Services is a fully functioning (albeit intensely weird)
> UNIX running on top of the mainframe OS in an LPAR.

If you haven't already, see this url:

http://www.s390.ibm.com/products/oe/python.html

It is intended for python 1.4.

  david.

-- 
To whom i and fate and you.  To feel, poor worlds must merely do, which
then are done. -- e.e.cummings in a blender



From wjdandreta at worldnet.att.net  Sun Jun 18 15:55:00 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Sun, 18 Jun 2000 19:55:00 GMT
Subject: Regular expressions vs find?
References: <sy535.11434$Xx5.530580@bgtnsc06-news.ops.worldnet.att.net> <01bfd93c$66c03180$2ace8ad1@daves-dell>
Message-ID: <oO935.12054$Xx5.557215@bgtnsc06-news.ops.worldnet.att.net>

Hi David,

The message is titled Python GREP(... you can take a look if you like. It
was suggested that reg expr might be faster so give it a try.
Bill

David C. Ullrich wrote in message <01bfd93c$66c03180$2ace8ad1 at daves-dell>...
>
>
>William Dandreta <wjdandreta at worldnet.att.net> wrote in article
><sy535.11434$Xx5.530580 at bgtnsc06-news.ops.worldnet.att.net>...
>> I recently read a message that suggested that using regular expressions
>> might be faster than the find function.
>
> I would have guessed that if anything find would be faster, although
>I certainly could be wrong. Did someone say that using a regular expression
>would be _faster_, or did they actually say it was more _powerful_ or some
>such?
>









From digitig at cix.co.uk  Tue Jun 27 16:00:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Tue, 27 Jun 2000 21:00 +0100 (BST)
Subject: boolean true and false values.
References: <slrn8lgsjq.us.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <memo.20000627210059.56821A@digitig.compulink.co.uk>

In article <slrn8lgsjq.us.hat at se-46.wpa.wtb.tue.nl>, 
hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:

The other comments have been addressed, but:

> >But that's /exactly/ right! If you execute 'total="kjerg"' then your 
> >invariant is indeed violated!
> 
> So, why is it then that Python allows this ?
> For a language aimed at professionals, I understand (although I wouldn't
> introduce it), but for a clean language such as Python ?

If you execute 'total = total - rand()' your invariant will probably be 
broken too, in a typed language or an untyped language. No language will 
stop you writing code that violates invariants! Some help you find out 
that you've done it, though; Python's (and C's and C++'s) assert() helps. 
Eiffel's contracts help even more. But this is not a typing issue.


From jcm at bigskytel.com  Sat Jun 10 06:33:06 2000
From: jcm at bigskytel.com (David Porter)
Date: Sat, 10 Jun 2000 04:33:06 -0600
Subject: Recruiting python programmer
In-Reply-To: <8hsv8f$2dj$1@wanadoo.fr>; from yannig.gourmelon@wanadoo.fr on Sat, Jun 10, 2000 at 10:43:09AM +0200
References: <8hsv8f$2dj$1@wanadoo.fr>
Message-ID: <20000610043306.B9887@bigskytel.com>

* Daniel Morton <yannig.gourmelon at wanadoo.fr>:
> I am looking to hire on a contract basis for 6 months a python programmer in
> Paris, France. Does anyone have someone to recommend or know where I can
> post the add? thanks.

Well, I don't live in France, nor am I much of a programmer yet, but you
should submit your offer to the Python job board:

  http://www.python.org/Jobs.html


  david.

-- 
Look my fingers. -- e.e.cummings in a blender



From wjdandreta at worldnet.att.net  Wed Jun 14 23:32:31 2000
From: wjdandreta at worldnet.att.net (William Dandreta)
Date: Thu, 15 Jun 2000 03:32:31 GMT
Subject: Hex to decimal
Message-ID: <j7Y15.2020$Uw3.137442@bgtnsc04-news.ops.worldnet.att.net>

I have an ascii string like 0C44DH. Ids there a Python function to convert
it to decimal?

Bill




From paul at prescod.net  Sun Jun  4 10:54:59 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 04 Jun 2000 09:54:59 -0500
Subject: xml and python
References: <3937CC38.FD1360E@ut.ee>
Message-ID: <393A6DC3.FA478109@prescod.net>

Kaimar Karu wrote:
> 
> Hi,
> 
> Where can I find some kind of helping material (tutorials etc) about how
> to use xml in python?

http://www.python.org/sigs/xml-sig should help!

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Simplicity does not precede complexity, but follows it. 
	- http://www.cs.yale.edu/~perlis-alan/quotes.html



From donn at u.washington.edu  Wed Jun 28 14:13:39 2000
From: donn at u.washington.edu (Donn Cave)
Date: 28 Jun 2000 18:13:39 GMT
Subject: Controlling an external program with Python
References: <slrn8lj6pa.l52.thomas@centurion.xs4all.nl> <3958C648.1D76@rd.francetelecom.fr> <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu> <39588741.FB7EB5FE@hursley.ibm.com> <3958A7AE.7DAA@rd.francetelecom.fr> <3958C134.7AAEF37B@hursley.ibm.com> <8jc28s$8eb$0@216.39.151.169>
Message-ID: <8jdf8j$8tm4$1@nntp6.u.washington.edu>

Quoth thomas at xs4all.nl (Thomas Wouters):
|>| In unix, AFAIK the only nonportable part is the *naming* of the pty
|>| devices. That kind of decision is fairly trivial to make in
|>| sh/awk/perl/python/tcl...
|>
|> The Berkeley pty, with the kind of naming scheme you allude to, is
|> already supported in the "pty" module.  It's supposedly tested on
|> Linux, and I'd expect it could work on FreeBSD/NetBSD/etc., and
|> probably lots of others like Digital UNIX.
|
|> But it's far from universal, and some of the alternatives do use
|> special functions that would need to be supported in C.
|
|> On the bright side, I am optimistic that the horrors of past
|> pseudotty implementations are fading into oblivion and the
|> survivors are fewer and cleaner.
|
| I'd be interested in hearing about some of these pty-implementations. I know
| about the SGI one, for which there is support in the sgi module, and I know
| about openpty() (currently used by at least BSDI, FreeBSD and (glibc
| 2.1-based) Linux) -- in fact I submitted a patch to make the pty module use
| 'openpty()' and 'forkpty()' if it's available. And I know of the old
| BSD/Linux way of 'just opening a pty' of course.
|
| Are there any other 'mainstream' methods of opening a pty ? Could they be
| implemented in terms of openpty() and/or forkpty() ? It would be pretty neat
| if the 'emulation' could stay inside the posixmodule, without the need for X
| different OS-specific modules and Y different function names for each method
| of getting a pty ;)
|
| If someone can send me some manpages/infopages/whatever on 'other methods'
| and be willing to test my code, I'd be happy to add it to the posix/pty
| module.

The MIT Kerberos 5 pty library I mentioned uses a getpty function that
opens the master pseudoty and returns its file descriptor and the name
of its slave terminal device.  Where it uses openpty() underneath, it
doesn't supply any termios settings and closes the slave file descriptor
immediately.  It doesn't use forkpty() for anything.  My reflex reaction
is that's the right level of support for a C module:  export all possible
semi-standard low level interfaces, and put convenience routines in a
Python module.

Some of these interfaces may be appropriate for the posix module, but
I doubt it!  Aside from the narrow-minded question of whether any of it
is covered by POSIX 1003.1, from the user side it's inevitable that on
some platform that does build the posix module, the pty functions won't
be supported.  They'll need another module that can be optionally
supported, like termios, fcntl et al.

The worst mess is that collection of hacks and tricks to open the
slave terminal without security exposure and in a way that allows
features like Berkeley job control to work correctly.  It's a mess
because it's poorly documented and, at least in the case of security,
it deals with gaps rather than features.  Someone will have to sort
those out and incorporate them into a few platform-independent functions
that cover the necessary ground without taking too much onto themselves -
I mean like forkpty() takes too much on itself, in my opinion, to be
something we would want at the C level.

The more we can steal, the better.  There's too much platform specific
folklore in here to write a good C implementation from scratch with
man pages.

	Donn Cave, donn at u.washington.edu


From mogul at primus.ca  Sat Jun 24 11:00:03 2000
From: mogul at primus.ca (Robin Porter)
Date: Sat, 24 Jun 2000 11:00:03 -0400
Subject: dot com success (unrelated PYHTON question)
Message-ID: <cY355.1801$u3.13961@news2.tor.primus.ca>

Could someone please let me know if they know of anyone from the Kingston
Ontario Canada area that has gone on to set up a successful dot company.
Please e-mail me with your response to mogul at primus.ca

Thank you!




From rupole at compaq.net  Sun Jun 25 17:40:45 2000
From: rupole at compaq.net (Roger Upole)
Date: Sun, 25 Jun 2000 21:40:45 GMT
Subject: Most Recently Used file list
References: <961776536.37.0.nnrp-06.c3ad12d2@news.demon.co.uk>
Message-ID: <x%u55.16571$i9.1082991@news.easynews.com>

You can use COM to get the recent files folder (Usually c:\windows\recent,
but not always.)
import win32com.client
sh = win32com.client.Dispatch('Shell.Application.1')
rcnt = sh.NameSpace(8)
for rcnt_file in rcnt.Items():
    print rcnt_file

Note that some applications keep their own recent files list.
            Roger Upole

"Toby" <t.harris at robinsons.co.uk> wrote in message
news:961776536.37.0.nnrp-06.c3ad12d2 at news.demon.co.uk...
> I am writing a Python 1.5.2 application which requires a "Most recently
> Used" file list, are there any library routines to help me, or do I have
to
> code from scratch using the registry ? (the app runs under Windows NT 4.0)
>
> Thanks in Advance
>
> Toby
>
>




From dsavitsk at e-coli.net  Fri Jun 23 16:59:27 2000
From: dsavitsk at e-coli.net (dsavitsk)
Date: Fri, 23 Jun 2000 20:59:27 GMT
Subject: verifying date values
Message-ID: <PcQ45.16280$dF.645882@news1.rdc1.il.home.com>

is there a quick way to see if a string is a valid date? something like

>>> isDate('5/22/1999')
1

thanks





From petersc at stud.ntnu.no  Sun Jun  4 07:16:42 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Sun, 04 Jun 2000 13:16:42 +0200
Subject: Print formatting - same line printing?
References: <393A38E9.MD-1.4.4.angband@blueyonder.co.uk>
Message-ID: <393A3A9A.FB9F2E37@stud.ntnu.no>

Bablos wrote:
> 
> Is there a way that I can get Python to print this all on the same
> line, so that it looks as if the line is being updated constantly, as
> opposed to having a ridiculously long list of 'uploaded...' lines?

For unix you could just write:

print "\ruploaded",x,"bytes",

don't-know-if-it-works-on-other-platforms-though-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From pshinners at mediaone.net  Sat Jun  3 14:57:47 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Sat, 03 Jun 2000 18:57:47 GMT
Subject: file.seek misbehaving?
References: <203da3d8.c492c8fd@usw-ex0103-019.remarq.com>
Message-ID: <Lyc_4.33464$WS3.397965@typhoon.we.rr.com>

"cheesecake" wrote
> I've run into a problem that is perhaps not a Python-specific
> bug. I was processing some log files using Python, when I
> noticed some mysterious bugs in certain files. I tracked down
> the problem to the seek method misbehaving.

you might get better results if you open the file in BINARY mode





From aahz at netcom.com  Wed Jun  7 09:58:07 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jun 2000 13:58:07 GMT
Subject: What is Python3000??
References: <393e426f.77267284@news.online.no>
Message-ID: <8hlkdf$nun$1@slb7.atl.mindspring.net>

In article <393e426f.77267284 at news.online.no>,
Thomas Weholt <thomas at cintra.no> wrote:
>
>I want to know more about Python3000. How is it going to be different
>from Python as we know it? More features, new syntax ?? 

Nobody -- repeat, NOBODY -- has the answers to these questions.  Py3K is
a codename for a project THAT HAS NOT YET STARTED.  Py3K is predicated
to a certain extent -- to what extent, NOBODY KNOWS -- on the Fred
Brooks maxim, "Build one to throw away."

There are certain design decisions that were made for Python 1.x that
Guido would like to revisit.  Some of these decisions *MAY* be changed
in ways that break 1.x code.  Py3K is the name of the version that *MAY*
have these incompatibilities.

Should I make this a friggin' FAQ?

(Thomas, if it seems like I'm pounding on you, it's because I know
you've been reading comp.lang.python -- if you weren't aware of this,
you haven't been paying attention.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"You could make Eskimos emigrate to the Sahara by vigorously arguing --
at hundreds of screens' length -- for the wonder, beauty, and utility of 
snow."  -- PNH to rb in r.a.sf.f


From gmcm at hypernet.com  Fri Jun 30 08:38:47 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 30 Jun 2000 12:38:47 GMT
Subject: Passing variable-length keyworded arguments list to base class constructor?
References: <962352947.103057778@news.libertysurf.fr>
Message-ID: <8F635367Cgmcmhypernetcom@199.171.54.154>

Jerome Quelin wrote:

>I was wondering how to pass variable-length keyworded arguments list
>to my base-class constructor? Look at the example code below:
>
>-- File test.py
>from Tkinter import *
>class F(Frame):
>    def __init__(self, **args):
>        Frame.__init__(self, args)
>        # Here my initialisation stuff
>F(bg='blue').mainloop()
>-- End of file
>
>When doing this, I got an error in the Tkinter module, because it does
>not expect a dictionnary for an argument, but rather a list of
>keyworded arguments.
>
>If I write:
>    def __init__(self, *args):
>
>it doesn't work because I create the class instance with keyworded
>arguments:
>    F(bg='blue')
>
>and the * (one star) expects just arguments. Therefore, I must use the
>** (double star) construct to fetch a dictionnary.
>But how to pass this dictionnary to the parent class constructor?

class F(Frame):
  def __init__(self, *args, **kwargs):
    apply(Frame.__init__, (self,)+args, kwargs)

Generally you have to muck with the kwargs dict before passing it to the 
base class __init__.

- Gordon


From hjwidmai at foxboro.com  Fri Jun 16 08:19:03 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Fri, 16 Jun 2000 14:19:03 +0200
Subject: Python segfault
Message-ID: <20000616141903.A23985@foxboro.com>

Thomas Wouters wrote:

> The problem is not the 'height' of this graph, or the total number of 'nodes' 
> (which is limited only by memory), but the 'width' of the first tree.
> The 'node' struct is defined like this: 
> 
> typedef struct _node { 
> short n_type; 
> char *n_str; 
> short n_lineno; 
> short n_nchildren; 
> struct _node *n_child; 
> } node; 
> 
> and the 'n_nchildren' member is the number of immediate subexpressions
> (not grandchildren, only children) -- and it is limited to a short, which is
> 16 bits on most systems. And the short is signed, which means the maximum number
> of children is 32767. This includes the operators itself, so you can operate on
> just 16384 items in a single expression. If you try to operate on more, the short
> wraps, and the parser tries to index node->n_child[-32768], which usually segfaults. 

Why not simply make 'n_nchildren' an (32 bit) int? And, while at it, 'n_lineno' also?
It would cost some memory, though.

-- 
Hans-Joachim Widmaier             E-Mail:  hjwidmaier at web.de



From mwh21 at cam.ac.uk  Fri Jun 30 16:50:37 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 30 Jun 2000 21:50:37 +0100
Subject: Test results of linuxaudiodev.c
In-Reply-To: Ka-Ping Yee's message of "Thu, 29 Jun 2000 17:44:35 -0700 (PDT)"
References: <Pine.SGI.3.96.1000629173646.621134u-100000@happy>
Message-ID: <m3r99e7v1e.fsf@atrus.jesus.cam.ac.uk>

To python-dev: Sorry for the hiatus.

To python-list: does someone who knows about audio and can run linux
want to play with this (Ping and I fall into just one of these
categories each!).

Ka-Ping Yee <pingster at ilm.com> writes:

> On 30 Jun 2000, Michael Hudson wrote:
> > Yup, that works fine.  Don't understand the details - and as I have my
> > graduation ceremony tomorrow I'm going to go to bed and leave learning
> > them until some other occasion!
> 
> Cool.
> 
> Okay, i just discovered sunau.py.  Here's my real shot at
> test_linuxaudiodev.py.  Can you give this a try?
> 
> 
> ---- test_linuxaudiodev.py ----
> from test_support import verbose, findfile, TestFailed
> import linuxaudiodev
> import os, sunau
> 
> formats = {("ULAW", 1): linuxaudiodev.AFMT_MU_LAW,
>            ("NONE", 1): linuxaudiodev.AFMT_S8,
>            ("NONE", 2): linuxaudiodev.AFMT_S16_BE}
> 
> def play_au_file(path):
>     au = sunau.open(path, "r")
>     data = au.readframes(sunau.AUDIO_UNKNOWN_SIZE)
>     au.close()
> 
>     spec = (au.getcomptype(), au.getsampwidth())
>     if not has_key(formats, spec):
             ^^^^^^^
Oops?

>         raise "audio format not supported by linuxaudiodev"

Maybe you mean:

        raise TestFailed, "audio format not supported by linuxaudiodev"

?  I thought string exceptions were deprecated...
         
>     dsp = linuxaudiodev.open("w")
>     dsp.setparameters(au.getframerate(), au.getsampwidth() * 8,
>                       au.getnchannels(), formats[spec])
>     dsp.write(data)
>     dsp.close()
> 
> def test():
>     play_au_file(findfile('audiotest.au'))
> 
> test()

This doesn't work.  I don't really understand why.

For audiotest.au, |spec| is ("ULAW",2), which I don't think is right;
file(1) says 

audiotest.au: Sun/NeXT audio data: 8-bit ISDN u-law, mono, 8000 Hz

So I think this could be a bug in sunau.

The other data (au.getframerate(), au.getnchannels()) agrees with
file(1).

Anybody have a better idea?

Cheers,
Michael Hudson, BA (as of today! woohoo!)




From paul at prescod.net  Tue Jun 20 19:32:04 2000
From: paul at prescod.net (Paul Prescod)
Date: Tue, 20 Jun 2000 18:32:04 -0500
Subject: Why is Python so slow ?- revisited.
References: <H_c35.12336$Xx5.574993@bgtnsc06-news.ops.worldnet.att.net> <1ecgntk.22kamk6lgef4N%bparsia@email.unc.edu> <20000619191457.A26436@xs4all.nl> <1echq6c.1433wr2isx3qgN%bparsia@email.unc.edu>
Message-ID: <394FFEF4.4B7A2765@prescod.net>

Actually, I think that a method cache for Python has been implemented. I
think that the main reason that it isn't "standard" is that Guido wanted
to preserve the simplicity of the Python *implementation*. In general
this is important because we want many people to poke around in Python's
internals and not just a tiny inner circle. 

Of course, any particular optimization could be so attractive that it
overwhelms this concern...but in general Guido does not take the
approach where you try to cram in every cool optimization you can find.
-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
"Music is the stuff between the notes." - Claude Debussy



From dieter at handshake.de  Mon Jun 26 16:48:33 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 26 Jun 2000 22:48:33 +0200
Subject: How to control Pmw.ScrolledText scrollbar position?
References: <8ire9f$rs4$1@nnrp1.deja.com>
Message-ID: <x7vgyww4m6.fsf@lindm.dm>

Maytag Repairman <maytagrepairman at yahoo.com> writes on Wed, 21 Jun 2000 22:06:46 GMT:
> I have this program that I've written which uses Pmw.ScrolledText.
> There are buttons on my GUI which take data from entry fields and
> insert, replace, and append data in a data structure.  The ScrolledText
> is cleared and the updated data structure is displayed in a user-
> friendly format.  Unfortunately, the scrollbar always ends up at the
> top after this operation but the area of most interest to the user is
> the point where the operation on the data structure took place.  I
> can't figure out how to control the position of the scrollbar to move
> it to the area of interest.
You can use the TextWidget's (! not the scrollbar's) "yview"
command to adjust the viewport of the widget. It will
automatically adjust the scrollbar.


Dieter


From porter at et.byu.edu  Thu Jun 29 14:37:10 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Thu, 29 Jun 2000 12:37:10 -0600 (MDT)
Subject: Constants and Enums
Message-ID: <Pine.GHP.4.21.0006291234020.9839-100000@don.et.byu.edu>

I am sure that this topic has been beat to death, but I can't seem to find
the answer to my question.  I am used to programming in C/C++, and using
many enums and #define ed constants.  In the absense of these, I have
resorted to using tons of global variables, which makes me
uncomfortable.  I know that the general concensus is to use dictionaries
for all this, but I can't figure out how a dictionary would be usefule to
use in place of an enum.  I am trying to think in a more python way, but I
need a little nudging to get my brain to go that way.

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------




From bragib at my-deja.com  Mon Jun 12 14:20:11 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Mon, 12 Jun 2000 18:20:11 GMT
Subject: telnetlib strange behavior?
Message-ID: <8i39kg$fla$1@nnrp1.deja.com>

I am trying to telnet from a UNIX box to an NT to get the doskeys.
The NT that I login to runs a couple of .bat scripts which clear the
screen set up some stuff etc.  When I try to do a read_all() it fails
the first time with the exception:
Traceback (innermost last):
  File "telnet.py", line 17, in ?
    tn.read_all()
  File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
229, in read_all
    self.fill_rawq()
  File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
367, in fill_rawq
    buf = self.sock.recv(50)
socket.error: (131, 'Connection reset by peer')


That is why I put the first try in a try: except: clause.  The second
time I try to do a read_all() it works fine but I don't get all the keys
I am missing some of the first keys.  I assume that the first read_all()
actually did read some stuff and then the second read got the rest.

Can anyone see anything obviously wrong with this.


Thanks, Bragi

import sys
import telnetlib
HOST = 'shadow'
USER = 'bragi'
tn = telnetlib.Telnet(HOST)
tn.open(HOST)
tn.read_until("login: ")
tn.write(USER+'\r')
tn.read_until("password: ")
tn.write('my_password\r')
tn.write('doskey /macros\r')
tn.write('exit\r')
print 'done'
try:
    all1 = tn.read_all()
except:
    pass
all = tn.read_all()
print all









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


From eric at estinc.com  Fri Jun  9 11:39:30 2000
From: eric at estinc.com (Eric Lee Green)
Date: Fri, 09 Jun 2000 15:39:30 GMT
Subject: Why is Python so slow?
References: <r7%%4.1754$iy.137191@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <39410FF7.848F1661@estinc.com>

William Dandreta wrote:
> I am just learning Python and I wrote a little program to modify a text file
> with 120,000 lines. It took 1hr and 30 mins. I wrote a C program that
> created the text file from a .dbf and it only took 2 mins.

> Why does it take Python 50 times longer than C?

This is actually a reasonable example of a problem set where you're using
Python when you should be using "C".

My guess is that you are attempting to do text parsing with raw Python (i.e.,
by walking down a string array) rather than using the built-in "C" tools to do
the same (string.split, string.replace, the whole regexp library, etc.). 

All that aside, there do exist problem domains where Python is clearly
significantly slower than a "C" implementation. At the Twofish for Python site
( http://twofish-py.sourceforge.net ) you will also find a "C" utility called
'aescrypt'. Despite the fact that Twofish for Python uses the exact same
underlying "C" module to do its encryptions, it takes 13 seconds to encrypt a
test file with aescrypt and over a minute to encrypt the same file with
encrypt.py (I think that's the name of the utility) -- or roughly 8 times
longer. It appears that most of that time is being spent allocating and
de-allocating strings (aescrypt uses a static encryption buffer). I wrote
aescrypt using the Python code as my guide, so I doubt that there's any other
major difference between the two programs that could account for the
difference in execution time. I guess the final reality is that yes, Virginia,
there do exist applications where Python is not appropriate -- such as
encrypting massive amounts of data. Thankfully I only intend to encrypt
passwords using Twofish for Python, rather than massive amounts of data.
(Note: If you are a Python programmer looking for a job, drop me some EMAIL --
especially if you've done GUI work in the past, you're hired :-). 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      Visit our Web page:
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From cut_me_out at hotmail.com  Sun Jun 25 10:49:30 2000
From: cut_me_out at hotmail.com (Alex)
Date: 25 Jun 2000 10:49:30 -0400
Subject: Python with braces.
References: <395683ae.299281703@nntp.interaccess.com>
Message-ID: <etdsnu1byth.fsf@w20-575-101.mit.edu>

I don't think there's an interpreter that can handle Python with braces,
but there is a preprocessor that can take Python with control structures
delimited by braces, and format it with the appropriate indentation for
the vanilla Python interpreter.  I've never tried it, because it seems
like a terrible way to code, for me, even though it's a cool idea.  But
you can find it here:

http://www.python.org/pipermail/python-list/1999-June/066547.html

Alex.


From moshez at math.huji.ac.il  Wed Jun 14 01:41:41 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 14 Jun 2000 08:41:41 +0300 (IDT)
Subject: no SyntaxError?
In-Reply-To: <Pine.LNX.4.10.10006131359020.17071-100000@localhost.localdomain>
Message-ID: <Pine.GSO.4.10.10006140841040.25977-100000@sundial>

On Tue, 13 Jun 2000, Robert W. Bill wrote:

> Just curious- I didn't find any references to this in a quick search of 
> the faq and elsewhere...
> 
> This snippet generates "SyntaxError: invalid syntax"
> 
> def test():
>     else: print "whoops"
> test()
> 
> However, the following snippet generates "0 1 whoops" when it seems like
> it should also generate "SyntaxError: invalid syntax" because there is no 
> if before the else. 
>  
> def test1():
>     for i in range(2):
>         print i,
>     else:
>         print "whoops"

for/else is just as valid as if/else (and so are while/else).
Read more about it in the tutorial.
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From ben at co.and.co  Sun Jun 11 17:16:10 2000
From: ben at co.and.co (ben at co.and.co)
Date: Sun, 11 Jun 2000 21:16:10 GMT
Subject: spaces in filenames
References: <000001bfd3d1$001a1c40$27a2143f@tim>
Message-ID: <ukT05.173292$Qp2.446315@afrodite.telenet-ops.be>

Tim Peters <tim_one at email.msn.com> wrote:
> [ben . de . rydt at pandora . be]
>> Try opening a "some bitmap with a space.bmp" in Paint, by
>> double-clicking (Win95/98).
> 
> OK ... and it worked fine (Win95a -- the first, albeit up-to-date with
> service packs), whether double-clicking from within Paint -> Open, or
> double-clicking from within Explorer, or even typing the name (spaces and
> all) by hand from within Paint -> Open.  It's only from a DOS box that I
> need to slap quotes around the name.

Stop abusing that time machine. You retro-actively changed 
the behaviour on my computer too. Anyway, the computers I was talking
about, had been through the Win3.11, Win95, Win98 upgrade path, and I
suppose there's some 16 bit Paint lingering around. And of course, I
can't check them anymore, unless you lend me that time machine.

Anyway, the argument is moot because I suspect 99% of all UNIX-scripts
fail on filenames with spaces too. And no, the time machine isn't
that powerful.

Greetings,
-- 
ben . de . rydt at pandora . be ------------------ your comments
http://users.pandora.be/bdr/ ------- inl. IPv6, Linux en Pandora



From warlock at eskimo.com  Sun Jun 11 02:41:50 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Sat, 10 Jun 2000 23:41:50 -0700
Subject: Python Productivity over C++
References: <Pine.LNX.4.10.10006091252070.10098-100000@naxos.skunk.org>
Message-ID: <slrn8k6cde.1if.warlock@troll.myth>

On Fri, 9 Jun 2000 13:07:25 -0400 (EDT), 
 Drew Csillag, in the persona of <drew.csillag at starmedia.net>,
 brought forth the following words...:


(I snipped all the stuff  comparing python with C++ because C++ makes
my teeth ache.)

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

This is about the cutest sig I have seen in a long time :)
 As soon as I can come up with something like it, I will :)


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



From damonNOdaSPAM at unispace.com.invalid  Sun Jun 18 12:50:40 2000
From: damonNOdaSPAM at unispace.com.invalid (damonr)
Date: Sun, 18 Jun 2000 09:50:40 -0700
Subject: Python!
Message-ID: <0ca08531.70307f7b@usw-ex0105-037.remarq.com>

Hi,

I am running a new site(http://www.devcritic.com) and i was
wondering if you people knew some good Python resource sites
that i could add to the Python part of the site directory:
http://www.devcritic.com/sites/Python/. Feel free to add any
sites by clicking "Add a site" on the side menu or tell me here
and i will add em to there.

Thanks in advance.


Got questions?  Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com



From gansevle at cs.utwente.nl  Tue Jun 20 10:01:31 2000
From: gansevle at cs.utwente.nl (Fred Gansevles)
Date: Tue, 20 Jun 2000 14:01:31 GMT
Subject: How to prevent illegal definition of a variable in objects?
References: <8i8ujt$hop$1@news2.isis.unc.edu> <8ia4tv$e19$1@nnrp1.deja.com> <8idr2g$fvi$1@news2.isis.unc.edu>
Message-ID: <8intfj$7t0$1@nnrp1.deja.com>

In article <8idr2g$fvi$1 at news2.isis.unc.edu>,
  "Daehyok Shin" <sdhyok at email.unc.edu> wrote:
> How can I get the DCI module?
>
> Peter
>
My posting can be found at http://www.deja.com/getdoc.xp?AN=620649658.1
--
-----------------------------------------------------------------------
----          Linux/Windows-NT/IntraNetware Administrator          ----
-- Whenever it sounds simple, I've probably missed something! ... Me --
-----------------------------------------------------------------------


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


From 12klat at sightreader.com  Tue Jun  6 09:24:07 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 06:24:07 -0700
Subject: Pausing Loop
References: <FpR_4.6466$gr4.252142@news1.rdc1.il.home.com>
Message-ID: <393CFB77.C13E6F61@sightreader.com>

dsavitsk wrote:

> is there a way to conditionally pause a loop until some event (button press)
> takes place?  opening a modal window is not my first choice, but i don't
> know another way of doing it.
> (this needs to be using tkinter on win32 if that makes a difference)
> thanks again
> doug

Sounds like you need to use threads.  Read about the 'threading' module.

The idea is that you would run your loop in a separate thread from the GUI
thread.  This has the additional benefit of keeping your GUI running while the
loop is running.  When you pause, create a lock which gets released when the
desired event occurs in the GUI thread..

- Ken Seehof
kens at sightreader.com
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!


-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/25e4649d/attachment.vcf>

From jmg at ecs.soton.ac.uk  Thu Jun  1 03:42:56 2000
From: jmg at ecs.soton.ac.uk (Jacek Generowicz)
Date: Thu, 01 Jun 2000 08:42:56 +0100
Subject: ANNOUNCEMENT: Python Development Team Moves to BeOpen.com
References: <200005301905.OAA07357@cj20424-a.reston1.va.home.com> <3935937F.CCF87258@stud.ntnu.no>
Message-ID: <39361400.5E86F678@ecs.soton.ac.uk>

Peter Schneider-Kamp wrote:

> Guido van Rossum wrote:
> >
> > Oh, and to top it all off, I'm going on vacation.  I'm getting married
> > and will be relaxing on my honeymoon.  For all questions about
> > PythonLabs, write to pythonlabs-info at beopen.com.
>
> Gefeliciteerd bij de eerste stap naar de echtscheiding!

So cynical for one so young !




From PClaerhout at CREO.BE  Thu Jun 22 06:26:27 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 12:26:27 +0200
Subject: win32print.GetPrinter()
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130120F@msgeuro1.creo.be>

Hi,

can anyone tell me which information the win32print.GetPrinter() function
returns exactly? Any direction on where I can find this. It's not in the docs
for the win32 extensions.

Kind regards,


Pieter



From tom-main at REMOVEME.home.com  Tue Jun 27 11:57:23 2000
From: tom-main at REMOVEME.home.com (Tom)
Date: Tue, 27 Jun 2000 15:57:23 GMT
Subject: more python evangelising...
References: <200006271406.AAA06218@mbuna.arbhome.com.au>
Message-ID: <D9465.71358$7o1.1704294@news2.rdc1.on.home.com>

But you're falling into the old trap of thinking that one language is best
for everything.

Python is excellent for what it was designed for, but this does not include
large scale app development.

On the other hand, maybe it will be possible to add 'optional' features to
Python 3000 that will make it suitable for that kind of development without
interfering with its simplicity and (development) speed.

Tom.

"Anthony Baxter" <anthony at interlink.com.au> wrote in message
news:200006271406.AAA06218 at mbuna.arbhome.com.au...
> in a mainstream-ish news article, no less...
>
> http://www.theregister.co.uk/content/1/11606.html
>
> when talking about C#, and it's relationship to Java...
>
>
> [snippity]
>
> And an even more cunning route was at hand, as the redoubtable maintainer
> of the Java site Cafe Au Lait, Rusty Harold Elliot points out. Elliot,
> a long time critic of Sun's licensing tactics, notes: "If Microsoft
> wanted to really challenge Java, they should have gone with Python. I
> just don't believe it's possible for any major advances in language
> design to be made while restricting oneself to the mistakes Kernighan
> and Ritchie made 30 years ago."
>
> Well, quite. The open source language Python has not only proven itself
> macho enough in heavy duty production environment, it's simple enough
> to be taught as a first language to kids. And it's interfaced with COM
> interface for several years now. Maybe someone in Redmond thought it
> was time to unleash some "innovation" at the world, and cometh the hour,
> cometh the new programming language.
>
> [snippity]
>
> --
> Anthony Baxter     <anthony at interlink.com.au>
> It's never too late to have a happy childhood.
>




From matthias.huening at univie.ac.at  Tue Jun  6 02:27:48 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Tue, 6 Jun 2000 08:27:48 +0200
Subject: Python's RE
References: <393B7A18.D0F1AA78@barak-online.net> <960229654.1963595132@news.libertysurf.fr>
Message-ID: <8hi5fc$5ies$1@www.univie.ac.at>

Jerome Quelin <jerome.quelin at insalien.org> schrieb in im Newsbeitrag:
960229654.1963595132 at news.libertysurf.fr...
> Evgeny Zemlerub <oblom at barak-online.net> wrote:
> >How exectly do I something that reassembles perl's :
> >$string = "qwerwe kkljlksdfg (first string) asdkjflkjg
> >(secondstring)jlskjlfdkg(one more string)";
> >@words = ($string =~ /\((.+?)\)/smg);
> >foreach(@words){
> >        print "$_\n";
> >}
> Python does not support non-greedy quantifiers (ie, *?, +? and {,}?)...
> In fact, Perl is the only language that does, I think. You are to do it
> another way in Python.

You are wrong. From the library reference:

*?, +?, ??
The "*", "+", and "?" qualifiers are all greedy; they match as much text as
possible. Sometimes this behaviour isn't desired; if the RE <.*> is matched
against '<H1>title</H1>', it will match the entire string, and not just
'<H1>'. Adding "?" after the qualifier makes it perform the match in
non-greedy or minimal fashion; as few characters as possible will be
matched. Using .*? in the previous expression will match only '<H1>'.

Matthias






From richard at insight20.freeserve.co.uk  Sun Jun 11 10:27:11 2000
From: richard at insight20.freeserve.co.uk (richard at insight20.freeserve.co.uk)
Date: 11 Jun 2000 15:27:11 +0100
Subject: MySQL module
References: <8hult4$t1a$0@216.39.141.21> <3942DC44.4A85EEA7@rubic.com> <39430A2E.99CE3FC@yahoo.com> <8hv4r9$pnt$0@216.39.141.21>
Message-ID: <864s70s37k.fsf@insight20.freeserve.co.uk>

dana_booth <dana at oz.net> writes:

> Sam Wun <samwun at yahoo.com> wrote:
> 
> SW: I've tried to use command python build.py to build the module, but it
> SW: complains the _mysqlmodule.so is not found...
> 
> I got ahold of this module, and it compiled just fine. Just make sure that
> in the "Setup" file, the paths to mysql are okay. For instance, on mine, I
> had to change from "/usr/include" and "/usr/lib" to "/usr/local/..."

Yes, I've trodden this path in the last few days!!!

Expanding Dana's advice - the compiler needs to be able to "see" the
MySQL data structures, which is why you have to give a path to
mysql.h, and link to its libraries, hence the path to them also
(that's technically correct, isn't it?)

I ended up with
_mysql _mysqlmodule.c -L/usr/local/mysql-3.22.32-pc-linux-gnu-i686/lib -I/usr/local/mysql-3.22.32-pc-linux-gnu-i686/include -lmysqlclient

in replacement for the original
"_mysql _mysqlmodule.c -L/usr/lib/mysql -I/usr/include/mysql -lmysqlclient"

Richard Smith


From PClaerhout at CREO.BE  Thu Jun 22 08:27:15 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 14:27:15 +0200
Subject: List of all running processes under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B0301301213@msgeuro1.creo.be>

> Hello,
> 
> anyone knows a fuction in Python to get a list off all currently running
> processes under NT, so that we can get something similar like the
> task manager?
> 
> Pieter



From mark at chem.uwa.edu.au  Wed Jun 28 00:09:22 2000
From: mark at chem.uwa.edu.au (Mark C Favas)
Date: 28 Jun 00 04:09:22 GMT
Subject: Controlling an external program with Python
References: <nz855.1583$iN5.265679@ptah.visi.com> <slrn8l87is.dpi.keating@mars.rcode.com.au> <kuzoobue0g.fsf@lasipalatsi.fi> <8j7vc3$6b6q$1@nntp6.u.washington.edu>
Message-ID: <mark.962165362@declan>

Donn Cave <donn at u.washington.edu> writes:

>Quoth ge at nowhere.none (Grant Edwards):
>| In article <kuzoobue0g.fsf at lasipalatsi.fi>, Erno Kuusela wrote:
>|>>>>>> "Tony" == Tony Keating <keating at mech.uq.edu.au> writes:
>|>
>|>    Tony> start the program, then send various strings to the program,
>|>    Tony> read some output from the program then send some more
>|>    Tony> strings to the program.
>|>
...
<snip>
...
>|
>| Or you can roll your own using the popen2 module.

>OK, but check to see if your application will work reliably on
>a pipe, before you put too much work a pipe solution.  I append
>a short sample program that works with the shell as a remote
>application, but deadlocks with awk.  The awk behavior is what
>I would expect from most programs, so normally you're sunk.

If it won't work with pipes, have a look at the pty library module...

Cheers,
	Mark
--
Email  - mark at chem.uwa.edu.au      ,-_|\                           Mark C Favas
Phone  - +61 9 380 3482           /     \               Department of Chemistry
Fax    - +61 9 380 1005      ---> *_,-._/   The University of Western Australia
                                       v                               Nedlands
Loc    - 31.97 S, 115.81 E                               Western Australia 6009


From kens at sightreader.com  Thu Jun 15 23:18:41 2000
From: kens at sightreader.com (Ken Seehof)
Date: Thu, 15 Jun 2000 20:18:41 -0700
Subject: How Fast Does Python Run?
References: <8ib1bs$1pv@autodesk.autodesk.com> <8ib2e6$ens$1@slb1.atl.mindspring.net>
Message-ID: <39499C91.3A1586BE@sightreader.com>

Aahz Maruch wrote:

> In article <8ib1bs$1pv at autodesk.autodesk.com>,
> Akira Kiyomiya <akira.kiyomiya at autodesk.com> wrote:
> >
> >I wonder if Python runs faster than Perl.
>
> Python almost never runs "faster" than Perl.  And there are some things
> for which Perl is indisputably faster than Python (such as regular
> expressions on a text file).  Where you gain in Python, though, is that
> it's a much cleaner language, so you stand a greater chance of writing
> an efficient algorithm in the first place -- and faster algorithms are
> what get you faster programs.
> --
>                       --- Aahz (Copyright 2000 by aahz at netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "real love can't be explained by simplistic platitudes."  --piranha

And python will get faster.  There is no intrinsic reason for perl to be faster
than python; it's just been around longer and has been optimized more.

--
Ken Seehof
kens at sightreader.com
starship.python.net/crew/seehof
Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kens.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000615/40218158/attachment.vcf>

From s2mdalle at titan.vcu.edu  Wed Jun  7 21:26:42 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Wed, 07 Jun 2000 20:26:42 -0500
Subject: toString()
Message-ID: <8hmspg$r7n$1@bob.news.rcn.net>

Sorry if this is a FAQ - I can't seem to find the answer anywhere.

Is there such thing as a toString() for objects like java has that
will be automatically called when an object is used "as a string"?

I.e. when you create an object, __init__ gets called automatically,
so if I say:

x = Spam()
print x

can I put a method in Spam such that it will print out something
useful based off of the class fields rather than the ugly data 
definition?

Thanks,
-- 
David Allen
http://opop.nols.com/



From rcc at nospamthanks_jennic.com  Tue Jun 13 06:10:59 2000
From: rcc at nospamthanks_jennic.com (Robert Cragie)
Date: Tue, 13 Jun 2000 10:10:59 GMT
Subject: tkinter question
References: <39454A0E.1A630514@uniserve.com> <yad15.25$x_6.8854@news1.rdc2.tx.home.com>
Message-ID: <newscache$rb83wf$404$1@jenpc07.jennic.com>

An example might have been useful...

q = Button( text="Quit", command = lambda text="sure?":quitit(text) )

I'm not exactly sure what you're trying to do in your first (and second)
questions. It looks like there might be some scope issues there though.

I would recommend buying John Grayson's book "Python and Tkinter
Programming", ISBN 1-884777-81-3.

Robert Cragie

Rainer Deyke <root at rainerdeyke.com> wrote in message
news:yad15.25$x_6.8854 at news1.rdc2.tx.home.com...
| Bob van der Poel <bvdpoel at uniserve.com> wrote in message
| news:39454A0E.1A630514 at uniserve.com...
|
| > Third, how can I define a widget command with an arg. For example:
| >
| > q = Button( text="Quit", command = quitit("sure?") )
|
| One word: lambda.
|
|
| --
| Rainer Deyke (root at rainerdeyke.com)
| Shareware action/role-playing games      -      http://rainerdeyke.com
| "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor
|
|




From alex at magenta.com  Tue Jun 27 09:43:59 2000
From: alex at magenta.com (Alex Martelli)
Date: Tue, 27 Jun 2000 15:43:59 +0200
Subject: boolean true and false values.
References: <slrn8lf3q9.us.hat@se-46.wpa.wtb.tue.nl> <memo.20000626221037.43263C@digitig.compulink.co.uk> <slrn8lgsjq.us.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <8jabbe0148h@news2.newsguy.com>

Albert Hofkamp <hat at se-46.wpa.wtb.tue.nl> wrote in message
news:slrn8lgsjq.us.hat at se-46.wpa.wtb.tue.nl...
    [snip]
> Until now, I like the basic idea of Python of variables not having a type,
but
> I don't really see the advantage.

Polymorphism based on exactly those parts of the signature that
one happens to use.  A BREEZE to explain to smart non-programmers.


> >As can any tool, of course. The dynamic typing is indeed never /needed/
> >(probably one of the things Dijkstra proved!), but it's often very /very/
> >convenient indeed. If it's likely to be a problem to you -- or to the
>
> Hmm, why is it convenient then ?
> Without declarations you can simply pick another identifier when you need
a new
> variable, rather than re-use existing identifiers, which may cause all
kind of
> bugs and/or hamper readability.

    a = someKindOfThing()
    a.prebalooze()
    if someCondition():
        a = someDifferentThing()
    a.foomeegate()

As long as the objects returned from someKindOfThing() and
someDifferentThing()
both have a .foomeegate() method that does the right thing, this just works.
Of course it's an error if .foomegate() does not exist or does inappropriate
stuff.  This is "intuitive" to a smart person who is not a programmer and
has
never been exposed to "type" as a concept.

Having to declare a as holding a reference to "something which has the
following methods respecting the following invariants" (a type or signature
characterized type) seems to be much less intuitive and therefore much
less convenient.  Such elegant conveniences are part of what makes Python
such a pleasure to learn, and to teach.


Alex





From dbroadwell at mindspring.com  Sun Jun  4 03:05:24 2000
From: dbroadwell at mindspring.com (David Broadwell)
Date: Sun, 04 Jun 2000 03:05:24 -0400
Subject: Palm Python?
References: <8h8jmc$t9i$1@gaia.cdg.acriter.nl> <14649.14534.184930.984180@malibu.aero.org> <FvLsJJ.CAC.0.bloor@torfree.net>
Message-ID: <3939FFB4.A7824367@mindspring.com>

> Not that this exactly belongs in a python ng but I'm told the 
> only usable onboard programming environment is quartus, an ansi 
> forth system by Neal Bridges.

Actually forth is the bios level language of the chip, before palm os
loads. Just like a sparc, only smaller. Nope it dosen't really belong
here, but is an option for the palm people. Still given the option of
python of forth, i'll take python. (i used to know forth)


From Cedric.Lefeuvre at tcomhp43.epfl.ch  Wed Jun 21 04:11:44 2000
From: Cedric.Lefeuvre at tcomhp43.epfl.ch (Cedric Lefeuvre)
Date: Wed, 21 Jun 2000 10:11:44 +0200
Subject: an help for every function
Message-ID: <395078C0.5105D26E@tcomhp43.epfl.ch>

i am a beginner in python, and i would be happy if anyone can answer me
if there is an help or a man function like in unix-like OS that we can
use to describe any function of python.



From petersc at stud.ntnu.no  Wed Jun 14 17:57:23 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Wed, 14 Jun 2000 23:57:23 +0200
Subject: tkinter/windows/starting html
References: <39476C94.DA0AA2AD@stud.ntnu.no> <028398e6.cd93152e@usw-ex0108-063.remarq.com> <3947D56E.4FCD93CA@NameConnector.com>
Message-ID: <3947FFC3.83EF9FA9@stud.ntnu.no>

Geoff Talvola wrote:
> 
> This works for me (assuming you have the win32 extensions installed):
> 
> import win32api
> win32api.ShellExecute(0, 'open', r'c:\junk.html', None, None, 0)

Thanks! I'll remember that for next time. This time I just exited from
Tkinter before calling the program. That's fine for this case, though
not generally useful at all.

the-biggest-fun-was-making-python-8.3-conform-because-the-CD-
had-to-be-like-that-ly y'rs Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From herzog at online.de  Thu Jun 22 07:13:25 2000
From: herzog at online.de (Bernhard Herzog)
Date: 22 Jun 2000 13:13:25 +0200
Subject: Problem in _tkinter.c with extending both tk and python with C++
References: <3951C0DC.8D2F15C0@iki.fi>
Message-ID: <m34s6mklyi.fsf@greebo.nodomain.de>

Jyrki Alakuijala <Jyrki.Alakuijala at mail.dosetek.varian.com> writes:

> I have an old python/tkinter application that used to work fine
> with Python 1.4 and 1.5. Roughly one year ago it got
> broken. The application is mostly C++ and Python with some
> "embedded" tcl. Now, with 1.5.2 the program crashes
> similarly in both Windows NT and Linux environments.
> 
> The problem occurs with the following pattern:
> First, I load a (self-made) tk widget as a dynamic library.
> 
> Second, I load the (self-made) python extensions as dynamic
> libraries.
> 
> Third, I make an operation in the user interface that
> starts as a Tk-event, going to python. Python updates
> my widget, which calls C++ code, which calls
> (Tcl_VarEval) which tries to visit Python code.
> This is when the program crashes.
> 
> The error message it gives before termination is:
> Fatal Python error: PyEval_RestoreThread: NULL tstate
> 
> I believe this has something to do with the thread
> safety of tkinter (ENTER_PYTHON and
> LEAVE_PYTHON). Perhaps I should take these
> macros into my code for visiting Tcl code from C++.
> However, I cannot get access the Tcl-lock declared in
> _tkinter.c, because it is static.

That's exactly what happens in _tkinter in Python 1.5.2. 

My drawing program Sketch was broken by this change when 1.5.2 came out
last year, too. Unfortunately, by the time I figured out what was
happening it was too late to make any more changes to Python.

> What should I do to get the code working with
> newer releases of Python? Should I give up with
> Tk and move to one of the newer UI libraries?
> Should I write my own _tkinter.c?

My solution was to distribute a modified _tkinter with Sketch and import
it instead of the standard one.


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


From cut_me_out at hotmail.com  Tue Jun 20 14:36:58 2000
From: cut_me_out at hotmail.com (Alex)
Date: Tue, 20 Jun 2000 18:36:58 GMT
Subject: Deja is useless?! [was: Dr. Dobb's Python-URL! - weekly Python news and   links (Jun 19)]
References: <5FFA24BF1B51EF65.C00EF894C61B0276.0559D51D92EAC74D@lp.airnews.net> <394E96D2.5787F2A3@gmx.de>
Message-ID: <ETD8ZW15MVD.FSF@W20-575-109.MIT.EDU>

Down the bottom of the page, you will see a link saying 'View original
usenet format.'  If you click on that, you can get the proper formatting
for python scripts.

http://www.deja.com/getdoc.xp?AN=635109618&fmt=text

But I wish the Dr. Dobb's weekly would give these links instead of the
original Deja links.

Alex.


From darrell at dorb.com  Wed Jun 28 14:21:21 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 28 Jun 2000 11:21:21 -0700
Subject: float, '%.2f' help... please!
References: <39596682.108805343@news1.on.sympatico.ca>
Message-ID: <07f501bfe12d$a9ac6a10$6401a8c0@home>

I've had a bad record lately :)
But maybe I can get this one right.

>>> print round.__doc__
round(number[, ndigits]) -> floating point number

Round a number to a given precision in decimal digits (default 0 digits).
This always returns a floating point number.  Precision may be negative.
>>> print '%.2f' % round(1.032, 2)
1.03
>>>

--Darrell


----- Original Message ----- 
From: <chibaA at TinterlogD.Tcom>
> 
> I'm just kinda stuck for answers...  I need to display a price -
> rounded up to the proper cent.  I was using this line:
> 
> print '%.2f' % (price+0.005)
> 





From david_ullrich at my-deja.com  Wed Jun 28 12:53:28 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Wed, 28 Jun 2000 16:53:28 GMT
Subject: what is lambda?
References: <8jd8uu$gnh$1@nnrp1.deja.com>
Message-ID: <8jdaht$i43$1@nnrp1.deja.com>

In article <8jd8uu$gnh$1 at nnrp1.deja.com>,
  kpmurphy at my-deja.com wrote:
> I am a beginner at python and I would like to know what lambda does.
> Could someone post an in-your-own-words definition and possibly an
> example?  Thanks,

   I actually think it's covered in the docs. What it does
is allow you to construct "anonymous functions".

   Oops, not my own words. The syntax

lambda x: [expression in x]

is itself an expression - the _value_ of the expression
"lambda x: [expression in x]" is a function which returns
[expression in x] when passed x. For example "lambda x: x+x"
is a function that returns x+x; saying

f = lambda x: x+x

has the same effect as saying

def f(x):
  return x + x

   But you don't usually use it that way - people use
lambda when they want to pass a function to some
routine without having to make up a name for the
function. For example map takes a function as a
parameter; saying

print map(lambda x: x+x, [1,2,3])

is the same as saying

def f(x):
  return x + x

print map(f, [1,2,3]])

   You should note there's varying opinions on whether
lambda is a good thing - if you're a beginner at programming
as well as with Python it's not the first thing you should
worry about.

DU

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


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


From david_ullrich at my-deja.com  Tue Jun 13 12:58:21 2000
From: david_ullrich at my-deja.com (david_ullrich at my-deja.com)
Date: Tue, 13 Jun 2000 16:58:21 GMT
Subject: Does Python help with the no-Unix handicap?
References: <ltq8ksc1a4jnpb54bkgeho8sdu32k561bf@4ax.com> <8i34pv$2ruu$1@nntp6.u.washington.edu>
Message-ID: <8i5p73$6eo$1@nnrp1.deja.com>

In article <8i34pv$2ruu$1 at nntp6.u.washington.edu>,
  Donn Cave <donn at u.washington.edu> wrote:
> [...]
> there are places where platform neutrality isn't
> really feasible, and there it seems to me UNIX often prevails.  The
> main thing I notice is the documentation,

    One of the very few things that bugs me about the Python
docs... let's say one of the _incredibly_ few things that
bugs me about the Python docs is where the explanation for
how something works refers to a UNIX manpage.

DU

> e.g., invoking the python
> interpreter in the tutorial, but I really only use it on UNIX and
> similar (BeOS), you may notice more functional issues.
>
> 	Donn Cave, donn at u.washington.edu
>


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


From scarblac-spamtrap at pino.selwerd.nl  Thu Jun 29 09:16:21 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 29 Jun 2000 13:16:21 GMT
Subject: How to convert a string to hex?
References: <3952EC0D.D5CF9066@esec.com.au> <wdD45.7813$_55.169378@news2-win.server.ntlworld.com> <395B4C56.19EFB075@wanadoo.dk>
Message-ID: <slrn8lmiqt.7hd.scarblac-spamtrap@flits104-37.flits.rug.nl>

Jens Arnfelt wrote in comp.lang.python:
> Is there any easy way to go the other way, like convert hex number
> "0x26100511FFFF" to a string with the the following value: "26100511FFFF"
> ?

x = 0x26100     # (your example is too large for my computer)

y = "0x%x" % x  # String formatting "%x" formatter makes hex numbers

y is now "0x26100".

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  3:13pm  up 115 days,  3:26,  5 users,  load average: 0.52, 0.52, 0.48


From jsantaniello at my-deja.com  Tue Jun  6 23:03:11 2000
From: jsantaniello at my-deja.com (jsantaniello at my-deja.com)
Date: Wed, 07 Jun 2000 03:03:11 GMT
Subject: SSL and POST with httplib or urllib
References: <8hhjpn$c7$1@nnrp2.deja.com> <3daegzpjsr.fsf@newcnri.cnri.reston.va.us>
Message-ID: <8hke17$44c$1@nnrp1.deja.com>

That was it! Besides not using the httplib right, which I only started
using as an experiment after I couldn't get urllib to work, I'm in
business.

Thanks everybody.

Joe

In article <3daegzpjsr.fsf at newcnri.cnri.reston.va.us>,
  Andrew Kuchling <akuchlin at mems-exchange.org> wrote:
> jsantaniello at my-deja.com writes:
> > cgi I'm trying to access requires POST. Are there
> > any particular strangenesses to using POST that
> > anyone knows about? Does someone have a
>
> There were bugs in the SSL support in the urllib.py included in
> 1.6a2; they've been fixed in the CVS tree.  You can download
> the CVS version from SourceForge:
> http://cvs.sourceforge.net/cgi-
bin/cvsweb.cgi/python/dist/src/Lib/urllib.py?cvsroot=python
>
> --amk
>


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


From garry at sage.att.com  Wed Jun 14 11:42:06 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 14 Jun 2000 15:42:06 GMT
Subject: SV: Python Productivity over C++
References: <20000613225554.27650.qmail@hotmail.com>
Message-ID: <3947A7CE.92B83BBC@sage.att.com>

Hung Jung Lu wrote:

> And guess what? No matter whether
> it is Perl or C++, I always do proper indentation and match up all the
> statements and parentheses neatly by hand, so do all the dozens programmers
> that I have worked with. So the parenthesis and the semicolons sit there for
> ... I guess, to be pretty? :)

makes the parser easy to write.
not that this helps the users at all.

-- 
Garry Hodgson                   Every night 
garry at sage.att.com                a child is born
Software Innovation Services        is a Holy Night.
AT&T Labs                         - Sophia Lyon Fahs


From Norman_Shelley-RRDN60 at email.sps.mot.com  Fri Jun 30 19:54:37 2000
From: Norman_Shelley-RRDN60 at email.sps.mot.com (Norman Shelley)
Date: Fri, 30 Jun 2000 16:54:37 -0700
Subject: UserDict and UserList issue (bug?_
Message-ID: <395D333D.9240593C@email.sps.mot.com>

in UserDict and UserList there is one or more places where a call to
self's __class__() method is made, e.g.

def __getslice__(self, i, j):
  i = max(i, 0); j = max(j, 0)
  # This class and its subclasses need more arguments then 0
  userlist = self.__class__()
  ....


This blows causes major problems when either of these two classes are
subclassed and the class which subclasses them requires creation
arguments.  I think maybe a copy of self instead of a __class__() may be
a more robust solution.






From sblakey at freei.net  Thu Jun  1 14:07:54 2000
From: sblakey at freei.net (Sean Blakey)
Date: Thu, 1 Jun 2000 11:07:54 -0700
Subject: Python to call commands, e.g. "find , perl scripts  , nmap , and others" is this possible?
In-Reply-To: <393692FB.8610EBCA@dallas.net>; from jovecka@dallas.net on Thu, Jun 01, 2000 at 11:44:43AM -0500
References: <393692FB.8610EBCA@dallas.net>
Message-ID: <20000601110754.A25527@freei.com>

On Thu, Jun 01, 2000 at 11:44:43AM -0500, Jim wrote:
>   
>   
> Dear list, 
> 
> I'm new to Python, and I'm sorry if I'm asking about anything that has
> been 
> already discussed. 
> 
> 
> I would like to use Python to call commands, e.g. "find , perl scripts
> , nmap , and others" is this possible? 
> 
> 
> I do not need GUI. All I need is to link the out put and work with it.
> Is perl or shell scripts my only 
> answer? 
> 
> 
> jim 
> 
> -- 
> 
> Proprietary & Confidential.  The information transmitted is intended
> only for the person or entity to whom it is addressed and may
> 
> contain confidential and/or privileged material.  Any review,
> retransmission, dissemination, or other use of, or taking of any
> 
> action in reliance upon, this information by persons or entities other
> than the intended recipient is prohibited.  If you received this in
> error, please
> 
> contact the sender and delete the material from any computer.
>  

Jim,

There are many ways to call other programs from python.

First, and simplest, is os.system.  This simply fires off a sub-shell to
run the specified program, than returnsthe exit code of that program.
For example:
>>>import os
>>>os.system('wget ftp://ftp.python.org/pub/python/src/py152.tgz')
will download the python source distrobution.

If you want the output of a command (or want to write to a command on
stdin) you can use os.popen, which returns a python filehandle.  For
example, to get a listing of allfiles ending in .mp3 on your system, you
can do:
>>import os
>>import string
>>fd = os.popen('locate *.mp3')
>>filenames = string.split(fd.read, '\n')
You can also use popen to write to a process:
>>>import os
>>>fd = os.popen('/usr/sbin/sendmail nobody at localhost', 'w')
>>>fd.write('''Subject: test

this is a test.
''')
>>>fd.close()
Since you are working on a linux box, you probably also have access to
the popen2 module.  This allows you to use coprocesses (giving you
access to both stdin and stdout of a process).

The popen2 module is documented at
http://www.python.org/doc/current/lib/module-popen2.html

Note that os.popen works unreliablyon Windows boxen, and the popen2
odule is unix-specific.
    -Sean

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
"Everyone is entitled to an *informed* opinion."
-- Harlan Ellison



From paul.magwene at yale.edu  Thu Jun 22 21:28:47 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Thu, 22 Jun 2000 21:28:47 -0400
Subject: url directory listing
References: <39518289.D4175618@bioeng.ucsd.edu> <slrn8l3qfm.nol.scarblac-spamtrap@flits104-37.flits.rug.nl> <39529682.A3F00A25@bioeng.ucsd.edu>
Message-ID: <3952BD4F.51308D93@yale.edu>

Curtis Jensen wrote:
> 
> Remco Gerlich wrote:
> >
> > Curtis Jensen wrote in comp.lang.python:
> > > I little while ago, there was a post about websucker.  I'm writing a
> > > similar script. What is the socket command or url command to get a
> > > directory listing?  Especialy if there is an index.html file in the
> > > directory.  Thanks.
> >
> > This is not possible with HTTP.
> >
> > --
> > Remco Gerlich,  scarblac at pino.selwerd.nl
> > Hi! I'm a .sig virus! Join the fun and copy me into yours!
> 
> It is possible to get the files off of a site.  softbytelabs did it:
> http://www.softbytelabs.com/
> Their black widow program does a fine job of copying site files.  Maybe
> it's not done with HTTP, but does anyone know how it's done?  What does
> WebSucker do?  Thanks.
> 

websucker.py is just a convenient utility for sucking up websites.
Look in the Tools/webchecker subdirectory of your python installation. 
You can check out the code for yourself.

--Paul


From alan.gauld at gssec.bt.co.uk  Fri Jun  9 12:39:33 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Fri, 09 Jun 2000 17:39:33 +0100
Subject: GUI editor
References: <393f4568$0$25329@wodc7nh0.news.uu.net>
Message-ID: <39411DC5.48217B32@gssec.bt.co.uk>

Blu Dragoon wrote:
> Is there a gui editor written in python for tkinter?

SpecTCL with its Python bolt-on SpecPy produces 
reasonable Python Tkinter code. Personally I find 
Tkinter easy enough to use that a GUI designer is
hardly necessary but some people like 'em and 
SpecTCL works with Python/Tkinter... with Perl/Tk, 
Java/AWT, HTML and of course Tcl/Tk too.

Find it on www.scriptics.com or do a search via 
your favouraite engine.

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


From ralf at silicon.org  Tue Jun 20 20:04:28 2000
From: ralf at silicon.org (Ralf Canis)
Date: Wed, 21 Jun 2000 02:04:28 +0200
Subject: Dictionary vs if-elif?
References: <UlS35.15101$Xx5.781176@bgtnsc06-news.ops.worldnet.att.net>
Message-ID: <rg10ls89rd67ufpp48s2pr8j78q657kbdm@4ax.com>

On Tue, 20 Jun 2000 22:36:36 GMT, "William Dandreta"
<wjdandreta at worldnet.att.net> wrote:

Don't know if this is really better but how about:

bogusCodes = ( '1.', '31', '90', 'AZ' ...... )
...
if item in bogusCodes:
  do something useful



From mwh21 at cam.ac.uk  Fri Jun 16 09:03:49 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 16 Jun 2000 14:03:49 +0100
Subject: cmd, readline, and rlcompleter problems
References: <slrn8kihdh.jok.pete-python-5-17@view.kazmier.com>
Message-ID: <m3puphwzey.fsf@atrus.jesus.cam.ac.uk>

BTW: your followups-to header is broken, I think.

pete-python-5-17 at kazmier.com (Peter Kazmier) writes:

> I need help debugging a very strange problem.  I'm not a python expert
> so it could be something very simple, on the other hand, I did use the
> debugger already, but the problem disappears miraculously when I use
> the debugger.  It makes no sense to me.
> 
> So, I'm basically trying to write my own cmd and rlcompleter.  I've
> attached both the module and a small example of its use.  Its almost
> identical to the standard cmd however I am trying to offer command
> line completion.  
[snip]
> Everything works fine IF I comment out the get_line_buffer() line.
> Unfortunately, that would only do completion on the current word.
> When I uncomment the get_line_buffer() line, everything seems to work,
> but during the 'for word in self.__class__.__dict__.keys()' loop, half
> of the keys mysteriously disappear!  The DEBUG output, shows only some
> of the keys, not all of them.  Comment out the get_line_buffer() line
> and all of the keys() are back!  When I stepped through this in the
> debugger, all of the keys() were also there!!  It appears that the
> get_line_buffer() call is somehow munging my environment in a weird
> way.  Oddly enough, whwen debugging, his munging does not appear.

If I understand the situation properly, you have several problems.

The first is to understand that when your completer function is
called, the terminal is in some unusual mode (cbreak mode? something
like that).

This means you really don't want to mess with it, eg. by printing to
it.  You'll only confuse yourself; I'd try writing debugging output to
some random file and then 'tail -f'-ing it from another xterm.

It also means that trying to run it under pdb is just hopeless, I'm
afraid.

If I comment out the prints, it sort of works for me, but you still
have other problems.  The simplest is that when you type

# show vers

and bang TAB, the completer returns "show version", so readline
completes "vers" as "show version" and the buffer becomes

# show show version

Also, wouldn't you like

# show     vers

to work?  It won't currently.

Your best be may be to change the "completer_delims" to not include
space, but I think you're going to have to work a bit harder than it
would seem you hope to in your completer function.

HTH,
Michael

-- 
  Or here's an even simpler indicator of how much C++ sucks: Print
  out the C++ Public Review Document.  Have someone  hold it about
  three feet  above your head and then drop it.  Thus  you will be
  enlightened.                                        -- Thant Tessman


From dick_streefland at tasking.com  Tue Jun 20 12:28:45 2000
From: dick_streefland at tasking.com (Dick Streefland)
Date: 20 Jun 2000 16:28:45 GMT
Subject: Why whitespace denotation of blocks doesn't work.
References: <394ac7ea.609024015@nntp.interaccess.com>
    <ted.drain-2006000833210001@mjordan.jpl.nasa.gov>
Message-ID: <8io63t$fj66$1@empoli.tasking.nl>

ted.drain at jpl.nasa.gov (Ted Drain) wrote:
| I'll write some code, and as I start debugging/prototyping/etc etc, I find
| that I have to enclose some existing code in a control block (if,etc). 
| With C/C++, I just put the code in brace, tell xemacs to re-indent the
| block, and everything's good.  With Python, I have to hand edit everything
| because the editor can't figure out how to do the indentation after I've
| inserted the control block.  Once you get a few nested indentation blocks,
| this gets very confusing and leads to bugs that are hard to find (code
| that did work suddenly doesn't).

This is a limitation of your editor, not Python. With vi/vim, it's
actually quite easy to reformat. For instance, to indent a paragraph,
you can use >}, indent to the bottom of the screen >L, etc. I don't
want to start yet another editor flame-war here, but perhaps someone
should make the indenting code of emacs aware of Python syntax.

Personally, I like the use of indenting for scoping. It makes it
impossible to have a mismatch between layout and syntax. For example,
look at the following C fragment:

  if (condition);
  {
    doit();
  }

The layout suggests that the compound statement is conditionally
executed, while it is not. And yes, I've seen these things happen.

-- 
Dick Streefland                      ////            TASKING Software BV
dick_streefland at tasking.com         (@ @)         http://www.tasking.com
--------------------------------oOO--(_)--OOo---------------------------


From aahz at netcom.com  Mon Jun 19 19:57:35 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jun 2000 23:57:35 GMT
Subject: os.path.isdir strange behaviouring
References: <394e3d29.0@katana.legend.co.uk>
Message-ID: <8imc1f$181$1@slb6.atl.mindspring.net>

In article <394e3d29.0 at katana.legend.co.uk>,
Thomas Mangin <thomasm at legend.co.uk> wrote:
>
>I notice that if a directory is root owned and chmoded 700 (let say /test)
>and if it contain a directory called test2
>os.path.isdir("/test/test2") return 0 if called by a "normal user"
>I was expecting an exception.

Why were you expecting an exception?
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"The only problem with Microsoft is they just have no taste." --Steve Jobs
(From _Triumph of the Nerds_ PBS special)


From dillera at isc.upenn.edu  Sun Jun 25 15:41:39 2000
From: dillera at isc.upenn.edu (Andrew Diller)
Date: Sun, 25 Jun 2000 15:41:39 -0400
Subject: Python for Palm ?
References: <8j56ir$7fr$1@ctb-nnrp2.saix.net>
Message-ID: <8j5n9o$grt$1@netnews.upenn.edu>

"D De Villiers" <ddevilliers at lando.co.za> wrote in message
news:8j56ir$7fr$1 at ctb-nnrp2.saix.net...
> Hello, Everyone
>
> I just started programming my Palm device and is looking for a version of
> Python for the Palm. Does anyone know please know if such a product exist
> and where I can find it ?
>
> Lennie De Villiers

someday..... it will be here:

http://www.tismer.com/research/stackless/




From akuchlin at mems-exchange.org  Mon Jun 26 14:28:34 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 26 Jun 2000 13:28:34 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jun 26)
Message-ID: <78AC4F307AACF027.B40E3092A9337E7D.7FEE57BA69DC986E@lp.airnews.net>

    Digital Creations releases the first public alpha of ZEO,
    version 0.2.  In essence, ZEO and ZODB form a language-specific
    object database for Python objects
        http://www.zope.org/Products/ZEO/

    Your editor writes a draft of an article about using ZEO and
    ZODB
        http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

    PythonWare releases an implementation of SOAP 1.1.  "This
    version is somewhat experimental, and has a couple of known
    shortcomings that we plan to fix asap. It has been tested against
    Userland's SOAP implementation, and it's likely that it will work
    with IBM's Java library (we'll know more about that later this
    week)."
        http://www.pythonware.com/products/soap/

    Les Smithson mentioned having some Python code for X10 home
    automation; contact him if you want to help out.
        http://www.deja.com/getdoc.xp?AN=636731298&fmt=text          	

    The 1.6 CVS tree has been hopping in preparation for a new release.
    New stuff: Greg Ward announces autoconf-like features in the
    Distutils
        http://www.python.org/pipermail/distutils-sig/2000-June/001636.html
    An example config
        http://www.python.org/pipermail/distutils-sig/2000-June/001637.html
          
    And Fredrik Lundh makes a bit of 1.6's Tkinter significantly
    faster:
        http://www.python.org/pipermail/python-dev/2000-June/006165.html


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

Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      eff-bot's complements this digest with his daily python url
          http://hem.passagen.se/eff/url.htm

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Python FAQTS
          http://python.faqts.com/

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

      Archive probing trick of the trade:
          http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From tim_one at email.msn.com  Sun Jun  4 17:23:39 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 4 Jun 2000 17:23:39 -0400
Subject: Questions regarding design decisions in listobject.c
In-Reply-To: <393ABA63.18E2F46@san.rr.com>
Message-ID: <000401bfce6b$26a91160$b02d153f@tim>

[Courageous]
> 1. For slice operations, ilow/ihigh indices which undeflow
> or overflow the list in question are reset to the edges
> of the list. Why was it that PyExc_IndexError wasn't
> thrown instead?

Because it's more convenient more often the way it is.

> 2. In slice expressions of the form s[i:j], what was the
> design decision that lead to s[i:i] always returning an
> empty list and s[i:i+1] selecting one item. At first
> glance, it would appear that s[i:i] is the obvious and
> intuitive correct expression to return just one item on
> a slice, where s[i:i+1] ought to select two items and so
> forth. Was there some particular situation which arose
> which required this slightly counterintuitive idiom?

It's highly intuitive after you understand the key point:  in Python,
indices conceptually point *between* elements, not *at* them.  Everything
follows from that.  Once you understand that, it's almost impossible to make
an off-by-1 error, and that len(x[i:j]) = j-i (ignoring pathologies) has a
lot to do with that.  So does that for any k, x[i:j] == x[i:k] + x[k:j]
(again ignoring pathologies).  The "between elements" idea has been
reinvented many times in many languages (e.g., see Icon and Java).  As far
as he knows, Guido thought it up on his own, but I routinely accuse him of
stealing it from Icon <wink>.

whatever-it's-a-major-feature-ly y'rs  - tim





From michael.drumheller at boeing.com  Sat Jun 17 17:48:33 2000
From: michael.drumheller at boeing.com (Michael Drumheller)
Date: Sat, 17 Jun 2000 21:48:33 GMT
Subject: writing (Gnu)MAKE in Python
References: <NEBBIOGBDMDPEANLFKGHCEPACDAA.jschmitt@vmlabs.com>
Message-ID: <394BF231.8B5BB387@boeing.com>

You might want to look at cons for ideas too.
http://www.baldmt.com/cons-faq/current.html

John Schmitt wrote:
 
> I'd love to take the diversion of writing Make using Python.  The things

> John

-- 
Michael Drumheller--Boeing Mathematics & Computing Technology
PO Box 3707 M/S 7L-40 Seattle 98124-2207 (425)865-3520 Fax:2965
michael.drumheller at boeing.com


From ge at nowhere.none  Thu Jun 22 15:21:22 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 22 Jun 2000 19:21:22 GMT
Subject: More control on screen output and input
References: <DZs45.19$zP.21280@news1.onlynews.com>
Message-ID: <SGt45.1331$iN5.178903@ptah.visi.com>

In article <DZs45.19$zP.21280 at news1.onlynews.com>, Jeremy Honl wrote:

>Is there some better way to control output to the screen than the print
>command, such as putting in cordinates where the text should be printed?
>How about the same for input?

There are two modules that provide that sort of thing:

  curses (uses curses)
  snack  (uses newt/slang)

I think you can find the curses module in the usual places
(www.python.org).

The docs for snack/newt are pretty skimpy.  It comes with RH,
and there are some example programs in the newt distribution.
I don't know if it's current, but you can get version 0.50 it
from my ftp site:

  ftp://ftp.visi.com/users/grante/stuff/newt-0.50.tar.gz


-- 
Grant Edwards                   grante             Yow!  Yow!! "Janitor
                                  at               trapped in sewer uses ESP
                               visi.com            to find decayed burger"!!


From aahz at netcom.com  Thu Jun  1 19:59:27 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Jun 2000 23:59:27 GMT
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com>
Message-ID: <8h6tcv$o5g$1@slb0.atl.mindspring.net>

In article <8h6rbt$ksv$1 at nnrp1.deja.com>,  <t_e_sanders at my-deja.com> wrote:
>
>> Permission to use, copy, modify, and distribute this software and its
>> documentation for any purpose and without fee is hereby granted,
>> provided that the above copyright notice appear in all copies and
>> that both that copyright notice and this permission notice appear in
>> supporting documentation...
>
>Do the terms of Python's distribution license require me to allow other
>users to copy and distribute my software? I would guess not, but
>inclusion of this notice in my package explicitly states so. There's
>obviously something I'm missing, (one of the downsides of being a
>developer, rather than a lawyer? :) Can anyone shed some light on this?

IANAL, but I know something about this subject (whether what I know is
correct is another issue ;-):

What this license means is that your customers need to be free to
distribute Python, but not your Python scripts.  As an example, use of
gcc does not force you to allow your users access to your source code.
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

Have you coined a word today?  --Aahz


From nospam at bitbucket.com  Wed Jun 21 21:41:02 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Wed, 21 Jun 2000 18:41:02 -0700
Subject: Access violation when no memory
References: <olD35.82$_%.4481@newsfeed.avtel.net> <slrn8kvpml.flp.perly@typhoon.xnet.com>
Message-ID: <o9e45.18$FG1.1680@newsfeed.avtel.net>

Pat Szuta wrote in message ...
>I think that this might be a problem with the python GC.  Basically you
>create circular lists,  and because their reference count never goes down
>to zero, the GC never frees them.  For example:
>ython 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux
>(egcs- on linux-i386
>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>> while 1 == 1:
>...     x = [1]
>...     y = [2]
>...     x[0] = y
>...     y[0] = x
>...
>
>This will eat up your memory like there is no tomorrow.  I think that the
>problem that you're describing my be similar.

No, I understand that - my program is explicitly designed to consume
all memory by progressively consuming smaller and smaller blocks,
throwing exceptions at each size.  My point is that eventually Python
/crashes/ by attempting to write to low memory instead of failing with
a MemoryError.  (Running your code on Win98 produced mixed results for
me, sometimes throwing MemoryError and sometimes crashing, probably
because the memory allocation is less controlled than in my sample.)

So I need a ceval.c expert to vet my change.  Let me reiterate that
this is not an academic case, as beta-testing with users has produced
crashes on Win98 machines with little virtual memory.

>On Mon, 19 Jun 2000 22:20:46 -0700, Phil Mayes <nospam at bitbucket.com>
wrote:
>>The following program progressively allocates all memory:
>>    size = 1 << 20
>>    list = [None,]
>>    while size:
>>        try:
>>            mem = [None] * size
>>            mem[0] = list
>>            list = mem
>>        except:
>>            size = size / 2
>>
>>It fails with an access violation.  (Turn off virtual memory
>>to provoke it.)  This is not a theoretical situation for me:
>>I have a beta release in the field that is crashing like this.
>>The immediate cause is this:
>>
>>1. we are handling a MemoryError
>>2. the following call cannot allocate a traceback object
>>     ceval.c line 1800:    PyErr_Fetch(&exc, &val, &tb);
>>3. the NULL pointer at tb is pushed here:
>>     ceval.c line 1816:    PUSH(tb);
>>     ceval.c line 1817:    PUSH(val);
>>     ceval.c line 1818:    PUSH(exc);
>>4. later it is popped here and its ref count is decremented:
>>     ceval.c line 682:    case POP_TOP:
>>     ceval.c line 683:        v = POP();
>>     ceval.c line 684:        Py_DECREF(v);
>>     ceval.c line 685:        continue;
>>
>>This is with 1.5.2 but the crash still happens with 1.6a2.
>>At this stage I am way out of my depth, and  the only way I
>>see to fix this is to change Py_DECREF to Py_XDECREF, which
>>is a (minor?) inefficiency. Can anyone recommend a better way?
>>
>>And should I report this at http://www.python.org/python-bugs?
>>I do not have access to the current CVS tree.
>>
>>TIA, Phil Mayes
>>--
>>pmayes AT olivebr DOT com
>>
>>
>>
>
>
>--
>---------------------------------------------------------------------------
----
>Pat Szuta www.sourcelight.com
>Software Engineer "Guiding you to
>Sourcelight Technologies great choices"(TM)
>(847)-475-1000 ext. 209
--
Phil Mayes    pmayes AT olivebr DOT com





From pecora at anvil.nrl.navy.mil  Thu Jun  8 07:48:31 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Thu, 08 Jun 2000 11:48:31 GMT
Subject: Some questions about C extensions
References: <etdu2f69rhz.fsf@w20-575-109.mit.edu>
Message-ID: <080620000748310603%pecora@anvil.nrl.navy.mil>

In article <etdu2f69rhz.fsf at w20-575-109.mit.edu>, Alex
<cut_me_out at hotmail.com> wrote:

> Hi.  I'm trying to write some C extensions, and have run across a couple
> of mysterious things.  I was wondering whether anyone could explain
> them:
> 
> -- Deallocation of memory for strings returned by python functions:
>    Do I have to do it, or is it somehow taken care of?  If I have
>    something like 
> 
>       if (!PyArg_ParseTuple (args, "s", &sequence)) {
>            return NULL;}
> 
>    do I have to do a free (sequence); at some point?  I would have
>    thought so, but it seems to be giving me a Seg fault.

I will try these out:

ParseTuple just sets pointers to data already allocated in Python and
hence handled automatically there.

> -- Reference counts for elements of brand new tuples.  Why do I have to
>    do a Py_INCREF on an object before I assign it to a new tuple with
>    PyTuple_SetItem?  Since it seems as though you have to do a Py_INCREF
>    every time anyway, is there a function like PyTuple_SetItem that will
>    do it for you?  The comment in the C API manual about this is very
>    cryptic, BTW.

Hmmm...not sure about that one.  Sorry I can't be more helpful.


From horst at proceryon.at  Wed Jun  7 07:52:35 2000
From: horst at proceryon.at (Horst Gassner)
Date: Wed, 07 Jun 2000 13:52:35 +0200
Subject: Tkinter: treeview widget
Message-ID: <393E3782.FF7A5DCA@proceryon.at>

Hi all!

I am looking for a treeview widget. Does anyone know if such a widget is
available somewhere?
At the moment I am using PMW which has a very good collection of meta
widgets but unfortunately it does not contain a treeview widget.

Thanx
- Horst


From loredo at spacenet.tn.cornell.edu  Thu Jun 15 15:54:16 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 15 Jun 2000 15:54:16 -0400
Subject: Where are the math functions?
References: <owen-97F32C.10015712062000@news.u.washington.edu> <memo.20000614223200.31901D@digitig.compulink.co.uk> <20000615134357.A1087@better.net>
Message-ID: <642A954DD517D411B20C00508BCF23B001296C8A@mail.sauder.com>

Howdy-

Below are some pure Python special functions, some adapted from
Numerical Recipes, some from formulas in standard references.  They are
mostly related to the Gamma function.  This
is old stuff, practically my first Python, so don't expect much!  8-)
I didn't know about cephes at the time, and mostly use cephes now.
But these have the virtue of being pure python, so they may be helpful
to you if you can't compile cephes on your platform.

I've done some other Python translations of NR routines, soon
to appear on a web page.  Others are already on the web:

http://www.python.org/topics/scicomp/recipes_in_python.html

Peace,
Tom loredo

PS: One of the Numerical Recipes coauthors is in the office next
door, and knows I've been up to this stuff.  When I first told him
I'd been translating some NR routines into Python, his response
was, "Why?!"  8-)

##### sp_funcs.py --- Not too thoroughly tested as yet!
from math import *
import Numeric
N = Numeric

# Exceptions:

max_iters = 'Too many iterations: '

#============= Globals ===============

rt2 = sqrt(2.)
gammln_cof = N.array([76.18009173, -86.50532033, 24.01409822,
	-1.231739516e0, 0.120858003e-2, -0.536382e-5])
gammln_stp = 2.50662827465

#============= Gamma, Incomplete Gamma ===========

def gammln(xx):
	"""Logarithm of the gamma function."""
	global gammln_cof, gammln_stp
	x = xx - 1.
	tmp = x + 5.5
	tmp = (x + 0.5)*log(tmp) - tmp
	ser = 1.
	for j in range(6):
		x = x + 1.
		ser = ser + gammln_cof[j]/x
	return tmp + log(gammln_stp*ser)
	
def gser(a, x, itmax=700, eps=3.e-7):
	"""Series approx'n to the incomplete gamma function."""
	gln = gammln(a)
	if (x < 0.):
		raise bad_arg, x
	if (x == 0.):
		return(0.)
	ap = a
	sum = 1. / a
	delta = sum
	n = 1
	while n <= itmax:
		ap = ap + 1.
		delta = delta * x / ap
		sum = sum + delta
		if (abs(delta) < abs(sum)*eps):
			return (sum * exp(-x + a*log(x) - gln), gln)
		n = n + 1
	raise max_iters, str((abs(delta), abs(sum)*eps))


def gcf(a, x, itmax=200, eps=3.e-7):
	"""Continued fraction approx'n of the incomplete gamma function."""
	gln = gammln(a)
	gold = 0.
	a0 = 1.
	a1 = x
	b0 = 0.
	b1 = 1.
	fac = 1.
	n = 1
	while n <= itmax:
		an = n
		ana = an - a
		a0 = (a1 + a0*ana)*fac
		b0 = (b1 + b0*ana)*fac
		anf = an*fac
		a1 = x*a0 + anf*a1
		b1 = x*b0 + anf*b1
		if (a1 != 0.):
			fac = 1. / a1
			g = b1*fac
			if (abs((g-gold)/g) < eps):
				return (g*exp(-x+a*log(x)-gln), gln)
			gold = g
		n = n + 1
	raise max_iters, str(abs((g-gold)/g))


def gammp(a, x):
	"""Incomplete gamma function."""
	if (x < 0. or a <= 0.):
		raise ValueError, (a, x)
	if (x < a+1.):
		return gser(a,x)[0]
	else:
		return 1.-gcf(a,x)[0]

def gammq(a, x):
	"""Incomplete gamma function."""
	if (x < 0. or a <= 0.):
		raise ValueError, repr((a, x))
	if (x < a+1.):
		return 1.-gser(a,x)[0]
	else:
		return gcf(a,x)[0]

#======== Error function, normal CDF and inverse ================

def ncdf_inv(p):
	"""Inverse of the normal CDF."""
	c0 = 2.515517
	c1 = 0.802853
	c2 = 0.010328
	d1 = 1.432788
	d2 = 0.189269
	d3 = 0.001308
	
	sign = -1.
	if (p > 0.5):
		sign = 1.
		p = 1. - p
	arg = -2.*log(p)
	t = sqrt(arg)
	g = t - (c0 + t*(c1 + t*c2)) / (1. + t*(d1 + t*(d2 + t*d3)))
	return sign*g


def erfcc(x):
	"""Complementary error function."""
	z = abs(x)
	t = 1. / (1. + 0.5*z)
	r = t * exp(-z*z-1.26551223+t*(1.00002368+t*(.37409196+
		t*(.09678418+t*(-.18628806+t*(.27886807+
		t*(-1.13520398+t*(1.48851587+t*(-.82215223+
		t*.17087277)))))))))
	if (x >= 0.):
		return r
	else:
		return 2. - r


def ncdf(x):
	"""Cumulative normal dist'n."""
	global rt2
	return 1. - 0.5*erfcc(x/rt2)


def ncdf_sig (nsig):
	"""Cummulative normal dist'n inside nsig sigmas.
	ncdf_sig = 1 - 2 * (upper tail) = 1 - erfc(sigfac/rt(2))"""
	global rt2
	return 1. - erfcc(nsig/rt2)

#=============== Chi squared dist'n ==============

def pchisq(chisq, nu):
	"""Lower tail area of the chi**2 dist'n with nu dof.
	Note that chisq is *not* the reduced chis**2!"""
	hnu = 0.5 * nu
	hchi = 0.5 * chisq
	return gammp(hnu, hchi)

def qchisq(chisq, nu):
	"""Upper tail area of the chi**2 dist'n with nu dof.
	Note that chisq is *not* the reduced chis**2!"""
	hnu = 0.5 * nu
	hchi = 0.5 * chisq
	return gammq(hnu, hchi)


def chisq_crit(nu, p, tol=1.e-5):
	"""Critical chi**2 with lower tail area of p for nu dof."""

	#  For the first guess, use the assyptotic normal limit of the
	# chi**2 distribution:  chi**2 ~ N(nu,sqrt(2*nu)).
	chi = nu + ncdf_inv(p)*sqrt(2.*nu)
	pcur = pchisq(chi,nu)
	
	# Now do a Newton-Raphson loop...
	while 1:
		dfdc = (pchisq(1.001*chi,nu) - pcur) / (0.001*chi)
		chi = chi - (pcur - p)/dfdc
		pcur = pchisq(chi,nu)
		if (abs(pcur-p) <= tol):
			return chi

# Allow it to run as a script	
if __name__ == "__main__":
	print 'gser: ', gser(5., 5.)
	print 'gcf:  ', gcf(5., 7.)
	print 'gammp, gammq: ', gammp(5.,5.), gammq(5.,5.)
	print 'erfcc: ', erfcc(.5)
	print 'ncdf_inv: ', ncdf_inv(0.977), ncdf_inv(0.5)
	print 'ncdf: ', ncdf(1.)
	print 'ncdf_sig: ', ncdf_sig(2.)
	print 'q & p chisq: ', qchisq(4.,1), pchisq(4.,1)
	print 'chisq_crit: ', chisq_crit(1.,0.954)



From dgrisby at uk.research.att.com  Fri Jun  9 06:19:24 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 9 Jun 2000 10:19:24 GMT
Subject: Using python with Corba
References: <eieJ4.8408$H7.704991@brie.direct.ca> <871z4awiz0.fsf@parus.parus.no> <39410efe.18494828@nntp.interaccess.com>
Message-ID: <8hqgbc$iga$1@pegasus.csx.cam.ac.uk>

In article <39410efe.18494828 at nntp.interaccess.com>,
 Thaddeus L. Olczyk <olczyk at interaccess.com> wrote:

>>> Anyone know TAO can be used in Python ?
>>
>>Don't think so. But a Python ORB can interoperate with TAO.

>They are supposed to but (sigh ) most orbs have many parts of the
>standard unimplemented yet. Do you know if Fnorv,ILU, OmniORb
>or ORBit (didn't know ORbit supported python) can interoperate with
>TAO?

omniORB can definitely interoperate with TAO. I am not aware of any
problems at all. ORB interoperability is now at the stage that there
are very few difficulties between ORBs, and those that remain are
obscure cases which are unlikely to crop up in real systems. I don't
imagine you'll encounter any problems with any of the Python ORBs.

Cheers,

Duncan (author of omniORBpy).

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From eviltofu at rocketmail.com  Sat Jun 10 10:05:00 2000
From: eviltofu at rocketmail.com (Jerome Chan)
Date: Sat, 10 Jun 2000 14:05:00 GMT
Subject: Checking socket type?
References: <Pine.GSO.4.10.10006101618110.5325-100000@sundial>
Message-ID: <eviltofu-029D7F.10045510062000@news.earthlink.net>

In article <Pine.GSO.4.10.10006101618110.5325-100000 at sundial>, Moshe 
Zadka <moshez at math.huji.ac.il> wrote:

> On Sat, 10 Jun 2000, Jerome Chan wrote:
> 
> > How would I check to see if an object is an instance of a socket?
> 
> Use the type() function:
> 
> >>> type(socket.socket(socket.AF_INET, socket.SOCK_STREAM))
> <type 'socket'>
> 
> --
> Moshe Zadka <moshez at math.huji.ac.il>
> http://www.oreilly.com/news/prescod_0300.html
> http://www.linux.org.il -- we put the penguin in .com
> 
> 

Oh I found it in the socket documentation.

if type(s)==socket.SocketType: print "hello"


From donn at u.washington.edu  Thu Jun 29 14:17:08 2000
From: donn at u.washington.edu (Donn Cave)
Date: 29 Jun 2000 18:17:08 GMT
Subject: Deleting mails with imapib
References: <Pine.LNX.4.21.0006291419090.25358-100000@lade.src.no> <Pine.LNX.4.21.0006221223060.22204-100000@lade.src.no> <8itd35$a7lc$1@nntp6.u.washington.edu>
Message-ID: <8jg3r4$6m7g$1@nntp6.u.washington.edu>

Quoth "Morten W. Petersen" <morten at src.no>:
[quoting me]

|>     ok, stuff = self.service.store(number, 'FLAGS', '(\Deleted)')
|>     if ok != 'OK':
|>         raise self.error(stuff)
|
| Where did you take the self.service from?

Didn't take it from anywhere, "service" is just the IMAP4 instance,

    self.service = imaplib.IMAP4(host, port)

Here "self" is the BeOS message dispatch interface for the thread
that handles the IMAP service connection.

	Donn Cave, donn at u.washington.edu


From hinsen at cnrs-orleans.fr  Wed Jun  7 06:44:09 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 07 Jun 2000 12:44:09 +0200
Subject: PyOpinion: Does Python Programming Marginalize You?
References: <393D2797.4E4EE7CC@wag.caltech.edu>
Message-ID: <m3u2f5hiti.fsf@chinon.cnrs-orleans.fr>

"Richard P. Muller" <rpm at wag.caltech.edu> writes:

> I love Python. Of the 10-15 programming languages I know, Python is
> the only language which I actually look forward to using. I wake up in
> the morning, ask myself, "Will I be using Python today?" and if the
> answer is "yes" I know that it's going to be a good day.

We seem to have something in common...

> And yet, I worry that using Python marginalizes one. Python is hard to
> beat for writing applications that run on a single computer. But
> that's yesterday's computing model. The future of computing, heck, the
> present of computing, is in writing programs that run through a web
> browser. And the fact of the matter is that 99% of web browsers
> surfing the internet run Java and JavaScript, which gives these

I don't agree. Any beginner's book on the Internet warns you about the
security risks with any form of active content, and many people
disable Java and JavaScript in their browsers.

Of course future technology may change this, but then future
technology could be Web browsers running Python!

Moreover, I don't expect Web-browser applets to become the main
programming model any time soon. There is always a need for things
that just can't be done inside a browser, things that need maximum
performance, and things that need powerful server hardware.

> I'm aware of CGI scripting with Python. It's great. But JavaScript

Try Zope, it's better!

> What else should we be doing to make Python *the* programming language
> rather than *a* programming language?

Write good Python code, and make it OpenSource. This will show what
Python is good for, and provide an incentive to learn Python for all
those who want to extend these programs.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From moshez at math.huji.ac.il  Fri Jun  9 14:41:13 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 9 Jun 2000 21:41:13 +0300 (IDT)
Subject: Debuggers.
In-Reply-To: <39483789.94409484@nntp.interaccess.com>
Message-ID: <Pine.GSO.4.10.10006092140280.28574-100000@sundial>

On Fri, 9 Jun 2000, Thaddeus L. Olczyk wrote:

> Basic question what are the debuggers available for python.
> ( pdb is OK, but I wish I could restart a project. )
> Everyone seems to say that pythonwin comes with one, but I don't see
> it. OTOH I have v 1.5.1

I've used both PythonWin's and IDLE's. Personally, I can't stand
debuggers, but they both seem good enough.

Try upgrading?
--
Moshe Zadka <moshez at math.huji.ac.il>
http://www.oreilly.com/news/prescod_0300.html
http://www.linux.org.il -- we put the penguin in .com




From hzhu at mars.localdomain  Sat Jun 17 16:28:53 2000
From: hzhu at mars.localdomain (Huaiyu Zhu)
Date: Sat, 17 Jun 2000 20:28:53 GMT
Subject: Python vs. Matlab
References: <724s6sa0k5.fsf@rogn.ii.uib.no>
Message-ID: <slrn8kno2k.vh.hzhu@mars.localdomain>

Hi,

There are several options you might consider:

If you want a Matlab clone, try Octave (http://www.che.wisc.edu/octave/). It
is GPL'ed, and source compatible to a large degree.  It uses the gnuplot
program for plotting.  If you run Linux you can also install both from rpm.

But Python has distinctive advantages: You can do OO programming with
powerful built-in data structures (list, dict).  I find that as program
structures grow they become unmanageable quickly in Matlab/Octave, but
remain as easy as ever in Python.

NumPy can do most of the matrix stuff, plus all the power of Python.
However, the expressions is really painful for someone who loves Matlab,
because it was geared towards multiarrays instead of matrices. Several
people claim that after some initial investment it is still convenient to
use, but this is not the case for me, as I found I was constantly tripped
over issues like matrix and element multiplication, and transpose behave
differently on shapes of (n,1), (1,n) and (n,), etc.

So, there also came the MatPy package which I am developing, which provides
wrappers around the packages Numpy, Gnuplot, and SpecialFunc.  It is an open
source project hosted on SourceForge.  It is currently classified as in
alpha stage, but it is usable in the features that I used or tested.  I no
longer need to use NumPy directly now and I'd claim the syntax for linear
algebra is much easier than NumPy.  However, for additional features you
need to either add them yourself (not difficult I'd say), or write to me and
wait for my free time. :-) Additional developers are very welcome.

I can't compare raw speed but for complicated programs Python is much faster
because I could use better algorithms.  For Matlab/Octave often I had to
write intermediate disk files, use sed, awk or perl to transform them, and
read in by Matlab programs, all managed by bash scripts.  For Python there's
never the need to do anything like that.

And, conclusions?  I'm obviously biased given what I'm doing.  But I'd
suggest you spend a few hours giving all three a try.  It's not so difficult
on Linux anyway.

Hope this helps.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


On 17 Jun 2000 15:39:54 +0200, Helge Avlesen <avle at rogn.ii.uib.no> wrote:
>
>I have become almost addicted to Matlab after using it extensively for
>some years, however, the licencing feels very strict after getting
>used to free and very good quality software like Emacs, Linux, etc.
>
>My main use of Matlab is reading in large binary files into
>arrays(output from simulations), do some processing with various
>matrix routines, then extract sections of the 3D arrays for plotting.
>
>Spotted Python some months ago, and liked it!  Can Python, Numpy and
>the many plotting extensions be used as a replacement for Matlab, is
>Python slower/faster etc.?  
>
>It would be nice to hear some experiences!
>
>-- 
>Helge Avlesen


From thomas at xs4all.net  Sat Jun 10 15:53:52 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sat, 10 Jun 2000 21:53:52 +0200
Subject: No strptime in Windows?
In-Reply-To: <8hu4tc$6m0$1@nnrp1.deja.com>; from david_ullrich@my-deja.com on Sat, Jun 10, 2000 at 07:28:56PM +0000
References: <8hu4tc$6m0$1@nnrp1.deja.com>
Message-ID: <20000610215351.L20700@xs4all.nl>

On Sat, Jun 10, 2000 at 07:28:56PM +0000, david_ullrich at my-deja.com wrote:

>     The docs say that time.strptime may not be available
> on all platforms. It doesn't seem to be available under MS
> Windows. I don't see why not - there's nothing Windological
> about strptime but the same applies to the other time
> functions that _are_ included.

time.strptime() is only included if the underlying platform supplies it.
Apparently, the Windows platform does not. The issue is certainly not wether
the underlying OS is UNIXy enough: BSDI, which is derived fairly directly
from 'the' original UNIX, does not supply strptime(), so there is no
time.strptime() on that platform.

A replacement strptime() (probably based on the glibc sources)
has been fairly high on my wishlist, but fairly low on my TODO ;) I'm sure
it's quite a big and complicated piece of code :P

>     Also: Is Monday really day 0? I woulda thought
> Sunday was day 0. (Doesn't seem to matter in the
> subtract-five-hours thing, I get the same results
> either way. Evidently mktime() ignores the day-of-week
> if it's inconsistent with the day-of-month field?)

The strptime manpage on Linux says this:

              %w     weekday number  (0-6)  with  Sunday  as  the
                     first day of the week

But perhaps this is different on Windows ? :-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From dgoodger at bigfoot.com  Fri Jun 23 21:52:41 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Fri, 23 Jun 2000 21:52:41 -0400
Subject: writelines() or write()
In-Reply-To: <3953C7B9.E32E8599@seas.upenn.edu>
References: <3953C7B9.E32E8599@seas.upenn.edu>
Message-ID: <B5798CA8.5512%dgoodger@bigfoot.com>

on 2000-06-23 16:27, dave white (dwhite2 at seas.upenn.edu) wrote:
> When I use writelines() and my strings have \r characters at the end
> somehow my output file only contains the last line.  What's the correct
> way to write multiple lines to a file?
> 
> strings = ['one \r', 'two \r', 'three \r}
> outfile.writelines(strings)
> ---
> outfile looks like this:
> 'three '
> ---
> What I want to have happen of course is:
> 'one '
> 'two '
> 'three '

Could it be, perhaps, that the last line is simply overwriting the first
two? You're specifying '\r', carriage return, which may not start a new line
first. On MacOS, \r is the line separator, but on Unix/Linux it's \n, and on
DOS/Windows it's both. Try this in the interactive interpreter:

>>> strings = ['one \r', 'two \r', 'three \r']
>>> f=open('tmpfile','w')
>>> f.writelines(strings)
>>> f.close()
>>> open('tmpfile','r').read()
'one \015two \015three \015'

(Don't use "print"! The interpreter will automatically escape all control
characters in expression statements, a handy debugging feature.) Do you get
the same result as above? Try using \n instead of \r in your code.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From digitig at cix.co.uk  Sun Jun 25 18:07:00 2000
From: digitig at cix.co.uk (Tim Rowe)
Date: Sun, 25 Jun 2000 23:07 +0100 (BST)
Subject: boolean true and false values.
References: <slrn8l69kp.geg.hat@se-46.wpa.wtb.tue.nl>
Message-ID: <memo.20000625230712.5257C@digitig.compulink.co.uk>

In article <slrn8l69kp.geg.hat at se-46.wpa.wtb.tue.nl>, 
hat at se-46.wpa.wtb.tue.nl (Albert Hofkamp) wrote:

>      I personally use the idea of 'first statement on a line messes up 
> an
>      invariant, the remainder of the line restores the invariant' in 
> C++, which
>      works great.

It works well in very simple cases. In more complicated cases you need 
something like Eiffel's structures for maintaining contracts and 
invariants. Basically where there's a lot of work to be done between 
breaking the invariant and having it fully fixed.

> Yes, I know, but why is it not in the language ?
> Quite a lot of algorithms rely on having sets. If you put them in a 
> list, there
> is an order which is not always wanted.

There is always an order, but you don't always know what it is. When you 
write your set class you just document that if the user iterates over the 
elements then the order in which they will be returned is undefined -- 
that's what happens in the C++ STL. The fact that the implementation has a 
natural ordering is no longer any use to the user. If you want to ram that 
home you can always shuffle the list from time to time :-)

> Also, implementing a set in anything else than a set is conceptually 
> not clean.
> (to be fair, the list operations are quite extensive, so living without 
> sets is
> doable).

I would consider implementing a set as a bag ("A set is a bag with no 
element duplicated") is conceptually clean. Of course, Python doesn't have 
bags as such, though they're easy to build...

Seriously, how do you think Python could implement it if not as a list or 
a tree or some ordered data structure? Even the Pascal set type is usually 
represented as a bit vector, which is every bit as ordered as a list...

> >delimiters on my keyboard are already used -- what would you delimit 
> it >with? :-)
> 
> {x,y,z} ?

That's going to get me confused...


From thomas at cintra.no  Wed Jun  7 09:13:13 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Wed, 07 Jun 2000 13:13:13 GMT
Subject: Why not a "Deluxe" Python-distribution
Message-ID: <393f42e4.77383471@news.online.no>

Hi,

Wouldn`t it be nice to have a deluxe edition of Python, with lots of
the adopted modules not in the standard distro? Like PIL, PST, code
from the sigs, database drivers etc. 

Alot of you might say its not needed since modules are easy to install
etc., but I think compiling code etc., which is necessary for some of
the available modules, scare newbies away from expanding python. Then
python might not do the job for them, and they chose another language
instead. 

Take the sigs xml-package for instance, I`ve had serious problems
getting this one to work, and by the looks of postings here, so have a
few others. This should be a standard part of the python-distro in my
opionon, being one of the areas where python is really getting
attention, in press and online.

What are the main obstacles or objections for making either a add-on
package containing the most popular modules or adding more modules to
the standard python distro as default?

I know I`ve asked a similar question before, but I`ve looked into
doing it myself but don`t have time right now, so I was hoping
somebody else out there just lost there job, got divorced or for some
other reason had time to kill ;->

Thomas


From ghost_man at my-deja.com  Thu Jun 22 13:20:00 2000
From: ghost_man at my-deja.com (ghost_man at my-deja.com)
Date: Thu, 22 Jun 2000 17:20:00 GMT
Subject: running pyton on NT command lines
Message-ID: <8ithrb$cj1$1@nnrp1.deja.com>

I have written several scripts that I want to run at some set time.  I
have been trying to get them to fire with the AT command line
scheduler, but keep getting errors.  I have tried several different
variations to get it to run even putting the python executable in a dos
batch file and trying to run the dos batch file.  I also have the
python registered in my Windows 2000 environment path.

my command line looks something like this

at 16:30 "myscript.py"

it fails everytime.  I have also tried putting in a complete path to
the python.exe.


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


From invalid.address at 127.0.0.1  Sun Jun 25 12:30:03 2000
From: invalid.address at 127.0.0.1 (David)
Date: Sun, 25 Jun 2000 16:30:03 GMT
Subject: Gadfly Singleton (was Re: Gadfly Question
References: <395408ae.75890752@news.telus.net> <3954CC36.EE2D4CC4@arrakis.es>
Message-ID: <395632a6.38770667@news.telus.net>

This seems to work nicely.  It creates a singleton instance and connection
to the database.  One can completely bugger it by supplying a different
dbase name/path, I suppose.
 
Feel free to fix/improve/extend this and re-post.  Thx!

======

import gadfly, os, os.path

class _DBase:
	instance = None
	connection = None
	def __init__(self, dbase="dbname", path="dbpath"):
		if not os.path.isdir(path):
			print "creating database directory: "+path
			os.mkdir(path)

		if not os.path.isfile(path+"/"+dbase+".gfd"):
			print "creating database file: "+dbase
			connection = gadfly.gadfly()
			connection.startup(dbase, path)

			cursor = connection.cursor()
			cursor.execute("CREATE TABLE spam (foo integer)")
			cursor.execute("DROP TABLE spam")

			connection.commit()

	def extendedFunctionality(self):
		print "Wowsa!"

def ConnectDB(dbase, path):
	if _DBase.instance == None:
		_DBase.instance = _DBase(dbase, path)

	_DBase.connection = gadfly.gadfly(dbase, path)

	return _DBase.instance


# stand-alone testing
if __name__ == "__main__":
	dbase = ConnectDB("testdb", "testdb")
	connection =  dbase.connection

	c = connection.cursor()
	c.execute("SELECT * FROM __table_names__")
	print c.pp() 

	print dbase.extendedFunctionality()
	


From akuchlin at mems-exchange.org  Wed Jun  7 22:05:21 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 07 Jun 2000 22:05:21 -0400
Subject: CPAN for python?
References: <393fbb15.100053125@nntp.interaccess.com> <393ECB55.4AC016A4@uab.edu>
Message-ID: <3dog5dhqqm.fsf@newcnri.cnri.reston.va.us>

Shae Erisson <shapr at uab.edu> writes:
> If you did it IETF[1] style, I think people would go with it.
> [1] IETF motto: 'rough consensus, running code'

I was thinking about this the other day.  The distutils work seems to
be freezing down nicely; the discussions have gone from requirements
to design to implementation, and are now doing the i-dotting and
t-crossing fixes, with the occasional slightly larger upheaval.  But
the basic principle of running setup.py seems stable. 

So, what would people want on top of distutils?  A Debian-style "do
apt-get, upgrade" to update currently installed packages?  A
CPAN-style "fetch and install this module for me, please"?  Simple
version checking?

--amk


From gmcm at hypernet.com  Fri Jun  2 17:14:12 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 02 Jun 2000 21:14:12 GMT
Subject: file.seek misbehaving?
References: <203da3d8.c492c8fd@usw-ex0103-019.remarq.com>
Message-ID: <393823a4$0$25427@wodc7nh6.news.uu.net>

cheesecake <vigenereNOviSPAM at leland.stanford.edu.invalid> 
wrote:
>
>I've run into a problem that is perhaps not a Python-specific
>bug. I was processing some log files using Python, when I
>noticed some mysterious bugs in certain files. I tracked down
>the problem to the seek method misbehaving.

[big snip]

Nope, it's not Python's problem. You can't reliably seek on a file 
opened in text mode (on Windows). You'll get the same (bad) results 
working in C.

If I were you, I'd open the file in binary mode (so seek and tell work 
properly) and deal with line endings explicitly.

- Gordon



From bjorn at roguewave.com  Thu Jun 29 13:23:19 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 29 Jun 2000 11:23:19 -0600
Subject: Yours in namelessness (was: How to concatenate external files ..?)
References: <slm1odapjev11@corp.supernews.com> <8jf6lf$v2j$1@nnrp1.deja.com> <slrn8lmb4t.lpi.thantos@brimstone.mecha> <B28AE9FD56BBFCF2.E102567A6FD85989.73C3BA8E8149CDA9@lp.airnews.net>
Message-ID: <395B8607.2FD17DB3@roguewave.com>

Cameron Laird wrote:
> 
> In article <slrn8lmb4t.lpi.thantos at brimstone.mecha>,
> Alexander Williams <thantos at gw.total-web.net> wrote:
> >On Thu, 29 Jun 2000 09:59:15 GMT, richard_chamberlain at my-deja.com
> ><richard_chamberlain at my-deja.com> wrote:
> >
> >>               file=open(x,'r')
> >
> >Teach me to post while still mostly asleep here at the office.  I
> >forgot entirely about opening the iterated filename for reading.
> >Bugger.
> >
> >To try and make up for it, how about this wee gem?
> >
> >> # "fls" is the globbed list of filenames
> >> # "outfile" is the file object already opened to write to
> >> flsHandles = map(lambda f: open(f, 'r').read, fls)
> >> for fH in flsHandles:
> >>     outfile.write(fH())
> >
> >Note that you never explicitly give each opened file its own name;
> >instead, you create a list of all the read methods of all the file
> >objects created dynamically from the list of filenames.  (That refered
> >to by "open(FILENAME, 'r').open" is just a function/method reference,
> >it can still be invoked.)
>                         .
>                         .
>                         .
> Well, let's get even more anonymous, and suggest
>   for fH in = map(lambda f: open(f, 'r').read, fls):
>       outfile.write(fH())

You'd be eating up a lot of file descriptors (is that still a real
issue?)  Here's my version :-)

  import glob

  pthstring = r'C:\wxpython\wxPython-2.1.16\docs\wx\wx*.htm'

  outfile = open('temp.txt,'w')
  for name in glob.glob(pthstring):
      outfile.write( open(name).read() )
  outfile.close()

-- bjorn



From garabik-news at spam.melkor.dnp.fmph.uniba.sk  Thu Jun 22 02:52:07 2000
From: garabik-news at spam.melkor.dnp.fmph.uniba.sk (Radovan Garabik)
Date: 22 Jun 2000 06:52:07 GMT
Subject: Game programming and python
References: <Kbf45.1986$az.236648@dfiatx1-snr1.gtei.net>
Message-ID: <8isd2n$cr8$1@sunnews.cern.ch>

jin choung <jin.choung at gte.net> wrote:
 : howdy!

 : are there any game programming sites specializing in python?  is there
such
 : a thing?

not that I know of (except of games section on parnassus)

 : any kind of game will do... graphics, ascii, etc...

try
<shameless advertising plug>
http://melkor.dnp.fmph.uniba.sk/~garabik/pytris.html
</shameless advertising plug>



--
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!





From petersc at stud.ntnu.no  Thu Jun 22 17:10:55 2000
From: petersc at stud.ntnu.no (Peter Schneider-Kamp)
Date: Thu, 22 Jun 2000 23:10:55 +0200
Subject: boolean true and false values.
References: <000d01bfdc23$c8aefae0$f0c809c0@lslp7o.lsl.co.uk> <0b0801bfdc5f$b3cca3c0$1906a8c0@fc.fenx.com>
Message-ID: <395280DF.7C76C1A4@stud.ntnu.no>

Emile van Sebille wrote:
> 
> Only when I want to know about the bits.  It's so
> much easier to recognize 122 octal as having
> alternate bits set than the equivalent in decimal
> or hex.

much easier than decimal - of course! But what is
wrong about 52? By the way 122 is not alternating...
125 is. But then 55 is probably much better to read.

or-did-you-mean-something-else-ly y'rs
Peter
--
Peter Schneider-Kamp          ++47-7388-7331
Herman Krags veg 51-11        mailto:peter at schneider-kamp.de
N-7050 Trondheim              http://schneider-kamp.de



From PClaerhout at CREO.BE  Mon Jun 19 09:52:40 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Mon, 19 Jun 2000 15:52:40 +0200
Subject: Getting remote memory configuration
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03013011F8@msgeuro1.creo.be>

Forgot to mention that I was talking about Windows NT machines.

Pieter

Pieter Claerhout
Application Support - CreoScitex Europe

> -----Original Message-----
> From:	Pieter Claerhout 
> Sent:	Monday, June 19, 2000 3:40 PM
> To:	'python-list at python.org'
> Subject:	Getting remote memory configuration
> 
> Anyone an idea on how you can find out how much memory a remote computer
> has installed? The computer is available on the same network and has the same
> login as the computer on which you are running the command.
> 
> Pieter



From hungjunglu at hotmail.com  Tue Jun 13 17:24:01 2000
From: hungjunglu at hotmail.com (Hung Jung Lu)
Date: Tue, 13 Jun 2000 14:24:01 PDT
Subject: ANNOUNCEMENT: The PythonWare py152 distribution (june 5, 2000)
Message-ID: <20000613212401.48393.qmail@hotmail.com>

--- In python-list at egroups.com, "PythonWare" <info at p...> wrote:
> > > Second: It will not register Python in the registry. which then causes
> > > win32_all not to install.
>
>It's done this way on purpose -- lots of people (including us) like to
>be able to install things without messing up the registry.

To put in some good words for PythonWare. I still remember from 3 years ago 
how hard it was for newbies to install Python. It was very confusing, you 
have to figure out what goes into where, and you have to handle the path 
information. And in the middle of all the chaos, there was Pythonware: you 
download it, click, and done. Back then, PythonWare saved the day for many a 
newbie.

Just as PythonWare, I am not a strong registry believer. I would like to be 
able to fully control my software programs, so I can build things, and move 
it to another computer (or even another operating system) and be sure that 
it will work. So I would not put things in the registry. (This perhaps has 
something to do with my background: I used to write cross-platform game 
software that had to run both on Mac and PC. So how good was Windows 
registry to me on a Mac? It's clear that the registry is just another tool 
in the pocket of Bill to subjugate programmers into the Windows Evil Empire. 
OK, maybe it's not so bad, but it is still really nice to be able to make 
things not dependent on the registry, if they need to be moved around.)

Hung Jung





________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com




From 12klat at sightreader.com  Tue Jun  6 08:17:36 2000
From: 12klat at sightreader.com (Ken Seehof)
Date: Tue, 06 Jun 2000 05:17:36 -0700
Subject: indirect references
References: <8hha1n$phl$1@nnrp1.deja.com>
Message-ID: <393CEBE0.8CEAF61D@sightreader.com>

I'm not sure exactly what the effect is that you are after, so I assume
that what you are looking is something like call by reference in a
fuction, e.g.

>>> a = "eggs"
>>> spam(a)
>>> print a
spam and eggs

Unfortunately the margin is too small to contain my implementation.  :-)

shaleh at debian.org wrote:

> How do I do this in python:
>
> foo = 'string'
> bar = foo
>
> somehow assign to bar and change the value of foo.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 12klat.vcf
Type: text/x-vcard
Size: 343 bytes
Desc: Card for Ken Seehof
URL: <http://mail.python.org/pipermail/python-list/attachments/20000606/b13c0b72/attachment.vcf>

From michael at stroeder.com  Mon Jun 26 19:09:38 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Tue, 27 Jun 2000 01:09:38 +0200
Subject: chroot'ed demon process
References: <395727E3.D4C1388A@stroeder.com> <39575bc1$0$18726@personalnews.de.uu.net>
Message-ID: <3957E2B2.C6A8CFEA@stroeder.com>

Johannes Stezenbach wrote:
> 
> Michael Str?der <michael at stroeder.com> wrote:
> >Is it possible to call a chroot() in a demon process running under
> >Python after reading all configuration data?
> 
> If python would have a chroot(), you would either have to call it
> after any modules have been imported or make the Python library
> and any shared libraries and support files needed for them
> available in the chroot "jail".

Yes. Other applications calling chroot() also have to take care of
getting shared libs etc. before "going into jail".

> You cannot access files outside of
> the chroot environment (that's the whole point of chroot).

Yes.

> Swigging chroot() or adding it to module posix would be trivial.

Hmm.

> Also see the chroot utility:
> man 1 chroot

With the chroot utility I have to put all files needed into the
chroot jail including the Python interpreter. Well, it's a start but
not really what I was looking for.

Ciao, Michael.


From thiele at muc.das-werk.de  Tue Jun 13 13:57:35 2000
From: thiele at muc.das-werk.de (Thomas Thiele)
Date: Tue, 13 Jun 2000 19:57:35 +0200
Subject: SV: Python Productivity over C++
References: <7BD10B680501D411B9DF009027E06F32011AC8@EXCHANGE> <394625A3.4489C696@muc.das-werk.de> <8i5fsu$t7$1@nntp9.atl.mindspring.net> <39464CD8.38FB435B@muc.das-werk.de> <8i5lnd$rsu$1@nntp9.atl.mindspring.net>
Message-ID: <3946760F.C5E46FD3@muc.das-werk.de>

>   I do have to wonder whether you've yet learned to think in
> Python; it's possible that you're trying to write C++ code in Python,
> the way some Perl people find writing Python difficult.

May be. But what are typical mistakes?
How should a REAL python program looks like.
I design classes like in C++, thats true. But I also use typical python features
like lists, dicts, map etc. I'm not trying to write everything by myself.

How have I to think as a pythonprogrammer?

(P.S. don't misunderstand me: I like python!)





From mwh21 at cam.ac.uk  Thu Jun 22 06:31:56 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jun 2000 11:31:56 +0100
Subject: Manipulating specific bits
References: <Brl45.80022$WS3.716680@typhoon.we.rr.com>
Message-ID: <m3snu63t2b.fsf@atrus.jesus.cam.ac.uk>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> writes:

>     Okay, this is a Very Newbie-ish question.
> 
>     What I need to do is manipulate specific bits in a byte, for
> transmission. I have absolutely *no* idea how to do this. :)
>     For example, say I have my 1 byte: 0 0 0 0 0 0 0 0. Say I want to set
> the 2nd and 5th bit to 1.. how the heck do I do that? :)

byte = 0
byte = byte | (1<<4) | (1<<1)

and so on...

>     I've never actually even /blinked/ in the direction of bit-ish
> manipulation, so am utterly clueless. Next i'm gonna have to stick it into a
> struct I think, before transmission.. I think. dunno. Will look into that
> afterwards.

bitfiddling is emphatically not what Python was designed for; it's
possible, but it ain't pleasant.  Thomas Wouters's augmented
assignment patch would make it more bearable.

It's also dog slow compared to C.  This may or may not be relavent...

Cheers,
M.

-- 
  This is an off-the-top-of-the-head-and-not-quite-sober suggestion,
  so is probably technically laughable.  I'll see how embarassed I
  feel tomorrow morning.            -- Patrick Gosling, ucam.comp.misc


From gmcm at hypernet.com  Tue Jun 27 15:47:05 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 27 Jun 2000 19:47:05 GMT
Subject: Threading +  embedding = System Dump.
References: <8j7udf$15au$1@news2atm.raleigh.ibm.com> <8F5FCDB88gmcmhypernetcom@199.171.54.154> <8jadre$k2c$1@news2atm.raleigh.ibm.com>
Message-ID: <8F60A0459gmcmhypernetcom@199.171.54.154>

Arint? wrote: 

[Gordon]
>> That sounds very much like there is no multithreading.
>>
>> So why doesn't "diz[0].write(mx)" block? Or does it?

>It doesn't need too.  The data is queued in a seperate subsystem.  My
>app only cares if the subsystem couldn't receive the data, so the
>response is quick no need to block.  This specific \x1b\x47 command
>tells when the device actually finished, I don't always want to wait for
>that . 

OK, so Python calls C++ (through "diz[0].write(mx)") which does it, then 
returns to Python. The Python script then goes on and finishes.

>> Is the real problem here that you're using one of the PyRun_ calls
>> when 
>you
>> need to go down a level and use PyImport_ etc. so the C++ has some
>> control over what's going on?

>Seemed like I can only run a script once.  If I tried to run the same
>script two times in a row the second time it did nothing, all the import
>calls return successfully both times.  I don't Py_Finalize until the dll
>is unloaded.

Well, you didn't answer the question, but you provided enough clues. You 
need to use a module, not a script. You need to move your top-level logic 
into functions. You need to PyImport the module, then fish the functions 
out of the module dict and call them. One function does the 
"diz[0].write(mx)" and then returns. Later, when the C++ has caught the 
relevant event, it can call the other Python function.

http://www.mcmillan-inc.com/embed2.html demonstrates most of this.

>Why, is it that if use python Event.wait function, the app crashes, as
>oppossed to popping up a messagebox and manually closing that box when I
>see the message come thru?  Does wait shut down the thread the script is
>running on in some sense?  I would try to do a notify from c++ on the
>python Event, but it seems I can't get any access to python when it is
>in the wait state? 

You have only one (Python) thread going. With only one thread, Python 
optimizes away the tstate, hence your crash. Multithreading Python in an 
embedded app with callbacks is a nightmare - you will have to have valid 
tstates for all C++ threads that call into (or callback into) Python. 
Review the 100s of messages on the threads SIG concerning making this part 
of the API sane before attempting it.

- Gordon


From richard_chamberlain at ntlworld.com  Tue Jun  6 10:58:23 2000
From: richard_chamberlain at ntlworld.com (RS Solutions)
Date: Tue, 6 Jun 2000 15:58:23 +0100
Subject: drawing graphs in python
References: <393d0c07$0$17334@news.denver1.Level3.net>
Message-ID: <8hj3g1$93l$1@gxsn.com>

There is an extension to Tkinter called Blt which has some graphing stuff,
look at:

http://www.ifi.uio.no/~hpl/Pmw.Blt/doc/

which is a tutorial on using the Pmw.Blt wrapper.

Of course if you don't know Tkinter then I recommend purchasing a book - in
Python and Tkinter programming by John Grayson there is a chapter on graphs.

Richard


Drew Fisher <drew at level3.nospam.net> wrote in message
news:393d0c07$0$17334 at news.denver1.Level3.net...
> Hello.  I am still learning Python, but I came across a need for
generating
> dynamic graphs (network load average, uptime, boring stuff like this :).
>
> Is the a module in Python that allows this?  If so, could someone point me
> to it?
>
> Thanks in advance,
> Andrew Fisher
>
>




From frankc at colconsulting.com  Fri Jun  9 11:34:10 2000
From: frankc at colconsulting.com (Frank V. Castellucci)
Date: Fri, 09 Jun 2000 11:34:10 -0400
Subject: Python Productivity over C++
References: <m130PPw-000wcEC@swing.co.at>
Message-ID: <39410E72.3DDE39B7@colconsulting.com>

Before you peruse this response please understand that I am not
advocating C++ over Python. I am having a tremendously educational and
productive time using Python. My responses are in regards to C++
comments which I feel are either not true, or that the arguments being
presented are based on a subjective desire, or lack thereof, to achieve
the functionality expressed as lacking or inferior.

Christian Tanzer wrote:
> 
> "Frank V. Castellucci" <frankc at colconsulting.com> wrote:
> 
> > > There are lots of reasons. The ones first springing to my mind are:
> > >
> > > - The tools you mention notwithstanding, C++ still forces you to write
> > >   lots of low-level code by hand (e.g., memory management, the
> > >   `orthodox canonical class form').
> > >
> > > - C++ doesn't offer any means of meta-programming. If you want to add
> > >   some mechanism (like serialization, introspection, ...) to a class
> > >   hierarchy, you'll have to add member functions to every class in the
> > >   hierarchy. In Python you can implement the complete mechanism in the
> > >   root class of the hierarchy.
> >
> > This is crap:
> >
> > 1. There are gc, SmartPointer, heap management, etc. class libraries out
> > there.
> 
> Guess what. Something like SmartPointer was one of the first classes I
> implemented. In C++, that doesn't relieve you of thinking about memory
> management, though.

This is kind of funny because in my experience memory management
thinking usually precludes SmartPointer implementation.

> 
> > 2. There is the type information in standard conforming compilers.
> 
> Nowadays, there is.

Agreed, but the language had the ability for you to write your own, or
use someone else's library which provided the same.
 
> RTTI doesn't offer any mechanisms remotely approaching Python features
> like `__getattr__'/`__setattr__' though. Not to speak of `eval' and
> `exec'.
> 
> > 3. There is, has been, and will be more, C++ class Class libraries which
> > are powerful, easy to use, and free. I was using NIHs Class class about
> > 5 years ago.
> 
> So what? Your pains start as soon as you want to combine libraries
> from different sources which happen to use incompatible idioms (like
> overloading global operators, ...).

Every language has a painful aspect to itself. Some take more time to
correct than others.

> I've used C++ and Python for projects of comparable size/complexity
> and based my statements on that experience.
> 
> That's not scientific research, and of course it depends on my style
> of using the two languages.
> 
> Your experience might be different.

I try not to pick a language for implementation until I am happy that
problem space is adequately defined, a reasonable degree of requirement
specification have been put forth, a sound analysis and design has been
done, and a language is available which satisfies the solution space. As
a contract developer this time alloted to perform these tasks may span
from one weekend by myself to many months with teams of people.
Sometimes the people who hire me decide the language before these steps.
I do the best I can.

Hands down, a interpretive language with a reasonably rich functional
capability will allow someone to "be more productive" than one which
requires intervening steps to get you into the unit test/debug cycle. I
don't go for the "one language cures all" position personally because if
it were true, we wouldn't have so many to choose from.

When any language has been around for awhile and supported through
developers like you and I by ADDING capability that the native grammar
ALLOWS us to, then the language becomes useful. Python is useful.
 
-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package


From kirill at xyz.donetsk.ua  Fri Jun 30 15:17:59 2000
From: kirill at xyz.donetsk.ua (Kirill Simonov)
Date: Fri, 30 Jun 2000 22:17:59 +0300
Subject: Bug in math.frexp?
Message-ID: <20000630221759.A1965@xyz.donetsk.ua>

$ python
Python 1.6a2 (#2, Jun 24 2000, 12:18:54)  [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import math
>>> print math.frexp.__doc__
frexp(x)

Return the matissa and exponent for x. The mantissa is positive.
>>> math.frexp(-1)     
(-0.5, 1)

In this example the mantissa is negative.


--
Kirill



From mhammond at skippinet.com.au  Wed Jun 14 15:25:54 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 14 Jun 2000 19:25:54 GMT
Subject: Windows: COM Automation problem with german VB localisation
References: <8i7klg$23i$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <8i8m7c$bra$1@nnrp1.deja.com>

In article <8i7klg$23i$1 at desig-bs01-s04.adtranzsig.de>,
  "Dirk-Ulrich Heise" <hei at adtranzsig.de> wrote:

> The problem is: It's one of the MS programs where they localized
> the Visual Basic inside it, unfortunately (and needlessly) german
> Umlaute (?,?,?)
> are part of the function names the MS program understands.
> Python chokes on them, it seems.
>
> Two questions:
> - Does anybody know a workaround?

You could try a getattr on the object - eg:

func = getattr(com_object, "whatever_with_funny_chars")
func(...)

> - Is there a more elegant way of finding out the COM interface
>   then recording the steps you want to automate with the
>   VB macro recorder?

You could try running makepy, and view the generated file.

Otherwise, look for one of the free COM viewers (or Pythonwin's COM
viewer, or (best of all) MS' COM viewer that comes with the SDK/VC)

Mark.


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


From the_wrights at onaustralia.com.au  Thu Jun 15 08:47:01 2000
From: the_wrights at onaustralia.com.au (the_wrights)
Date: Thu, 15 Jun 2000 22:47:01 +1000
Subject: 8 queen/game of life
References: <20000615104403.A1240@sirius.ftsm.ukm.my> <20000614200609.A323@activestate.com> <39484ED4.2D3CBCD7@micron.net>
Message-ID: <ld325.16426$N4.617712@ozemail.com.au>

Peter Immarco <android at micron.net> wrote in message
news:39484ED4.2D3CBCD7 at micron.net...
> Prolog's excellent for this kind [8 queens]of problem too.  (Sorry about
the null-answer on
> doing it in Python. :)
>
> Peter.
>

Someone not experienced in prolog might be seduced by the backtracking, only
to find that the default control structures in the naive implementation of 8
queens led to hopeless inefficiencies.

Prolog is indeed a good language for this type of problem, but those new to
the language should get O'Keefe's book (The Craft of Prolog) and Sterling
and Shapiro (The Art of Prolog) so that the folk lore is available to them.

cheers

chris




From jepler.lnk at lnk.ispi.net  Fri Jun  2 19:17:38 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Fri, 02 Jun 2000 23:17:38 GMT
Subject: Currency format for floats?
References: <Pine.LNX.4.21.0006011916380.17663-100000@lade.src.no>
Message-ID: <slrn8jf8ge.3pv.jepler.lnk@potty.housenet>

On Thu, 1 Jun 2000 19:20:31 +0200, Joseph Santaniello
 <joseph at src.no> wrote:
>Hi,
>
>I have an SQL query which I can format to give me a nice value like
>23,342.40 which Python squashes into 23342.4 when I stick it into a
>list element. Is there a nice way to format floats so they look like
>(US) currency?

According to the Linux manpage for printf(3), the following flag
can be included in a format specifier:
              '      specifying that in a numerical argument  the
                     output is to be grouped if the locale infor-
                     mation indicates any.  Note that  many  ver-
                     sions  of  gcc  cannot parse this option and
                     will issue a warning.
The manpage doesn't note that this is a nonstandard extension to printf
(as it does for the use of q, ll, and L), so I am lead to understand
that this conforms to ANSI C3.159-1989 (``ANSI C'').

However, 1.6a1 gives:
	>>> print "%'f" % 66666.666
	Traceback (most recent call last):
	  File "<stdin>", line 1, in ?
	ValueError: unsupported format character ''' (0x27)
i.e., no go---this isn't directly supported yet.

If you have locale enabled, and I don't, it looks like locale.format
will return a string in the desired format, and local.ato[if] will make
that string into an integer/float again.

These techniques, to the extent that they work, will make your program
behave correctly in *all* locales (or at least, more than one), for
instance those which use . for the thousands separator and , for the
decimal point.

Jeff


From moeller at dophnic.yi.org  Wed Jun 14 18:59:18 2000
From: moeller at dophnic.yi.org (Derek Moeller)
Date: Wed, 14 Jun 2000 17:59:18 -0500
Subject: Which linux distros have Python by default?
In-Reply-To: <%TS15.655$i06.206156@ptah.visi.com>; from ge@nowhere.none on Wed, Jun 14, 2000 at 09:34:51PM +0000
References: <Pine.GSO.4.10.10006142334560.3119-100000@sundial> <%TS15.655$i06.206156@ptah.visi.com>
Message-ID: <20000614175918.A5625@dophnic.yi.org>

On Wed, Jun 14, 2000 at 09:34:51PM +0000, Grant Edwards wrote:
> In article <Pine.GSO.4.10.10006142334560.3119-100000 at sundial>, Moshe Zadka wrote:
> >On Wed, 14 Jun 2000, Grant Edwards wrote:
> >
> >> I'm going to distribute a utility written in Python/Tkinter as
> >> a configuration/administration utility that goes with a Linux
> >> device driver.  I was just wondering how many customers are
> >> likely to have Python/Tkinter available.  I doubt that anybody
> >> is going to install Python just to run this program.
> >
> >Pretty soon, everyone will want to have Python to use ESR's CLM2, a
> >configuration utility for the the kernel written in Python/Curses. 
> >Tkinter is a different matter though
> 
> Well, people who compile kernels will, anyway.  There are
> getting to be a lot of users who never do a kernel build...
> 

Yeah, but the distribution maintainers will be forced to include it as
a default to be considered seriously.

-- 
Derek Moeller

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \



From corg at copernic.com  Wed Jun 21 11:44:20 2000
From: corg at copernic.com (Gaetan Corneau)
Date: Wed, 21 Jun 2000 11:44:20 -0400
Subject: mailing list duplicates?
Message-ID: <C5A7720BDEB7D3118A5200E018C200A05D40EA@EXCHANGE>

Yes, a lot.

______________________________________________________
   Gaetan Corneau
   Software Developer 
   Copernic.com 
   http://www.copernic.com
   E-mail: corg at copernic.com                         
______________________________________________________
"Golf is a game whose sole aim is to hit a very small
ball into an even smaller hole, with weapons
singularly ill designed for that purpose." -- Winston Churchill



> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Bill Scherer
> Sent: Wednesday, June 21, 2000 8:34 AM
> To: python-list at python.org
> Subject: mailing list duplicates?
> 
> 
> Is anyone else seeing a lot of duplicates on the list?
> 
> --
> William K. Scherer
> Sr. Member of Applications Staff
> Verizon Wireless
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From bvdpoel at uniserve.com  Mon Jun 12 16:37:34 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Mon, 12 Jun 2000 13:37:34 -0700
Subject: tkinter question
Message-ID: <39454A0E.1A630514@uniserve.com>


I'm trying to set up a few simple things using tkinter and have run into
a couple of roadblocks...probably because I'm new to python, don't have
a good reference to tkinter...

First, I'd like to have a function which saves some of it's own data
(privately). I tried to do this with a class, but essentially run into
the same problem with a function or a class. I've ended up doing
something like this:

	def myfunc( stats=[] ):
		if len( stats ) > 0:
			.. do already 'done'
		else:
			do first time...
			stats.append("done")

Okay, this works. Is this a good way to handle this situation?


Second, if I try to append to stats with something like:

	stats.append ( Toplevel() )

I get an error. So I end up with:

	w = Toplevel()
	stats.append(w)

which works. I have no idea why the first doesn't.

Third, how can I define a widget command with an arg. For example:

	q = Button( text="Quit", command = quitit("sure?") )

Of course, 

	q = Button( text = "Quit", command = quitit )

works fine.

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



From rhicks at rma.edu  Sun Jun 25 15:30:01 2000
From: rhicks at rma.edu (Robert)
Date: Sun, 25 Jun 2000 19:30:01 GMT
Subject: books????
References: <sla4ago0jev12@corp.supernews.com>
Message-ID: <961961617.1744372225@news.rma.edu>

On Sat, 24 Jun 2000, SugarKing wrote:
>Are there any good beginner books for Python? I'm itching to get my hands on
>any. Thanks

There is:

Learning Python
Learning Python in 24 hours
Python and Tkinter programming

There should be links to these books in the library portion of the python
website. I have not used the LP in 24 hours but have heard good things about it.

Robert


From kra at monkey.org  Fri Jun 23 16:43:37 2000
From: kra at monkey.org (Karl Anderson)
Date: 23 Jun 2000 13:43:37 -0700
Subject: using emacs pdb with a running python process?
Message-ID: <m0k8fg86x2.fsf@dclxvi.foobar.not>

Has anyone been able to attach a pdb-mode buffer to a python process
that is already running?  As in, typing this into an existing py-shell
buffer:

>>> import pdb; import foo
>>> pdb.run('foo.function()')

and having all of the pdb/gud features - the foo.py buffer follows
where the debugger is, etc.

pdb.el, and the gud commands that they're on top of, want to accept a
command line & run that in a shell.  I got frustrated trying to take
out the command parts so I could just turn a py-shell buffer into pdb
mode.

I want to do this so I can debug zope, which has the pdb.run command
buried in a script, so I'd like to be able to switch to pdb mode
manually.

-- 
Karl Anderson      kra at monkey.org           http://www.pobox.com/~kra/


From tismer at tismer.com  Sun Jun  4 20:47:48 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 05 Jun 2000 02:47:48 +0200
Subject: Starship ever gonna fly?
References: <f9oljso6hl7r7blv64an77h8lfj6dhr6f9@4ax.com> <20000605012821.V469@xs4all.nl>
Message-ID: <393AF8B4.44B5EAC@tismer.com>


Thomas Wouters wrote:
> 
> On Sun, Jun 04, 2000 at 04:09:58PM -0700, Kirby Urner wrote:
> 
> > The loss of Starship was a real tragedy.  Glad to
> > read it's recovering.  But is it?  The same not-
> > so-reassuring text has remained unchanged for some
> > time.
> 
> > Part of what made the Python community seem so vital
> > was all these add-ons (like PyOpenGL).  But with
> > Starship down for the count, so many of the pointers
> > in Vaults of Parnassus are just dead, stone dead.
> 
> If you search the list archive or browse the newsgroup for the last few days,
> you'll see that Starship is moving, or has been moved, this weekend. It'll
> take a few days for the DNS changes to settle, though. After the move, it
> should be more actively maintained. (By BeOpen, if I'm not mistaken.)

Yes. We are moving to the same place where Guido and
his crew are moving.

We got a very nice machine, in a very good maintained
environment. Starship will soon be better than ever before.

> > Whoever killed Starship did real and lasting damage
> > it seems.  Did anyone call the FBI?
> 
> I was under the impression that the Big Crash was hardware-related, not a
> malicious attack.

So it is. ciao - chris

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



From root at rainerdeyke.com  Tue Jun 20 12:57:26 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 20 Jun 2000 16:57:26 GMT
Subject: nest class name problem
References: <394E67CE.9E29D310@uniserve.com> <riu35.36$JM5.4708@news.uswest.net> <j8y35.1071$ul.67488@news1.rdc2.tx.home.com> <394EB843.EEB3B68F@uniserve.com>
Message-ID: <WnN35.2100$ul.243704@news1.rdc2.tx.home.com>

Bob van der Poel <bvdpoel at uniserve.com> wrote in message
news:394EB843.EEB3B68F at uniserve.com...
>
> Yes, but I'm using __name to keep the variable private/static to a
> particular instance of the Goption().

If you want each instance of Goption to have its own __name, you need to
totally change your design.


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




From contactXXX at redspice.com  Tue Jun  6 01:35:27 2000
From: contactXXX at redspice.com (contactXXX at redspice.com)
Date: Tue, 06 Jun 2000 05:35:27 GMT
Subject: E-Commerce solution using Python and Zope.
References: <Pine.SGI.4.05.10005241745340.25496-100000@lemur.magnet.com> <392CD741.E7241339@maxnm.com> <8gjde8$dsk$1@violet.singnet.com.sg>
Message-ID: <393C8B2D.FB367644@redspice.com>

Ng Pheng Siong wrote:
> 
> According to Itamar Shtull-Trauring  <itamar at maxnm.com>:
> > John Mitchell wrote:
> >> Zope (http://www.zope.org) has a module or something for
> >> Cybercash and/or
> >> OpenMarket's stuff.  Report back your findings!
> >
> > The CyberCash product (Wampum) works nicely (well, queries - I'm
> > testing it
> > with a credit-card this afternoon.).  There's a nice store solution at
> > http://e-tailer.adroit.net.
> 
> <shameless plug>
> 
> I've written an S/MIME-generator for Zope called ZSmime.
> Your Zope site can accept credit card numbers, say, and have
> ZSmime send them out as S/MIME-signed/encrypted mail to your
> Netscape Messenger-using back offce.
> 
> http://www.post1.com/home/ngps/zope/zsmime
> 
> </shameless plug>
 
who process the actual transaction ?. Does netscape have a secure
transaction gateway. we have a python wrapper for the signio libs if
anybody is interested


From bvdpoel at uniserve.com  Mon Jun 19 12:26:31 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Mon, 19 Jun 2000 09:26:31 -0700
Subject: tkinter: option_add() problem
References: <394BF3CF.450F776E@uniserve.com> <cDi35.2869$_55.58179@news2-win.server.ntlworld.com>
Message-ID: <394E49B7.4AB9BE2@uniserve.com>

Thanks! Works perfectly (of course). Ummm, how in the world is someone
to know this from reading the tk and tkinter docs?

richard_chamberlain wrote:
> 
> Hi Bob,
> 
> try
> 
> root.option_add('*Font','Times 14 bold')
> 
> Richard
> 
> Bob van der Poel <bvdpoel at uniserve.com> wrote in message
> news:394BF3CF.450F776E at uniserve.com...
> >
> > Using tkinter to modify the option database I CAN do things like:
> >
> > root.option_add( '*activeBackground', 'white')
> >
> > However,
> >
> > root.option_add( '*Font', "family times -weight bold  -size 14" )
> >
> > gives me the error message:
> >
> > TclError: expected integer but got "times"
> >
> > the equivalent??? tcl:
> >
> > option add *Font {-family times -weight bold  -size 14}
> >
> > works fine.
> >
> > Help!
> >
> >
> > --
> >    __
> >   /  )      /         Bob van der Poel
> >  /--<  ____/__        bvdpoel at uniserve.com
> > /___/_(_) /_)         http://users.uniserve.com/~bvdpoel
> >

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


From cut_me_out at hotmail.com  Wed Jun 14 12:30:27 2000
From: cut_me_out at hotmail.com (Alex)
Date: 14 Jun 2000 12:30:27 -0400
Subject: tuple creation in C extensions
References: <3942AFEE.5155481@stud.ntnu.no> <3942B267.C769395@stud.ntnu.no> <110620000925578570%pecora@anvil.nrl.navy.mil> <etd4s7015zr.fsf@w20-575-109.mit.edu> <m3g0qh5pe2.fsf@greebo.nodomain.de> <etdk8ftm55x.fsf@w20-575-109.mit.edu> <140620000716512992%pecora@anvil.nrl.navy.mil>
Message-ID: <etdya48z0m4.fsf@w20-575-109.mit.edu>

> And who could have seen all that so clearly without an expert on hand. 
> Even then I couldn't understand all the reasoning.  In Python we are
> all happy, but the ugly side is when we return to C.

I agree with you about needing a list of reference count effects, but I
think the tutorial is otherwise about as clear as it could be on the
subject.  It's just a boring, unavoidably complex topic, and my eyes
would have just glazed over if someone had tried to carefully explain
the issues Bernhard raised before I had written that code.

I was thinking it ought to be possible to write a script that would pull
the signatures of the API functions from the docs, and write some C code
to determine their refence count effects.  You might be able to get a
long way to the sort of table we need that way.

Alex.


From sridhar at mediaone.net  Fri Jun 30 21:49:06 2000
From: sridhar at mediaone.net (Sridhar Bidigalu)
Date: Fri, 30 Jun 2000 21:49:06 -0400
Subject: Subscribe
Message-ID: <395D4E11.EBAEA908@mediaone.net>


-------------- next part --------------
A non-text attachment was scrubbed...
Name: sridhar.vcf
Type: text/x-vcard
Size: 386 bytes
Desc: Card for Sridhar Bidigalu
URL: <http://mail.python.org/pipermail/python-list/attachments/20000630/6ec58387/attachment.vcf>

From bragib at my-deja.com  Thu Jun 15 12:33:08 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Thu, 15 Jun 2000 16:33:08 GMT
Subject: Anyone experienced in telnetlib?
Message-ID: <8ib0g4$2ao$1@nnrp1.deja.com>

This is a repost as I am still searching for help...

Bragi


I am trying to telnet from a UNIX box to an NT to get the doskeys. The
NT that I login to runs a
 couple of .bat scripts which clear the screen set up some stuff etc.
When I try to do a read_all() it
 fails the first time with the exception:
 Traceback (innermost last):
   File "telnet.py", line 17, in ?
     tn.read_all()
   File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
229, in read_all
     self.fill_rawq()
   File "/usr/abaqus60/5-0_copy/source/Python/Lib/telnetlib.py", line
367, in fill_rawq
     buf = self.sock.recv(50)
 socket.error: (131, 'Connection reset by peer')


 That is why I put the first try in a try: except: clause.  The second
time I try to do a read_all() it
 works fine but I don't get all the keys I am missing some of the first
keys.  I assume that the first
 read_all() actually did read some stuff and then the second read got
the rest.

 Can anyone see anything obviously wrong with this.


 Thanks, Bragi

 import sys
 import telnetlib
 HOST = 'shadow'
 USER = 'bragi'
 tn = telnetlib.Telnet()
 tn.open(HOST)
 tn.read_until("login: ")
 tn.write(USER+'\r')
 tn.read_until("password: ")
 tn.write('my_password\r')
 tn.write('doskey /macros\r')
 tn.write('exit\r')
 print 'done'
 try:
     all1 = tn.read_all()
 except:
     pass
 all = tn.read_all()
 print all


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


From paul.magwene at yale.edu  Sat Jun 10 21:23:53 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Sat, 10 Jun 2000 21:23:53 -0400
Subject: PATCH: Augmented assignment
References: <m130q7t-000wcDC@swing.co.at> <3942956B.2CAB4C9A@yale.edu> <20000610214319.J20700@xs4all.nl>
Message-ID: <3942EA29.25501B82@yale.edu>

Thomas Wouters wrote:
> 
> On Sat, Jun 10, 2000 at 03:22:19PM -0400, Paul Magwene wrote:
> > Christian Tanzer wrote:
> 
> > > It doesn't obfuscate code looking like:
> 
> > >    nested_structure [some_index] [another_index] = nested_structure [some_index] [another_index] + 1
> 
> > Please tell me you're not REALLY writing code that looks like that, when
> > something like:
> >
> >       nested[i][j] = nested[i][j] + 1
> 
> > will do just as well (and is a lot easier to read).
> 
> *Please* tell me you aren't really using variable names like 'i' and 'j' in
> large programs ? Sure, they're fine for loop counters, but you dont always
> index large structures with those.
> 

If I have a bunch of way-too-long-variables-names that I need to use to
index structures I'll setup aliases --- see Moshe Zadka's post...

--Paul

-- 

Paul Magwene
paul.magwene at yale.edu



From jovecka at dallas.net  Sun Jun 18 12:30:47 2000
From: jovecka at dallas.net (Jim)
Date: Sun, 18 Jun 2000 11:30:47 -0500
Subject: Using Python instead of Perl?
Message-ID: <394CF937.D0BDB2B4@dallas.net>


I am trying to use nmap to test my own network and automate the process.

I keep seeing perl being used for this type of work , but was hopping
that
 Python could do this with out all the lines of code needed for perl. Is

perl a better choice than Python for this?

Any help would be welcomed.
Thanks.





From ge at nowhere.none  Fri Jun 23 18:28:53 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 23 Jun 2000 22:28:53 GMT
Subject: adding mail attachments
References: <8ivdl4$90k$1@supernews.com>
Message-ID: <FwR45.1449$iN5.226321@ptah.visi.com>

In article <8ivdl4$90k$1 at supernews.com>, geoff munday wrote:

>OK, I can send mail via the smtplib but am at a loss as how to
>attach files to the Email. Can anyone advise where to look for
>instructions or examples??

Take a look at <ftp://ftp.visi.com/users/grante/stuff/mfetch.py>.

There's a class called SmtpWriter that has a "Message" method
that probably does what you want...

def Message(self,sender="", subject="", recipients=[], body="", attachments=[]):
    if self.__debugLevel > 2:
        sys.stderr.write("SmtpWriter: Building RFC822 message From: %s; Subject: %s; (Length=%d with %d attachments)\n" % 
                          (sender, subject, len(body),len(attachments)))
[...]

The MimeWriter library module is where most of the magic
happens...

-- 
Grant Edwards                   grante             Yow!  My nose feels like a
                                  at               bad Ronald Reagan movie...
                               visi.com            


From sean at digitome.com  Wed Jun 28 17:44:07 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Wed, 28 Jun 2000 21:44:07 GMT
Subject: XML-code examples - mini-review of Python XML
References: <394f7bc2.365514882@news.online.no> <00062201025104.08285@quadra.teleo.net> <8jbteo$q3v$1@netnews.upenn.edu>
Message-ID: <395a68f6.107291957@news.iol.ie>

I want to make a couple of quick points about my book
"XML Processing with Python".

1) Its target audience is developers working with XML who
are new to Python. If you are already a dab hand at Python
this book is *not* for you.

On the other hand, if you are knew to Python and caught up in the
growing wave of interest in Python as an excellent language for
developing XML applications then this book should be of interest.

2) The Pyxie library is a XML processing library specifically
written for use withPython. It is intended to provide an alternative
to the more pervasive - but also more complex - language neutral
XML APIs. As with any Python specific library, your mileage
will vary.

3)  PYX is a new name for an old, tried and trusted
technique for processing hierachical data structures that
dates back at least as far as 1986. See ISO 8879:1986
and also appendix B, attachment 1, The SGML Handbook
by Charles F. Goldfarb (0-19-8853737-9)

4) The PYX notation and PYX generators such as
XMLN and XMLV (available on http://www.pyxie.org),
can be used independently of Python.

regards,
Sean McGrath

On Wed, 28 Jun 2000 00:03:27 -0400, "Andrew Diller"
<dillera at isc.upenn.edu> wrote:

>His book is out there! We got ours last week.
>
>I like the book, but I also found it dissapointing. Most of it is an intro
>into the Python language itself. What's left was mostly about his Pyxie and
>PYX. I'm not sure why I'd want to learn PYX, when what I want is a solid
>book on manipulating XML with python.
>
>I didn't think I got that. I'm interested in learning about DOMs, SAX and
>things that I'm reading and hearing about. PYX seems like a bizzare
>off-shoot.
>
>As for the into to Python part of the book (about 3/4 of it) that was a good
>intro if you've never used python.
>
>-andy diller
>
>
>--
>Andrew Diller       dillera at isc.upenn.edu
>Information Systems & Computing
>University of Pennsylvania
>x3-7360
>"Patrick Phalen" <python-list at teleo.net> wrote in message
>news:00062201025104.08285 at quadra.teleo.net...
>> [Thomas Weholt, on Wed, 14 Jun 2000]
>> :: Is there more simple XML code out there? I mean the HowTos took about
>> :: ten minutes, and it didn`t cover much.
>> ::
>> :: If somebody has some tips or code on how to traverse a DOM-tree,
>> :: processing nodes based on tagname, and especially, based on relations
>> :: with other nodes ( i.e. who`s your daddy, here`s your sister etc ),
>> :: that would be great.
>>
>>
>> While waiting for Sean McGrath's book to hit the shelves, you may find
>> that David Mertz's nice Python XML introductory article gives you a leg
>> up:
>>
>> http://www-4.ibm.com/software/developer/library/python1/
>>
>
>



From parkw at better.net  Wed Jun 21 17:52:58 2000
From: parkw at better.net (William Park)
Date: Wed, 21 Jun 2000 17:52:58 -0400
Subject: Speed with Python
In-Reply-To: <QH645.94$QA1.13929@news.uswest.net>; from jmassung@magpiesystems.com on Wed, Jun 21, 2000 at 11:07:25AM -0600
References: <QH645.94$QA1.13929@news.uswest.net>
Message-ID: <20000621175258.A1512@better.net>

On Wed, Jun 21, 2000 at 11:07:25AM -0600, Jeff Massung wrote:
> Many threads with this kind of name have popped up, comparing the speed of
> Python to C, Java, etc. and trying to determine how fast some of the
> algorithms in Python are.
> 
> My question however, is how many of you who are Python gurus/junkies (no
> offence, I myself am a C and Forth junkie) :) use Python for speed intensive
> processing (not graphics, but databases and such)? Does efficiency of
> developement outweight the eventual speed of use?

Yes.  I'm not going to spend 5 months trying to make programs run 5
seconds faster.

--William



From karl_python at sub.ulbrich.org  Thu Jun 15 21:36:22 2000
From: karl_python at sub.ulbrich.org (Karl G. Ulbrich)
Date: Thu, 15 Jun 2000 21:36:22 -0400
Subject: Suggestion: "Completing" HTTP support in Python.
In-Reply-To: <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru>; from phd@phd.russ.ru on Thu, Jun 15, 2000 at 04:39:50PM +0000
References: <3949028F.3B512AF9@darmstadt.gmd.de> <Pine.LNX.4.21.0006151637480.8145-100000@fep132.fep.ru>
Message-ID: <20000615213621.F30050@exeter.exeter.org>

Quoting Oleg Broytmann (phd at phd.russ.ru):
> > How do you add HTTP parameters?
> > How do you specify that the connection should be POST or GET?
> 
> GET:  filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript?%s" % urllib.urlencode({"param": value}))
> POST: filename, headers = urllib.urlretrieve("http://my.host.name:8080/loginscript", urllib.urlencode({"param": value}))

Note that POST support in urllib lacks support for posting form contents
that include files to be uploaded.  That means you're back to down and 
dirty with the HTTPlib and some ugly http and mime boundary syntax.

What I need (and am writing as a 1st Python project) is something that'll
let me post an arbitrary mix of normal fields and file-upload-fields+files
to a URL.  (usage: POST a file, title, and description field).  

I'd love to see this support (and some of the other stuff in this thread)
make it into urllib as easy-to-use methods.

Karl Ulbrich






From thomasm at legend.co.uk  Tue Jun 13 10:12:21 2000
From: thomasm at legend.co.uk (Thomas Mangin)
Date: Tue, 13 Jun 2000 15:12:21 +0100
Subject: No Boolean type in "types"
References: <XFMail.000613145513.mikael@isy.liu.se>
Message-ID: <394640df.0@katana.legend.co.uk>

>  if {}:
>    print 'Shouldn't happen!'
>
>Defining a single boolean type is not only unnecessary, it's plain wrong.

Argh, ok I will fight again my masochist tendencies .. , but you meant
>    print "Shouldn't happen!"
;*)

Thanks for the enlightenment
Thomas




From nbecker at fred.net  Fri Jun  9 11:37:39 2000
From: nbecker at fred.net (nbecker at fred.net)
Date: 09 Jun 2000 11:37:39 -0400
Subject: Python Productivity over C++
References: <8F4E9C028stevemulozemailcomau@139.130.250.4> <sk1sq867h5126@corp.supernews.com>
Message-ID: <x88hfb2j264.fsf@adglinux1.hns.com>

>>>>> "Bill" == Bill McClain <wmcclain at salamander.com> writes:

    Bill> What frustrates me about C++ is the enormous declarative overhead required
    Bill> to get anything done. Every class requires me to instruct the compiler on
    Bill> how to copy, assign, compare and reference objects, and how subclasses
    Bill> are allowed to inherit from the parent. STL works well for data structures,
    Bill> but the advanced features are very cumbersome.

    Bill> Much of this burden simply vanishes with Python.

I rarely need to do this for my classes.  Using STL heavily is a big
help.


From bvdpoel at uniserve.com  Sat Jun 17 00:07:13 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Fri, 16 Jun 2000 21:07:13 -0700
Subject: tkinter: making multiple command bindings
Message-ID: <394AF971.CCA0C410@uniserve.com>

In tcl/tk I can bind a sequence of commands to an event-binding:

	bind $w.e <Return> {destroyWin $w; func1; func2 $parm}

Is it possible to do the same in tkinter? I'm sure it has to do with the
use of lambda (learned that last week in creating a function call with
an arg...thanks!).


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



From aahz at netcom.com  Tue Jun 20 19:28:36 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jun 2000 23:28:36 GMT
Subject: List without duplicates?
References: <VlS35.15102$Xx5.781310@bgtnsc06-news.ops.worldnet.att.net> <394FFA1A.FBF8F81D@roguewave.com>
Message-ID: <8ioun4$o7g$1@slb7.atl.mindspring.net>

In article <394FFA1A.FBF8F81D at roguewave.com>,
Bjorn Pettersen  <bjorn at roguewave.com> wrote:
>
>This should probably be a faq by now...:

Yup.

>def unique(lst):
>    d = {}
>    for item in lst:
>        d[item] = None
>    return d.keys()

The Timbot recommends

        d[item] = 1

because there's no lookup involved on 1, so it's fractionally faster.
(Any other constant should probably work the same.)
--
                      --- Aahz (Copyright 2000 by aahz at netcom.com)

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

"This is Usenet.  We're all masturbating in public places."  -DH


From richard_chamberlain at ntlworld.com  Thu Jun 29 02:11:21 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Thu, 29 Jun 2000 07:11:21 +0100
Subject: entry widget and rotor module
References: <8jdff7$mc0$1@nnrp1.deja.com>
Message-ID: <MLB65.43$zo2.907@news2-win.server.ntlworld.com>

Hi Keith,

If you ignore the fact your using the rotor module and just print your
string you get the following:

>>> print '.\314q\217'
.?q?

If you do this:

>>> print r'.\314q\217'
.\314q\217

So it's nothing to do with rotor but just escape characters in the string
i.e. \ being converted.

Richard




<kpmurphy at my-deja.com> wrote in message news:8jdff7$mc0$1 at nnrp1.deja.com...
> can anyone help out... what's going on below?
>
> anna% python
> Python 1.5.2 (#12, Jun 26 2000, 13:56:56)  [GCC 2.95.2 19991024
> (release)] on sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import rotor
> >>> rt = rotor.newrotor('qwerty', 20)
> >>> rt.encrypt('boat')
> '.\314q\217'
> >>> foo = rt.encrypt('boat')
> >>> type(foo)
> <type 'string'>
> >>> foo
> '.\314q\217'
> >>> print foo
> .?q
> >>>
>
> ...it appears that the string foo has the right data in it, but prints
> out weird characters.  Thanks,
>
> keith
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From m.faassen at vet.uu.nl  Tue Jun 20 11:26:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jun 2000 15:26:04 GMT
Subject: How to sell Python ( and OpenSoruce )
References: <39565119.358692532@news.online.no> <8inq3u$p0i$1@newshost.accu.uu.nl> <39586f3d.366407626@news.online.no>
Message-ID: <8io2ec$jcg$1@newshost.accu.uu.nl>

Thomas Weholt <thomas at cintra.no> wrote:
>>> Linux with Windows  2000/ NT 4.0.
>>
>>Why switch from Linux to Windows NT/2000, though? ;)

> Well, first off it`s my OS of choice,

Windows 2000 or Linux? I was reading what you typed, not reading
what you meant. :)

> but mostly it *was* because it
> costs money. I`ve realized later that the cost of an os is scraps in a
> huge project so that`s not a very important issue anymore. But I would
> still go for Linux. It has remote-administration out-of-the-box with
> telnet/ssh, ready-and-running web-server ( Apache ) , dns, ftp - well,
> do I need to go on. AND PYTHON by default.

Oh, I'm typing this on a Linux box, so I am inclined to agree.

[snip linux advocacy]
> Oh, the big-ol` Raving-Linux-Fanatic took over, sorry.

So why replace Linux with Windows 2000 then? ;)

Okay-okay-I'm-meanly-exploiting-braino's-ly yours,

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


From mfletch at tpresence.com  Fri Jun  9 09:04:34 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 9 Jun 2000 09:04:34 -0400 
Subject: Python Productivity over C++
Message-ID: <CC59868A0A76D311B1F50090277B321C1F339E@VRTMAIL>

Just a minor note, having spent all of last night (unsuccessfully) trying to
track down a memory leak.  Yes, memory leak tools would be nice to have in
Python.  Hidden memory leaks (cycles) are rare, but when they happen in a
complex project they are often a nightmare to track down.  An optional
garbage collector with a debug mode saying "hey, I found a cycle that wasn't
cleaned up, the objects were: object types, object cycle description" would
be really cool.  Yes, I'd probably remove it for the final proggy, but as a
development tool it would be very, very, useful, and is currently (to my
memory), not available.

0.02
Mike(y)

-----Original Message-----
From: Drew Csillag [mailto:drew.csillag at starmedia.net]
Sent: Friday, June 09, 2000 1:07 PM
To: Steve Mullarkey
Cc: python-list at python.org
Subject: Re: Python Productivity over C++




On Fri, 9 Jun 2000, Steve Mullarkey wrote:
...

Python's tools are very good (and don't need them a lot of the time
anyhow).  The main thing missing is a GUI builder (but glade and PyGlade
are getting there quickly).  Memory leak tools just don't apply (Python
has reference counted garbage collection).  Programmers editors are
numerous and one is included (IDLE).

...



From ge at nowhere.none  Wed Jun 14 17:33:58 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 14 Jun 2000 21:33:58 GMT
Subject: Which linux distros have Python by default?
References: <uKs15.437$i06.133733@ptah.visi.com> <8i86sf$e7c$0@216.39.141.36> <1VO15.607$i06.193222@ptah.visi.com> <3947C5E8.4CE1032E@estinc.com>
Message-ID: <aTS15.654$i06.206149@ptah.visi.com>

In article <3947C5E8.4CE1032E at estinc.com>, Eric Lee Green wrote:
>Grant Edwards wrote:
>> I'm going to distribute a utility written in Python/Tkinter as
>> a configuration/administration utility that goes with a Linux
>> device driver.  I was just wondering how many customers are
>> likely to have Python/Tkinter available.  I doubt that anybody
>> is going to install Python just to run this program.
>
>Red Hat and Caldera certainly will, since their system
>administration tools are written in Python.
>
>But I would strongly suggest using TCL/Tk instead if it is a
>simple little configuration program, because I don't know of
>any Linux distribution that does not come with TCL/Tk --
>amongst other things, 'make xconfig' needs TCL/Tk in order to
>work. TCL is brain-dead, but for simple stuff it's usable. 

Nah, I wrote a TCL program once and swore never again.  Since
this is just a GUI front-end for some command line utilities,
the user who doesn't have Python/Tkinter can get along fine
without it.

-- 
Grant Edwards                   grante             Yow!  HOORAY, Ronald!! Now
                                  at               YOU can marry LINDA
                               visi.com            RONSTADT too!!


From thomas at cbs.dtu.dk  Wed Jun 28 03:52:20 2000
From: thomas at cbs.dtu.dk (Thomas Sicheritz-Ponten)
Date: 28 Jun 2000 09:52:20 +0200
Subject: Can anyone solve this?
References: <DD768E0BEA95D211B27A0008C74C86250A6B8E@BEKEXCH1> <20000628124201.184982@210.9.21.253>
Message-ID: <y9vaeg6qm2z.fsf@genome.cbs.dtu.dk>

Matthew Schinckel <matt at null.net> writes:

> <rdudfield at my-deja.com> wrote in message
> news:<8j737m$t5a$1 at nnrp1.deja.com>...
> > Hi,
> > 
> > I've got a problem here which I just can't figure out.
> > 
> > a = [1,2,3] b = ["c",32] c = [1,2]
> > 
> > inputlist = [a,b,c]
> > 
> > Inputlist can have any number of lists in it. The lists in it can have
> > any number of elements in them.
> > 
> > I need the code to give every possible combination of the lists in
> > inputlist.
> > 
> > This is what the answer should be for the inputlist above.
> > 
> > outputlist = [[1, 'c', 1], [1, 'c', 2], [1, 32, 1], [1, 32, 2], [2,
> > 'c', 1], [2, 'c', 2], [2, 32, 1], [2, 32, 2], [3, 'c', 1], [3, 'c',
> > 2], [3, 32, 1], [3, 32, 2]] []
> 
> I did find something about combinatorics on starship yesterday, but I
> can't remember where it was.

something I found on dejanews a couple of months ago ...

def cartesian(listList):
    if listList:
        result = []
        prod = cartesian(listList[:-1])
        for x in prod:
            for y in listList[-1]:
                result.append(x + (y,))
        return result
    return [()]

a = [1,2,3]
b = ["c",32]
c = [1,2]

inputlist = [a,b,c]

print cartesian(inputlist)

[(1, 'c', 1), (1, 'c', 2), (1, 32, 1), (1, 32, 2), (2, 'c', 1), (2, 'c',
2), (2, 32, 1), (2, 32, 2), (3, 'c', 1), (3, 'c', 2), (3, 32, 1), (3, 32,
2)]



From PClaerhout at CREO.BE  Thu Jun 22 06:24:17 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 22 Jun 2000 12:24:17 +0200
Subject: Labels of remote volumes under NT
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130120D@msgeuro1.creo.be>

Hello,

is it possible with Python under NT, to find out the type and the labels
of the drives of a remote computer?

Kind regards,


Pieter

Pieter Claerhout
Application Support - CreoScitex Europe




From morpheus at despair.rpglink.com  Fri Jun  2 04:03:02 2000
From: morpheus at despair.rpglink.com (Steve Lamb)
Date: Fri, 02 Jun 2000 08:03:02 GMT
Subject: Copyright Notice in Commercial Software
References: <8h6rbt$ksv$1@nnrp1.deja.com> <8h7blm$j1b$1@slb2.atl.mindspring.net>
Message-ID: <slrn8jeqfv.n4g.morpheus@teleute.rpglink.com>

On Thu, 1 Jun 2000 22:04:22 -0600, Andrew Dalke <dalke at acm.org> wrote:
>The GPL, which you are probably thinking of, does not allow similarly
>unfettered use of their code.  Packages compiled with GPL code
>restrict their use to only GPL-compatible licenses.

    That's a new one to me.  I've never heard of the GPL requiring things that
I compile with a GPLed compiler to be distributed also under that license.
I've only seen it where the distributed /code/, if modified and recompiled
must also be distributed under the GPL.  

-- 
         Steve C. Lamb         | I'm your priest, I'm your shrink, I'm your
         ICQ: 5107343          | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------


From franck at soleil.serma.cea.fr  Wed Jun 14 07:18:34 2000
From: franck at soleil.serma.cea.fr (Franck Goubioud - SERMA/LENR)
Date: Wed, 14 Jun 2000 13:18:34 +0200
Subject: Static python on AIX
References: <3C1077AC64755A57.0FB8B24AA3155646.CE4465C98A055096@lp.airnews.net> <8gvvtt$s2n$1@slb2.atl.mindspring.net>
Message-ID: <39476A09.E9D5E89C@soleil.serma.cea.fr>

I'm trying to use Python on AIX.
I'm trying to build a static executable with my modules and with the
Numerical extension.
The makefile builds it, but with an error message wich is:

The file access permissions do not allow the specified action.
/usr/local/lib/python1.5/config/makexp_aix[41]:
/usr/local/lib/python1.5/config/python.exp: 0403-005 Cannot create the
specified file.

And the executable fails on the "import Numeric" instruction.
On other platforms, there is no problems.






From mjackson at wc.eso.mc.xerox.com  Sat Jun  3 18:13:52 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 3 Jun 2000 22:13:52 GMT
Subject: float does not parse '0.000000E+00'
References: <39397EFF.E1B76E6E@inc.it>
Message-ID: <8hbvv0$40h$1@news.wrc.xerox.com>

Paolo Redaelli <redaelli at inc.it> writes:
> Peter Schneider-Kamp wrote:
> > 
> > Paolo Redaelli wrote:
> > >
> > > but if I run a program that does x=float(x) where x = "0.000000E+00 "
> > > (being a string) here's the output.
> > 
> > Can you give some code? I tried the following:
> > 
> > === BEGIN dumm.py ===
> > x = "0.000000E+00 "
> > x = float(x)
> > === END dumm.py ===
> > 
> > it worked just fine. What version of Python are you using?
> 1.5.2 from RedHat 6.2
> It works for me if I use the interactive version.
> I took that very code, placed into a program and whooops... it hangs!
> That's driving me crazy..... 

Have you checked to see if you might have rebound float in that namespace?

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	You can never solve all difficulties at once.
				- Paul Dirac




From slay3241 at bright.net  Thu Jun 15 14:15:25 2000
From: slay3241 at bright.net (bill slaybaugh)
Date: Thu, 15 Jun 2000 14:15:25 -0400
Subject: newbie needs select-files tool
Message-ID: <39491D3D.536F0CFF@bright.net>

I've written a few workable python tools to assist my file and project 
management.  I work with multiple app's and projects using InstallShield
to set up installations.  What with the repetition of numerous
InstallShield
projects, I wrote several python scripts which "automate" pulling files
off
the network, build InstallShield projects, and copy media-output to a
unique
location where a CD gets burned.
I would like to add a GUI to the front of this.  But, short of
reinventing the
wheel, I haven't found a toolset comparable to VB's DRIVE and DIRECTORY
list box.
I've played with examples under Tkinter and PMW, but I don't see a very
direct
route to this sort of listbox.
Any suggestions?